1. Alberuni's India_vol2

ALBERUNI'S INDIA.

AN ACCOUNT OF THE RELIGION, PHILOSOPHY, LITERATURE, GROGRAPHT, CHRONOLOGY, ASTRONOMY, CUSTOMS, LAWS AND ASTROLOGY OF INDIA ABOUT A.D. 1030.

PM.

117.828

Hin Engifeb Edition, witb Rotes and 3ndices.

BY

DR. EDWARD C. SACHAU, Profemor in the Royal Univeraity of Betiin, and Principal of the Seminary for Oriental LAnguages; Member of the Royal Academy of Berlin, and Corresponding Member of the Imperial Academy of Vienna ; Honorary Member of the Aslatie Society of Great Britain and Ireland, London, and of the American Orlental Society, Cambrige, U.S.A.

IN TWO VOLUMES.

VOL. II.

LONDON: TRUBNER & CO., LUDGATE HILL 1888. [Al righte reserved.]

ALBÊRUNÎ'S INDIA.

CHAPTER XLIX.

A SUMMARY DESCRIPTION OF THE ERAS.

THE eras serve to fix certain moments of time which are Page 203. mentioned in some historical or astronomical connection. tion of some Enumera- The Hindus do not consider it wearisome to reckon with of the eras huge numbers, but rather enjoy it. Still, in practical dun use, they are compelled to replace them by smaller (more handy) ones. Of their eras we mention- I. The beginning of the existence of Brahman. 2. The beginning of the day of the present nychthe- meron of Brahman, i.c. the beginning of the kalpa. 3. The beginning of the seventh manvantara, in which we are now. 4 The beginning of the twenty-eighth caturyuga, in which we are now. 5. The beginning of the fourth yuga of the present caturyuga, called kalikala, i.e the time of Kali. The whole yuga is called after him, though, accurately speaking, his time falls only in the last part of the yuga. Notwithstanding, the Hindus mean by kalikala the beginning of the kaliyuga. 6. Pundava-kala, ie. the time of the life and the wars of Bharata. All these eras vie with each other in antiquity, the VOL. II.

2 ALBERUNPS INDIA.

one going back to a still more remote beginning than the other, and the sums of years which they afford go beyond hundreds, thousands, and bigher orders of num- bers. Therefore not only astronomers, but also other people, think it wearisome and unpractical to use them. The aathor In order to give an idea of these eras, we shall use ndopta the as a first genge or point of comparison that Hindu stat yor. year the great bulk of which coincides with the year Ydsfird M

400 of Yazdajird. This namber consists ooly of hun- dreds, not of nnits and tens, and hy this peculiarity it is distinguished from all other years that might possibly be chosen. Besides, it is a memorable time; for the breaking of the strongest pillar of the religion, the decease of the pattern of a prince, Mahmud, the lion of the world, the wonder of his time -- mey God have mercy upon him I took place only s short time, less than a year, before it. The Hindu year precedes the Nanroz or new year's day of this year only by twelve deye, and the death of the prince occurred pre- cisely ten complete Persian months before it. Page 204 Now, presnpposing this our gauge as known, we shall compute the years for this point of junction, which is the beginning of the corresponding Hindu year, for the end of all years which come into question coincides with it, and the Nsuroz of the year 400 of Yazdajird falls only s little latter (viz. twelve days). How much The book Vishnu-Dharma says: " Vajra asked Mar- of the life of Brabman ksndeys how mach of the life of Brahman had elspsed; baa eiapeed somding to whereupon the sage answered : 'That which has elapsed tho Fishpu- Dharne. is 8 years, 5 months, 4 days, 6 manvantaras, 7 sarndhi, 27 caturyugas, and 3 yugas of the twenty-eighth catur- yuga, and 10 divya-years up to the time of the asvamedha which thon hast offered' He who knows the details of this statement and comprehends them duly is a sage man, and the sage is he who serves the ouly Lord and strives to reach the neighbourhood of his place, which is called Paramapada."

CHAPTER XLIX. 3

Presupposing this statement to be known, and refer- ring the reader to our explanation of the various mea- sures of time which we have given in former chapters, we offer the following analysis. Of the life of Brahman there have elapsed before our gange 26,215,732,948,132 of our years. Of the nych- themeron of Brahman, i.e. of the kalpa of the day, there have elapsed 1,972,948,132, and of the seventh manvan- tara 120,532,132. The latter is also the date of the imprisoning of the King Bali, for it happened in the first caturyuga of the seventh manvantara. In all chronological dates which we have mentioned already and shall still mention, we only reckon with complete years, for the Hindus are in the habit of dis- regarding fractions of a year. Further, the Vishnu-Dharma says : "Markandeya The time of says, in answer to a question of Vajra, 'I have already cording to Hama ac-

lived as long as 6 kalpas and 6 manvantaras of the Dharma. Vishnu-

seventh kalpa, 23 tretayugas of the seventh manvantara. In the twenty-fourth tretayuga Rama killed Ravana, and Lakshmana, the brother of Rams, killed Kumbha- karna, the brother of Ravana. The two subjugated all the Rakshasas. At that time VAlmiki, the Rishi, com- posed the story of Rama and Ramayana and eternalised it in his books. It was I who told it to Yudhishthira, the son of Pandn, in the forest of Kamyakavana.'" The author of the Vishnu-Dharma reckons here with tretayugas, first, because the events which he mentions ocenrred in a certain tretayuya, and secondly, because it is more convenient to reckon with a simple unit than with snch a unit as requires to be explained by reference to its aingle quarters. Besides, the latter part of the „tretâyuga is a more suitable time for the events men- tioned than its beginning, because it is so much nearer to the age of evil-doing (v. i. pp. 379, 380). No doubt, the date of Rama and Ramayana is known among the

4 ALBERUNI'S INDIA.

Hindus, but I for my part have not been able to ascer- tain it. Twenty-three caturyugas are 99,360,000 years, and, together with the time from the beginning of a caturyuga till the end of the tretdyuga, 102,384,000 years. If we subtract this number of years from the number of years of the seventh manvantara that have elapsed before our gauge-year, viz. 120,532,132 (v. p. 3), we get the remainder of 18,148,132 years, i.e. so many years before our gauge-year as the conjectural date of Râma; and this may suffice, as long as it is not supported by a truatworthy tradition. The here-mentioned year corresponds to the 3,892,132d year of the 28th catur- yuga. How much All these compntations rest on the measures adopted time has elapaod be- by Brahmagupta He and Pulisa agree in this, that fore o of the present the number of kalpas which have elapsed of the life of Lalpa, ne- cording ro Brahman before the present kalpa is 6068 (equal to 8 Pnlisa and Brahma- years, 5 months, 4 days of Brahman). But they differ gupts. from ecch other in converting this number into catur- yugas. According to Pulisa, it is eqnal to 6,116,544; according to Brahmagupta, only to 6,068,000 catur- yugas Therefore, if we adopt the system of Pulisa, reckoning I manvantara as 72 caturyugas without samdhi, I kalpa as 1008 caturyugas, and each yuga as the fourth part of a caturyuga, that which has elapsed of the life of Brahman before our gauge-year is the sum of 26,425,456,204,132 (!) years, and of the kalpa Page 205 there have elapsed 1,986,124,132 years, of the manvan- tara 119,884,132 years, and of the caturyuga 3,244,132 years. How much Regarding the time which has elapsed since the tme bas elapaed of beginning of the kaliyuga, there exists no difference the currant baliyngo. amounting to whole years. According to both Brahma- gupta and Pulisa, of the kaliyuga there have elapsed before our gange-year 4132 years, and between the

CHAPTER XLIX. 5

wars of Bharata and our gauge-year there have elapsed 3479 years. The year 4132 before the gauge-year is the epoch of the kalikdla, and the year 3479 before the gange-year is the epoch of the Pandavakala. The Hindus have an era called Kalayavana, regard- The era ing which I have not been able to ohtain full infor- Kalayavana. mation. They place its epoch in the end of the last dvdparayuga. The here-mentioned Yavana (JMN) severely oppressed both their country and their religion. To date by the here-mentioned eras requires in any case vast numbers, since their epochs go back to a most remote antiquity. For this reason people have given up using them, and have adopted instead the eras of- (1.) Śrt Harsha. (2.) Vikramâditya. (3.) Saka. (4) Valabha, and (5.) Gupta.

The Hindus believe regarding Sri Harsha that he Em of drl used to examine the soil in order to see what of hidden Harsba.

treasures was in its interior, as far down as the seventh earth; that, in fact, he found such treasures; and that, in consequence, he could dispense with oppressing his snbjects (by taxes, &e.) His era is used in Mathura and the country of Kanoj. Between Śri Harsha and Vikra- maditya there is an interval of 400 years, as I have been told by some of the inhabitants of that region. How- ever, in the Kashmirian calendar I have read that Śri Harsha was 664 years later than Vikramaditya. In face of this discrepancy I am in perfect uncertainty, which to the present moment has not yet been cleared up by any trustworthy information. Those who use the era of Vikramaditya live in the Era of Vik. southern and western parts of India. It is used in the rmAditya following way: 342 are multiplied by 3, which gives

6 ALBERUNPS INDIA.

the prodnet 1026. To this number you add the years which have elapsed of the current shashtyabda or sexa- gesimal samvatsara, and the sum is the corresponding year of the era of Vikramaditya. In the book Srad- hava by Mahadeva I find as his name Candrablja. As regards this method of calcnlation, we must first say that it is rather awkward and unnatural, for if they began with 1026 as the basia of the calculation, as they begin-withont any apparent necessity-with 342, this would serve the same purpose. And, secondly, admit- ting that the method is correct as long as there is only one shashtyabda in the date, how are we to reckon if there is e number of shashtyabdas! The Saka- The epoch of the era of Saka or Sakakala falls 135 kAla. years later than that of Vikramaditya. The here-men- tioned Saka tyrannised over their country between the river Sindh and the ocean, after he had made Ârya- varta in the midst of this realm his dwelling-place. He interdicted the Hindus from considering and repre- senting themselves as anything but Sakas. Some main- tain that he was a Sudra from the city of Almangura; others maintain that he was not a Hindu at all, and that he had come to India from the west The Hindus had mnch to suffer from him, till at last they received help from the east, when Vikramaditya marched against him, put him to flight and killed him in the region of Kartr, between Multan and the castle of Loni. Now this date became famons, as people rejoiced in the news of the death of the tyrant, and was used as the epoch of an era, especially by the astronomers. They honour the conqueror by adding Sri to his name, so as to say Sri Vikramâditya. Since there is & long interval between the era which is called the era of Vikramaditya (v. p. 5) and the killing of Saka, we think that thet Vik- ramâditya from whom the era has got its name is not identical with that one who killed Sake, but only a namesake of his.

CHAPTER XLIX. 7

The era of Valabha is called so from Valabha, the ruler Em of of the town Valabht, nearly 30 yojanas south of Anhil- Valbha vara. The epoch of this era falls 241 years later than Page m06. the epoch of the Saka era. People use it in this way. They first put down the year of the Sakakala, and then subtract from it the cube of 6 and the square of 5 (216 + 25 = 241). The remainder is the year of the Valabha era. The history of Valabha is given in its proper place (cf chap. xvii.) As regards the Guptakala, people say that the Guptas OuptakAla. were wicked powerful people, and that when they ceased to exist this date was used as the epoch of en era It seems that Valabha was the last of them, be- canse the epoch of the era of the Guptas falls, like that of the Valabha era, 24t years later than the Saka- kâla. The era of the astronomers begins 587 years later than Era of the the Sakakala. On this era is based the canon Khanda- mer. khadyaka by Brahmagupta, which among Muhammadans is known as Al-arkand. Now, the year 400 of Yazdajird, which we have Comparieon chosen as a gauge, corresponds to the following years of she lu- ofthe epooue

of the Indian eras :- dien oras with the tost-year. (1) To the year 1488 of the era of Śri Harsha, (2) To the year 1088 of the era of Vikramâditya, (3) To the year 953 of the Sakakala, (4) To the year 712 of the Valabha era, which is identical with the Guptakâla, (5) To the year 366 of the era of the canon Khanda- khadyaka, (6) To the year 526 of the era of the canon Panca- siddhantika by Varâhamihira, (7) To the year 132 of the era of the canon Kara- nasdra; and (8) To the year 65 of the era of the canon Karana- tilaka.

4

8 ALBERUNPS INDIA.

The eras of the here-mentioned canones are such as the authors of them considered the most suitable to be used as cardinal points in astronomical and other cal- cnlations, whence calculation may conveniently extend forward or backward. Perhaps the epochs of these eras fall within the time when the authors in question them- selves lived, but it is also possible that they fall within a time anterior to their lifetime. Oa the popu- Common people in India date hy the years of a cen- of dating by tennium, which they call samvatsara, It a centennium lr mode

marteru is finished, they drop it, and simply begin to date by a cntexnia or new one. This era is called lokakdla, i.c. the era of the nation at large. But of this era people give such totally different accounts, that I have no means of making ont the truth. In a similar manner they also differ among themselves regarding the beginning of the year. On the latter anbject I shall communicate what I have heard myself, hoping meanwhile that one day we shall be able to discover a rule in thie apparent confusion. Digerent Those who use the Saka era, the astronomers, begin oi the yen. the year with the month Caitra, whilst the inhabitants of Kanir, which is conterminous with Kashmir, begin it with the month Bhadrapada. The same people count our gauge-year (400 Yazdajird) as the eighty-fourth year of an era of theirs. All the people who inhabit the country between Bardari and Marigala begin the year with the month Karttika, and they count the gauge-year as the 11oth year of an era of theirs. The author of the Kashmirian calendar maintains that the latter year corresponds to the sixth year of a new centennium, and this, indeed, is the usage of the people of Kashmtr. The people living in the country Nirahara, behind Marigala, as far as the utmost frontiers of Takeshar and Lohavar, begin the year with the month Margafirsha, and reckon our gange-year as the 1o8th year of their

CHAPTER XLIX. 9

era. The people of Lanbaga, i.e. Lamghan, follow their example. I have been told by people of Multan that this syatem ie peculiar to the people of Sindh and Kanoj, and that they used to begin the year with the new moon of Margasirsha, bnt that the people of Multan only a few years ago had given up this syatem, and had adopted the syatem of the people of Kashmir, and followed their example in beginning the year with the new moon of Caitra. I have already before excused myself on account of Poplar the imperfection of the information given in this chap- dating in mode of ter. For we cannot offer a atrictly scientific account of the Hindus, the eraa to which it is devoted, simply because in them cisms thero- wa have to reckon with periods of time far exceeding a on. centennium, (and becanse all tradition of events farther back than a hundred years is confused (v. p. 8).) So 1 have myself seen the roundabont way in which they compute the year of the destruction of Somanath in the year of the Hijra 416, or 947 Sakakala. First, they write down the number 242, then under it 606, then under this 99. The sum of these numhers is 947, or the year of the Śakakâla. Now I am inclined to think that the 242 years have elapsed before the beginning of their centennial system, and that they have adoptad the latter together with the Guptakala; further, that the number 606 represents complete samvatsaras or centennials, each of which they Page 207. must reckon as IO1 years; lastly, that the 99 years represent that time which has elapsed of the current centennium. That this, indeed, is the nature of the calculation is confirmed by a leaf of a canon composed by Durlabha of Multan, which I have found by chance. Here the anthor says: " First write 848 and add to it the laukika- kdla, ic. the era of the people, and the sum is the ŚakakAia" If we write firat the year of the Sakakala correspond-

10 ALBERUNI'S INDIA.

ing to our gange-year, viz. 953, and subtract 848 from it, the remainder, 105, is the year of the laukika-kala, whilst the destruction of Somanath falls in the ninety- eighth year of the centennium or laukika-kala. Durlabha says, besides, that the year begina with the month Margasirsha, but that the astronomers of Multan begin it with Caitra, Orlgin of the The Hindus had kings residing in Kabul, Turks who dynasty of the Shins of were said to be of Tibetan origin. The first of them, KAbul. Barhatakin, came into the country and entered a cave in Kabul, which none could enter except by creeping on hands and knees. The cave had water, and besides he deposited there victuals for a certain number of daya. It is still known in our time, and is called Var. People who consider the name of Barhatakin as & good omen enter the cave and bring out some of ita water with great trouble. Certain troops of peasants were working before the door of the cave. Tricks of this kind can only be carried out and become notorious, if their author has made a secret arrangement with somebody else-in fact, with confederates. Now these had induced per- sons to work there continnally day and night in turns, so that the place was never empty of people. Some days after he had entered the cave, he began to creep out of it in the presence of the people, who looked on him as & new-born baby. He wore Turkish dress, a short tunic open in front, & high hat, boots and arms. Now people honoured him as & being of mira- culous origin, who had been destined to be king, and in fact he brought those countries under his sway and ruled them under the title of a shdhiya of Kabul. The rule remained among his descendants for gene- rations, the number of which is said to be about aixty. Unfortunately the Hindus do not pay much attention to the historical order of things, they are very careless

CHAPTER XLIX. 11

in relating the chronological succession of their kings, and when they are pressed for information and are at a loss, not knowing what to say, they invariably take to tale-telling. But for this, we should com- mnnicate to the reader the traditions which we have received from some people among them. I have been told that the pedigree of this royal family, written on silk, exists in the fortress Nagarkot, and I much desired to make myself acquainted with it, but the thing was impossible for various reasons. One of this series of kings was Kanik, the same who The story of is said to have built the vihdra (Buddhistic monastery) Kanik.

of Purushavar. It is called, after him, Kanik-caitya, People relate that the king of Kanoj had presented to him, among other gifts, a gorgeous and most singular piece of cloth. Now Kanik wanted to have dresses made out of it for himself, but his tailor had not the courage to make them, for he said, "There is (in the embroidery) the figure of a human foot, and whatever trouble I may take, the foot will alwaya lie between the shoulders." And that meana the same as we have already mentioned in the story of Bali, the son of Virocana (i.e. a eign of subjugation, cf, i. p. 397). Now Kanik felt convinced that the ruler of Kanoj had thereby intended to vilify and disgrace him, and in hot haste he set out with his troops marching against him. When the rdf heard this, he was greatly perplexed, for he had no power to resist Kanik. Therefore he consulted his Vaztr, and the latter said, "You have roused a man who was quiet before, and have done un- becoming things. Now cut off my nose and lips, let me be mutilated, that I may find a cunning device; for there is no possibility of an open resistance." The rdt did with him as he had proposed, and then he went off to the frontiers of the realm.

12 ALBERUNTS INDIA.

There he was found by the hostile army, was recog- nised and brought before Kanik, who asked what was the matter with him. The Vaztr said, " I tried to dissuade him from oppoaing yon, and sincerely advised him to be obedient to you. He, however, conceived a suspicion against me and ordered me to be mutilated. Since then he has gone, of his own accord, to a place which a man can only reach by a very long journey when he marches on the highroad, bnt which he may easily reach by undergoing the trouble of crossing an intervening desert, supposing that he can carry with himself water for so and so many days." Thereupon Kanik answered: "The latter is easily done." He ordered water to be carried along, and engaged the Vazir to show him the road. The Vazir marched be- fore the king and led him into a bonndless desert. After the nnmber of days had elapsed and the road did not come to an end, the king asked the Vazir what was now to be done. Then the Vazir said, "No blame attaches to me that I tried to save my master and to destroy his enemy. The nearest road leading ont of this desert is that on which you have come. Now do with me as you like, for none will leave this desert alive." Then Kanik got on bis horse and rode round a de- pression in the soil, In the centre of it he thrust his spear into the earth, and lo! water poured from it in eufficient quantity for the army to drink from and to draw from for the march back. Upon this the Vazir said, "I had not directed my ennning scheme against powerful angels, but against feeble men. As things etand thus, accept my intercession for the prince, my benefactor, and pardon him." Kanik answered, "I Fur sol. march back from this place. Thy wish is granted to thee. Thy master has already received what is dne to him." Kanik retnrned out of the desert, and the Vaztr went back to his master, the rdt of Kanoj. There he

CHAPTER XLIX. 13

found that on the same day when Kanik had thrast his spear into the earth, both the hands and feet had fallen off the body of the raf. The last king of this race was Lagaturman, and his Eind of the Vazir was Kallar, a Brahman. The latter had been for- nasty, and Tibetan dy. tunate, in so far as he had fonnd by accident hidden Brahmen ortrio of the treasures, which gave him mtch influence and power. dynasty. In consegnence, the last king of this Tibetan house, after it had held the royal power for so long a period, let it by degrees slip from his hands. Besides, Laga- turman had bad manners and a worse behaviour, ou account of which people complained of him greatly tc the Vazir. Now the Vazir put him in chains aud imprisoned him for correction, but then he himself found ruling sweet, his riches enabled him to carry out his plans, and so he occupied the royal throne. After him ruled the Brahmnan kings Samand (Samanta), Kamalû, Bhîm (Bhima), Jaipal (Jayapâla), Ânanda- pala, Tarojanapala (Trilocanapala). The latter was killed A.II. 412 (A.D. 1021), and his son Bhimspala five years later (A.D. 1026). This Hindn Shahiya dynasty is now extinct, and of the whole house there is no longer the slightest rem- nant in existence. We must say that, in all their grandeur, they never slackened in the ardent desire of doing that which is good and right, that they were men of noble sentiment and noble bearing. I admire the following passage in a letter of Auandapala, which he wrote to the prince Mabmud, when the relations be- tween them were already strained to the utmost: "I have learned that the Turks have rebelled against you and are spreading in Khurasao. If you wish, I shall come to you with 5000 horsemen, 10,000 foot-soldiers, and 100 elephants, or, if you wish, I shall send you my son with donble the number. In acting thus, I do not speculate on the impression which this will make on you. I have been conquered by you, and

14 ALBERUNrS INDIA.

therefore I do not wish that another man should conquer you." The same prince cherished the bitterest hatred against the Muhammadans from the time when his son was . made a prisoner, whilst his son Tarojanapala (Triloca- napAla) was the very opposite of his father.

( 15 )

CHAPTER L.

HOW MANY STAR-CYCLES THERE ARE BOTH IN A " KALPA" AND IN A "CATURYUGA."

Ir is one of the conditions of a kalpa that in it the planets, with their apsidea and nodes, must unite in o° of Aries, i.e. in the point of the vernal equinox. Therefore each planet makes within a kalpa a certain number of complete revolutions or cycles. These atar-cycles as known through the canon of The tradi- Alfazart and Ya'kub Ibn Tarik, were derived from a zarl and tiun of Alfa-

Hindu who came to Bagdad as a member of the politi- Tank. cal mission which Sindh sent to the Khalif Almansur, AH. 154 (=A.D. 771). If we compare these secondary statements with the primary statements of the Hindus, we discover discrepancies, the cause of which is not known to me. Is their origin dne to the translation of Alfazari and Ya'kub? or to the dictation of that Hindn? or to the fact that afterwards these computa- tions have been corrected by Brahmagupta, or some one else ? For, certainly, any acholar who becomes aware of mistakes in astronomical computations and takes an interest in the subject, will endeavour to correct them, Muhammad as, e.g. Muhammad Ibn Ishak of Sarakhs has done. Sarakha ' jbu labåk of

For he had discovered in the computation of Saturn a . falling back behind real time (i.e., that Saturn, accord- ing to this compntation, revolved alower than it did in reality). Now he assiduonsly atudied the subject, till at last he was convinced that his fault did not originate

16 ALBERUNTS INDIA.

from the equation (ie. from the correction of the places of the stars, the computation of their mean places). Then he added to the cycles of Saturn one cycle more, and compared his calculation with the actnal motion of the planet, till at last he found the calculation of the cycles completely to agree with astronomical observa- tion. In accordance with this correction he states the star-cycles in his canon. Aryabhets Brahmagupta relates a different theory regarding the ganted by cycles of the apsides and nodes of the moon, on the anthority of Aryabhata We quote this from Brah- magupta, for we could not read it in the original work of Aryabhata, but only in a quotation in the work of Nenber ol the rota- Brahmagupta. tiơng o the plaets in a The following table contains all these traditions, which will facilitate the study of them, if God will !

The planets. Number of their rovolutioms in a Number of tho revolationa of Nutober of the ro-

Kalpa, thoir apeides. volationa of their nodea.

Sen Brahmagupta 4,320,000,000 480 Has no node.

The translition 488,105,858 232,311,168 232,312,138

Aryabhats .

The anomalistic 488,219,000 232,316,000

revolution of 57,265,194,142 The anomalistic

the moon ac- revolution of the Moon.

cording moon is here to 57.753,300,000 Brahmagupta treated as if it were the apsis, being the differ- ence between the motion of the moon and that of the apsis. (See

Mars Mercury 2,296,828,522 the notes.) 292 267 . Jupitor . 17,936,998,984 364,226,455 332 855 521 . 63 . Venus Brahmagupta 7,022,389.492 653 893

Thetranglation 146,567,298

of Alfasirt The correctioa . 146,569,238 41 584

of Alsarakhal Sataro. <

The fixed stars 120,000 according to the translatinn of Alfaztrt.

CHAPTER L. I7

The computation of these cycles rests on the mean Oycles of motion of the planets. As a caturyuga is, according to in a calur the planeta

Brahmagupta, the one-thousandth part of a kalpa, we taliyuga. have only to divide these cycles by 1000, and the quotient is the number of the star-cycles in one catur- yuga. Likewise, if we divide the cycles of the table by 10,000, the qnotient is the number of the atar-cycles in a kaliguga, for this is one-tenth of a caturyuga. The fractions which may occor in those quotients are raised to wholes, to caturyugas or kaliyugas, by being mnlti- plied by a number equal to the denominator of the fraction. The following table represents the star-cycles speci- ally in a caturyuga and kaliyuga, not those in a man- vantara. Although the manvantaras are nothing but multiplications of whole caturyugas, still it is difficult to reckon with them on account of the samdhi which is attached both to the beginning and to the end of them.

The names of the planeta. Their revolutions in & Caturyuga. Their revolutions in a Kullyuga Page 210.

Son His psis . 4,320,000 432,000

Moon 57.753,300 5,775-330 . . Brahmagupta Aryabhata 488,105188 488,219 48,8102129 48,821r6 Her anomalistic revolutioo Brahmagupta 57,265,194775 5,726,5191834

The translation of 232,31 L11s opou 23,2311855 Her

Alfazart Aryabhata 232,312r8s 23,2311842 Her

Mars 232,316 23,2312

His apsis 2,296,828?11 229,6821111

Hig node .

Mercury 0.347

His apsis 17.936,998138 1,793,6991123

His node Jupiter His apsis 364.226,5 36,4225681 His node

VOL II. B

18 ALBERUNrS INDIA ..

The names of tho pluaots, Thofr revolutions in a Caturyors Thetr roolutions in a Kallyags.

Venns Her apeis 7,022,3893H 702.2383114

Her node OTTHT

Satarn His apais 146,567111 14.656111

His nodo The translation of Alfazart 146,569.34 OTLIT 14,6561111

The correction of Alsarakhsf . 146,569111 14.6561118 Saturn : The fired stars . 120 12 .

Pago att. After we hsve stated how many of the star-cycles of Biar-oyelen of a talja a kalpa fall in a caturyuga and in a kaliyuga, according nad catar- ynga, ac- to Brahmagupta, we shall now derive from the number cording to Pultan, of star-cycles of a caturyuga according to Pulisa the number of star-cycles of a kalpa, first reckoning a kalpa = 1000 caturyugas, and, secondly, reckoning it as 1008 caturyugas. These numbers are contained in the following table :-

The Fugas according to Pulisa.

The names of tho Nutnber of Number of their Number of thotr

plabetz their revolu- tions in & ravolutions in rorolutiona: in. a Kalpa of Caturyuyt Kalpa of Icoo Caturyngas, 1008 Caturyugas.

Son Moon . 4,320,000 4,320,000,000 4,354-560,000

Her apsis. 57.753-336 58,215,362,688

Her node 488,219 57753-336,000

232,226 488,219,000 492,124,752

Mars 232,226,000 234,083,808

Mercury 2,296,824 2,296.824,000 2315,198.592

Jupiter 17,937,000 17,937,000,000 18,080,496,000

Venus 364,220 364,220,000 367,133.760

Saturn 7,022.388 146,564 7,022,388,c00 146,564,000 7,078,567,104 147.736,512

Transforma- We meet in this context with a curious circumstance. ton of tho word Aryt- Evidently Alfazari and Ya'kub sometimes heard from bhata among tha their Hindu master expressions to this effect, thst his Arabi. calculation of the star-cycles was that of the great Sid- dhanta, whilst Aryabhata reckoned with one-thousandth

CHAPTER L. 19

part of it. They apparently did not understand him properly, and imagined that dryabhata (Arab. arjabhad) meant a thousandth part. The Hindus pronounce the d of this word something between a d and an r. So the consonant became changed to an r, and people wrote drjabhar. Afterwards it was still more mutilated, the first r being changed to a z, and so people wrote drja- thar. If the word in this garb wanders back to the Hindus, they will not recognise it. Further, Abt-alhasan of Al'ahwâz mentions the revo- Star-epeles lutions of the planets in the years of al-arjabhar, i.c. in Aba-alpasan sccording to

caturyugas. I shall represent them in the table auch of Al'ahwiz

as I have found them, for I guess that they are directly derived from the dictation of that Hindu. Possibly, Pago aia. therefore, they give us the theory of Aryabhata. Some of these numbers agree with the star-cycles in a catur- yuga, which we have mentioned on the authority of Brahmagupta; others differ from them, and agree with the theory of Pulisa; and a third class of numbers differs from those of both Brahmagupta and Pulisa, as the examination of the whole table will show.

The names of the Their Tugas as parts

plemeta. of a Caturyuga sccording to Abo-alposan Afahwis.

Sun Moon . 4,320,000

Her apsjs 57,753,336

Her node 488,219

Mars 232,226

Meroury 2,296,828

Jnpiter 17,937,020

Venus . 364,224

Saturn 7,022,388 146,564

( 20 )

CHAPTER LI.

AN EXPLANATION OF THE TERMS " ADHIMASA," " CNA- RATRA," AND THE " AHARGANAS," AS REPRESENTING DIFFERENT SUMS OF DAYS.

On the lap THE months of the Hindus are lunar, their years solar ; month. therefore their new year's day must in each solar year fall by so much earlier as the lunar year is shorter than the solar (ronghly speaking, by eleven days). If this precession makes up one complete month, they act in the same way as the Jews, who make the year a leap year of thirteen months by reckoning the month Adar twice, and in a similar way to the heathen Arabs, who in a so-called annus procrastinationis postponed the new year's day, thereby extending the preceding year to the duration of thirteen months. The Hindus call the year in which a month is repeated in the common language malamâsa. Mala means the dirt that clings to the hand. As snch dirt is thrown away, thus the leap month is thrown away out ef the calculation, and the number of the months of a year remains twelve. However, in the literature the leap month is called adhimasa. That month is repeated within which (it being con- sidered as a solar month) two lunar months finish. If the end of the lunar month coincides with the beginning of the solar month, if, in fact, the former ends before any part of the latter has elapsed, this month is re- peated, because the end of the lunar month, althongh

CHAPTER L 21

it has not yet run into the new solar month, still does no longer form part of the preceding month. If a month is repeated, the first time it has its ordinary name, whilst the second time they add before the name the word durd to distinguish between them. If, eg. the month Âshadha is repeated, the first is called Pago 113. Âshadha, the second Durdshddha. Tha first month is that which is disregarded in the calculation. The Hin- dus consider it as unlucky, and do not celebrate any of the festivals in it which they celebrate in the other months. The most nnlucky time in this month is that day on which the lunation renches its end. The author of the Vishnu-Dharma says: " Candra Quotation (mana) is smaller than sdrana, i.e. the lunar year is Fishyr- from the smaller than the civil year, by six days, i.e. unaratra. Dkarma. Ona means decrease, deficiency. Saura is greater than candra by elevan days, which gives in two years and seven months the supernumerary adhimasa month. This whole month is unlucky, and nothing must be done in it." This is a rough description of the matter. We ahall now describe it accurately. The lunar year has 360 lunar days, the solar year has 3712c lunar days. This difference sums up to the thirty days of an adhimasa in the course of 976,776 lunar days, i.e. in 32 months, or in 2 years, 8 months, 16 days, plus the fraction: er lunar day, which is nearly = 5 minutes, 15 seconds. As the religious reason of this theory of intercala- Quotation tion the Hindus mention a passage of the Veda, which Vola. from tho

they have read to us, to the following tenor: " If the day of conjunction, i.e. the first lunar day of the month, passes without the aun'e marching from one zodiacal aign to the other, and if this takes place on the following day, the preceding month falls out of the calculation." The meaning of this passaga is not correct, and the CHticiema fault must have risen with the man who recited and thereon.

22 ALBBRUNTS INDIA. trenslated the passage to ma For a month has thirty lunar days, and a twelfth part of the solar year has 3oifH lunar days This fraction, reckoned in day- minntes, is equal to 55' 19" 22" 30". If we now, for example, suppose a conjunction or new moon to take place at o° of a zodiacal sign, we add this fraction to the time of the conjunction, and thereby we find the times of the sun's entering the signs successively. As now the difference between a lunar and a solar month is only a fraction of s day, the son's entering a new sign may natnrally take place on any of the days of the month. It may even heppen that the aun enters two consecutive signs on the same month-day (e.g. on the second or third of two consecutive months) This is the case if in one month the sun enters & sign before 4'40" 37t 30tr have elapsed of it; for the next follow- ing entering a sign falla later by 55' 19" 23"1 30h, and both these fractions (ie. less than 4' 40" 37" 30'r plus the last-mentioned fraction) added together are not sufficient to make np one complete day. Therefore the quotation from the Veda is not correct. Tropored 'explanation I sappose, however, that it may have the following the Vedie correct meaning :- If a month elapses in which the sun does not march from one sign to another, this month is disregarded in the calculation. For if the sun entera a sign on the 29th of a month, when at least 4' 40" 37!" 30" have elapsed of it, this entering takes place before the beginning of the sncceeding month, and therefore the latter month is withont an entering of the sun into a new sign, because the next following entering falls on the first of the next but one or third month. If you compute the consecutive enterings, beginning with a conjunction taking place in o° of a certain sign, you find that in the thirty-third month the sun enters s new sign at 30' 20" of the twenty-ninth day, and that ho enters the next following sign at 25' 39" 22" 30' of the first day of the thirty-fifth month,

CHAPTER LI. 23 Hence also becomes evident why this month, which is disregarded in the calenlation, is considered as un- Incky. The reason is that the month misses just that moment which is particularly adapted to earn in it a beavenly reward, viz. the moment of the snn's entering a new sign. As regards adhimdsa, the word means the first month, for AD means beginning (i.c. ddi). In the books of Ya'kub Ihn Tarik and of Alfazart this name is written padamdsa. Pada (in the orig. P-Dh) means end, and it is possible that the Hindus call the leap month by both names; bnt the reader must be aware that these two authors frequently misspell or disfigure the Indian words, and that there is no reliance on their tradition. I only mention this becanse Pulisa explains the latter of the two months, which are called by the same name, as the supernumerary one. The month, as the time from one conjunction to the Erplanation following, is one revolution of the moon, which revolves uaiocroel or of the terms

through the ecliptic, but in a course distant from that monthe and of the sun. This is the difference between the motions dayı. of the two heavenly luminaries, whilst the direction in which they move is the same. If we snhtract the revolutions of the sun, i.e. the solar cycles of & kalpa, from its lunar cycles, the remainder shows how many more lunar months a kalpa has than solar months. All months or days which we reckon as parts of whole kalpas we call here universal, and all months or daya which we reckon as parta of a part of a kalpa, e.g. of a caturyuga, we call partial, for the purpose of sim- plifying the terminology. ' The year has twelve solar months, and likewise Univeral twelve lunar months. The lunar year is complete with moutu twelve months, whilst the solar year, in consequence of the difference of the two year kinds, has, with the addition of the adhimdsa, thirteen months. Now evi- dently the difference between the universal solar and

ALBERUNPS INDIA.

lunar months is represented by these supernumerary months, by whioh a single year is extended to thirteen months. These, therefore, are the universal adhimdsa months. The unirersal solar months of a kalpa are 51,84C, 000,000; the universal lunar months of a kalpe are 53.433,300,000. The difference between them or the adhimdsa months is 1,593,300,000. Multiplying each of these numbers by 30, we get days, viz. solar days of a kalpa, 1,555,200,000,000; Innar days, 1,602,999,000,000; the days of the adhimasa months, 47.799,000,000. In order to reduce these numbers to emaller ones we divide them by a common divisor, viz 9,000,000. Thus we get as the sum of the days of the solar months 172,800; as the sum of the days of the lunar months, 178,111; and as the sum of the days of the adhimdsa months, 5311. If we further divide the universal solar, civil, and noh, lanss, lunar days of a kalpa, each kind of them separately, by How many and eivil deya are ro- quired for the unicersal adhimdsa months, the quotient represents the forasa- Hơn of an the number of days within which a whole adhimdsa month sums up, viz. in 976444, solar days, in 1006151r lunar days, and in 99014g civil days. This whole compntation rests on the measures which Brahmagopta adopts regarding a kalpa and the star- oycles in a kalpa. The compo- According to the theory of Pulisa regarding the tatlon of caturyuga, a caturyuga has 51,840,000 solar months, copording to Pullm. 53433,336 lunar months, 1,593,336 adhimdsa months. Accordingly a caturyuga has 1,555,200,000 solar days, r,603,000,080 lunar days, 47,800,080 days of adkimdsa months. If we rednce the numbers of the months by the common divisor of 24, we get 2,160,000 solar months, 2,226,389 lunar months, 66,389 adhimdsa months. If . we divide the nombers of the day by the common

CHAPTER LI. 25

divisor of 720, we get 2,160,000 solar daya, 2,226,389 lunar days, 66,389 days of the adhimdsa months. If we, lastly, divide the universal solar, lunar, and civil days of a caturyuga, each kind separately, by the uni- versal adhimdsa months of a caturyuya, the qnotient represents the numbers of days within which a whole adhimasa month sums up, viz. in 976655a solar days, in 1006455 lnnar days, and in 990311g: civil days. These are the elements of the computation of the adhimasa, which we have worked ont for the benefit of the following investigations. Regarding the cause which necessitates the dnardtra, Explanstion lit. the days of the decrease, we have to consider the fol- dnardtra. of the term

lowing. If we have one year or a certain number of years, and reckon for each of them twelve months, we get the corresponding number of solar months, and by mnlti- plying the latter by 30, the corresponding number ... of solar days, It is evident that the number of the lunar months or days of the same period is the same, plus an increase which forms one or several adhimdsa months. If we reduce this increase to adhimdsa months due to the period of time in question, according to the relation between the universal solar months and the universal adhimdsa months, and add this to the months or days of the years in question, the sum represents the partial lunar days, i.c. those which correspond to the given number of yesrs This, however, is not what is wanted. What we want ie the number of civil days of the given number of years which are less than the lunar days; for one civil day ia greater than one lunar day. Therefore, in order to find that which is sought, we must subtract some- thing from the number of Innar days, and this element which must be subtracted is called unardtra. The tnardtra of the partial lunar days stands in the same relation to the universal lunar days as the uni-

16 : ALBERUNPS INDIA.

versal civil days are leas than the universal lunar deys. The univeraal lunar days of a kalpa tre 1,602,999,000,000. This number is larger than the number of universal civil days hy 25,082,550,000, which represents the uni- versal Anardtra. Both these numbers may be diminished by the com- mon divisor of 450,000. Thus we get 3,562,220 uni- versal lunar daya, and 55.739 universal unardtra days. Coapte- According to Pulisa, a caturyuga has 1,603,000,080 tiơn uf the Caurátra Innar days, and 25,082,280 Anardtra dsys. The com- Cocording to Puttee mon divisor by which both numbera may be reduced is 360. Thus we get 4,452,778 lunar days and 69,673 unaratra days. These are the rules for the computation of the ana- ratra, which we ahall hereafter want for the compu- tation of the ahargana. The word means sum of days; for dh means day, and argana, sum. Criticiams Ya'kub Ibn Tarik has made a mistake in the compu- on Takab Ibo Tirik. tation of the solar days; for he maintains that you get

Pago $16. them by subtracting the solar cycles of & kalpa from the civil days of a kalpa, i.c. the universal civil daya. But this is not the case. We get the solar daya by multiplying the solar cycles of a kalpa by 12, in order to rednce them to months, and the product by 30, in order to reduce them to days, or by multiplying the number of cycles by 360. In the computation of the lunar daya he has first taken the right course, multiplying the lunar months of a kalpa by 30, but afterwards he again falls into & mistake in the computation of the daya of the unardtra. For he maintains that you get them by aubtracting the solar days from the lunar days, whilst the correct thing is to subtract the civil daye from the lunar days.

( 27 )

CHAPTER LII.

ON THE CALCULATION OF " AHARGANA" IN GENKRAL THAT IS, THE RESOLUTION OF YEARS AND MONTHS INTO DAYS, AND, VICE VERSA, THE COMPOSITION OF YEARS AND MONTHS OUT OF DAYS.

THE general method of resolution is as follows: The General ruls complete years are multiplied by 12; to the product are the sdrond how to find

added the months which have elapsed of the current hargaya.

year, [and this sum is multiplied by 30;] to this produet are added the days which have elapsed of the current month. The sum represents the saurdhargana, ie the sum of the partial soler days. You write down the number in two places. In the one place you multiply it by 5311, i.c. the number which represents the universal adhimasa months. The product yon divide by 172,800, ie. the number which represents the universal solar months. The quotient you get, as far as it contains complete days, is added to the number in the second place, and the sum represents the candrahargana, ie. the sum of the partial lunar days. The latter number is again written down in two different places. In the one place you multiply it by 55,739, i.a. the number which represents the universal unardtra days, and divide the product by 3.562,220, i.e. the number which represents the oniversal lunar days. The quotient you get, as far as it represents complete days, is subtracted from the number written in the second place, and the remainder is the sdrandhargana, i.a the sum of civil days which we wanted to find.

28 ALBBRUNS INDIA.

However, the reader must know that this computa- talled rule for the cume tion applies to dates in which there are only complete purpora. adhimdsa and unardira daya, without any fraction. If, therefore, a given nnmber of years commences with the beginning of a kalpa, or a caturyuga, or a kaliyuga, this computation is correct. Bnt if the given years begin with some other time, it mnay by chance happen that this compntation is correct, bnt posaibly, too, it may result in proving the existence of adhimdsa time, and in that case the compntation would not be correct. Also the reverse of these two eventualities may take place. However, if it is known with what particnlar moment in the kalpa, caturyuga, or kaliyuga a given nnmber of years commences, we use a apecial method of com- pntation, which we shall hereafter illustrate by some examples. The latter We ahall carry out this method for the begin- mathod DarTled out ning of the Indian year Sakakala 953, the same year for daka- kAla 953- which we nse as the gauge-year in all these computa- tions. First we compnte the time from the beginning of the life of Brahman, according to the rules of Brahma- gupta We have already mentioned that 6068 kalpas have elapsed before the present one. Multiplying this by the well-known number of the days of a kalpa (1,577.916,450,000 civil days, vide i. p. 368), we get 9,574.797,018,600,000 as the anm of the days of 6068 kalpas. Dividing this number by 7, we get 5 as a remainder, and reckoning five days backwards from the Saturday which is the last day of the preceding kalpa, we get Tuesday as the first day of the life of Brahman. We have already mentioned the sum of the days of a caturyuga (1,577,916,450 daya, v. i. p. 370), and have explained that a kritayuga is equal to four-tenths of it, i.a 631,166,580 daya A manvantara has seventy-one Pare 117. times as mnch, i4 112,032,067,950 days. The days of

CHAPTER LII. 29

six manvantaras and their samdhi, consisting of seven kritayuga, are 676,610,573,760. If we divide this number by 7, we get a remainder of 2. Therefore the six manvantaras end with a Monday, and the seventh begina with a Tuesday. Of the seventh manvantara there have already elapsed twenty-seven caturyugas, i ... 42,603,744,150 days. If we divide this number by 7, we get a remainder of 2. Therefore the twenty-eighth caturyuga begins with a Thursday. The days of the yugas which have elapsed of the present caturyuga are 1,420,124,805. The division by 7 gives the remainder I. Therefore the kaliyuga begins with a Friday. Now, returning to our gauge-year, we remark that the years which have elapsed of the kalpa up to that year are 1,972,948,132. Multiplying them by 12, we get as the number of their months 23,675,377,584 In the date which we have adopted as gauge-year there is no month, but only complete years; therefore we have nothing to add to this number. By multiplying this number by 30 we get days, viz, 710,261,327,520. As there are no days in the normal date, we have no daya to add to this number. If, therefore, we had multiplied the number of years by 360, we should have got the same result, viz. the partial solar days. Multiply this number by 5311 and divide the pro- duct by 172,800. The quotient is the number of the adhimsa days, viz. 21,829,849,018198. If, in multi- plying and dividing, we had used the months, we ahould have found the adhimasa months, and, multi- plied by 30, they would be equal to the here-meutioned number of adhimasa days. If we further add the adhimasa days to the partial solar days, we get the sui of 732,091,176,538, ie. the partial lunar days. Multiplying them by 55.739, and

30 ALBERUNS INDIA. dividing the product by 3,562,220, we get the partial unardtra days, vis, 11,455,224,5751:141:41+ This sum of days without the fraction is subtract- ed from the partial lunar days, and the remainder, 720,635,951,963, represents the number of the civil daya of our gange-date. Dividing it by 7, we get as remainder 4, which means that the last of these days is a Wednesday. Therefore the Indian year commences with a Thursday. If we further want to find the adhimdsa time, we . divide the adhimasa days by 30, and the quotient is the number of the adhimdsas which have elapsed, viz. 727,661,633, plus a remainder of 28 days, 51 minntes, 30 seconds, for the carrent year. This is the time which has already elapsed of the adhimasa month of the current year. To become a complete month, it only wants : day, 8 minutes, 30 seconds more. We have here used the solar and Innar days, the applisd tos adhimdsa and fnardtra days, to find a certain past caleolation

sccordiug to portion of a kalpa. We shall now do the same to find the theor the past portion of a coturyuga, and we may use the same elements for the compntation of & caturyuga which we have used for that of a kalpa, for both methods lead to the same result, as long as we adhere to one and the same theory (eg. that of Brahmagupta), and do not mix up different chronological systems, and as long as each gunakdra and its bhagabhara, which we here mention together, correspond to each other in the two compntations. The former term means a multiplicator in all kinds of calculations. In our (Arahic) astronomical hand- books, as well as those of the Persians, the word occura in the form guncdr. The second term means each divisor. It occurs in the astronomical handbooks in the form bahcar. It wonld be useless if we were to exemplify this com- potetion on a caturyuga according to the theory of Brah-

CHAPTER LII. 31

magupta, as according to him a caturyuga is simply one- Page art. thousandth of a kalpa. We ahould only have to ahorten the above-mentioned numbers by three ciphers, and in every other respect get the same results. Therefore we shall now give this computation according to the theory of Pulisa, which, though applying to the caturyuga, is similar to the method of computation used for a kalpa. According to Pulisa, in the moment of the beginning of the gange-year, there have elapsed of the years of the caturyuga 3,244,132, which are equal to 1,167,887,520 solar days. If we multiply the numher of months which corresponds to this number of days with the number of the adhimdsa months of a caturyuga or a corresponding multiplicator, and divide the product by the number of the solar months of a caturyuga, or a corresponding divisor, we get as the number of adhi- masa months 1, 196,52514851 Forther, the past 3,244,132 years of the caturyuga are. 1,203,783,270 lnnar days. Maltiplying themn by the number of the unardtra days of a caturyuga, and dividing the product by the lunar days of a caturyuga, we get as the number of unardtra days 18,835,70088.058 Accordingly, the civil days which have elapsed since the beginning of the cuturyuga nte 1,184.947,570, and this it was which we wanted to find We shall here communicate a passage from the A similar Pulisa-siddhanta, describing a similar method of com- computation pntation, for the purpose of rendering the whole subject the Pulid- clearer to the mind of the reader, and fixing it there riddkanta. more thoroughly. Pulisa says: "We first mark the kalpas which have elapsed of the life of Brahman before the present kalpa, i.c. 6068. We multiply this number by the number of the caturyugas of a kalpa, i.c. 1008. Thus we get the product 6,116,544. This number we multiply by the number of the yugas of a caturyuga, ic. 4, and get the produet 24,466,176. This nnmber we multiply by the number of years of a yuga,

32 ALBERUNIS INDIA.

ic. 1,080,000, and get the prodnct 26,423,470,080,000. These are the years which have elapsed before the present kalpa. We further multiply the latter number by 12, so as to get months, viz. 317,081,640,960,000. We write down this number in two different places. In the one place, we multiply it by the number of the adhimasa months of a caturyuga, i.e. 1,593.336, or a corresponding number which has been mentioned in the preceding, and we divide the product by the num- ber of the solar months of a caturyuga, ie. 51,840,000. The quotient is the number of adhimdsa months, viz. 9,745.709,750,784 This number we add to the number written in the second place, and get the snm of 326,827,350,710,784- Multiplying this number by 30, we get the product. 9,804,820,521,323,520, viz lnnar days. This number is again written down in two different places. In the one place we multiply it by the unaratra of a caturyuga, i.e. the difference between civil and Innar days, and divide the product by the lunsr days of a caturyuga. Thus we get as quotient 153,416,869,240,320, ic. unarâtra days. We subtract this nnmber from that one written in the second place, and we get as remainder 9.651,403,652,083,200, ie. the days which have elapsed of the life of Brahman before the present kalpa, or the days of 6068 kalpas, each kalpa having 1,590,541,142,400 days. Dividing this sum of days by 7, we get no remainder. This period of time ends with a Saturday, and the present kalpa commences with a Sunday. This shows that the beginning of the life of Brahman too was a Sunday. Of the current kalpa there have elapsed six manvan- Page 219 taras, each of 72 caturyugas, and each caturyuga of 4,320,000 years. Therefore six manvantaras have 1,866,240,000 years. This number we compute in the

CHAPTER LII. 33

same way as we have done in the preceding example, Thereby we find as the number of days of six complete manvantaras, 681,660,489,600. Dividing this nnmber by 7, we get as remainder 6. Therefore the elapsed manvantaras end with & Friday, and the seventh man- vantara begins with & Saturday. Of the current manvantara there have elapsed 27 caturyugas, which, according to the preceding method of computation, represent the number of 42,603,780,600 days. The twenty-seventh caturyuga ends with a Monday, and the twenty-eighth begins with a Tnes- day. Of the current caturyuga there have elapsed three yugas, or 3,240,000 years. These represent, according to the preceding method of computation, the number of 1,183,438,350 days. Therefore these three yugas end with a Thursday, and kaliyuga commences with a Friday. Accordingly, the snm of days which have elapsed of the kalpa is 725,447,708,550, and the sum of days which have elapsed hetween the beginning of the life of Brahman and the beginning of the present kaliyuga is 9,652,129,099,791,750. To judge from the quotations from Aryabhata, as we The method have not seen a book of his, he seems to reckon in the employed of akargaŅa

following manner :- by Arya- bhata. The sum of days of a caturyuga is 1,577,917,500. The time between the beginning of the kalpa and the beginning of the kaliyuga is 725,447,570,625 days. The time between the beginning of the kalpa end our gauge-date is 725,449,079,845. The number of days which have elapsed of the life of Brabman before the present kalpa is 9,651,401,817,120,000. This is the correct method for the resolution of years into days, and all other measures of time are to be treated in accordance with this. We have already pointed out (on p. 26) a mistake VOL. II. c

34 ALBERUNIS INDIA.

The akar- of Ya'kub Ibn Tarik in the calculation of the universal given by solar and dnardtra days. As he translated from the Yn Tab Ibn Indian language a calculation the reasons of which he did not understand, it would have been his duty to examine it, and to check the various numbers of it one by the other. He mentions in his book also the method of ahargana, ie. the resolution of years, but his descrip- tion is not correct; for he says :- " Multiply the months of the given number of years by the number of the adhimdsa months which have elapsed up to the time in question, according to the well-known rules of adhimdsa. Divide the product by the solar months. The quotient is the number of complete adhimdsa months plus its fractions which have elapsed up to the date in question." The mistake is here so evident that even a copyist would notice it; how much more a mathematician who makes a computation according to this method; for he multiplies by the partial adhimasa instead of the universal. A cocond Besides, Ya'kth mentions in his book another and method given by perfectly correct method of resolution, which is this: Ya kub. "When you have found the number of months of the years, multiply them by the number of the lunar months, and divide the product by the solar months The quotient is the number of adhimdsa months toge- ther with the number of the months of the years in question. "This number you multiply by 30, and yon add to the product the days which have elapsed of the current month. The eum represents the lunar days. "If, instead of this, the first number of months were mnltiplied by 30, and the past portion of the month were added to the product, the sum would represent the partial solar days; and if this number were further computed according to the preceding method, we should get the adhimasa days together with the solar days."

CHAPTER LII. 35

The rationale of this calculation is the following :- If Expllostion we multiply, as we have done, by the number of the mentloned of the Laat

universal adhimasa months, and divide the product by mothod.

the universal solar months, the quotient represents the portion of adhimsa time by which we have multiplied. As, now, the lunar months are the sum of solar and adhimdsa months, we multiply by them (the lunar months) and the division remains the same. The quo- tient is the sum of that namber which is mnltiplied and that one which is songht for, i.e. the lunar days. We have already mentioned in the preceding part that by multiplying the lunar daye by the universal Paye m2a unardtra days, and by dividing the product by the universal lunar days, we get the portion of unaratra days which belongs to the number of luner days in question. However, the civil days in a kalpa are less than the lunar days by the amount of the unaratra days. Now the lunar days we have stand in the same relation to the lunar days minus their due portion of unaratra days as the whole number of lunar days (of a kalpa) to the whole number of lunar days (of a kalpa) minus the complete number of unaratra days (of a kalpa); and the latter number are the universal civil days. If we, therefore, multiply the number of lunar days we have by the universal civil days, and divide the product by the universal lunar days, we get as quotient the number of civil days of the date in ques- tion, and that it was which we wanted to find In- stead of multiplying by the whole sum of civil days (of a kalpa), we multiply by 3,506,481, and instead of dividing by the whole number of lunar daya (of a kalpa), we divide by 3,562,220. The Hindus have etill another method of calculation. Another. It is the following :- "They multiply the elepsed years ahargana of method of

of the kalpa hy 12, and add to the product the com- the Hindus.

plete months which have elapsed of the current year. The sum they write down above the number 69,120,

36 ALBERUNIS INDIA.

and the number they get is snbtracted from the num- (Lacuna)

ber written down in the middle place. The donble of the remainder they divide by 65. Then the quotient represents the partial adhimdsa montha This number they add to that one which is written down in the uppermost place. They multiply the sum by 30, and add to the prodnct the days which have elapsed of the current month. The sum represents the partial solar days. This number is written down in two different places, one under the other. They multiply the lower number by 11, and write the prodnct under it. Then they divide it by 403,963, and add the quotient to the middle number. They divide the sum by 703, and the quotient represents the partial Anardtra days. This number they subtract from the number written in the nppermost place, and the remainder is the number of civil days which we want to find." Expliostion The rationale of this compntation is the following :- of the latter mothid. If we divide the nniversal solar monthe by the uni- versal adhimdsa months, we get as the measure of one adhimdsa month 3211835 sist solar months. The double of this is 657i3s solar months. If we divide by this nnmber the double of the months of the given years, the quotient is the number of the partial adhimdsas. How- ever, if we divide by wholes plus a fraction, and want to subtract from the number which is divided a certain portion, the remainder being divided by the wholes only, and the two snbtracted portions being equal por- tions of the wholes to which they belong, the whole divisor stands in the same relation to its fraction as the divided number to the subtracted portion. The lattar If we make this computation for our gauge-year, we method applied to tho gnupc- get the fraction of r.ost.soo, and dividing both num- bers by 15, we get $51so It would also be possible here to reckon by single adhimasas instead of donble ones, and in that case it

CHAPTER LII. 37 would not be necessary to double the remainder. Bnt the inventor of this method seems to have preferred the rednplication in order to get smaller numbers; for if we reckon with single adhimasas, we get the fraction stog, which may be rednced by 96 as a common divisor. Thereby we get 89 as the multiplicator, and 5400 as the divisor. In this the inventor of the method has shown his sagacity, for the reason for his computation is the intention of getting partial lunar daye and smaller multiplicators. His method (i.e. Brahmagupta's) for the computation Method for of the anardira days is the following :- the compu- tation of the If we divide the universal lunar days by tho uni- days accord- Anarâtra

versal unardtra days, we get as quotient 63 and a Brahma- ing to

fraction, which may be reduced by the common divisor Pare zar. gupta

450,000. Thus we get 6352:953 lunar days as the period of time within which one dnaratra day sums up. If we change this fraction into eleventh parts, we get 1'r and a remainder of 43:1$1, which, if expressed in minutes, is equal to o' 59" 54"". Since this fraction is very near to one whole, people have neglected it, and use, in a rough way, 2i instead. Therefore, according to the Hindus, one unardtra day sums np in 6319 or lunar days. If we now multiply the number of unardtra days, which corresponds to the number of lunar daye by 6318:933, the product is less than that which we get by multiplying by 6319. If we, therefore, want to divide the lunar days by T', on the supposition that the quotient is equal to the first number, a certain portion must be added to the lunar days, and this portion he (the author of Pulisa-Siddhanta) had not computed accu- rately, bnt only approximatively. For if we multiply the universal unardtra days by 703, we get the product 17,633,032,650,000, which is more than eleven times the nniversal lunar days. And if we multiply the universal lunar days by I1, we get the produot 17,632,989,000,000.

38 ALBERUNIS INDIA.

The difference between the two numbers is 43,650,000. If we divide by this number the product of eleven timas the universal lnnar days, we get as quotient 403,963. Sritichaas of thia method. This is the number used by the inventor of the method. If there were not a emall remainder beyond the last-mentioned quotient (403,963 + a fraction), hia method would be perfectly correct. However, there remains a fraction of or ir, and this is the amount which is neglected. If he uses this divisor without the fraction, and divides by it the produet of eleven times the partial lunar days, the quotient would be by so much larger as the dividendum has increased. The other details of the calculation do not require comment. Method for Because the majority of the Hindus, in reekoning adhimdsa for their years, require the adhimasa, they give the pre- Anding tbe the yeara of a kalpa, ference to this method, and are particularly painstaking caturyuga, or kallywga. in describing the methods for the compatation of the adhimdsa, disregarding the methods for tha compu- tation of the dnardtra days and the sum of the days (ahargana). One of their methods of finding the ad- himasa for the years of a kalpa or caturyuga or kaliyuga is this :- They write down the years in three different places. They multiply the upper number by 10, the middle by 2481, and the lower by 7739. Then they divide the .middle and lower numbers by 9600, and the quotients are days for the middle number and avama for the lower number. The sum of these two qnotients is added to the number in the npper place. The sum represents the number of the complete adhimdsa days which have elapsed, and the sum of that which remains in the other two places is the fraction of the current adhimdsa. Dividing the days hy 30, they get months. Yakub Ibn Tarik states this method quite correctly. We shall, as an example, carry out this computation for ourgauge-year. The years of thekalpa which haveelapsed

CHAPTER LII. 39 till the moment of the gange-date are 1,972,948,132. The Intter We write down this number in three different places. plied to the The. upper number we multiply by ten, by which it ngoyear. gets a cipher more at the right side. The middle number we multiply by 2481 and get the product 4,894,884315.492. The lower number we multiply by Page a3a. 7739, and get the product 15,268,645,593,548. The latter two numbers we divide by 9600; thereby we get for the middle number as quotieut 509,883,782 and a remainder of 8292, and for the lower number a quo- tient of 1,590,483,915 and a remainder of 9548. The sum of these two remainders is 17,840. This fraction (i.e. 17.81°) is reckoned as one whole. Thereby the sum of the numbers in all three places is raised to 21,829,849,018, i.e. adhimdsa days, plus 388 day of the current adhimdsa day (i.e. which is now in course of summing up). Reducing these days to months, we get 727,661,633 months and a remainder of twenty-eight days, which is called Sh-D-D. This is the interval between the beginning of the month Caitra, which is not omitted in the series of months, and the moment of the vernal equinox. Further, adding the quotient which we have got for the middle number to the years of the kalpa, we get the sum of 2,482,831,914. Dividing this number by 7, we get the remainder 3. Therefore the aun has, in the year in question, entered Aries on a Tuesday. The two numbers which are used as multiplicators Explanatory for the nombers in the middle and lower places are to latter mp- note to the

. be explained in the following manner: thod.

Dividing the civil days of a kalpa by the solar cycles of a kalpa, we get as quotient the number of days which compose a year, i.e. 3651:318:680:888. Reducing this fraction by the common divisor of 450,000, we get 3651181. The fraction may be further reduced by being divided by 3, bnt people leave it as it is, in order

40 ALBERUNPS INDIA.

that this fraction and the other tractione whiob occur. in the furthar course of this computation should have the same denominator. Dividing the universal dnardtra days hy the solar years of a kalpa, the quotient is the nnmber of anardira days which belong to a solar year, viz. 51.320.000.000 daya. Reducing this fraction by the common divisor of 450,000, we get 51138 days. The fraction may fur- ther be rednced by being divided by 3. The measnres of solar and lunar years are abont 360 days, as are also the civil years of aun and moon, the ona being a little larger, the other a little ahorter. The one of these measnres, the lunar year, is used in this computation, whilst the other measure, the solar year, is aought for. The sum of the two quetients (of the middle and lower number) is the diffarence between the two kinds of years. The upper number is multiplied by the sum of tha complete days, and tha middle and lower numbers are multiplied by each of the two fractions. If we want to abbreviate the computation, and do tion of the not, like the Hindus, wisl to fiud the mean motiona of thod. sun and moon, wa add the two mnltiplicators of the middla and lower numbers together. This gives tha sum of 10,220. To this sum we add, for the upper place, tha prodnct of the divisor X 10 = 96,000, and we get 86o. Reducing this fraction by the half, wa get 4311. In this chapter (p. 27) we have already explained that by multiplying the daya by 5311, and dividing the product by 172,800, we get tha number of tha adhimdsas. If we now multiply the number of years instead of tha days, tha product is 3tr of the product which we should get when multiplying by the number of days. If wa, therefore, want to have the same qnotient which we get by the first diviaion, we must divide by Tage 223. ydo of the diviaor by which we divided in the first case, viz. 480 (for 360 X 480 = 172,800).

CHAPTER LII. 41

. Similar to this method is that one prescribed by A meoond Pulisa: "Write down the number of the partial months fnding the method for

in two different places. In the one place multiply aocordiug to adkimdsa,

it by 11t1, and divide the product by 67,500. Sub- tract the quotient from the number in the other place, and divide the remainder by 32. The quotient is the number of the adhimdsa months, aud the fraction in the quotient, if there is one, represents that part of an adhimdsa month which is in course of formation. Mul- tiplying this amount by 30, and dividing the product by 32, the quotient represents the days and day-frac- tions of the current adhimdsa month." The rationale of this method is the following :- If you divide the solar months of a caturyuga by the Explication adhimasa months of a caturyuga, in accordance with the thod of of the me-

theory of Pulisa, you get as quotient 3288:5g8. If you Puliaa

divide the months by this nnmber, yon get the com- plete adhimasa months of the past portion of the catur- yuga or kalpa. Pulisa, however, wanted to divide by . wholes alone, without any fractions. Therefore he had to subtract something from the dividendum, as has already been explained in a similar case (p. 36). We lave found, in applying the computation to our gauge- year, as the fraction of the divisor, 5.188.886, which may be rednced by being divided by 32. Thereby we get 1111 87.300. Pulisa has, in this calculation, reckoned by the solar days into which a date is resolved, instead of by months. For he says: "You write this number of daya in two Farther different places. In the one place yon multiply it by from Pulisa. 271 and divide the product by 4,050,000. The quo- tient yon aubtract from the number in the other place and divido the remainder by 976. The quo- tient is the number of adhimasa months, days, and day-fractions." Further he saya: "The reason of this is, that by dividing the days of a caturyuga by the adhimasa

42 ALBERUNPS INDIA.

months, you get as quotient 976 days and a remainder of 104,064 The common divisor for this nnmber and for the divisor is 384. Reducing the fraction thereby, we get r.uld.dos days." Here, however, I auspect either the copyist or the on the pee- cp trom translator, for Pulisa was too good a scholar to commit similar blunders. The matter is this: Those days which are divided by the adhimdsa months are of necessity solar days. The quotient con- tains wholes and fractions, as has been stated. Both denominator and numerator have as common divisor the namber 24 Reducing the fraction thereby, we get

If we apply this rule to the months, and reduce the nnmber of adhimdsa months to fractions, we get 47,800,000 as denominator. A divisor common to both this denominator and its numerator is 16. Redncing the fraction thereby, we get 1.300.000 If we now muitiply the number which Pulisa adopts as devisor by the just-mentioned common divisor, i.c. 384, we get the product 1,555,200,000, viz solar daya in a caturyuga. Bnt it is quite impossible that this nomber should, in this part of the calculation, be used as a divisor. If we want to base thia method on the rules of Brahmagupta, dividing the universal solar months by the adhimdsa months, the result will be, according to the method employed by him, double the amount of the adhimasa. Motbod for Further, a similar method may be used for the com- the compn- tation of the putation of the unaratra daya. daşa. Write down the partial lanar daya in two different places. In the one place, multiply the number by 50,663, and divide the product by 3,562,220. Sub- tract the quotient from the nomber in the other place, and divide the remainder by 63 without any fraction. In the further very lengthy speculations of the

CHAPTER LII. 43 Hindas there is no use st all, especially as they require the avama, i.c. the remainder of the partial unardtra, for the remainders which we get by the two divisions have two different denominators. He who is perfectly acquainted with the preceding Rale how to rules of resolution will also be able to carry out the chronologi- construet

opposite function, the composition, if a certain amount a certatn cal datefroin

of past days of a kalpa or caturyuga be given. given onm- To ber of dass. make aure, however, we sball now repeat the necessary of the akar- Theconverso

rules. gana.

If we want to find the years, the days being given, the latter must necessarily be civil daya, i.e. the differ- ence between the lunar days and the nardtra days. This difference (i.e. the civil days) stands in the same relation to their unaratra aa the difference between the universal lunar days and the universal unardtra days, viz. 1,577,916,450,000, to the universal unardtra days. The latter number (i.e. 1,577,916,450,000) is represented by 3,506,481. If we multiply the given daya by 55,739, and divide the product by 3,506,481, the quotient repre- sents the partial unaratra days. Adding hereto the civil days, we get the number of lunar days, viz. the sum of the partial solar and the partial adhimdsa days. These Innar days stand in the same relation to the adhimasa days which belong to them as the sum of the uni- versal solar and adhimdsa days, viz. 160,299,900,000, to the universal adhimdsa days, which number (i.e. 160,299,900,000) is represented by the number 178,111. If you, further, multiply the partial lunar days by 5311, and divide the product by 178,111, the quotient is the number of the partial adhimasa days. Subtract- ing them from the lunar daya, the remainder is the number of solar days. Thereupon you reduce the days to montha by dividing them by 30, and the months to years by dividing them by 12. This is what we want to find. E.g. the partial civil days which have elapsed up to

ALBERUNPS INDIA.

onr gauge-year are 720,635,951,963. This number is n given, and what we want to find is, how many Indian years and months are equal to this sum of days. First, we multiply the number by 55.739, and divide the product by 3,506,481. Thequotient is 11,455,224.575 Enardtra days. We add this number to the civil days. The aum is 732,091,176,538 lunar days. We multiply them by S311, and divide the product by 178,111. The quotient is the number of adhimdsa daya, viz. 21,829,849,018. We aubtract them from the lunar days and get - the remainder of 710,261,327,520, i.e. partial solar days. We divide these by 30 and get the quotient of 23,675,377,584, ie solar mouths, Dividing them by 12, we get Indian years, viz. 1,972,948,132, the same number of years of which our gange-date consiats, as we have already mentioned in a previous passage. Yakûb lbn Țariķ has a note to the same effect: Rule for - "Multiply the given civil daya by the universal lunar tha mint purporo days and divide the prodnct by the universal civil given by bItn days. Write down the quotient in two different places. In the one place multiply the number by the universal adhimdsa days and divide the product by the universal lunar days. The qnotient gives the adhimdsa months. Multiply them by 30 and subtract the product from the number in the other place. The remainder is the number of partial solar days. You further reduce them to montha aud years." The rationale of this calculation is the following :- Eplanstin We Lave already mentioned that the given number of of the lattor mothod. days are the difference between the lunar daya and their unardtra, as the universal civil days are the dif- ference between the universal lunar days and their universal tnardtra. These two measnres atand in a constant relation to each other.' Therefore we get the partial lunar days which are marked in two different places. Now, these are equal to the sum of the solar

CHAPTER LII. 45 and adhimdsa days, as the general lunar daye are equal to the sum of universal solar days and universal adhi- mdsa days. Therefore the partial and the universal adhimdsa days stand in the same relation to each other as the two numbers written in two different places, there being no difference, whether they both mean months or days. The following rule of Ya'kuh for the computation of Yskub's the partial unardtra days by meana of the partial adhi- the compu- method for

masa months is found in all the manuscripts of his partial ung- tation of the

book :- rátra days

"The past adhimdsa, together with the fractions of the current adhimasa, are multiplied by the universal una- > rdtra days, and the product is divided by the universal solar months. The quotient is added to the adhimasa. The sum is the number of the past unardtras." This rule does not, as I think, show that its author Critieiem knew the subject thoronghly, nor that he had mach thereon.

confidence either in analogy or experiment For the adhimdsa months which have passed of the caturyuga up to our gange-date are, according to the theory of Pulisa, 1,196,52511837. Multiplying this nnmber by the unardtra of the caturyuga, we get the prodnct 30,011,600,068,426 -. Dividing this number by the solar months, we get the quotient 578,927. Adding this to the adhimasa, we get the sum 1,775.452. And this is not what we wanted to find. On the contrary, the nnmber of dnaratra days ia 18,835,700. Nor is the product of the multiplication of this number by 30 that which we wanted to find. On the contrary, it is 53,263,560. Both numbers are far away irom the truth.

( 46 )

CHAPTER LIII,

ON THE AHARGANA, OR THR RESOLUTION OF YEARS INTO MONTHS, ACCORDING TO BPECIAL RULES WHICH ARE ADOPTED IN THE CALENDARS FOR CERTAIN DATES OR XOMENTS OF TIME.

Nothod of NoT all the eras which in the calendars are resolved ckarggha al applied to into days have epochs falling at such moments of time special dstor. when jast an adhimdsa or unardtra happens to be com- plete. Therefore the suthors of the calendars require for the calculation of adhimdsa and unardtra certain numbers which either must be added or subtracted if the calculation is to proceed in good order. We shall communicate to the reader whatever of these rules we happened to learn by the study of their calendars or astronomical handbooka. First, we mention the rule of the Khandakhadyaka, because this calendar is the best known of all, and pre- ferred by the astronomers to all others. Method of Brahmagupta says: " Take the year of the Sakakala, tho Khan- dakbad-" subtract therefrom 587, multiply the remainder by 12, yakı., and add to the product the complete months which hsve elapsed of the year in question. Multiply the sum by 30, and add to the product the days which have elapsed of the current month. The sum represents the partial solar days. " Write down this number in three different places. Add 5 both to the middle and lower numbers, and divide the lowest one by 14,945. Subtract the quotient

CHAPTER LIII. 47

from the middle number, and disregard the remainder which you have got by the division. Divide the middle number by 976. The quotient is the number of com- plete adhimdsa months, and the remainder is that which has elapsed of the enrrent adhimasa month, " Multiply these months by 30, and add the product to the upper number. The sum is the number of the partial lunar days. Let them stand in the npper place, and write the same mamber in the middle place. Mul- tiply it by II, and add thereto 497. Write this sum in the lower place. Then divide the sum by 111,573- Subtract the quotient from the middle number, and dis- regard the remainder (which you get by the division). Further, divide the middle number by 703, and the quotient represents the unarâtra days, the remainder the avamas. Subtract the unardtra days from the upper number. The remainder is the number of civil days." Page 226. This is the ahargana of the Khandakhadyaka. Divid- ing the number by 7, the remsinder indicates the week- day on which the date in question falls. We exemplify this rule in the case of our gauge-year. Application The corresponding year of the Sakakdla is 953. We thod to the of this me-

snbtract therefrom 587, and get the remainder 366. gauge-year.

We multiply it by the product of 12 x 30, since the date is without months and days. The product is 131,760, i.c. solar days. We write down this number in three different places. We add 5 to the middle and lower nnmbers, whereby we get 131,765 in both places. We divide the lower number by 14,945. The quotient is 8, which we sub- tract from the middle number, and here we get the remainder 131,757. Then we disregard the remainder in which the division has resulted. Further, we divide the middle nnmber by 976. The quotient 134 represents the number of months. There is besides a remainder of 878. Multiplying the months by 30, we get the product 4020, which we add to the

48 ALBERUNPS INDIA.

solar days. Thereby we get lunar days, viz. 135.780. We write down this number below the three numbers, multiply it by II, and add 497 to the prodnct. Thus we get the sum 1,494,077. We write this number below the four numbers, and divide it by 111,573. The qnotient is 13, and the remainder, i.e. 43,628, is dis- regarded. We enbtract the quotient from the middle number. Thus we get the remainder, 1,494,064. We divide it by 703. The quotient is 2125, and the re- mainder, ie. avama, is 789. We subtract the quotient from the lnnar days, and get the remainder 133,655- These are the civil days which we want to find. Divid- ing them by 7, we get 4 as remainder. Therefore the Ist of the month Caitra of the gauge-year falls on a Wednesday. The epoch of the era of Yazdajird precedes the epoch of this era (v. era nr. 5, p. 7) by 11,968 days. There- fore the sum of the days of the era of Yazdajird np to our gange-date is 145,623 days. Dividing them by the Persian year and months, we get as the corresponding Persian date the year of Yazdajird 399, the 18th Isfan- darmadh. Before the adhimdsa month becomes com- plete with 30 days, there must etill elapse five ghatt, ia. two hours. In consequence, the year is a leap year, and Caitra is the month which is reckoned twice in it. Mathod of The following is the method of the canon or calendar the Arablo book Ai-ar- Al-arkand, according to a bad translation : " If you Land. want to know the Arkand, ie. ahargana, take 90, mul- tiply it by 6, add to the product 8, and the years of the realm of Sindh, ie. the time till the month Safar, A.H. 117, which corresponds to the Caitra of the year 109. Subtract therefrom 587, and the remainder re- presents the years of the Shakh. An easier method is the following: "Take the com- plete years of the Aera Yazdagirdi, and anbtract there- from 33. The remainder represents the years of the Shakh. Or you may also begin with the original ninety

CHAPTER LIII. 49

years of the Arkand. Multiply them by 6, and add 14 to the product. Add to the sum the years of the Aera Yazdajirdi, and subtract therefrom 587. The remainder represente the years of the Shakh." I believe that the here-mentioned Shakh is identical Crittonl with Saka. However, the result of this calculation does iatter notes on the not lead us to the Saka era, but to the Gupta era, which method here is resolved into days. If the suthor of the Arkand began with 90, multiplied them by 6, added thereto 8, which would give 548, and did not change this number by an increase of years, the matter would come to the same result, and would be more easy and simple. The first of the month Safar, which the author of the latter method mentions, coincides with the eighth Daimab of the year 103 of Yazdajird. Therefore he makes the Page ar7. month Caitra depend npon the new moon of Daimah. However, the Persian months have since thst time been in advance of real time, because the day-quarters (after the 365 complete days) have no longer been inter- calated. According to the author, the era of the realm of Sindh which he mentions must precede the era of Yazdajird by eix years. Accordingly, the years of this era for our gauge-year would be 405. These together with the years of the Arkand, with which the author begins, viz. 548, represent the sum of 953 years as the year of the Sakakala. By the subtraction of that amount which the author has mentioned, it is changed into the corresponding year of the Gupta- kâla. The other details of this method of resolution or ahargana are identical with those of the method of the Khandakhadyaka, as we have described it. Sometimes you find in a maruseript such a reading as prescribes the division by 1000 instead of by 976, but this is simply a mistake of the manuscripts, as such a method is without any foundation. Next follows the method of Vijayanandin in his VOL IL D

ALBRRUNTS INDIA.

Motbod of canon called Karanatilaka: "Take the years of the the capon Śakakala, subtract therefrom 888, multiply the re- mainder by 12, and add to the product the complete months of the current year which have elapsed. Write down the sum in two different places. Multiply the one number by 900, add 661 to the product, and divide the sum by 29,282. The quotient represents adhimasa months. Add it to the number in the second place, multiply the sum by 30, and add to the product the days which have elapsed of the current month. The sum represents the lunar days. Write down this num- ber in two different places. Multiply the one number by 3300, add to the product 64,106, divide the sum by 210,902. The qnotient represents the unardtra days, and the remainder the avamas Subtract the dnaratra days from the lunar days. The remainder is the ahar- gana, being reckoned from midnight as the beginning." Appllcation We exemplify this method in the use of onr gauge- of this mothod to year. We subtract from the corresponding year of the the gaugo- Joar. Śakakâla (953) 888, and there remains 65. This num- ber of years is eqnal to 780 months. We write down this number in two different places. In the one place we multiply it by 900, add thereto 661, and divide the product by 29,282. The quotient gives 2318115 adhi- masa months. The multiplicator is 30. By being multiplied by it, the months are changed into days. The product, how- ever, is again multiplied by 30. The divisor is the pro- duct of the multiplication of 976 plus the following fraction by 30, the effect of which is that both numbers belong to the same kind (i.c. that both represent days). Further, we add the resulting number of months to those months which we have previously found. By multiplying the sum by 30, we get the product of 24,060 (read 24,090), i.e. lunar days. We write them down in two different places. The one number we multiply by 3300 and get the product

CHAPTER LIII .. 51

79,398,000 (read 79,497,000). Adding thereto 64,106 (read 69,601), we get the eum 79,462,104 (read 79,566,601). By dividing it by 210,902, we get the quotient 376 (read 307), i.c. anardtra days, and a re- mainder of 41895; (read Hrossr), i.e. the avamas. We snbtract the dnardtra daye from the lunar days, written in the second place, and the remainder is the civil ahargana, ic. the sum of the civil days, viz. 23,684 (read 23,713). The method of the Paica-Siddhantikd of Varahami- Mothod of hira is the following: "Take the years of the Sakakdla, siddatild. the Pafca-

subtract therefrom 427. Change the remainder into months by multiplying it by 12. Write down that number in two different places. Multiply the one number by 7 and divide the product by 228. The quotiont is the number of adhimdsa months. Add them to the number written down in the second place, multiply the eum by 30, and add to the product the days which have elapsed of the current month. Write down the sum in two different places. Multiply the lower number by 11, add to the product 514, and divide Page 228, the sum by 703. Subtract the quotient from the num- ber written iu the upper place. The remainder you get is the number of the civil days." This, Varahamihira says, is the method of the Sid- dhânta of the Greeks. We exemplify this method in one of our gauge-years. Applicatlon From the years of the Sakakala we subtract 427. The method to remainder, i.e. 526 years, is equal to 6312 months. yewr. the gauge- The corresponding number of adhimdsa months is 193 and a remainder of 15. The aum of these months together with the other months is 6505, which are equal to 195,150 lunar daye. The additions which occur in this method are required on account of the fractions of time which adhere to the epoch of the era in question. The multiplication by 7 is for the purpose of reducing the number to seventh parts, LI *

ALBERUNTS INDIA.

The divisor is the number of sevanths of the time of one adhimdsa, which he reckons as 32 months, 17 days, 8 ghatt, and about 34 cashaka. Further, we write down the lunar days in two diffe- rent places. The lower number we multiply by 11, and add to the product 514. The sum is 2,147,164- Dividing it by 703, we get the quetient 3054, i.e. the Anardtra days, and a remainder of 481. We subtract the days from the number in the second place, and get the remainder 192,096, i.c. the civil days of the date on which we base the chronological computations of this book. The theory of Varahamihira comes very near that of Brahmagupta; for here the fraction at the end of the number of the adhimdsa days of the gauge date is 35, whilst in the calculations which we have made, starting from the beginning of the kalpa, we found it to be 1f8, which is nearly equal to 1+ (cf. p. 29). Method of In a Muhammadan canen or calendar called the canon the Arable cabon. Al-harkan we find the same method of calculation, but All-Aartan, applied to and starting from another era, the epoch of which must fall 40,081 (days) after that of the era of Yazdajird. According to this book, the beginning of the Indian year falls on Sunday the 21st of Daimah of the year IIO of Yazdajird. The method may be tested in the following manner :- "Take seventy-two years, change them into months by multiplying them by 12, whioh gives the product 864 Add thereto the months which have elapsed between the Ist of Shaban of the year 197, and the Ist of the month in which yon happen to be. Write down the sum in two different places. Multiply the lower number by 7 and divide the produot by 228. Add the quotient to the upper number and multiply the sum by 30. Add to the product the number of days which have elapsed of the month in which you are. Write down this number in two different places.

CHAPTER LIII. 53 Add 38 to the lower number and multiply the sum by 11. Divide the product by 703, and subtract the quo- tient from the upper number. The remainder in the upper place is the number of the civil days, and the remainder in the lower place is the numher of the avamas. Add I to the number of days and divide the sum by 7. The remainder ehows the day of the week on which the date in question falls." This method would be correct if the months of the seventy-two years with which the calculation begins were lunar. However, they are solar months, in which nearly twenty-seven months must be intercalated, so that these seventy-two years are more than 864 months. We shall again exemplify this method in the case of Applicatlon our gange-date, i.c. the beginning of Rabi' I., A.H. 422. method to of the

Between the above-mentioned Ist of Sha'ban and the dnte. the gauge-

latter date there have elapsed 2695 months. Adding these to the number of months adopted by the anthor of the method (864), you get the sum of 3559 months. Write down this number in two places. Multiply the one by 7, and divide the product by 228. The quotient represents the adhimdsa months, viz. 109, Page 229. Add them to the number in the other place, and you get the sum 3668. Multiply it by 30, and you get the product 110,040. Write down this number in two different places. Add to the lower number 38, and yon get 110,078. Multiply it by I1 and divide the prodnct by 703. The quotient is 1722 and a remain- der of 292, i.e. the avamas. Suhtract the quotient from the upper number, and the remainder, 108,318, repre- sents the civil days. This method is to be amended in the following way : Emendation Yon must know that between the epoch of the era here method. nsed and the first of Sha'ban, here adopted as a date, there have elapsed 25,958 days, i.e. 876 Arabic months, or seventy-three years and two months. If we further

:54 ALBBRUNPS INDIA.

add to this number the months which have elapsed between that ist Sha'ban and the ist Rabi' L of the gange-year, we get the sum of 3571, and, together with the adhimasa months, 3680 months, ie. 110,400 days. The corresponding number of anardtra days is 1727, and a remainder of 319 avamas. Subtracting these days, we get the remainder 108,673. If we now sub- tract I and divide the remainder by 7, the computation is correct, for the remainder is 4, ic. the day of the gange-date is a Wednesday, as has above (p. 48) been stated. Method of The method of Durlabha, a native of Multan, is the Durlabha of MultAn. following :- He takes 848 years and adds thereto the Laukika-kala. The sum is the Sakakala. He subtracts therefrom 854, and changes the remainder of years into months. He writes them down together with the nast months of the current year in three different places. The lower number he multiplies by 77, and divides the product by 69,120. The quotient he snbtracts from the middle number, doubles the remainder, and adds thereto 29. The sum he divides by 65, so as to get adhimasa months. He adds them to the upper number and multiplies the sum by 30. He writes down the prodnct together with the past days of the current month in two different places. He multiplies the lower number by II and adds to the prodnct 686. The sum he writes underneath, He dividea it by 403,963, and adds the quotient to the middle number. He divides the sum by 703. The quotient represents the unaratra days. He aubtracts them from the upper number. The remainder is the civil ahargana, ie. the eum of the civil days of the date in question. We have already in a former place mentioned the .ontlines of this method. After the author, Durlabha, had adopted it for a particular date, he made some additions, whilst the bulk of it is unchanged. How- ever, the Karanasara forbide introducing any innovations

CHAPTER LIII. 55

which in the method of ahargana deviate to some other process. Unfortunately that which we possess of the book is badly translated. What we are sble to quote from it is the following :- He subtracts 821 from the years of the Sakakala, The remainder is the basis. This would be the year 132 for our gauge-year. He writes down this number in three different places. He multiplies the first num- ber by 132 degrees. The product gives the number 17.424 for our gauge-date. He multiplies the second number by 46 minutes, and gets the product 6072. He multiplies the third number by 34, and gets the product 4488. He divides it by 50, and the quotient represents minutes, seconds, &c., viz. 89' 46". Then he adds to the sum of degrees in the upper place J12, changing the seconds to minutes, the minutes to degrees, the degrees to circles. Thus he gets 48 circles 358' 41' 45". This is the mean place of the moon when the sun enters Aries. Further, he divides the degrees of the mean place of the moon by 12. The quotient represents days. The remainder of the division he multiplies by 60, and adds thereto the minutes of the mean place of the moon. He divides the sum by 12, and the quotient represents Page 230. ghatis and minor portions of time. Thus we get 27° 23' 29", ic. adhimasa days. No doubt this number represents the past portion of the adhimasa month, which is at present in the coorse of formation. The suthor, in regard to the manner in which the measure of the adhimdsa month is found, makes the following remark :-- He divides the lunar number which we have men- tioned, viz. 132° 46' 34", by 12. Therehy he gets as the portio anni 11° 3' 52" 50", and as the portio mensis 0 55 19 24" 1O". By means of the latter portio he computes the duration of the time in which 30 days sum up as 2 years, 8 months, 16 days, 4 ghatt, 45

ALBERUNTS INDIA. caskaka .:. Then he multiplies the basis by 29 and gets the product 3828. He adds thereto 20, and divides the sum by 36. The quotient represents the dnardtra days, viz. 106g. However, as I have not been able to find the proper explanation of this method, I simply give it as I find it, but I must remark that the amount of dnardira days which corresponds to a single adhimdsa month is

( 57 )

CHAPTER LIV.

ON THE COMPUTATION OF THE MEAN PLACES OF THE PLANETS.

Ir we know the number of cycles of the planets in & General kalpa or caturyuga, and further know how many cycles the deter- method for

have elapsed at a certain moment of time, we also the mean mination of know that the sum-total of the days of the kalpa or planet at caturyuga stands in the same relation to the sum-total time any given of the cycles as the past days of the kalpa or caturyuga to the corresponding amount of planetary cycles. The most generally nsed method is this :- The past days of the kalpa or caturyuga are multi- plied by the cycles of the planet, or of its apsis, or of its node which it describes in a kalpa or caturyuga. The product is divided by the sum-total of the days of the kalpa or caturyuga accordingly as you reckon by the one or the other. The quotient represents complete cycles. These, however, because not wanted, are dis- régarded. The remainder which you get by the division is mul- tiplied by 12, and the product is divided by the sum- total of the days of either kalpa or caturyuga by which we have already once divided. The qnotient repre- sents signs of the ecliptic. The remainder of this divi- sion is multiplied by 30, and the product divided by the same divisor. The quotient represents degrees. The remainder of this division is multiplied by 60, and is divided by the same divisor. The quotient represents minutes.

58 ALBERUNTS INDIA.

This kind of computation may be continued if we want to have seconds and minor values, The quotient represents the place of that planet according to its mean motion, or the place of that apsis or that node which we wanted to find. Nethod of The same is also mentioned by Pulisa, but his Palim for the mmo method differs, as follows :- " After having found porpoia the complete cycles which have elapsed at a cer- tain moment of time, he divides the remainder by 131,493,150. The quotient represents the mean signs of the ecliptic. "The remainder is divided by 4,383,105. The quo- tient represents degrees. The fourfold of the remainder is divided by 292,207. The quotient represents minntes. The remainder is multiplied by 60 and the product divided by the last-mentioned divisor. The quotient represents seconds. "This calculation may be continued, so as to give third parts, fourth parts, and minor values. The quo- tient thus found is the mean place of the planet which we want to find." Erplans- The fact is that Pulisa was obliged to multiply the tory Dotes therson. remainder of the cycles by 12, and to divide the pro- duct by the days of a caturyuga, because his whole computation is based on the caturyuga. But instead of doing this, he divided by the quotient which you get if you divide the number of days of a caturyuga by 12. This quotient is the first number he mentions, viz. 131,493,150. Further, he was obliged to multiply the remainder :

of the signs of the ecliptic by 30, and to divide the product by the first divisor; but instead of doing this, he divided by the quotient which you get if you divide the first number by 30. This quotient is the second number, viz. 4,383,105. According to the same analogy, he wanted to divide the remainder of the degrees by the quotient which

CHAPTER LIV. 59 you get if yon divide the second number by 60. How- ever, making this division, he got as quotient 73,051 and & remainder of . Therefore he moltiplied the whole by 4, in order that the fractions should be raised to wholes. For the same reason he also multiplies the following remainder by 4; bnt when he did not get wholes, as has been indicated, he returned to mul- tiplying by 60. Page 231. If we apply this method to a kalpa according to the theory of Brahmagupta, the first number, by which tbe remainder of the cyeles is divided, is 131,493,037,500. The second number, by which the remainder of the signs of the ecliptic is divided, is 4,383,101,250. The third number, by which the remainder of the degrees is divided, is 73,051,687. In the remainder which we get by this division there is the fraction of }. There- fore we take the double of the number, viz. 146,103.375. and we divide by it the double of the remainder of minutes. Brahmagupta, however, does not reckon by the kalpa nrahma- and caturyuga, on account of the enormous sums of plles thie their days, but prefers to them the kaliyuga, in order tho kalivuga method to

to facilitate the calculation. Applying the preceding get atnaller in order to

method of ahargana to the precise date of the kaliyuga, bum bers.

we multiply its sum of days by the star-cycles of a kalpa. To the prodnct we add the basis, i.e. the remain- ing cycles which the planet had at the beginning of the kaliyuga. We divide the sum by the civil days of the kaliyuga, viz. 157,791,645. The quotient repre- sents the complete cycles of the planet, which are dis- regarded. The remainder we compute in the above-described manner, and thereby we find the mean place of the planet. The here-mentioned bases are the following for the -single planets :-

60 ALBERUNI'S INDIA.

For Mars, 4,308,768,000. For Mercury, 4,288,896,000. For Jopiter, 4,313,520,000. For Venus, 4,304.448,000. For Satorn, 4,305,312,000. For the Sun's apsis, 933,120,000. For the Moon's apsis, 1,505,952,000. For the ascending node, 1,838,592,000 (v. the notes). At the same moment, ie. at the beginning of the kali- yuga, sun and moon stood according to their mean motion in o° of Aries, and there was neither a plus nor a minus consisting of an adhimdsa month or of ana- rdtra days. Methods of In the above-mentioned canones or calendars we find the Khapda- khádyaka, the following method :- "The ahargana, i.e. the sum of : Karanali -. Laka, and the days of the date, is, for each planet respectively, Karapaidra multiplied by a certain number, and the prodnct is divided by another number. The quotient represents complete cycles and fractions of cycles, according to mean motion. Sometimes the computation becomes perfect simply by this multiplication and division. Sometimes, in order to get a perfect result, you are compelled once more to divide by a certain nnmber the days of the date, either such as they are, or multi- plied by some number. The quotient must then be combined with the result obtained in the first place. Sometimes, too, certein numbers are adopted, as e.g. the basis, which must either be added or subtracted for this purpose, in order that the mean motion at the beginning of the era shonld be compnted as beginning with o° of Aries. This is the method of the books Khandakhadyaka and Karanatilaka. However, the anthor of the Karanasdra compntes the mean places of the planets for the vernal equinox, and reckons the ahargana from this moment. But these methods are very subtle, and are so numerous, that none of them has

CHAPTER LIV. 6r

obtained any particular anthority. Therefore we refrain from reprodueing them, as this would detain us too long and be of no use. The other methods of the computation of the mean places of the planets and similar calculations have nothing to do with the subject of the present book.

62

CHAPTER LV.

ON THE ORDER OF THE PLANETS, THEIR DISTANCES AND SIZES.

Traditlonl WHEN speaking of the lokas, we have already given a viow.on tho. san being below the quotation from the Vishnu-Purdna and from the com- mentary of Patanjali, according to which the place of the sun is in the order of the planets below that of the moon. This is the traditional view of the Hindus. Compare in particular the following passage of the Matsya-Purdna :- "The distance of heaven from the earth is equal to the radius of the earth. The snn is the lowest of all . planets. Above him there is the moon, and above the moon are the lunar stations and their stars. Above them is Mercury, then follow Venus, Mars, Jupiter, Saturn, the Great Bear, and above it the pole. The pole is connected with the heaven. The stars can- not be counted by man. Those who impugn this view mnaintain that the moon at conjunction becomes hidden by the eun, as the light of the lamp becomes invisible in the light of the sun, and she becomes more visible the more she moves away from the aun." We shall now give some qnotations from the booka of this school relating to the eun, the moon, and the stars, and we shall combine herewith the views of the astronomers, althongh of the latter we have only a very Popabar slender knowledge. aatrocomy. The Vdyu-Purdna says: "The sun has globnlar Pago sjt. shape, fiery nature, and 1coo rays, by which he attracts ..

CHAPTER LV. 63

the water; 400 of these are for the rain, 300 for the Quotstions snow, and 300 for the air." from Vdyu- Purdna. In another passage it says : " Some of them (ie. the rays) are for this purpose, that the devas should live in bliss; others for the purpose that men should live in comfort, whilst others are destined for the fathers." In another passage the anthor of the Vayu-Purana divides the rays of the sun over the six seasons of the year, saying: "The sun illuminates the earth in that third of the year which commences with o° of Pisces by 300 rays; he causes rain in the following third by 400 rays, and he causes cold and snow in the remain- ing third by 300 rays." Another passage of the same book runs as follows : "The rays of the snn and the wind raise the water from the sea to the sun. Now, if the water dropped down from the sun, it would be hot. Therefore the sun handa the water over to the moon, that it should drop down from the moon cold, and thus refresh the world." Another passage: "The heat of the sun.and his light are one-fourth of the heat and the light of the fire. In the north, the sun falls into the water during the night; therefore he becomes red." Another passage: "In the beginning there were the earth, water, wind, and heaven. Then Brahman per- ceived sparks under the earth. He brought them forth and divided them into three parts. One third of them is the common fire, which requires wood and is extin- guished by water. Another third is the sun, and the last third is the lightning. In the animals, too, there is fire, which cannot be extinguished by water. The sun attracts the water, the lightning shines through the rain, but the fire in the animals is distributed over the moist snbstances by which they nourish themselves." The Hindus seem to believe that the heavenly bodies nourish themselves by the vapours, which also Aris- totle mentions as the theory of certain people. Thus

64 ALBERUNPS INDIA.

the author of the Vishnu-Dharma erplains that "the sun nourishes the moon and the stars. If the sun did not exist, there would not be a star, nor angel, nor man." On the The Hindus believe regarding the bodies of all the nature of the stara. stars that they have & globular shape, a watery essence, and that they do not shine, whilst the sun alone is of fiery essence, self-shining, and per accidens illnminates other stars when they stand opposite to him. They reckon, ac- cording to eyesight, among the stars also such luminous bodies as in reality are not stars, hnt the lights into which those men have been metamorphosed who have received eternal reward from God, and reside in the Quotatio height of heaven on thrones of crystal The Vishnu- from tho Dharma says: "The stars are watery, and the rays of Dharma. the sun illuminate them in the night. Those who by their pions deeds have obtained a place in the height sit there on their thrones, and, when shining, they are reckoned among the stars." All the stars are called tdra, which word is derived from tarana, i.c. the passage. The idea is that those saints have passed through the wicked world and have reached bliss, and that the stars pass through heaven in a circular motion. The word nakshatra is limited to the stars of the lunar stations. As, however, all of these are called fized stars, the word nakshatra also applies to all the fixed stars; for it means not increas- ing and not decreasing. I for my part am inclined to think that this increasing and decreasing refers to their number and to the distances of the one from the other, but the anthor of the last-mentioned book (Vishnu- Dharma) combines it with their light. For he adds, "as the moon iocreases and decreases." Further, there is a passage in the same book where Markandeya says: "The stars which do not perish be- fore the end of the kalpa are equal to a nikharra, i.e. 100,000,000,000. The number of those which fall down before the end of a kalpa is unknown. Only he can know it who dwells in the height duriog a kalpa."

CHAPTER LV. 65 Vajra spoke : "O Markandeya, thou hast lived during six kalpas. This is thy seventh kalpa. Therefore why dost thou not know them ?" He answered: " If they always remained in the same condition, not changing as long as they exist, I should uot be ignorant of them. However, they perpetually raise some pious man and bring another down to the earth. Therefore I do not keep them in my memory." Regarding the diameters of eun and moon and their On the dla- shadows the Matsya-Purana says: "The diameter of the planete. the body of the sun is gooo yojanas; the dismeter of the moon is the donble of it, and the apsis is as much as the two together," The same occurs in the Vdyu-Purdna, except that it says with regard to the apsis that it is equal to the sun when it is with the sun, and that it is equal to the moon when it is with the moon. Another anthor says: " The apsis is 50,000 yojanas." Regarding the diameters of the planets the Matsya- Purana says: "The circumference of Venus is one- sixteenth of the circumference of the moon, that of Page 933 Jupiter three-fourtha of the circumference of Venns, that of Saturn or Mars three-fourths of that of Jupiter, that of Mercury three-fourths of that of Mars." The same statement is also found in the Vayu-Purâna. The same two books fix the cirenmference of the On the cir great fixed stars as equal to that of Mercury. The next of the fxed smaller class have a circumference of 500 yojanas, the stars. following classes 400, 300, and 200. But there are no fixed stars with a smaller circumference than 150 yojanas. Thus the Vayu-Purana. But the Matsya-Purana says: "The next following classes have a circumference of 400, 300, 200, and 100 yojanas. But there is no fixed star with less circumference than a half yojana." The latter statement, however, looks suspicious to me, and is perbaps a fault in the manuscript. The anthor of Vishnu-Dharma says, relating the VOL. II, E

66 ALBERUNI'S INDIA.

words of Markandeya: "Abhijit, the Falling Eagle; Årdrd, the Sirius Yemenicus; Rohint, or Aldabaran; Punarvasn, ic. the Two Heads of the Twins; Pushya, Revatt, Agastya or Canopus, the Grest Bear, the master of Vayu, the master of Ahirbudhnya, and the master of Vasishtha, each of these stars has a circumfer- ence of five yojanas. All the other stars have each only a circumference of four yojanas, I do not know those stars, the distance of which is not measurable. They have & circumference between four yojanas and two kuroh, i.c. two miles. Those which have less cir- cumferenca than two kuroh are not seen by men, but only by the devas" The Hindus have the following theory regarding the magnitude of the stars, which is not traced back to any known authority : " The diameters of the sun and moon are each 67 yojanas; that of the spsis is 100; thst of Venus ro, of Jupiter 9, of Ssturn 8, of Mars 7, of Mer- cTy 7." Views of This is all we have been able to learn of the confused utmnomers notions of the Hindus regarding these subjects. We the Hindu on the same sabjocts shall now pass on to the views of the Hindu astro- nomers with whom we agree regarding the order of the planets and other topics, viz. that the sun is the middle of the planets, Saturn and the moon their two ends, and that the fixed stars are above the planets. Some of these things have already been mentioned in the preceding chapters. Quotation Varahamihira says in the book Samhitd: "The moon from tho Saikitd of is alwsys below the sun, who throws his rays npon her, VarAhart- htra, chap. and lits up the one half of her body, whilst the other iv. 1-3- half remains dark and shadowy like a pot which yon place in the sunshine. The one half which faces the sun is lit up, whilst the other half which does not face it remains dark. The moon is watery in her essence, therefore the rays which fall on her are reflected, as . they are reflected from the water and the mirror towards

CHAPTER LV. 67

the wall. If the moon is in conjunction with the sun, the white part of her turns towards the sun, the blaok part towards us. Than the white part sinks downward towards us slowly, as the sun marches away from the moon" Every edncated man among the Hindu theologians, and much more so among their astronomers, believes indeed that the moon is below the sun, and even below all the plenets. The only Hindu traditions we have regarding the Taknb Ihn distances of the stars are those mentioned by Ya'kub distances of Tarik on the

Ibn Tarik in his hook, The Composition of the Spheres, the stars.

and he had drawn his information from the well-known Hindn scholar who, A.H. 16t, accompanied an embassy to Bagdad. First, he gives a metrological statement: " A finger is eqnal to six barleycorne which are put one by the side of the other, An arm (yard) is eqnal to twenty-four fingers. A farsakh is equal to 16,000 yards." Here, however, we must observe that the Hindus do not know the farsakh, that it is, as we have already explained, equal to one half a yojana. Further, Ya'kub says: " The diameter of the earth is 2100 farsakh, its circumference 6596.', farsakh." On this basis he has computed the distances of the planets as we exhibit them in the following table. However, this statement regarding the size of the Puliss and earth is by no means generally agreed to by all the ta on the Brah wagnp-

Hindus. So, e.g. Pulisa reckons its diameter es 1600 ject. ame sub-

yojanas, and its circumference as 502611 yojanas, whilst Brahmagupta reckons the former as 1581 yojanas, and the latter as 5ooo yojanas If we double thesenumbers, they onght to be equal to the numbers of Ya'kub; bnt this is not the case. Now the yard and the mile are respectively identical accord- ing to the measurement both of us and of the Hindus. According to onr computation the radius of the earth is 3184 miles. Reckoning, according to the custom of our

68 ALBERUNPS INDIA.

country, I' farsakh = 3 miles, we get 6728 farsakh; and reckoning : farsakh = 16,000 yards, as is mentioned by Page 234. Ya'kub, we get 5046 farsakh. Reckoning 1 yojand = 32,000 yards, we get 2523 yojanas Distances of The following table is borrowed from the book of the planeta fròm the Ya kub Ibn Tarik :- contro of tho carth, and . tbetr dia- meters, a0- cording to iThe cooventional Ya kab Ibn Ineasures of tho. Tarik. Their distances from the contre of the carth, and their distances, difering Their constant measures, based dinmetors. aocording to timo and place, reckoned on the radiu of in ftratk, z farmth the carth =1. =16,000 yarda. 'The planeta,

Radins of the earth The smallest distance 1,050

The middle distance 37,500 35₺

The greatest distance 48,500

Digmeter of the moon 59,000 5,000 Moon.

The smallest diatance The middle distance 64,000 6o1f

The greatest distance . 164,000 1563

Diameter of Mercury 264.000 2514 5,000 The smallest distance The middle distance 269,000 2565

The greatest distance 709,500 675+

Diameter of Venus 1,150,000 J,0953T Venus. Mercury. The smallest distance 20,000 19

The middle distance 1,170,000 J,LI4F

Pago 235- The greatest distanco 1,690,000 1,60g}

Diameter of the Sun . 2,210,000 2,1041 Sun.

The smallest distance 20,000 191

The middle distance 2,230,000 21234F

The greatest distance 5.315,000 Diameter of Mars 8,400,000 5,0611 8,000 20,000 The smallest distance The middle distance 8,420,000 8,0191

The greatest distance 11,410,000 10,8661

Diameter of Jupiter 14,400,000 13,7147 20,000 The smallest distance The middle distance 14,420,000

The greatest distance 16,220,000 13.7331 I5.4471

Diameter of Saturn 18,020,000 20,000 The radins of the outside The radins of the inside 20,000,000 19,962,000 19,04771 Its circumference from 1,8664 (sic)

the outsido 125,664,000 Zodiacna. Saturn. Jupiter. Mars.

CHAPTER LV. 69

This theory differs from that on which Ptolemy has Ptolemy on based his computation of the distances of the planets tances of the in the Kitab-almanshurdt, and in which he has been Pago 236. planets.

followed hoth by the ancient and the modern astrono- mers. It is their principle that the greatest distance of a planet is equal to its smallest distance from the next higher planet, and that between. the two globes there is not a space void of action According to this theory, there is between the two globes a space not occnpied by either of them, in which there is something like an axia around which the rota- tion takes place. It seems that they attributed to the æther a certain gravity, in consequence of which they felt the necessity of adopting something which keeps or holds the inner globe (the planet) in the midst of the outer globe (the æther). It is well known among all astronomers that there on occalta- is no possibility of distinguishing between the higher parallax. tion and the

and the lower one of two planets except by means of the occultation or the increase of the parallax. However, the occultation occurs only very seldom, and only the parallax of a single planet, viz. the moon, can be ob- served. Now the Hindus helieve that the motions are equal, but the distances different. The reason why the . higher planet moves more slowly than the lower is the greater extension of its sphere (or orbit); and the reason why the lower planet moves more rapidly is that its sphere or orbit is less extended. Thus, e.g. one minute in the sphere of Saturn is equal to 262 minutes in the sphere of the moon. Therefore the times in which Saturn and the moon traverse the same space are dif- ferent, whilst their motions are equal. I have never found a Hindn treatise on this subject, bnt only numbers relating thereto acattered in various books-numbers which are corrupt. Somebody objected to Pulisa that he reckoned the circumference of the sphere of each planet as 21,600, and its radius as 3438,

70 ALBERUNI'S INDIA.

whilst Varahamihira reckoned the sun's distance from the earth as 2,598,900, and the distance of the fixed stars as 321,362,683 Therenpon Pulisa replied that the for- mer numbers were minutes, the latter yojanas; whilst in another passage he says that the distance of the fixed stars from the earth is sixty times larger than the distance of the sun. Accordingly he ought to have reckoned the distance of the fixed stars as 155.934,000. Hindn The Hindu method of the computation of the dis- th coapo- tances of the planets which we have above mentioned rortbod for

dinemoms of is based on a principle which is unknown to me in the taton of the tho plaort present etage of my knowledge, and as long as I have no facility in translating the books of the Hindus, The principle is this, that the extension of a minute in the orbit of the moon is equal to fifteen yojanas. The nature of this principle is not cleared up by the commentaries Qeotationa of Balabbadra, whatsoever trouble he takes. For he trom BaLc says: "People have tried to fix by observation the time of the moon's passing through the horizon, ie. the time between the shining of the first part of her body and the rising of the whole, or the time between the beginning of her setting and the completion of the act of setting. People have found this process to last thirty-two minntes of the circumference of the sphere" However, if it is dificult to fir by obser- vation the degrees, it is much more so to fix the minutes, Further, the Hindus have tried to determine by observation the yojanas of the diameter of the moon, and have found them to be 480. If you divide them by the minutes of her body, the quotient is 15 yojanas, as corresponding to one minute. If you multiply it by the minutes of the circumference, you get the product 324,000. This is the measure of the sphere of the moon which she traverses in each rotation. If you multiply this number by the cycles of the moon in a kalpa or caturyuga, the product is the distance which

CHAPTER LV.

the moon traverses in either of them. According to Brahmagupta, thia is in a kalpa 18,712,069,200,000,000 yojanas. Brahmagupta calls this number the yojanas of the ecliptic. Evidently if you divide this number by the cycles of each planet in a kalpa, the quotient represents the yojanas of one rotation, However, the motion of the planets is, according to the Hindus, as we have already mentioned, in every distance one and the same. Therefore the quotient represents the measure of the path of the ephere of the planet in question. As further, according to Brahmagupta, the relation of the diameter to the circumference is nearly equal to The radil of that of 12,959: 40,980, yon multiply the measure of or thedr dis- the planete,

the path of the sphere of the planet by 12,959, and the contre tances from

divide the product by 81,960. The quotient is the computed of the earth,

radius, or the distance of the planet from the centre of Brahma- e scoording to

the earth. gupta.

We have made this computation for all the planets according to the theory of Brahmagupta, and present the results to the reader in the following table :-

The efreumferenc of tha Their radil, which Pago 237.

The planets. sphere of each planet, are identical with!

reckoned in yojanas. their distances from the earth'e centre, reckoned in yojanat.

Moon Mercnry . . 324,000 51,229 . Vonus . . I,043,21011 164,947

Sun . . 2,664,6291/11 421,315 4,331,4973 684,869 Mars . . . 8, 146,916 1111 Jupiter 1,288,1 . . Satumn 51,374,8218111 8,123,064

The Fired Stars, . 127,668, 7873934412 20,186,186

their distance from the earth's. centre being sixty times the 259,889,850 41,092, 140 distance of the aun from the

72 ALBERUNI'S INDIA.

The cimo As Pulisa reckons by caturyugas, not by kalpas, he

the thoory woordiet multiplies the distance of the path of the sphere of the moon by the lunar cycles of a caturyuga, and gets the product 18,712,080,864,000 yojanas, which he calls the yojanas of heaven. It is the distance which the moon traverses in each caturyuga. Pulisa reckons the relation of the diameter to the circumference as 1250: 3927. Now, if you multiply the circumference of each planetary sphere by 625 and divide the prodnct by 3927, the quotient is the distance of the planet from the earth's centre. We have made the same compntation as the last one according to the view of Pulisa, and present the results in the follow- ing table. In computing the radii we have disre- garded the fractions smaller than }, and have reduced larger fractions to wholes. We have, however, not taken the same liberty in the calculation of the circum- ferences, but have calculated with the utmost accuracy, because they are required in the compntations of the. revolutions. For if you divide the yojanas of heaven in . Page sgt & kalpa or caturyuga by the civil days of the one or the other, you get the quotient 11,858 plus & remainder, which is $:f Tts according to Brahmagupta, and :09:154 according to Pulisa. This is the distance which the moon every day traverses, and as the motion of all planets is the same, it is the distance which every planet in a day traverses. It stands in the same relation to the yojanas of the circumference of its sphere as its motion, which we want to find, to the circumference, the latter being divided into 360 equal parts. If you therefore multiply the path common to all the planets by 360 and divide the product by the yojanas of the circumference of the. planet in question, the quotient represents its mean daily motion.

.CHAPTER LV. 73

The oireumferences of The distances of the The plabeta, the spheres of the: planets, reckoued in planets from the carth'e contre, yojanas. rockoned in yojanas.

Moon Mercury . 324,000 51,566

Venus 1,043,211818 166,033

Son 2,664,632-t111 424,089 . . . 4,331,500} 690,295 (sic) Mara . . 8,146,93738111 Jupiter I,296,624 (!)

Satorn 51,375,764-PRC 127,671,7391191 8,176,689 (!) . The Fixed Stars, the 20,319,542 (!)

sun's distance from the earth's centre 259,890,012 41,417,700 (sic) being ith of theirs.

As, now, the minutes of the diameter of the moon The'dis- stand in the same relation to the minutes of her cir- tha planets. meters of

cumference, i.c. 21,600, as the number of yojanas of the Page $39.

diameter, i.e. 480, to the yojanas of the circumference of the whole sphere, exactly the same method of calcnlation has been epplied to the minutes of the diameter of the sun, which we have found to be equal to 6522 yojanas according to Brahmagupta, and equal to 6480 according to Pulisa. Since Pulisa reckons the minutes of the body of the moon as 32, i.e. a power of 2, he divides this number in order to get the minutes of the bodies of the planets by 2, till he at last gets I. Thus he attributes to the body of Venus } of 32 minutes, i.e. 16; to that of Jupiter } of 32 minntes, i.e. 8; to that of Mercury } of 32 minutes, i.e. 4; to that of Saturn i's of 32 minutes, i.c. 2; to that of Mars 3r of of 32 minutes, i.e. I. This precise order seems to have taken his fancy, or he would not have overlooked the fact that the diameter of Venus ia, according to observation, not equal to the radius of the moon, nor Mars equal to rsth of Venus. Method for The following is the method of the compntation of tation of the the compu-

the bodies of sun and moon at every time, based on sun and bodies of

their distances from the earth, ic the true diameter given time. mooD at any

74 ALBERUNTS INDIA.

of ita orbit, which is found in the computations of the corrections of sun and moon. AB is the diameter of the body of the sun, CD is the diameter of the earth, CDH is the cone of the shadow, HL is its elevation. Further, draw CR parallel to DB. Then is AR the difference between AB and CD, and the normal line CT is the middle distance of the sun, ie. the radius of its orbit derived from the yojanas of heaven (v. p. 72) From this the true distance of the sun always differs, sometimes being larger, sometimes emaller. We draw CK, which is of course determined by the parts of the sine. It stands in the same relation to CT, this being the sinus totus (=radins), as the yojanas of CK to the yojanas of CT. Hereby the measure of the diameter is reduced to yojanas. The yojanas of AB stand in the same relation to the yojanas of TC as the minutes of AB to the minntes of TC, the latter being the sinus totus. Thereby AB becomes known and determined by the minntes of the sphere, because the sinus totus is determined by the Quotations measure of the circumference. For this reason Pulisa from Pulisa, Brahns- says: "Multiply the yojanas of the radius of the sphere gupta, and Bnhtbdn of the sun or the moon by the true distance, and divide the product by the sinus totus. By the quotient you get for the sun, divide 22,278,240, and by the qnotient you get for the moon, divide 1,650,240. The quotient then represents the minutes of the diameter of the body of either sun or moon" The last-mentioned two numbers are products of the multiplication of the yojanas of the diameters of sun and moon by 3438, which is the number of the minntes of the sinus totus. Likewise Brahmagupta says: "Multiply the yojanas of sun or moon by 3416, ie. the minntes of the sinus tofus, and divide the product by the yojanas of the radius of the sphere of sun or moon." But the Iatter rule of division is not correct, because, according to it,

CHAPTER LV. 75

the 'measure of the body would not vary (v. p. 74)- Therefore the commentator Balabhadra holds the same opinion as Palisa, viz. that the divisor in this division ehould be the true distance reduced (to the measure of yojanas). Brahmagupta gives the following rule for the com- Brahma- putation of the diameter of the shadow, which in our motlod for gupta's

canones is called the measure of the sphere of the dragon's tation of the the compn-

head and tail: "Subtract the yojanas of the diameter the shadow. diamater of

of the earth, i.e. 1581, from the yojanas of the diameter of the sun, i.e. 6522. There remains 4941, which is kept in memory to be used es divisor. It is represented in the figure by AR. Further multiply the diameter 'of the earth, which is the double sinus totus, by the yojanas of the true distance of the sun, which is found by the correction of the sun. Divide the product by the divisor kept in memory. The quotient is the true distance of the shadow's end. " Evidently the two triangles ARC and CDH are similar to each other. However, the normal line CT does not vary in size, whilst in consequence of the true distance the appearance of AB varies, though its size is constantly the same. Now let this distance be CK. Draw the lines AJ and RV parallel to each other, and JKV parallel to AB. Then the latter is equal to the divisor kept in memory. " Draw the line JCM. Then M is the head of the cone of the shadow for that time. The relation of JV, the divisor kept in memory, to KC, the true distance, is the same as that of CD, the diameter of the earth, to ML which he (Brahmagupta) calls a true distance (of Page 240. the shadow's end), and it is determined by the minutes of the sine (the earth's radius being the sinus totus). For KC -- " Now, however, I suspect that in the following some- Lacuna In thing has fallen ont in the manuscript, for the author soript copy the manu-

continues: " Then multiply it (i.e. the quotient of CK, gupta of Brahma-

26 ALBERUNES INDIA.

by the divisor kept in memory) by the diameter of the earth. The product is the distance between the earth's centre and the end of the shadow. Subtract there- from the true distance of the moon and multiply the remainder by the diameter of the earth. Divide the product by the true distance of the shadow's end. The quotient is the diameter of the shadow in the sphere of the moon. Further, we suppose the true distance of the moon to be LS, and FN is a part of the Iunar sphere, the radius of which is LS. Since we have found LM as determined by the minutes of the sine, it stands in the same relation to CD, this being the double sinus totus, as MS, measured in minutes of the sine, to XZ, measured in minutes of the sine." Here I suppose Brahmagupta wished to reduce LM, the true distance of the shadow's end, to yojanas, which is done by multiplying it by the yojanas of the diameter of the earth, and by dividing the product by the double sinus totus. The mentioning of this division has fallen out in the manuscript; for without it the multiplication of the corrected distance of the shadow's end by the diameter of the earth is perfectly superfluous, and in no way required by the computation. Further: " If the number of yojanas of LM is known, LS, which is the true distance, must also be reduced to yojanas, for the purpose that MS should be determined by the same measure. The measure of the diameter of the shadow which is thus found represents yojanas. Further, Brahmagupta says: "Then multiply the shadow which has been found by the sinus totus, and divide the product by the true distance of the moon. The quotient represents the minutes of the shadow which we wanted to find." Criticismns However, if the shadow which he has found were on Brabmat- gupta's determined by yojanas, he ought to have multiplied it by the donble sinus totus, and to bave divided the pro- duct by the yojanas of the diameter of the earth, in

CHAPTER LV .:. 77 order to find the minutes of the shadow. Bnt as he has not done so, this shows that, in his computation, he limited himself to determining the true diameter in minutes, without reducing it to yojanas. The anthor uses the true (sphuta) diameter without its having been rednced to yojanas. Thus he finds that the shadow in the circle, the radius of which is LS, is the true diameter, and this is required for the compu- tation of the cirele, the radius of which is the sinus totus. The relation of ZX, which he has already found, to SL, the true distance, is the same as the relation of ZX in the measure which is sought to SL, this being the sinus totus. On the basis of this equation the rednction (to yojanas) must be made. In another passage Brahmagnpta says: "The dia- Another meter of the earth is 1581, the diameter of the moon Brahma method of

480, the diameter of the sun 6522, the diameter of the computing shadow 1581. Snbtract the yojanas of the earth from thenhadow. the yojanas of the snn, there remains 4941. Multiply this remainder by the yojanas of the true distance of the moon, and divide the product by the yojanas of the true distance of the sun. Subtract the quotient you get from 1581, and the remainder is the measure of the shadow in the sphere of the moon. Multiply it by 3416, and divide the product by the yojanas of the middle radius of the sphere of the moon. The quotient represents the minutes of the diameter of the shadow. "Evidently if the yojanas of the diameter of the earth are subtracted from the yojanas of the diameter of the sun, the remainder is AR, i.e. JV. Draw the line VCF and let fall the normal line KC on O. Then the relation of the surplus JV to KC, the true distance of the sun, is the same as the relation of ZF to OC, the true distance of the moon. It is indifferent whether these two mean diameters are rednced (to yojanas) or not, for ZF is, in this case, found as determined by the measure of yojana. "Draw XN as equal to OF. Then ON is necessarily

78 ALBERUNTS INDIA.

equal to the diameter of CD, and ita sought-for part is ZX. The number which is thus found must be sub- tracted from the diameter of the earth, and the remainder will be Z" The sathor For such mistakes as occur in this compntation, the the corrupt author, Brahmagupta, is not to be held responsible, bnt stato of his panuirlpt we rather suspect that the fault lies with the mann- of Brahmna- gupta. script. We, however, cannot go beyond the text we Pago 341. have at our disposal, as we do not know bow it may be in a correct copy.

J

K T C

N

L R H M

D N

B

The measure of the shadow adopted by Brahma- gupta, from which he orders the reader to subtract, cannot be a mean one, for a mean measure stands in the midst, between too little and too mnch. Further, we cannot imagine that this measure should be the greatest of the measures of the shadow, including the plus (?); for ZF, which is the minus, is the base of a triangle, of which the one side, FC, cuts SL in the direction of the sun, not in the direction of the end of the shadow. Therefore ZF has nothing whatsoever to do with the shadow (conjectural rendering.)

CHAPTER LV. 79 Lastly, there is the possibility that the minus belongs to the diameter of the moon. In that case the relation of ZX, which has beon determined in yojanas, to SL, the yojanas of the true distance of the moon, is the same as the relation of ZX reckoned in minntes to SL, this being the sinus totus (conjectural rendering.) By this method is found what Brahmagupta wants to find, quite correctly, withont the division by the mean radius of the sphere of the moon, which is derived from the yojanas of the sphere of heaven (v. p. 72). (For the last three passages vide Notes.) The methods of the computation of the diameters of The compn- sun and moon, as given by the Hindn canones, such a5 diameters tation of tha

tne Khandakhadyaka and Karanasra, are the same as monn ac. of sun and

are found in the canon of Alkhwarizmi. Also the com- to other cording . putation of the diameter of the shadow in the Khanda- khadyaka is similar to that one given by Alkhwarizmi, whilst the Karanasara has the following method :-- "Multiply the bhukti of the moon by 4 and the bhukti of the sun by 13. Divide the difference between the two prodnets by 30, and the quotient is the diameter of the shadow." The Karanatilaka gives the following method for the Diameter of compntation of the diameter of the sun :- " Divide the of thesha- bhukti of the sun by 2, and write down the half in two dow aocordd- different places. In the one place divide it by IO, and taka. Raranati- add the qnotient to the number in the second place. The sum is the number of minutes of the diameter of the sun" In the computation of the diameter of the moon, he first takes the bhukti of the moon, adds thereto aoth of it, and divides the nnmber by 25. The quotient is the number of the minutes of the moon's diameter. In the computation of the diameter of the ehadow, lie multiplies the bhukti of the sun by 3, and from the prodnct he snbtracts fth of it. The remainder he aub- tracts from the bhukti of the moon, and the donble of

80 ALBERUNTS INDIA.

the remainder he divides by 15. The qnotient is the number of the minutes of the dragon's head and tail. If we would indulge in further quotatione from the canones of the Hindus, we should entirely get away from Pago 242. the subject of the present book. Therefore we restriot ourselves to quote from them only subjects more or less connected with the special subject of this book, which either are noteworthy for their strangeness, or which are unknown among our people (the Muslims) and in our (the Muslim) countries.

( 81 )

CHAPTER LVI.

ON THE STATIONS OF THE MOON.

TuR Hindus tise the lunar atations exactly in the same On the way as the zodiacal signs. As the ecliptic is, by the aven lunat zodiacal signs, divided into twelve equal parts, so, by statlona. the lunar stations, it is divided into twenty-seven equal parts. Each station occupies 13} degrees, or 8o0 minutes of the ecliptic. The planets enter into them and leave them again, and wander to and fro through their nor- thern and southern latitudes. The astrologers attribute to cach station a special nature, the quality of foreboding events, and other particular characteristic traits, in the same way as they attribute them to the zodiacal signs. The number 27 rests on the fact that the moon passes through the whole ecliptic in 27} days, in which nnm- ber the fraction of } may be disregarded. In a similar Lanar sta- way, the Arabs determine their lunar stations as begin- Arabs. tions of tho

ning with the moon's first becoming visible in the west till her ceasing to be visible in the east. Herein they use the following method :- Add to the circumference the amount of the revoln- tion of the sun in a lunar month. Subtract from the som the march of the moon for the two days called almihdk (i.e the 28th and 29th days of a lunation). Divide the remainder by the march of the moon for one day. The qnotient is 27 and a little more than i, which fraction must be connted as a whole day. However, the Arabs are illiterate people, who can neither write nor reckon. They only rely npon numbers and eyeaight. They have no other medinm of research than eyesight, and are not able to determine the lunar stations withont the fixed stars in them. If the Hindus VOL II,

83 ALBBRUNTS INDIA.

want to describe the single stations, they agree with the Arabs regarding certain atars, whilat regarding others they differ from them. On the whole, tha Arabe keep near to the moon's path, and use, in describing the stations, only those fixed stars with which the moon either stands in conjunction at certain times, or through the immediate neighbourhood of which she passes. Whether the The Hindus do not strictly follow the same line, but Hindus luntwaty. also take into account the various positions of one star twenty- with reference to the other, e.g. one star'e standing in ctatione, opposition or in the zenith of another. Besides, they reckon also the Falling Eagle among the stations, so as to get 28. It is this which has led our astronomere and the anthom of 'antd books astray; for they say that the Hindns have twenty-eight lunar stationa, but that they leave out one which is always covered hy the rays of the sun. Perhaps they may have beard that the Hindos call that station in which the moon is, the burning one; that station which it has just left, the left one after the embrace; and that station in which she will enter next, the smoking one. Some of our Muslim anthors have main- tained that the Hindus leave out the station Al-zubdnd, and account for it by declaring that the moon's path is burning in the end of Libra and the beginning of Scorpio. All this is derived from one and the same source, viz. their opinion that the Hindus have twenty-eight stations, and that under certain circumstances they drop one. Whilst just the very opposite is the case; they have twenty-seven stations, and under certain circumstances add one, A Vedie tra- Brahmagupta says that in the book of the Veda there dition from Brabins- is a tradition, derived from the inhabitants of Mount Meru, to this effect, that they see two suns, two moons, and fifty-four lunar stations, and that they have douhle the smount of days of ours. Then he tries to refute this theory by the argument that we do not see the fish (sic) of the pole revolve twice in a day, but only once. I for

CHAPTER LVI. 33

my part have no means of arraying this erroneous sen- tence in a reasonable shape. The proper method for the computation of the place Metbod for of a star or of a certain degree of a lunar atation is this :- the plaoo of compating Take its distance from o Aries in minntes, and divide degres of a them by 800. The quotient represents whole stations tou. Janer ate-

preceding that station in which the star in qnestion stands. Then remains to be found the particular place within the station in qnestion, Now, either star or degree is simply determined according to the 800 parts of the station, and reduced by a common denominator, or the degrees are reduced to minutes, or they are multiplied by 60 and the product is divided by 800, in which case the quotient represents that part of the atation which the moon has in that moment already traversed, if the station is reckoned as 1s. These methods of compntation suit as well the moon as the planets and other atara. The following, however, applies exclusively to the moon :- The product of the multiplication of the remainder (Le. the portion of the incomplete lunar atation) by 60 is divided by the bhukti of the moon. The quotient shows how mnch of the lunar nakshatra day has elspsed. The Hindus are very little informed regarding the Table of the fixed stars. I never came across any one of them who tions taken lunar eta-

knew the single stars of the lunar stations from eye- xhanda- from the

sight, and was able to point them out to me with his kAddyakd.

fingers. I have taken the greatest pains to investigate this subject, and to settle most of it by all sorts of com- parisons, and have recorded the results of my research in a treatise on the determination of the lunar stations. Of their theories on this subject I aball mention as much as I think suitable in the present context. But before, that I shall give the positions of the stations in longitnde and latitnde and their numbers, according to the canon Khandakhadyaka, facilitating the study of the subject by comprehending all details in the follow- ing table :-

ALBERUNTS INDIA. Atrint 8 o o Nortbern Albatali. . Rharaaf 0 Northorn . Krittika . 7 S Northarn AlthimyyL

4 Rohint S 19 Beutbern Aldabarta, torothar with the o stam of the besd of Taarma. .

3 Sontbern O Mrigattrba 3 S Alber'c

Årdra. Sogthern Unknowa. Most likely ideati-

Pnarvasn . 3 3 O 6 O Northorn onl with Canis Minot. Aldbirk.

Pushya Witbout any 3 16 O O 0 latitede Alaathra. Unkaown. Most likely identi- 6 6 o 9 3 Sonthern cal witk two stars of Canoer snd four star cotalde of it

Maghá Withont aay Aljabhs together witk two O 10 6 4 9 O Iatitade other itam. . ParraphAlguni . 4 37 O 12 O Northern Alsubra. Almifa, torother with the third 13 Uttaraphalgunt . 0 I3 Northern atar of Aldatira.

Harts. 5 Soatbern Consista of the stars of the Crow. . . 30 O 14 CitrA . Bouthern Alsimak Al's'al. . SvLtt 3 . 37 O Northern Alsimak Alramib. Vikha 6 . 0 . 7 5 I 30 Southern Unknown, .. 17 Anuradha . 3 Boathern S The Crown, torether with an- 4 7 5 other star. .

18 Jychtha . S The heart of Scorpio, together 3 7 19 4 O Sonthern .

19 Mila . Southern with the pericardium. 8 9 30 Alshauls 889 30 Parrisbagha 8 I4 S Southorn Alna'am Alwarid. Uttariabadha Southern Alna'Am Alşadir. CHAPTER LVI. 8 20 o 5 . Abhijit 8 Northern o 3 25 62 Alnasr Alwil. 33 23 Bravaza 3 9 8 O 30 o Northarn Alnaer Alta'ir. 24 Unknown. Most Hkaly it is 33 Dhanishtha 5 9 20 36 Northarn the Dolphin. 35 Bntabhishaj Unknown. Most likely idon- I 10 20 O Sonthern tical with the npper part of the hip-joint of Aquarins $6 Parrabbadrapada 2 10 26 24 Northern Unknown

Uttarabhadrapada 6 26 Northern Most likely identical with the stars of Arir Al'a'zam, Unknown. Most likely identi- 28 Revatt Withont any cal with some of the stars of o o O . I 0 O latitade the Cotton Thread betweon . the Two Fishes.

86 ALBERUNPS INDIA.

The notions of the Hindus regarding the stars are not free from confusion. They are only little skilled in practical observation and calculation, and hsve no under- standing of the motions of the fixed stars. So Varâha- On thepre- mihira says in his book Samhitd: " In six stations, tho cqui- beginning with Revati and ending with Mrigasiras, ob- tatles from servation precedes calculation, so that the moon enters hin, chap, each one of them earlier according to eyesight than tr. 7. according to calculation. " In twelve stations, beginning with Ardra and ending with Anuradha, the precession is equal to half a station, so that the moon is in the midat of a station according to observation, whilst she is in its first part according to caleulation. "In the nine stations, beginning with Jyeshtha and ending with Uttarabhadrapada, observstion falls back behind calculation, so that the moon enters each of them according to observation, when, according to cal- culation, she leaves it in order to enter the follow- ing." The suthor My remark relating to the confused notions of the Hindus regarding the stars is confirmed, though this is perhaps not apparent to the Hindus themselves, e.g. by the note of Varahamihira regarding Alsharatdn = Asvini, one of the first-mentioned six stations; for he says that in it observation precedes calculation. Now the two stars of Asvini stand, in our time, in two-thirds of Aries (i.c. between 1O°-20° Aries), and the time of Varahamihira precedes our time by about 526 years. Therefore by whatever theory you may compute the motion of the fized stars (or precession of the equinoxes), the Asvint: did, in his time, certainly not stand in less than one- third of Aries (i.e. they had not come in the precession of the eqninoxes farther than to 1°-10° Aries). Supposing that, in his time, Asvint really stood in this part of Aries or near it, as is mentioned in the Khandakhadyaka, which gives the computation of sun

CHAPTER LVI. 87

and moon in a perfectly correct form, we must state that at that time there was not yet known what is now known, viz. the retrograde motion of the star by the distance of eight degrees. How, therefore, could, in his time, obserzation precede calculation, since the moon, when standing in conjunction with the two stars, had already traversed nearly two-thirds of the first sta- tion ? According to the same analogy, also, the other etatements of Varahamihira may be examined. The stations occupy a smaller or larger space ac- Each sta- cording to their figures, i.c. their constellations, not pies the tion occu-

they themselves, for all stations occupy the same space on the same space on the ecliptic. This fact does not seem to be known ecliptic. to the Hindus, althongh we have already related similar notions of theirs regarding the Great Bear. For Brah- magupta says in the Uttara-khandakhadyaka, i.c. the emendation of the Khanda-khadyaka :- "The measure of some stations exceeds the measure Quotation of the mean daily motion of the moon by one half, maguptu. from Brah-

Accordingly their measure is 19° 45' 52" 18"". There are six stations, viz Rohini, Punarvasu, Uttarapbal- gunì, Visakhâ, Uttarâshâdha, Uttarabhadrapadâ. These together occupy the space of 118° 35' 13" 48". Fur- ther six stations are short ones, each of them ocenpying less than the mean daily motion of the moon by one half. Accordingly their measure is 6° 35' 17" 26'". These are Bharant, Ârdra, Âsleshâ, Svâti, Jyeshthâ, Satabhishaj. They together occupy the space of 39° 31' 44" 36"". Of the remaining fifteen atations, each occu- pies as much as the mean daily motion. Accordingly it occupies the space of 13° 10' 34" 52"". They to- gether ocenpy the space of 197° 38' 43". These three gronps of stations together occupy the space of 355° 45' 41" 24"", the remainder of the complete circle 4° 14' 18" 36", and this is the space of Abhijit, i.e. the Falling Eagle, which is left ont. I have tried to make the investigation of this snbject acceptable to the

88 ALBERUNI'S INDIA.

stndent in my above-mentioned special treatise on the lunar stations (v. p. 83). Quotation The scantiness of the knowledge of the Hindus re- from Vart hamfhira, garding the motion of the fixed stars is sufficiently ch. iiL 1-3- illustrated by the following passage from the Sanhitd of Varâhamihira :- " It has been mentioned in the books of the ancients that the summer solstice took place in the midet of Aslesha, and the winter solstice in Dha- nishtha. And this is correct for that time. Nowadays the summer solstice takes place in the beginning of Cancer, and the winter solstice in the beginning of Cap- Page 246. ricornus, If any one doubts this, and maintains that it is as the ancients have said and not as we say, let him go ont to some level country when he thinks that the summer solstice is near. Let him there draw a circle, and place in its centre some body which stands perpen- dicular on the plain. Let him mark the end of its shadow by some sign, and continne the line till it reaches the circumference of the circle either in east or west. Let him repeat the same at the same moment of the following day, and make the same observation. When he then finds that the end of the shadow deviates from the first sign towards the south, he must know that the ann has moved towards the north and has not yet reached its solstice. But if he finds that the end of the shadow deviates towards the north, he knowe that the sun has already commenced to move sonth- ward and has already passed its solstice. If a man continnes this kind of observations, and thereby finds the day of the solstice, he will find that our worde are true"

The sathor This passage shows that Varahamihira had no know- on the pre- cemion of ledge of the motion of the fixed stars towards the east. the equi- Ha considers them, in agreement with the name, as DOXAL fixed, immovable stars, and represents the solstice as moving towards the west. In conseqnence of this fancy, he has, in the matter of the Iunar stations, confounded

CHAPTER LVI. 89

two things, between which we shall now properly dis- tinguish, in order to remove donbt and to give the matter in a critically emended form. In the order of the zodiacal signs we begin with that twelfth part of the ecliptic which lies north of the point of intersection of the equator and the ecliptic according to the second motion, i.c. the precession of the equinoxes. In that case, the summer solstice always occurs at the beginning of the fourth sign, the winter solstice at the beginning of the tenth sign. In the order of the lunar ststions we begin with thst twenty-seventh part of the ecliptic which belongs to the first of the first zodiacal sign. In thst case the summer solstice falls always on three-fourths of the seventh station (ia. on 6o0' of the etation), and the winter solstice on one fourth of the twenty-first station (ie. on 200' of the station). This order of things wil remain the same as long as the world lasts. If, now, the Iunar stations sre marked by certain constellations, and are called by names peculiar to these constellations, the stations wander round together with the constellations. The stars of the zodiacal signs and of the stations have, in bygone times, oceupied earlier (i.e. more western) parts of the ecliptic. From them they have wandered into those which they occupy st present, and in future they will wander into other atill more eastern parts of the ecliptic, so that in the course of time they will wander through the whole ecliptic. According to the Hindus, the stars of the station Aslesha stand in 18° of Cancer. Therefore, according to the rate of the precession of the equinoxes adopted by the ancient astronomers, they stood 1800 years before our time in the o° of the fourth sign, whilst the con- stellation of Cancer stood in the third sign, in which there was also the solstice. The solstice has kept its place, but the constellations have migrated, just the very opposite of what Varahamihira has fancied.

( 90 )

CHAPTER LVII.

ON THE HELIACAL RISINGS OF THE STARS, AND ON THE CEREMONIES AND RITES WHICH THE HINDUS PRAC- TISE AT SUCH A MOMENT.

Bow far a THE Hindu method for the computation of the heliacal distant from risings of the stars and the young moon is, as we think, etar mast bo

order to bo- the same as is explained in the canones called Sindhind. the sun in come viri- ble. They call the degrees of a star's distance from the sun which are thought necessary for its heliacal rising kalamsaka. They are, according to the author of the Ghurrat-alzijat, the following :- 13° for Suhail, Alya- mâniya, Alwâki, Alayyûk, Alsimâkân, Kalb-al'akrab; 20° for Albutain, Alhak'a, Alnathra, Âśleshâ, Śata- bhishaj, Revati; 14° for the others. Evidently the etars have, in this respect, been divided Page 247- into three groups, the first of which seems to comprise the stars reckoned by the Greeks as stars of the first and second magnitude, the second the stars of the third and fonrth maguitnde, and the third the stars of the fifth and sixth magnitnde. Brahmagupta onght to have given this classification in his emendation of the Khandakhadyaka, but he has not done so. He expresses himself in general phrases, and simply mentions 14° distance from the sun as necessary for the heliacal risings of all lunar stations. Quotatiob Vijayanandin says: " Some stars are not covered by from Vijaya- aandin the rays nor impaired in their shining by the sun, viz. Al'ayyûk, Alsimak, Alramih, the two Eagles, Dhanish- thå, and Uttarabhadrapada, because they have so much

CHAPTER LVII. 91

northern latitude, and becanse also the country (of the observer) has so much latitude. For in the more northern regions they are seen both at the beginning and end of one and the same night, and never dis- appear." They have particular methods for the compntation On the of the heliacal rising of Agastya, ie. Suhail or Canopus, ing of Cano- heliacal rls-

They observe it firat when the sun enters the station pus

Hasta, and they lose it out of sight when he enters the station Rohint. Pulisa saye: " Take donble the apsis of the sun. If it is equalled by the corrected place of the sun, this is the time of the heliacal setting of Agastya." The apsis of the sun is, according to Pulisa, 23 zodiacal signs. The double of it falls in 1o° of Spica, which is the beginning of the station Hasta. Half the apsis falls on 10° of Taurus, which is the beginning of the station Rohiņi. Brahmagupts maintains the following in the emen- Quotation dation of the Khandakhadyaka :- from Brah- magupta. "The position of Snhail is 27° Orion, its southern latitude 71 perts. The degrees of its distance from the sun necessary for its heliacal rising are 12. "The position of Mrigavyadha, i.c. Sirius Yemenicus, is 26° Orion, its sonthern latitude 40 parts. The de- grees of ita distance from the sun necessary for its heliacal rising are 13. If you want to find the time of their risings, imagine the sun to be in the place of the star. That amount of the day which has already elspsed is the number of degrees of ita distance from the sun necessary for its heliacal rising. Fix the ascendens on this particular place. When, then, the sun reaches the degree of this ascendens, the star first becomes visible. "In order to find the time of the heliacal setting of a star, add to the degree of the star six complete zodiacal signs. Snbtract from the sum the degrees of its dis- tance from the sun necessary for its heliacal rising, and

92 ALBERUNTS INDIA.

fix the ascendens on the remainder. When, then, the sun enters the degree of the ascendens, that is the time of its setting." On tho The book Samhitd mentions certain sacrifices and coromonics pramteed at ceremonies which are practised at the heliacal risings rring of oer. of various atars. We shall now record them, translat- tho beliacal tain itara, ing also that which is rather chaff than wheat, since we have made it obligatory on ourselves to give the quota- tions from the books of the Hindus complete and exactly as they are Quotation Varâhamihira tays: " When in the beginning the sun from Vari- bamihira's had risen, and in his revolution had come to etand in the xil. pretses, zenith of the towering mountain Vindhya, the latter Samhusd, ch. end vr. 1-18, OH would not recognise his exalted position, and, actuated Armtraand by haughtineas, moved towards him to hinder his Canopus the merifics to him. march and to prevent his chariot from passing above it. The Vindhya rose even to the neighbourhood of Paradise and the dwellings of the Vidyadharas, the apiritual beings Now the latter hastened to it because it was pleasant and its gardens and meadows were lovely, and dwelt there in joy; their wives going to and fro, and their children playing with each other. When the wind blew against the white garments of their danghters, they flow like waving bannera. In its ravines the wild animals and the lions ap- pear as dark black, in consequence of the multitude of the animals called dhramara, which cling to them, liking the dirt of their bodies when they rub each other with the soiled claws. When they attack the rutting elephants, the latter become raving. The monkeys and bears are seen climbing up to the horns of Vindhys and to its lofty peaks; as if by instinct, they took the direction towards heaven. The anchorites are seen at its water-places, satisfied with nourishing themselves by its fruita. The further glorious things of the Vin- dhys are innumerable. When, now, Agastya, the son of Varuna (Le. Suhail,

CHAPTER LVII. 93 the son of the water), had observed all these proceed- ings of the Vindhya, he offered to be his companion in his aspirations, and asked him to remain in his place until he (Agastya) should return and shonld have freed him (Vindhys) from the darkness which was on him. V. I .- Then Agastya turned towards the ocean, de- vouring its water, so that it disappeared. There appeared the lower parts of the mountain Vindhya, whilst the makara and the water animals were clinging to it. They scratched the mountain till they pierced it and dug mines in it, in which there remained gems and pearls. V. 2 .- The ocean became adorned by them, further by trees which grew up, though it (the water) was feeble, and by serpents rushing to and fro in windings on its surface. V. 3 .- The mountain has, in exchange for the wrong done to it by Suhail, received the ornament which it has acquired, whence the angels got tiaras and crowns made for themselves. V. 4-Likewise the ocean has, in exchange for the sinking down of its water into the depth, received the sparkling of the fishes when they move about in it, the appearance of jewels at its bottom, and the rushing to and fro of the serpents and snakes in the remainder of its water. When the fishes rise over it, and the conch- shells and pearl-oysters, yon would take the ocean for ponds, the snrface of their water being covered with the white lotus in the season of sarad and the season of autamn. V. 5 .- Yon could scarcely distinguish between this water and heaven, because the ocean is adorned with jewels as the heaven is adorned with stars; with many- headed serpents, resembling threads of rays which come from the sun; with crystal in it, resembling the body of the moon, and with a white mist, above which rise the olonds of heaven, V. 6 .- How should I not praise him who did this

94 ALBERUNTS INDIA.

great deed, who pointed ont to the angels the beanty of the orowns, and made the ocean and the mountain Vindhya a treasure-house for them ! V. 7 .- That is Subail, by whom the water becomes clean from earthly defilement, with which the purity of the heart of the pious man is commingled, clean, I say, from that which overpowers him in the intercourse with the wicked. V. 8 .- Whenever Agastya rises and the water in- creases in the rivers and valleys during his time, you see the rivars offering to the moon all thet is on the surface of their water, the various kinds of white and red lotus and the papyrus; all that swima in them, the ducks and the geese (pelicans ?), as a sacrifice unto him, even as a young girl offers roses and presents when she enters them (the rivers). V. 9-We compare the standing of the pairs of red geese on the two ahores, and the ewimming to and fro of the white ducks in the midst while they sing, to the two lips of a beautiful woman, showing her teeth when she laughs for joy. V. IQ-Nay, we compare the black lotus, standing between white lotus, and the dashing of the bees against it from desire of the fragrancy of its smell, with the black of her pupil within the white of the ring, moving coquettishly and amorously, being surrounded by the hair of the eyebrows. V. 11 .- When you then see the ponds, when the light of the moon has risen over them, when the moon illu- minates their dim waters, and when the white lotus opens which was shut over the bees, yon would think them the face of a beantiful woman, who looks with a black eye from a white eyeball. V. 12 .- When a stream of the torrents of Varshakâla has flown to them with serpents, poison, and the impu- rities, the rising of Suhail above them cleans them from defilement and saves them from injury.

CHAPTER LVII. 95

V. 13 .-- As one momant's thinking of Suhail before the door of a man blots out his sins deserving of punish- Page 249- ment, how much more effective will be the fluency of the tongue praising him, when the task is to do away with sin and to acquire heavenly reward! The former Rishis have mentioned what sacrifice is necessary when Suhail rises. I shall make a present to the kings by relating it, and shall make this relation a sacrifice unto Him. So I say: V. 14 .- His rising takes place at the moment when some of the light of the sun appears from the east, and the darkness of night is gathered in the west. The beginning of his appearance is difficult to perceive, and not every one who looks at him nndertands it. There- fore ask the astronomer at that moment about the direc- tion whence it rises. V. 15, 16 .- Towerds this direction offer the sacrifice called argha, and apread on the earth what you hap- pen to have, roses and fragrant flowers as they grow in the country. Put on them what yon thiok fit, gold, garments, jewels of the sea, and offer incense, saffron, and sandalwood, musk and camphor, together with an ox and a cow, and many dishes and sweet- meats, V. 17 .- Know that he who does this during seven consecutive years with pions intention, atrong belief, and confidence, possesses at the end of them the whole earth and the ocean which surrounds it on the four sides, if he is a Kshatriya. V. 18 .- If he is a Brahman, he obtains his wishes, learns the Veda, obtains a beantifol wife, and gets noble children from her. If he is a Vaisya, he obtains much landed property and acquires a glorious lordship. If he is a Shdra, he will obtain wealth, All of them obtain health and safety, the cessation of injuries, and the realisation of reward." This is Varâhamihira's statement regarding the offering

96 ALBERUNI'S INDIA.

to Suhail. In the same book be gives also the rules rogarding Rohiņt: " Garga, Vasishtha, KMyape, and Partfara told their pupils that Mount Moru is built of planks of gold. an Baati Out of them there have risen trees with numerous sweet-swelling flowers and blossoms The bees already surround them with & humming pleasant to hear, end the nymphs of the Devas wander there to and fro with exhilarating melodies, with pleasant instruments and overlasting joy. This mountain lies in the plain Nan- danavana, the park of paradise. So they say. Jupiter was there at a time, and then Narada the Rishi asked him regarding the prognostics of Rohint, upon which Jupiter explained them to him. I shall here relate them as far as necessary. V. 4-Let & man in the black days of the month Ashadha observe if the moon reaches Rohint. Let him seek to the north or east of the town e high spot. To this spot the Brahman must go who has the charge of the houses of the kings. He is to light there a fire and to draw a diagram of tho various planets and lunar stations round it. Ho is to recite what is necessary for each one of them, and to give each its share of the roses, barley, and oil, and to make each planet propi- tious by throwing these things into the fire. Round the fire on all four sides there must be as mach as possible of jewels and jugs filled with the aweetest water, and whatever else there happens to be at hand at the moment, fruits, drugs, branohes of trees, and roots of planta Further, he is to spread there grass which is cut with a nickle for his night-quarters. Then he is to tako the different kinds of seeds and corns, to wash them with water, to put gold in the midst of them, and to deposit them in a jug. He is to place it towards & certain direction, and to prepare Homa, i.c. throw- ing barley and oil into the fire, at the same time re- citing certain passages from the Veda, which refer to

CHAPTER LVII. 97

different directions, viz. Varuņa-mantra, Vayava-mantra, Page aso. and Soma-mantra He raises a danda, i.c. a long and high epear, from the top of which hang down two straps, the one as long as the spear, the other thrice as long. He must do all this before the moon reaches Rohint, for this purpose, that when ahe reaches it, he should be ready to deter- mine the times of the blowing of the wind as well as its directions. He learns this by means of the straps of the spear. V. 10 .- If the wind on that day blowa from the centres of the four directions, it is considered propitious; if it blows from the directions between them, it is considered unlucky. If the wind remaina steady in the same direction, powerful and withont changing, this too is considered propitious. The time of its blowing is measured by the eight parts of the day, and each eighth part is considered as corresponding to the half of a month. V. 11 .- When the moon leaves the station Rohint, you look at the seeds placed in & certain direction. That of them which spronts will grow plentifully in that year. V. 12 .- When the moon comes near Rohint, you must be on the look-ont. If the sky is clear, not affected by any disturbance; if the wind is pure and does not cause a destructive commotion; if the melodies of the animals and birds are pleasant, this is considered pro- pitious. We shall now consider the clonds. V. 13, 14-If they float like the branches of the valley (? batn ?), and out of them the flashes of lightning appear to the eye; if they open as opens the white lotus; if the lightning encircles the cloud like the rays of the sun; if the cloud has the colour of stibium, or of bees, or of saffron ; V. 15-19-If the aky is covered with clonds, and ont of them flashss the lightning like gold, if the rain- VOL. II. G

ALBERUNPS INDIA.

bow shows its round form coloured with something liko the red of evening twilight, and with coloura like thoro of the garments of a bride; if the thunder roars like the screaming peacock, or the bird which cannot drink water except from falling rain, which then screams for joy, as the frogs enjoy the full water-places, so as to croak vehemently; if yon see the eky raging like the raging of elephants and buffaloes in the thicket, in the various parts of which the fire is blazing; if the clonds move like the limbs of the elephants, if they shine like the shining of pearls, conch-shells, snow, and even as the moonbeams, as though the moon had lent the clonds her lustre and splendour; V. 20-All this indicates much rain and blessing by a rich growth, V. 25 .- At the time when the Brahman sits amidst the water-jugs, the falling of stars, the flashing of the lightning, thunderbolta, red glow in the sky, tornado, earthquake, the falling of hail, and the screaming of the wild animals, all these things are considered as unincky. V. 26-If the water decreases in a jag on the north side, cither by itself, or by a hole, or by dripping away, there will be no main in the month Sravana. If it de- creases in a jug on the cast side, there will be no rain in Bhadrapada If it decroases in a jug on the south aide, there will be no min in Aśvayuja; and if it de- crenses in a jug on the west side, there will be no rain in KArttika. If there is no decrease of water in the jugs, the summer rain will be perfect. V. 27 .- From the jugs they also derive prognostics as to the different castes. The northern jug refera to the Brahman, the eastern to the Kshatriya, the sonthern to the Vaisya, and the western to the Shdra. If the names of people and certain circumstances are inscribed upon the jugs, all that happens to them if, eg. they break or the water in them decreases, is considered as

CHAPTER LVII. 99

prognosticating something which concerns those per- sons or circumstances." "The rules relating to the stationa Svatt and Sravana Samaitd, are similar to those relating to Rohint When yon are r. r., on chap. xxt.

in the white days of the month Ashadha, when. the gramne Svåti and moon atands in either of the two stations Ashadha, ia Pûrva-ashadha or Uttara-ashadha, select a spot as yon have selected it for Rohint, and take a balance Pagn agt. of gold. That is the best. If it is of silver, it is Samhitd, middling. If it is not of silver, make it of wood v.9 chap. zIvi.

called khayar, which seema to be the khadira tree (i.c. Acacia catechn), or of the head of an arrow with which already a man has been killed. The smallest measure for the length of its beam is a span. The longer it is, the better; the shorter it is, the less favourahle. V. 6 .- A scale has four strings, each ro digits long- Its two scales are of linen cloth of the size of 6 digits. Its two weights are of gold. V. 7, 8-Weigh by it eqoal quantities of each matter, water of the wells, of the ponds, and of the rivers, elephants' teeth, the hair of horses, pieces of gold with the names of kings written on them, and pieces of other metal over which the names of other people, or the names of animals, years, days, directions, or countries have been pronounced. V. 1 .- In weighing, turn towards the esst; put the weight in the right scale, and the things which are to be weighed in the left. Recite over them and speak to the balance: V. 2 .- 'Thou art correct; thon art Deva, and the . wife of a Deva. Thou art Sarasvati, the daughter of Brahman, Thon revealest the right and the truth. Thon art more correct than the soul of correctness. V. 3 .- Thou art like the son and the planets in their wandering from east to west on one and the same road. V. 4-Through thee stands upright the order of the

100 ALBERUNTS INDIA.

world, and in thee is united the truth and the correct- ness of all the angels and Brahmans. V. 5 .- Thou art the daughter of Brahman, and a man of thy house is Kafyapa.' V. I .- This weighing must take place in the even- ing. Then put the things aside, and repeat their weighing the next morning. That which has increased in weight will flourish and thrive in that year; that which bas decreased will be bad and go back. This weighing, however, is not only to be done in Ashadha, bat also in Rohini and Svatt. V. 1I .- If the year is a leap-year, and the weigh- ing happens to take place in the repeated month, the weighing is in that year twice done. V. 12 .- If the prognostics are identical, what they forebode will happen. If they were not identical, observe the prognostics of Rohint, for it is predomi- nant."

( 101 ) ..

CHAPTER LVIII.

HOW EBB AND FLOW FOLLOW EACH OTHER IN THE OCEAN.

WITH regard to the cause why the water of the ocean Qnotution always remains as it is, we quote the following pasgage Matya- from the

from the Matsya-Purdna :- " At the beginning there Purdpa.

were sixteen mountains, which had wings and could fly and rise np into the air. However, the rays of Indra, the ruler, burned their wings, so that they fell down, deprived of them, somewhere abont the ocean, fonr of them iu each point of the compass-in the east, Risha- bha, Balahaka, Cakra, Maioaka; in the north, Candra, Kanka, Drona, Suhma; in the west, Vakra, Vadhra, Nârada, Parvata; in the south, Jimûta, Dravina, Main- ka, Mahasaila (?). Between the third and the fourth of the eastern monntains there is the fire Samvartaka, which drinks the water of the ocean. But for this the ocean would fill up, eince the rivers perpetually flow to it. " This fire was the fire of one of their kings, called story of Aurva. He had inherited the realm from his father, King Aurva. who was killed while he was still an embryo. When he was born and grew up, and heard the history of his father, he became angry against the angels, and drew his sword to kill them, since they had neglected the guardianship of the world, notwithstanding mankind's : worshipping them and notwithstanding their being in close contact with the world. Therenpon the angels humiliated themselves before him and tried to con-

102 ALBERUNI'S INDIA.

ciliate him, so that he ceased from his wrath. Then he apoke to them: 'But what am I to do with the fire of my wrath ?' and they advised him to throw it into the ocean. It is this fire which absorbs the waters of the ocean. Others say : 'The water of the streams does not Pago 25z. increase the ocean, because Indra, the ruler, takes up the ocean in the shape of the cloud, and sends it down as rains.'" The man in Again the Matsya-Purdna says: "The black part in the moon. the moon which is called Sasalaksha, i.e. the hare's figure, is the image of the figures of the above-men- tioned sixteen mountains reflected by the light of the moon on her body." The Vishnu-Dharma says : " The moon is called Sasa- laksha, for the globe of her body is watery, reflecting the figure of the earth as a mirror reflects. On the earth there are mountains and trees of different shapes, which are reflected in the moon as a hare's figure. It is also called Mrigalancana, i.c. the figure of a gazelle, for certain people compare the black part on the moon's face to the figure of a gazelle." Story of the The lunar stations they declare to be the daughters loprosy of the moon. of Prajapati, to whom the moon is married. He was especially attached to Rohini, and preferred her to the others. Now her sisters, urged by jealousy, complained of him to their father Prejapati. The latter strove to keep peace among them, and admonished him, but with- out any success. Then he cursed the moon (Zunus), in consequence of which his face became leprous. Now the moon repented of his doing, and came penitent to Prajapati, who spoke to him: "My word is one, and cannot be cancelled; however, I shall cover thy shame for the half of each month." Thereupon the moon spoke to Prajapati: "But how shall the trace of the sin of the past be wiped off from me ?" Prajapati answered: " By erecting the shape of the linga of Mahadeva as an object of thy worship." This he did. The linga he

CHAPTER LVIII. 103 raised was the stone of Somanath, for soma means the The ldol of moon and ndtha means master, so that the whole word Bomanath.

means master of the moon. The image was destroyed by the Princa Mahmud-may God be merciful to him !- A.H. 416. He ordered the npper part to be broken and the remainder to be transported to his resi- deuce, Ghazntn, with all its coverings and trappings of gold, jewels, and embroidered garments. Part of it has been thrown into the hippodrome of the town, togetber with the Cakrasvamin, an idol of bronze, that had been brought from Taneshar. Another part of the idol from Somanath lies before the door of the mosque of Ghaznin, on which people rub their feet to clean them from dirt and wet. The linga is an image of the penis of Mahadeva. I Origin of have hesrd the following story regarding it :- "A Rishi, tho Linga.

on seeing Mahadevs with his wife, became suspicious of him, and cursed him that he should lose his penis. At once his penis dropped, and was as if wiped off. But afterwards the Rishi was in a position to establish the eigns of his innocence and to confirm them by the necessary proofs. The suspicion which had troubled his mind was removed, and he spoke to him: 'Verily, I shall recompense thee by making the image of the limb which thou hast lost the object of worship for men, who thereby will find the road to God, and come near him.'" Varâhamihira says about the construction of the The con- linga: "After having chosen a faultless stone for it, the Lngn struction of take it as long as the image is intended to be. Divide Vardhami- a according to it into three parts. The lowest part of it is quad- bre. rangular, as if it were a cube or quadrangular column. ivii. 53 Aitd, chan, The middle part is octagonal, its surface being divided by four pilasters. The upper third is round, rounded off so as to resemble the gland of a penis. . V. 54 .- In erecting the figure, place the qnadran- gular third within the earth, and for the octagonal third

104 ALBERUNIS INDIA. make a cover, which is called pinda, quadrangular from withont, but so as to fit also on the quadrangular third in the earth. The octagonal form of the inner side is to fit on to the middle third, which projects out Pago 253.] of the earth. The round third alone remains withont cover." Further he saye :- V. 55 .- " If you make the round part too small or too thin, it will hurt the country and bring abont evil among the inhabitants of the regions who have con- structed it. If it does not go deep enough down into the earth, or if it projects too little ont of the earth, Chapter Iz .: this causes people to fall ill When it ie in the course v. 6. of construction, and is struck by a peg, the ruler and his family will perish, If on the transport it is hit, and the blow leaves a trace on it, the artist will perish, and destruction and diseases will spread in that country." The worship In the sonth-west of the Sindh conntry this idol is of the idol of Somanth. frequently met with in the houses destined for the worship of the Hindus, but Somanath was the most famous of these places. Every day they bronght there a jug of Ganges water and a basket of fiowers from Kashmir. They believed that the linga of Somanath would cure persons of every inveterate illness and heal every desperate and incurable disease. The reason why in particular Somanath has become so famous is that it was a harbour for seafaring people, and a station for those who went to and fro between Sufala in the country of the Zanj and China. Popaler be- Now as regards ebb and flow in the Indian Ocean, Hef about therameof of which the former is called bharna (?), the latter the tides. vuhara (?), we state that, according to the notions of the common Hindus, there is a fire called Vadavdnala in the ocean, which is always blazing. The flow is caused by the fire's drawing breath and its being blown up by the wind, and the ebb is caused by the fire's exhaling

CHAPTER LVIII. 105 the breath and the cessation of its being blown up by the wind. Mant has come to a beliet like this, after he had heard from the Hindus that there is a demon in the sea whose drawing breath and exhaling breath causes the flow and the ebb. The educated Hindus determine the daily phases of the tides hy the rising and setting of the moon, the monthly phases by the increase and waning of the moon; but the physical cause of both phenomena is not understood by them. It is flow and ebb to which Somanath owes its name Origin of the (ia master of the moon); for the stone (or linga) of of soma- , eacredneas

Somanath was originally erected on the coast, a little nath less than three miles west of the mouth of the river Sarsuti, east of the golden fortress Baroi, which had appeared as a dwelling-place for Vasudeva, not far from the place where he and his family were killed, and where they were barned. Each time when the moon rises and sets, the water of the ocean rises in the flood so as to cover the place in question. When, then, the moon reaches the meridian of noon and midnight, the water recedes in the ebb, and the place becomes again visible. Thus the moon was perpetually oceupied in serving the idol and bathing it. Therefore the place was considered as sacred to the moon, The fortress which contained the idol and its treasures was not ancient, but was built only about a hundred years ago. The Vishnu-Purdna asys: "The greatest height of Quotation the water of the flow is 1500 digits." This etatement Vuknu. from tho

seems rather exaggerated; for if the waves and the Purdnd

mean height of the ocean rose to between sixty to seventy yards, the shores and the bays would be more overfiown than has ever been witnessed, Still this is not entirely improbable, as it is not in itself impossible on account of some law of nature. The fact that the just-mentioued fortress is said to

106 ALBERUNTS INDIA.

have appeared out of the ocean is not astonishing for that particnlar part of the ocean; for the Dtbajat The golden islands (Maledives and Laccadives) originate in e fortren aimilar manner, rising out of the ocean as sand-downs. Furllel of the Male They increase, and rise, and extend themselves, and dires and remain in this condition for e certain time. Then they become decrepit as if from old age; the single parta become dissolved, no longer keep together, and dis- appear in the water as if melting away. The inhabi- tants of the islands quit that one which apparently dies away, and migrate to a young and fresh one which is abont to rise above the ocean. They take their cocoa- nnt palms along with them, colonise the new ialand, and dwell on it. That the fortress in question is called golden may only be a conventional epithet. Possibly, however, this object is to be taken literally, for the islands of the Zabaj are called the Gold Country (Suvarnadvipa), because you obtain much gold as deposit if you wash only a little of the earth of that country.

( 107 )

CHAPTER LIX.

ON THE SOLAR AND LUNAR ECLIPSES.

IT is perfectly known to the Hindn astronomers that the moon is eclipsed by the shadow of the earth, and the sun is eclipsed by the moon. Hereon they have based their computations in the astronomical hand- books and other works. Varahamihira says in the Samhitd :-- V. I .- "Some scholars maintain that the Head he- Quotation longed to the Daityas, and that his mother was Simh- hamihfra's from Vara-

hika. After the angels had fetched the amrita out of ch. v. the ocean, they asked Vishnn to distribnte it among them. When he did so, the Head also came, resembling the angels in shape, and associated himself with them. When Vishnn handed him a portion of the amrita, he took and drank it. But then Vishnu perceived who it was, hit him with his round cakra, and eut off his bend. However, the head remained alive on acconnt of the amrita in its month, whilst the body died, since it had not yet partaken of the amrita, and the force of the latter had not yet spread through it. Then the Head, humbling itself, spoke: 'For what sin has this been done?' Therenpon he was recompensed by being raised to heaven and by being made one of its inha- bitants. V. 2 .- Others say that the Head has a body like sun and moon, bnt that it is black and dark, and cannot therefore be seen in heaven. Brahman, the first father,

308 ALBERUNPS INDIA.

ordered that he should never appear in heaven except at the time of an eclipse. V. 3 .- ' hers say that he has a head like that of a serpent, ar a tail like that of a serpent, whilst others say that he has no other body besides the black colour which is seen." After having finished the relation of these absurdities, Varâhamibira continues :- V. 4-"If the Head had a body, it wonld act by immediate contact, whilst we find that he eclipses from a distance, when between him and the moon there is an interval of six zodiacal signs. Besides, his motion does not increase nor decrease, so that we cannot imagine an eclipse to be caused by his body reaching the spot of the lanar eclipse. V. 5 .- And if a man commits himself to euch a view, let him tell us for what purpose the cycles of the Head'e rotation have been calculated, and what is the use of their being correct io consequence of the fact that his rotation is a regular one. If the Head is imagined to be a serpent with head and tail, why does it not eclipse from a distance less or more than six zodiacal signs ? V. 6 .- His body is there present between head and tail; both hang togethar by means of the body. Still it does not eclipse aun nor moon nor the fixed stars of the lunar stations, there being an eclipse ouly if there are two heads opposed to each other. V. 7 .- If the latter were the case, end the moon rose, being eclipeed by one of the two, the son would necessarily set, being eclipeed by the other. Likewise, Pogo a55. if the moon should set eclipsed, the sun would rise eclipsed. And nothing of the kind ever occurs, V. 8 .- As has been mentioned hy scholars who enjoy the help of God, an eclipse of the moon is her enter- ing the shadow of the earth, and an eclipse of the sun consista in this that the moon covers and hides the sun

CHAPTER LIX. 109 from us. Therefore the lunar eclipse will never revolve from the west nor the solar eclipse from the east. V. 9 .- A long shadow stretches awsy from the earth, in like manner as the shadow of a tree. V. 10 .- When the moon has only little Istitade, standing in the seventh sign of its distance from the sun, and if it does not stand too far north or south, in that case the moon enters the shadow of the earth and is eclipsed thereby. The first contact takes place on the side of the east. V. 11 .- When the sun is reached by the moon from the west, the moon covers the sun, as if & portion of s cloud covered him. The amount of the covering differs in different regions. V. 12 .-- Because that which covers the moon is large, her light wanes when one-half of it is eclipsed; and because that which covers the sun is not large, the rays are powerfnl notwithstanding the eclipse. V. 13 .- The nature of the Hcad has nothing what- ever to do with the lunar and solar eclipses. On this subject the scholars in their books agree." After having described the nature of the two eclipses, as he understands them, he complains of those who do not know this, and says: " However, common people are always very lond in proclaiming the Head to be the cause of an eclipse, and they say, 'If the Head did not appear and did not bring about the eclipse, the Brahmans would not at that moment undergo an obli- gatory washing.'" Varahamihira says :-- V. 14-"The reason of this is that the head humi- liated itself after it had been cut off, snd received from Brahman a portion of the offering which the Brahmane offer to the fire at the moment of an eclipse. V. 15 .- Therefore he is near the spot of the eclipse, searching for his portion. Therefore at that time people mention him frequently, and consider him as the cause

11O ALBERUNTS INDIA.

of the eclipse, although he has nothing whatsoover to do with it; for the eclipse depends entirely upon the nniformity and the declination of the orbit of the moon" Prim of The latter words of Varahamihira, who, in passages qnoted previously, has already revealed himself to us as a man who accnrately knows the shape of the world, are odd and surprising. However, he seems sometimes to side with the Brahmans, to whom he belonged, and from whom he could not separate himself, Still he does not deserve to be blamed, as, on the whole, his foot stands firmly on the basis of the truth, and he clearly speaks out the truth. Compare, e.g. his state- ment regarding the Samdhi, which we have mentioned above (v. i. 366). Strictures Would to God that all distinguished man followed mpta'swaat his example! But look, for instance, at Brahmagupta, d tincerity. who is certainly the most distinguished of their astro- nomers. For as he was one of the Brahmans who read in their Puranas that the sun is lower than the moon, and who therefore require e bead biting the sun in order that he should be eclipsed, he shirks the truth and lends his support to imposture, if he did not-and this we think by no means impossible-from intense disgust at them, speak as he spoke simply in order to mock them, or under the compulsion of some mental derangement, like a man whom death is about to rob of : his consciousness. The words in question are found in the first chapter of his Brahmasiddhanta ;- Gaatation "Some people think that the eclipse is not cansed by frota tho Brakmasid- the Head. This, however, is a foolish idea, for it is he in fact who eclipses, and the generality of the inbabi- Pago a56. tants of the world say that it is the Head who eclipses. The Veda, which is the word of God from the mouth of Brahman, says that the Head eclipses, likewise the book Smriti, composed by Manu, and the Samhitd, composed by Garga the son of Brahman. On the contrary, Vara-

CHAPTER LIX.

hamihira, Śrisheņa, Åryabhata, and Vishņucandra main- tain that the eclipse is not cansed by the Head, bnt by the moon and the shadow of the earth, in direct opposition to all (to the generality of men), and from enmity against the just-mentioned dogma. For if the Head does not cause the eclipse, all the usages of the Brahmans which they practise at the moment of an eclipse, viz. their rubbing themselves with warm oil, and other works of prescribed worship, would be illu- sory and not be rewarded by hesvenly bliss. If a man declares these things to be illusory, he stands ontside of the generally acknowledged dogma, and that is not allowed. Manu says in the Smriti: 'When the Head keeps the snn or moon in eclipse, all waters on earth become pure, and in purity like the water of the Ganges.' The Veda saya: 'The Head is the son of a woman of the daughters of the Daityas, called Sainakd' (? Simhika ?). Therefore people practise the well-known works of piety, and therefore those anthors must cease to oppose the generality, for everything which is in the Veda, Smriti, and Samhita is true." If Brahmagupta, in this respeet, is one of those of whom God says (Koran, Stra xxvii. 14), " They have denied our signs, although their hearts knew them clearly, from wickedness and haughtiness," we shall not argne with him, but only whisper into his ear: If people must nnder cirentstances give up opposing the reli- gious codes (as seems to be your case), why then do you order people to be pious if you forget to be so your- self? Why do you, after having apoken such words, then begin to calenlate the diameter of the moon in order to explain her eclipsing the sun, and the dia- meter of the shadow of the earth in order to explain its eclipsing the moon ? Why do you compute both eclipses in agreement with the theory of those heretics, and not according to the views of those with whom you think it proper to agree? If the Brahmans are ordered to

112 ALBERUNPS INDIA.

practise some act of worship or something else at the occqrrence of an eclipse, the eclipse is only the date of these things, not their cause, Thus we Muslims are bound to asy certain prayera, and prohibited from say- ing others, at certain times of the revolution of the sun and his light. These things are simply chronological dates for those acts, nothing more, for the sun has nothing whatever to do with our (Muslim) worship, Brahmagupta says (ii IIO), "The generality thinks thus" If he thereby means the totality of the inhabi- tants of the inhahitable world, we can only say that he would be very little able to investigate their opinions either by exact research or by means of historical tra- dition. For India itself is, in comparison to the whole inhabitable world, only a small matter, and the number of those who differ from the Hindus, both in religion and law, is larger than the number of those who agree with them, Pordbte Or if Brahmagupta means the generality of the Hindus, cTeuem for Brabano- we agree that the uneducated among them are much more numerous than the educated; but we also point out that in all our religious codes of divine revelation the uneducated crowd is blamed as being ignorant, always doubting, and ungrateful. I, for my part, am inclined to the belief that that which made Brahmagupta speak the above-mentioned words (which involve a sin against conscience) was something of a calamitous fate, like that of Socrates, which had befallen him, notwithstanding the abun- dance of his knowledge and the sharpness of his intel- Pago 157. lect, and notwithstanding his extreme yonth at the time. For he wrote the Brahmasiddhanta when he was only thirty years of age. If this indeed is his excuse, we accept it, and herewith drop the matter. As for the above-mentioned people (the Hindn theo- logians), from whom you must take care not to differ, how shonld they be sble to understand the astronomical

CHAPTER LIX. 113

theory regarding the moon's eclipsing the sun, as they, in their Purtnas, place the moon abore the sun, and that which is higher cannot cover that which is lower in the sight of those who stand lower than both. Therefore they required some being which devours moon and sun, as the fish devours the bait, and causes them to appear in those shapes in which the eclipsed parts of them in reality appear. However, in each nation there are ignorant people, and leaders still more ignorant than they themselves, who (as the Koran, Sura xxix. 12, says) " bear their own burdens and other burdens besides them," and who think they can increase the light of their minds; the fact being that the masters are as ignorant as the pupils. Very odd is that which Varahamihira relates of certain Quotations ancient writera, to whom we must pay no attention if hamthir' we do not want to oppose them, viz thst they tried to chap . 17, prognosticate the occurrence of an eclipse by pouring a 16, 63- small amouut of water together with the same amount of oil into a large vase with a flat bottom on the eighth of the lunar days. Then they examined the spots where the oil was united and dispersed. The united portion they considered as a prognostication for the be- ginning of the eclipse, the dispersed portion as a prog- nostication for its end. Further, Varahamihira saye that somebody used to think that the conjunction of the planets is the cause of the eclipse (V. 16), whilst others tried to prognosticate an eclipse from unlucky phenomena, as, e.g. the falling of stars, comets, halo, darkness, hurricane, landslip, and earthquake. "These things," so he says, " are not alwaye contemporary with an eclipse, nor are they its cause; the nature of an unlucky event is the only thing which these occurrences have in common with an eclipse. A reasonable explanation is totally different from such absurdities." The same man, knowing only too well the character FOL IL

114 ALBERUNPS INDIA. of his countrymen, who like to mir up peas with wolfs beans, pearls with dung, says, withont qnoting any anthority for his words (V. 63): " If at the time of an eelipse a violent wind blows, the next eclipse will be six months later. If a star falls down, the next eclipse will be twelve months later. If the air is dusty, it will be aighteen months latar. If there is an earthquake, it will be twenty-four months later. If the air is dark, it will be thirty months later. If hail falls, it will be thirty-six months later." To such things silence is the only proper answer. I shall not omit to montion that the different kinds of oclipees described in the canon of Alkhwarizmt, though correctly represented, do not agree with the rorulta of actual obeervation. More correct is a similar viow of the Hindus, vis. that the eclipse has the colour of amoke if it covers less than half the body of the moon ; that it is conl-black if it completely covers one half of ber; that it has a colonr between black and red if the eolipee covers more than half of her body; and, lastly, that it is yollow-brown if it cover the whole. body of the moon.

( 15 )

CHAPTER LX.

ON THE PARVAN.

THE intervals between which an eclipse may happen Pago a58. and the number of their Innations are anfficiently Esplanation demonstrated in the sixth chapter of Almagest. The pervan. Hindus call a period of time st the beginning and end of which there occur lunar eclipses, parvan. The fol- lowing information on the subject is taken from the Samhitd. Its author, Varahamihira, says: "Each six Quotation months form a parvan, in which an eclipse may happen. hamibira's from Vard-

These eclipses form a cycle of seven, each of which has chap. v. a particular dominant and prognostics, as exhibited in 19-23 the following table:

Norabor of tho 7 PArTaDA.

Doml- daiin, Indra, Kuborn, Varuns, Agni, tha Yama, Brahman. tho Pro- i the Pro Firo, also i.z. tho called tho of the the Parvand. Moon. Ruler. tector tector of tho North. of the Mitrt- Angel of Wstor. Ehya. Death

Thair prognostics, decliniog. general well-being and safoty; and this leads to famine. are fourishing. rioh people ruin their pro- perty. scholara are ilL but rain is scarco in it, and are rained. favourable to othors ; the crops olines, aod the automnal crops are flonrinhiug, and there is peatilence and mortality aro from each other, aafoty de- general woll-being and mafoty. the cattle is thriving, tho crops Rain is ucarce, the crops perinh, There is abundance and vealtb ; Not favourable to kings, but The same as in the drat Parran, There is muob water, fine crops, The kings become estranged Favourable to the ;Brahmans ;

116 ALBERUNS INDIA.

Ralen for The computation of the parvan in which you happen the compa-t tation of the to be is the following, according to the Khandakhddyaka: paroen from tho Khap -. "Write down the ahargana, as computed according to this canon, in two places. Multiply the one by 5o, and divide the product by 1296, reckoning a fraction, if it is not less than one-half, as a whole. Add to the quotient 1063. Add tbe sum to the number written in the second place, and divide the sum by 180. The quotient, as consisting of wholes, means the number of complete parvans. Divide it by 7, and the remainder under 7 which you get means the distance of the particular parvan from the first one, i.c from that of Brahman. However, the remainder under 180 which you get by the division is the elapsed part of the parvan in which Page 259- yon are. You subtract it from 180. If the remainder is less than 15, a lunar eclipse is possible or necessary ; if the remainder is larger, it is impossible. Therefore yon mnst always by a similar method compnte that time which has elapsed before the particular parvan in which you happen to be." Iu another passage of the book we find the following rule: "Take the kalpa-ahargana, i.c. the past portion of the days of a kalpa. Subtract therefrom 96,031, and write down the remainder in two different places. Subtract from the lower number 84, and divide the sum. by 561. Subtract the qnotient from the upper number and divide the remainder by 173- The quotient you disregard, bnt the remainder you divide by 7. The quo- tient gives parvans, beginning with Brahmadi" (sic). These two methods do not agree with each other. We are under the impression that in the second pas- sage something has either fallen out or been changed by the copyists. What Varahamihira says of the astrological portents Quotation of the parvans does not well suit his deep learning. from Vart- hamthira's He says: "If in a certain parvan there is no eclipse, chap. t. zgb but there is one in the other cycle, there are no rains, Sankita,

CHAPTER LX. 117

and there will be much hunger and killing." If in this passage the translator has not made & blunder, we can only say that this description applies to each parvan preceding such a one in which there occurs an eclipse. Stranger still is the following remark of his (V. 24) : "If an eclipse occurs earlier than has been calculated, there is little rain and the sword is drawn. If it occurs later than has been calculated, there will be pestilence, and death, and destruction in the corn, the fruit, and flowers. (V. 25.) This is part of what I have found in the books of the ancients and transferred to this place. If a man properly knows how to calculate, it will not happen to him in his calculations that an eclipse falls too early or too late. If the sun is eclipsed Chap. Ii. and darkened outside a parvan, you must know that an" angel called Tvashtri has eclipsed him." Similar to this is what he says in another passage: " If the turning to the north takes place before the sun Ibid. r. t. 5 enters the sign Capricornus, the south and the west will be ruined. If the turning to the routh takes place before the sun enters Cancer, the east and the north will be ruined. If the turning coincides with the sun's entering the first degrees of these two signs, or takes place after it, happiness will be common to all four sides, and bliss in them will increase." Such sentences, understood as they seem intended to be understood, sound like the ravings of a madman, bnt perhaps thera is an esoteric meaning concealed behind them which we do not know. After this we must continne to speak of the domini temporum, for these too are of a cyclical nature, adding such materials as are related to them,

( 118 )

CHAPTER LXI.

ON THE DOMINANTS OF THE DIFFERENT MEASURES OF TIME IN BOTH RELIGIOUS AND ASTRONOMICAL RELA- TIONS, AND ON CONNECTED SUBJECTS.

Whieh of the DURATION, or time in general, only applies to the measures of Creator as being his age, and not determinable by a different time bave dominanta beginning and an end. In fact, it is his eternity. and which Bot. They freqnently call it the soul, ie purusha. But as regards common time, which is determinable by mo- tion, the single parts of it apply to beings beside the Creator, and to natural phenomena beside the soul. Thus kalpa is always used in relation to Brahman, for it is his day and night, and his life is determined by it. Page 2ó0. Each manvantara has a special dominant called Manu, who is described by special qualities, already mentioned in & former chapter. On the other hand, I have never heard anything of dominants of the catur- yugas or yugas. Varâhamihira says in the Great Book of Nativities: " Abda, i.e. the year, belongs to Saturn; Ayana, half a year, to the sun; Ritu, the sixth part of a year, to Mer- cury; the month, to Jupiter; Paksha, half & month, to Venus; Vasara, the day, to Mars; Muhurta, to the moon." In the same book he defines the sixth parts of the year in the following manner: "The first, beginning with the winter solstice, belongs to Saturn; the second, to Venns; the third, to Mars; the fonrth, to the Moon the fifth, to Mercury ; the sixth, to Jupiter."

CHAPTER LXI. 119

We have already, in former chapters, described the dominants of the hours, of the muhertas, of the halves of the lunar days, of the single days in the white and black halves of the mouth, of the parvans of the eclipses, and of the single manvantaras. What there is more of the same kind we shall give in this place. In computing the dominant of the year, the Hindug Compata- use another method than the Western nations, who domtnant of tion ơf tho

compute it, according to certain well-known rules, from scoording to the year

the ascendens or horoscope of a year. The dominant of Ehddyalw. the Khanda-

the year as well as the dominant of the month are the rulers of certain periodically recurring parts of time, and are by a certain calculation derived from the domi- nants of the hours and the dominants of the days. If you want to find the dominant of the year, com- pute the sum of days of the date in question according to the rules of the canon Khandakhadyaka, which is the most universally used among them. Subtract there- from 2201, and divide the remainder by 360. Multiply the quotient by 3, and add to the product always 3- Divide the aum by 7. The remainder, a number under 7, you count off on the week-days, beginning with Sunday. The dominant of that day you come to is at the same time the dominant of the year. The remainders you get by the division are the days of his rule which have already clapsed. These, together with the days of his rule which have not yet elapsed, give the sum of 360. It is the same whether we reckon as we have just explained, or add to the here-mentioned anm of days 319, instead of subtracting from it. If you want to find the dominant of the month, sub- How to find] tract 71 from the sum of days of the date in question, nant of the the domi-

and divide the remainder by 30. Double the quotient month.

and add I. The sum divide by 7, and the remainder count off on the week-days, beginning with Sunday. The dominant of the day you come to is at the same

ALBERUNPS INDIA.

time the dominant of the month. The remaindor yon got by the divicion is that part of his rale which has already elapsed. This, together with that part of his rale which has not yet elapeed, gives the sum of 30 days. It is the same whether you reckon as we have just orplained, or add 19 to the days of the date, instead of subtracting from them, and then add 2 instead of 1 to the double of the sum. It is useless here to speak of the dominant of the day, for you find it by dividing the sum of the daye of a date by 7; or to speak of the dominant of the hour, for you find it by dividing the revolving sphero by 15. Those, however, who use the Spas raspta! divide by 15 the distance between the degree of the sun and the de- gree of the ascendens, it being measured by equal degrees. The book Sredhava of Mahddeva asys: "Each of Quatattsn the thirds of the day and night has a dominant. The fien MaM dominant of the first third of day and night is Brabman, that of the second Vishnu, and that of the third Rndra." This division is based on the order of the three prime- val forces (satva, rajas, tamas). The Hindus have still another custom, viz. that of Hon with mentioning together with the dominant of the year one the plaasta. of the Ndgas or serpents, which have certain names as they are used in connection with one or other of the planets. We have united them in the following table :-

Table of the sorpots.

Tho docatmant of The naises of the cwrpente which sooompany the Dominus Auxi, riven in twe dideront forms.

Saka (? Vasuki), Nanda. Moon. Pushkara, Piadaraki, Bharma (:), Citriagada. Man. Mereary. Takahaka.

Japitec. Cabrabasta (?). Kerkota. ElApatra, Karkotaka, Padmae'

Cakababbadra (?), MabApdma.

CHAPTER LXI.

The Hindus combine the planets with the sun be- The dewal- cause they depend upon the sun, and the fixed stars plenota so- ninta of the

with the moon because the stars of her stations belong Fitys- cording to

to them. It is known among Hindu as well as Muslim dharns.

astrologers that the planets exercise the rule over the zodiacal signs. Therefore they assume certain angelio beings as the dominants of the planets, who are ex- hibited in the following tahle, taken from the Vishņu- dharma :-

Table of the dominanta of the planeta.

The planets and the two boder. Thelr docinanta.

Sun. Agni. Moon. Vyana (?). Mars Kalmdsha (!). Mercury. Viahuu Jupitor. Śukri Vonus. Gaurt Saturn. Prajapati. The Head. Ganapati (!). The Tail. Viśvakarman.

The same book attribntes also to the lunar stations The dorsl- as to the planets certain dominants, who are contained luner nanta of the

in the following table :- stations.

The Lunar Stations Their domiaants: Page 36a.

KrittikA Agni. Rohint. Keśvara. Mrigaátraba. Inda, i.c. the moon. Ardri Rudra. Punarvasu. Aditi. Puahya, Gura, ic. Jupiter. Sarpts. MaghA. Pitares Parraphalgunt Bhaga. Uttaraphalgunt. Aryaman Hasts. Savitri, i.e. Savith. CitrA Trashtri. Svatt Vayu. ViskhA Indragnt.

I22 ALBERUNIS INDIA.

Anpridht. Mitra. Bakra. Mals Nimid. Apna. Uttartebidbs Abbijit Brahman. Bravana. Vishnu. Dhanisht. Vasavas. Satabhishaj. Parvabhidrapads [Aja okapld] Uttarabhadrapadt Ahir badhnys Revatt, Pashan. Atrint Atvin (!). Bharaat. Yama

( 123 )

CHAPTER LXII.

ON THE SIXTY YEARS-SAMVATSARA, ALSO CALLED Pago 263. "SHASHTYABDA,"

THE word samvatsara, which means the years, is a tech- Erplana- nical term for cycles of years constructed on the basis torms sam- Hion of the of the revolutions of Jupiter and the sun, the heliacal mharktyobda. ratsara and rising of the former being reckoned as the beginning. It revolves in sixty years, and is therefore called shash- fyabda, ie. sixty years. We have already mentioned that the names of the A reer la Innar stations are, by the names of the months, divided over by that presided into groups, each month having a namesake in the cor- which tho month in

responding group of stations. We have represented rising of heliacal

these things in a table, in order to facilitate the snbject oocura. Jupater (v. i 218). Knowing the station in which the heliacal rising of Jnpiter occurs, and looking up this station in the just-mentioned table, you find at the left of it the name of the month which rules over the year in question. Yon bring the year in connection with the month, and say, e.g. the year of Caitra, the year of Vai- sdkha, &c. For each of these years there exist astro- logical rules which are well known in their literature. For the computation of the lunar station in which How to find the heliacal rising of Jupiter occurs, Varahamihira statlon of the lunar

gives the following rule in his Samhitd :- Jupitor'e hellacal ria- " Take the Sakakala, multiply it by I1, and multiply tlon from ing. Quota-

the product by 4 Yon may do this, or you may also hira's Sam- VarAhami-

multiply the Sakakala by 44 Add 8589 to the product vi. so, st.

124 ALBERUNPS INDIA.

and divide the sum by 3750. The quotient represents yeara, months, days, &c. " Add them to the Sakakala, and divide the sum by 60. The quotient represents great sexagenarian yugas, ia complete shashtyabdas, which, as not being necessary, are disregarded. Divide the remainder by 5, and the quotient represents small, complete five-year yugas. That which remains being less than one yuga, is called samvatsara, i.e. the year. "V. 22 .- Write down the latter nomber in two diffe- rent places. Multiply the one by 9, and add to the pro- duct i of the number in the other place. Take of the sum the fourth part, and this number represents com- plete lunar stationa, ita fractions representing part of the next following current station. Count off this number of the stations, beginning with Dhanishth&. The station you arrive at is that one in which the heliacal rising of Jupiter takes place." Thereby you know the month of the years, as has above been ex- plained. The great yugas begin with the heliacal rising of Cootnined la Jupiter in the beginning of the station Dhanishth& and tha cyala of uxty yun. the beginning of the month Magha. The small yugas have within the great ones a certain order, being divided into groups which comprehend certain numbers . of years, and each of which has a special dominant. This division is represented by the following table. If you know what number in the great yuga the year. in question occupies, and you look up this number among the numbers of the year in the upper part of the table, you find under it, in the corresponding columns, both the name of the yesr and the name of ita dominant.

Nucabers Numbers Nambers Num bars Numbon Mumbers Numbers Nucsbers Numhm with the Vith the with the with the with tbe vith the with the with the with the without a antt I. anit 6 unit 7. uDit 3 ants 8. untt 4. unit o. unit 5.

6 8 $7 13 18 4 9 5 10 11 16 14 19 30 56 22 27 28 24 25 30 cyele

31 3ª 35 CHAPTER LXII. 36 37 33 38 39 40 sixty-yeara

41 46 42 47 34 43 44 49 45 șo The namber of each yesr of the 56 52 57 53 58 54 59

The names which cach dozen of years Samvatsara. Parivatsara. Idavatsara. Anuvatsara. Udvatsara. has in common,

Their Śitamayâkhamalin, Sailasutapat, ie. Agni, Arkı, Prajapati, the the hasband of the dominanta. i.e. the fire, i.c. i.e. having a cold father of the lunar stations. danghter of tho the sun. my, viz. the moon mountain, viz Mahadeva

125

126 ALBERUNTS INDIA.

Further, every single one of the sixty years has a name of tho rtd of ita own, and the yugas, too, have names which are the names of their dominants. All these names are exhi- bited in the following table This table is to be used in the same way as the preceding one, as yon find the name of each year of the whole cycle (of sixty years) under the corre- sponding number. It would be a lengthy affair if we were to explain the meanings of the single names and their prognostics. All this is found in the book Samhita,

L-Lustrom. J. 2 3- 5. Favoorable. Its lord is Manu, i.c. Nart- Prabhave. Vibhava, Bukla. Pramoda. Prajapati yana . . . .

IJ .- Lustram. 6. 7- 8 9. 10. Favourable. Ita lord Surejya, i ... Juplter Asglras. Śrtmukha. Bhava. Yuvan. Dhatri.

.' CHAPTER LXII. III .- Lustrum. II. 12. 14 I5. Favonrabla. Ita lord. Balabhit, i.e .? 13.

Indra Lévara. BabudbAnys. Pramsthin. Vikrama. Viaba (Vrishabba ?)

IV .- Lustrum. 16. 17 18 19 20 Favourable. Ita lord Huthis, i.c. the fre CitrabhAnu. Sabhtnn. PArthiva (1). Vyaya.

V .- Lustrnm 21. 22. 23 24 25. Indifferent. Its lord Tvashtri, the lord } of the lunar station Citra . Sarvajit SarvadhArin. Virodbin. Vikrita. Khara.

VI .- Lustrum. 26. 27 28 29 30. Indifferent. Its lord Proshthapada, the ) lord of the lunar station Uttarabha- Nandana. Vijaya. Jaya Manmaths Cadur (!). drapadå 137.

128 VIL-Lastrum. Iudifferent. Ita lord Pitaras, i.e. the} 31 33 33 34- 35. fathors Hemalambe. Vilambin. VikArin. Šarvart (?). Plevt. . .

VIII .- Lastrum. Indifferent. Its lord is siva, io. the ) 36. 37. 38. Šokakrit 39. 40. Bubhakrit. creatures Krodbin. Viivivasu, Paravasn, . ALBERUNTS INDIA. IX .- Lustram. 43. 42. 43- 44. 45 Unlncky. Its lord Soma, Le. the moon Plavaigt. Kilaka Sanmys. Sadbâraņt. Rodhakrit

X .- Lnstrum. Unlacky. Its lord Bakranals, i.e. Indra 46. 47 48. 49. 50. and the fire together . Paridhtrin. Pramadin, Vikrama. RAkehasa. Anala .

XL .- Lustrum. Unlucky. Its lord Asviu, the lord of 51. 52 53- 55- the lanar station Asvint PiAgala. Klayukta SiddhArtha. Randra Darmati,

XII,-Lustram. . Unlucky. Ita lord Bhaga, the lord of 56 57. 58. 59 60.

the lunar statiou Porvaphalgunt Dundubhi. Angara. Raktakaha (7) Krodha. Kshaya.

CHAPTER LXII. I29

This is the method for the determination of the Page a67. years of the shashtyabda, as recorded in their books. However, I have seen Hindus who subtract 3 from the era of Vikramaditya, and divide the remainder by 60. The remainder they count off from the begin- ning of the great yuga. This method is not worth anything. By-the-bye, it is the same whether you reckon in the manuer mentioned, or add 12 to the Śakakâla. I have come across some people from the country The samral- of Kanoj who told me that, with them, the cycle of peoplo of taras of the

samvatsaras has 1248 years, each single one of the Kanol.

twelve samvatsaras heving 104 years. According to this statement we must subtract 554 from the Sakakala, and with the remainder compare the following diagram. In the corresponding colnmn you see in which samvat- sara the year in qrestion lies, and how many years of the sampatsara have already elapsed :--

The year I. 105 209. 313. 417. 52I.

Their Rokmåksbs. Plnmant. Kadara. KAlavriDta. Naumand. Merd. (:)

The years 625 729. 833. 937. 104I. II45.

Their Barbara. Jamba. Kriti. Barps. Hindba. Sindbu.

When I heard, among these pretended names of sam- vatsaras, names of nations, trees, and mountains, I con- ceived a suspicion of my reporters, more particularly as their chief business was indeed to practise hocus- pocus and deception (as jugglers ?); and a dyed beard proves its bcarer to be a liar. I used great care in examining every single one of them, in repeating the same questions at different times, in a different order and context. But lo ! what different answers did I get! God is all-wise !

FOL 1I.

( 130 )

CHAPTER LXIII.

ON THAT WHICH ESPECIALLY CONCERNS THE BRAHMANS, AND WHAT THEY ARE OBLIGED TO DO DURING THEIR WHOLE LIFE.

First period THE life of the Brahman, after seven years of it have in the Brab- mab's life. passed, is divided into four parts. The first part begins with the eighth year, when the Brabmans come to him to instruct him, to teach him his dnties, and to enjoin him to adhere to them and to embrace them es long as he lives. Then they bind a girdle round his waist and invest him with a pair of yajnoparitas, ia one strong cord consisting of nine single cords which are twisted together, and with a third yajnopavtta, s single one made from cloth. This girdle runs from the left shoulder to the right hip. Further, he is presented with a stick which he has to wear, and with a seal- ring of a certain grass, called darbha, which he wears on the ring-finger of the right hand. This seal-ring is also called pavitra. The object of his wearing the ring on the ring-finger of his right hand is this, that it should be a good omen and a blessing for all those who receive gifts from that hand. The obligation of wearing the ring is not quite so stringent as that of wearing the yajnopavita, for from the latter he is not to separate himself under any circumstances whatever. If he takes it off while eating or fulfilling some want of nature, he thereby commits a sin which cannot be Prgt s68. wiped off save by some work of expiation, fasting, or almsgiving.

CHAPTER LXIII. 131

This first period of the Brahman'a life extends till the twenty-fifth year of his age, or, according to the Vishnu- Purdna, till his forty-eighth year. His dnty is to prac- tise abstinence, to make the earth his bed, to begin with the learning of the Veda and of its explanation, of the science of theology and law, all this being tanght to him by a master whom he serves day and night. He washes himself thrice a day, and performs a sacrifice to the fire both at the beginning and end of the day. After the sacri- fice he worships his master. He fasts a day and he breaks fast a day, but he is never allowed to eat meat. He dwells in the house of the master, which he only leaves in order to ask for a gift and to beg in not more than five houses once a day, either at noon or in the evening. Whatever alms he receives he places before his master to choose from it what he likes. Then the master allows him to take the remainder. Thus the pupil nourishes himself from the remains of the dishes of his master. Further, he fetches the wood for the fire, wood of two kinds of trees, palasa (Butea frondosa) and darbha, in order to perform the sacrifice; for the Hindus highly venerate the fire, and offer flowers to it. It is the same case with all other nations. They always thought that the sacrifice was accepted by the deity if the fire came down upon it, and no other worship has been able to draw them away from it, neither the worship of idols nor that of stars, cows, asses, or images. Therefore Bashshar Ibn Burd says: " Since there is fire, it is worshipped." The second period of their life extendsfrom the twenty- Second fifth year till the fiftieth, or, according to the Vishnu-Pur- Brabman's period in the

ana, till the seventieth, The master allows him to marry. lifo.

He marries, establishes a household, and intends to have descendants, but he cohabits with his wife only once in a month after she has become clean of the menstruation. He is not allowed to marry a woman above twelve years of age. He gains his sustenance either by the fee he

132 ALBERUNPS INDIA.

obtaina for teaching Brahmans end Kshatriyas, not as a payment, bnt as a present, or by presents which he receives from some one because he performs for him the sacrifices to the fire, or by asking a gift from the kings and nobles, there being no importunate pressing on his part, and no nnwillingness on the part of the giver. There is always a Brahman in the honses of those people, who there administers the affairs of reli- gion and the works of piety. He is called purohita. Lastly, the Brahman lives from what he gathers on the earth or from the trees. He may try his fortune in the trace of clothes and betel-nnts, but it is preferable that he should not trade himself, and that a Vaisya should do the business for him, because originally trade is for- bidden on account of the deceiving and lying which are mixed up with it. Trading is permitted to him only in case of dire necessity, when he has no other means of sustenance. .The Brahmans are not, like the other castes, bound to pay taxes and to perform services to the kings. Further, he is not allowed continuslly to busy himself with horses and cows, with the care for the cattle, nor with gaining by nsury. The blue colour Paga s6g. is impure for him, so that if it touches his body, he is obliged to wash himself. Lastly, he must always beat the drum before the fire, and recite for it the prescribed holy texts. The third The third period of the life of the Brahman extends poriod. from the fiftieth year to the seventy-fifth, or, according to the Vishnu-Purdna, till the ninetieth. He practises abstinence, leaves his household, and hands it as well as his wife over to his children, if the latter does not prefer to accompany him into the life in the wilderness. He dwells ontside civilisation, and leads the same life again which he led in the first period. He does not take ahelter under a roof, nor wear any other dress but some bark of a tree, simply sufficient to cover his loins. He sleeps on the earth without any bed, and only

CHAPTER LXIII. 133 nourishes himself by fruit, vegetables, and roots. He lets the hair grow long, and does not anoint himself with oil. The fourth period extends till the end of life. He Tho fourth wears a red garment and holds a stick in his hand Perlod. He is always given to meditation; he strips the mind of friendship and enmity, and roots out desire, and lust, and wrath. He does not converse with anybody at all. When walking to a place of a particular merit, in order to gain a heavenly reward, he does not etop on the road in a village longer than a day, nor in a city longer than five days. If any one gives him something, he does not leave a remainder of it for the following day. He has no other business but that of caring for the path which leads to salvation, and for reaching moksha, whence there is no return to this world. The universal dnties of the Brahman throughout his The dutles whole life are works of piety, giving alms and receiving in genoral. of Brahmana

them. For that which the Brahmans give reverts to the pitaras (is in reality a benefit to the Fathers). He must continually read, perform the sacrifices, take care of the fire which he lights, offer before it, worship it, and preserve it from being extinguished, that he may be burned by it after his death. It is called homa. Every day be must wash himself thrice: at the sarndhi of rising, i.e, morning dawn, at the sardhi of setting, i.e. evening twilight, and between them in the middle of the day. The first washing is on account of sleep, because the openings of the body have become lax during it. Washing is a cleansing from accidental impurity and a preparation for prayer. Their prayer consists of praise, glorification, and pros- tration according to their peculiar manner, viz. pros- trating themselves on the two thumbs, whilst the two palms of the hands are joined, and they turn their faces towards the eun. For the sun is their kibla, wherever he may be, except when in the south. For they do not

134 ALBERUNTS INDIA.

perform any work of piety with the face turned south- ward; only when occupied with something evil and unlucky they torn themselves towards the sonth. The time when the aun declines from the meridian (the afternoon) is well suited for acquiring in it a heavenly reward. Therefore at this time the Brahman must be clean. The evening is the time of supper and of prayer. The Brahman may take his supper and pray without having previously washed himself. Therefore, evidently, the rule as to the third washing is not as stringent as that relating to the first and second washings. A nightly washing is obligatory for the Brahman only at the times of eclipses, that he should be pre- pared to perform the rales and sacrifices prescribed for that occasion.' The Brahman, as long as he lives, eats only twice a day, at noon and at nightfall; and when he wants to take his meal, he begins by putting aside as mnch as Page 270 is enfiicient for one or two men as alms, especially for strange Brahmans who happen to come at evening- time asking for something. To neglect their mainten- ance would be a great sin. Further, he puts something aside for the cattle, the birds, and the fire. Over the remainder he says prayers and eats it. The remainder of his dish he places ontside his house, and does not any more come near it, as it is no longer allowable for him, being destined for the chance passer-by who wants it, be he a man, bird, dog, or something else. The Brahman must have a water-vessel for himself. If another one uses it, it is broken. The same remark applies to his eating-instruments. I have seen Brah- mans who allowed their relatives to eat with them from the same plate, but most of them disapprove of this. He is obliged to dwell between the river Sindh in the north and the river Carmanvati in the sonth. He is not allowed to cross either of these frontiers so as

CHAPTER LXIII. 135

to enter the country of the Turks or of the Karnata. Further, he must live between the ocean in the east and west. People say that he is not allowed to stay in a conntry in which the grass which he wears on the ring-finger does not grow, nor the black-haired gazelles graze. This is a description for the whole country within the just-mentioned boundaries. If he passes beyond them he commits a sin, In a country where not the whole spot in the house which is prepared for people to eat upon it is plastered with clay, where they, on the contrary, prepare a sepa- rate tablecloth for each person eating by pouring water over a spot and plastering it with the dung of cows, the shape of the Brahman's tablecloth must be square, Those who have the custom of preparing such table- claths give the following as the cause of this cnstom: -The spot of eating is soiled by the eating. If the eating is finished, the spot is washed and plastered to become clean again. If, now, the soiled spot is not distinguished by a separate mark, you would suppose also the other spots to be soiled, since they are similar to and cannot be distinguished from each other. Five vegetables are forbidden to them by the reli- gious code :- Onions, garlic, a kind of gourd, the root of a plant like the carrots called krnen (?), and another vegetable which grows round their tanks called nalt.

( 136 )

CHAPTER LXIV.

ON THE RITES AND CUSTOMS WHICH THE OTHER CASTES, BESIDES THE BRAHMANS, PRACTISE DURING THEIB LIFETIME.

Duties of THE Kshatriya reads the Veda and learns it, but does the aingle not teach it. He offere to the fire and acts according to the rules of the Puranas. In places where, as we have mentioned (v. p. 135), a tablecloth is prepared for eating, he makes it angular. He rules the people and defends them, for he is created for this task. He girds himeelf with a single cord of the threefold yajno- pavita, and a single other cord of cotton. This takes place after he has finished the twelfth year of his life. It is the duty of the Vaisya to practise agriculture and to cultivate the land, to tend the cattle and to remove the needs of the Brahmans. He is only allowed to gird himself with a single yajnopavita, which is made of two cords. The Stdra is like a servant to the Brahman, taking care of his affairs and serving him. If, thongh being poor in the extreme, he still desires not to be without a yajnoparita, he girds himself only with the linen one. Every action which is considered as the privilege of a Page 271. Brahman, such as saying prayers, the recitation of the Veda, and offering sacrifices to the fire, is forbidden to him, to snch a degree that when, e.g. a Sûdra or a Vaiya is proved to have recited the Veda, he is accused by the Brahmans before the ruler, and the latter will order his tongue to be ent off. However, the meditation on God,

CHAPTER LXIV. 137 works of piety, and almsgiving are not forbidden to him. Every man who takes to some occupation which is not allowed to his caste, as, e.g. a Brahman to trade, a Śudra to agriculture, commits s sin or crime, which they consider only a little less than the crime of theft. The following is one of the traditions of the Hindns: -In the days of King Rama human life was very long, Story of always of a well-defined and well-known length. Thus the Canddla ' King Hama.

a child never died before its fether. Then, however, Brahmen. end the it happened that the son of a Brahman died while the father was still alivs. Now the Brahman brought his child to the door of the king and spoke to him: "This innovation has sprung up in thy days for no other reason but this, that there is something rotten in the state of the country, and because a certain Vazir com- mits in thy realm what he commits." Then Rama began to inquire into the cause of this, and finally they pointed ont to him & Candala who took the greatest pains in performing worship and in self-torment. The king rods to him and found him on the banks of the Ganges, hanging on something with his head down- ward. The king bent his bow, shot at him, and pierced his bowels. Then hs spoke: "That is it! I kill thee on account of a good action which thou art not allowed to do." When he returned home, he fonnd the son of the Brahman, who had been deposited before his door, alive. All other men except the Candala, as far as they are not Hindus, are called mleccha, i.e. unclean, all thoss who kill men and slaughter animals and eat the flesh of cows. All these things originate in the difference of the Philosophic classes or castes, one set of people treating the others sbout all opinion

as fools. This apart, all men are equal to each other, equri things being as Vasudeva says regarding him who seeks salvation: "In the judgment of the intelligent man, the Brahman

138 ALBERUNIS INDIA.

and the Candala are equal, the friend and the foe, the faithful and the deceitful, nay, even the serpent. and the weasel. If to the eyes of intelligence all things are equal, to ignorance they appear as separated and different." Vasudeva speaks to Arjuna: "If the civilisation of the world is that which is intended, and if the direc- tion of it cannot proceed without our fighting for the purpose of snppressing evil, it is the dnty of us who are the intelligent to act and to fight, not in order to bring to an end that which is deficient within us, but because it is necessary for the purpose of healing what is ill and banishing destructive elements. Then the ignorant imitate us in acting, as the children imitate their elders, without their knowing the real aim and purport of actions. For their nature has an aversion to intellectual methods, and they use force only in order to act in accordance with the influences of lnst and passion on their senses. In all this, the intelligent and educated man is directly the contrary of them."

( 139 )

CHAPTER LXV.

ON THE SACRIFICES.

MOsT of the Veda treats of the sacrifices to the fire, and describes each one of them. They are different in extent, so that certain of them can only be performed by the greatest of their kings. So, e.g. the asvamedha. Afvamedha. A mare is let freely to wander about in the country grazing, without enybody's hindering her. Soldiers follow her, drive her, and cry out before her: "She is the king of the world. He who does not agree, let him come forward." The Brahmans walk behind her and perform sacrifices to the fire where she casts dung. When she thus has wandered about through all parts Paga 272. of the world, she becomes food for the Brahmans and for him whose property she is. Fnrther, the sacrifices differ in duration, so that only he could perform certain of them who lives a very long life; and such long lives do no longer occur in this our age. Therefore most of them have been abolished, and only few of them remain and are practised now- adays. According to the Hindus, the fire eats everything. On bro- Therefore it becomes defiled, if anything unclean is generl mixed np with it, as, e.g. water. Accordingly they are very punctilions regarding fire and water if they are in the hands of non-Hindus, becanse they are defiled by being tonched by them. That which the fire eats for its ahare, reverts to the Devas, because the fire comes ont of their months

140 ALBERUNPS INDIA.

What the Brahmans present to the fire to eat is oil and different cereals -- wheat, barley, and rice -which they throw into the fire Further, they recite the proscribed texta of the Veda in case they offer on their own behalf. However, if they offer in the name of somebody else, they do not recite anything. The Visinu-Dharma mentions the following tradi- tion :- Once upon a time there was a man of the class of the Daityas, powerful and brave, the ruier of a wide realm called Hiranyaksha, He had a danghter of the name of Dktsh (7), who was always bent upon worship and trying herself by fasting and abstinenca. Thereby ahe had earned as reward a place in heaven. She was married to Mabadeva, When he, then, was alone with her and did with her according to the custom of the Devas, ie. cohabiting very long and transferring the semen very slowly, the fire became aware of it and be- came jealous, fearing lest the two might procreate a fire similar to themselves. Therefore it determined to defile and to ruin them. When Maladeva saw the fire, his forehead became covered with aweat from the violence of his wrath, so that some of it dropped down to the earth. The earth drank it, and became in consequence pregnant with Mars, ie. Skanda, the commander of the army of the Devas. Rudra, the destroyer, seized a drop of the semen of Mahadeva and threw it away. It was scattered in the interior of the earth, and represents all atom-like sob- stances (?). The fire, however, became leprons, and felt so much ashamed and confounded that it plunged down into patdla, i.e. the lowest earth. As, now, the Devas missed the fire, they went ont to search for it. First, the frogs pointed it out to them, The fire, on seeing the Devas, left its place and concealed itself in the tree afvattha, laying a cume on the frogs, that they

CHAPTER LXV. 14I

should have a horrid croaking and be odious to all others. Next, the parrots betrayed to the Devas the hiding- place of the fire Thereupon the fire cursed them, that their tongues should be turned topsy-turvy, that their root should be where its tip ought to be. But the Devas spoke to them: " If your tongue is turned topsy- tarvy, you shall speak in human dwellings and eat delicate things." The fire fled from the asvattha tree to the tree damf. Therenpon the elephant gave & hint to the Devas re- garding its hiding-place. Now it cursed the elephant thst his tongue should be turned topsy-turvy. But then the Devas spoke to him: " If your tongue is turned topsy-turvy, you shall participate with man in his victuals and understand his speech." At last they hit upon the fire, but the fire refused to stay with them because it was leprous. Now the Devas restored it to health, and freed it from the leprosy. The Devas brought back to them the fire with all honour snd made it a mediator between them- selves and mankind, receiving from the latter the shares which they offer to the Devas, and making these shares reach them.

( 142 )

CHAPTER LXVI.

ON PILGRIMAGE AND THE VISITING OF SACRED PLACES.

Page 273 PILGRIVAGES are not obligatory to the Hindus, but facultative and meritorious. A man sets off to wander to some holy region, to some much venerated idol or to some of the holy rivers. He worships in them, worships the idol, makes presents to it, recites many hymns and prayers, fasts, and gives alins to the Brahmans, the priests, and others. He shaves the hair of his head and beard, and returns home. The holy much venerated ponds are in the cold mountains round Meru, The following information regarding them is found in both the Vayu and the Matsya Puranas :- An extract " At the foot of Mera there is Arhata (?), a very great pouda frota pond, described as shining like the moon. In it origi- and Matya nates the river Zanba (? Jambn), which is very pure, the Fays Pardpas flowing over the purest gold. "Near the mountain Sveta there is the pond Uttara- manasa, and around it twelve other ponds, each of them like a lake. Thence come the two rivers Sandi (?) and Maddhyanda (?), which flow to Kimpurusha. "Near the monntain Nila there is the pond pyvd (pitanda ?) adorned with lotuses. " Near the mountain Nishadha there is the pond Vish- nnpada, whence comes the river Sarasvatt, i.e., Sarsuti. Besides, the river Gandharvt comes from there. " In the mountain Kailasa there is the pond Manda, as large as a sea, whence comes the river Mandakini,

CHAPTER LXVI. 143

" North-east of Kailasa there is the monntain Can- draparvata, and at its foot the pond Acud (?), whence comes the river Acud. "Sonth-east of Kailasa there is the mountain Lohita, and at its foot a pond called Lohita Thence comes the river Lohitanadi "South of Kailasa there is the monntain Saraynsatt (?), and at its foot the pond Manasa. Thenca comes the river Sarayû. "West of Kailasa there is the mountain Aruna, always covered with anow, which cannot be ascended. At its foot is the pond Sailoda, whence comes the river Śailôdâ. "North of Kailasa there ia the mountain Gaura (?), and at its foot the pond C-n-d-sara(?), i.e. having golden sand, Near this pond the King Bhagiratha led his anchorite life. "His story is as follows :- A king of the Hindus Story of called Sagara had 60,000 sons, all of them bad, mean Bhagtratha

fellowe. Once they happened to lose a horse. They at once searched for it, and in searching they continu- ally ran abont so violently that in consequence the surface of the earth broke in. They found the horse in the interior of the earth standing before a man who was looking down with deep-sunken eyes. When they came near him he amote them with his look, in conseqnence of which they were burned on the spot and went to hell on account of their wicked actions. "The collapsed part of the earth became a sea, the great ocean. A king of the descendants of that king, called Bhagiratha, on hearing the history of his ances- tors, was much affected thereby. He went to the above-mentioned pond, the bottom of which was polished gold, and stayed there, fasting all day and Page 284. worshipping during the nights. Finally, Mahadeva asked him what he wanted; npon which he answered,

144 ALBERUNI'S INDIA.

'I want the river Ganges which fows in Paradise,' knowing that to any one over whom its water flows all his sins are pardoned. Mahadeva granted him his desire. However, the Milky Way was the bed of the Ganges, and the Ganges was very hanghty. for nobody had ever been able to stand against it. Now Mahadeva took the Ganges and put it on his hesd. When the Ganges could not move away, he became very angry and made a great uproar. How- ever, Mahâdeva held him firmly, so that it was not possible for anybody to plunge into it. Then he took part of the Ganges and gave it to Bhagiratha, and this king made the middle one of its seven branches flow over the bones of his ancestors, whereby they became liberated from punishment. Therefore the Hindus throw the burned bones of their dead into the Ganges. The Ganges was also called by the name of that king who brought him to earth, i.a Bhagtratha." On the onn- We have already quoted Hindn traditions to the boly poods. eflect that in the Dvipas there are rivers as holy as the struetion of Ganges. In every place to which some particular holi- ness is ascribed, the Hindus construct ponds intended for the ablutions. In this they have attained to a very high degree of art, so that our people (the Muslims), when they see them, wonder at them, and are unable to describe them, much less to construct anything like them. They build them of great stones of an enor- mous bulk, joined to each other by sharp and strong cramp-irons, in the form of steps (or terraces) like so many ledges; and these terraces run all around the pond, reaching to a height of more than a man's stature. On the surface of the stones between two terraces they construct staircases rising like pinnacles. Thus the first steps or terraces are like roads (leading round the pond), and the pinnacles are steps (leading up and down). If ever so many people descend to the pond whilst others ascend, they do not meet each other, and

CHAPTER LXVI. 145

the road is never blocked np, because there are so many terraces, and the ascending person can always turn aside to another terrace than that on which the descend- ing people go. By this arrangement all troublesome thronging is avoided. In Multan there is a pond in which the Hindus On single worship by bathing themselves, if they are not pre- boly ponds. vented. The Samhitd of Varahamihira relates that in Tane- shar there is a pond which the Hindus visit from afar to bathe in its water. Regarding the cause of this custom they relate the following :- The waters of all the other holy ponds visit this particular pond at the time of an eclipse. Therefore, if a man washes in it, it is as if he had washed in every single one of all of them. Then Varahamihira continues: "People say, if it were not the head (apsis) which causes the eclipse of sun and moon, the other ponds would not visit this pond." The ponds become particularly famous for holiness either because some important event has happened at them, or because there is some passage in the holy text cr tradition which refars to them. We have already quoted words spoken by Saunaka. Venus had related them to him on the authority of Brahman, to whom they had originally been addressed. In this text King Bali also is mentioned, and what he would do till the time when Narayana would plunge him down to the lowest earth. In the same text occurs the follow- ing passage :- " I do that to him only for this purpose On the in- that the equality between men, which he desires to created bo- equality of

realise, ahall be done away with, that men shall be origin of . ings and the different in their conditions of life, and that on this A tradition patriotism. difference the order of the world is to be based; further, from Sau. that people shall turn away from his worship and Page 275. worship me and believe in me. The mutual assistanca of civilised people presupposes a certain difference VOL IL K

146 ALBERUNI'S INDIA.

among them, in consequence of which the one requires the other. According to the same principle, God has created the world as containing many differences in itself. So the single countries differ from each other, one being cold, the other warm; one having good soil, water, and air, the other having bitter salt soil, dirty and bad smellirg water, and unhealthy air. There are still more differences of this kind; in some cases advantages of all kinds being numerous, in others few. In some parts there are periodically return- ing physical disasters; in others they are entirely unknown. All these things indnce civilised people carefully to select the places where they want to build towns. That which makes people do these things is ugage and custom. However, religious commands are mncl more powerful, and inflnence mnch more the nature of man than nsages and customs. The bases of the latter are investigated, explored, and accordingly either kept or abandoned, whilst the bases of the religions com- mands are left as they are, not inquired into, adhered to by the majority simply on trust. They do not argne over them, as the inhabitants of some sterile region do not argue over it, since they are born in it and do not know anything else, for they love the conntry as their fatherland, and find it difficult to leave it. If, now, besides physical differences, the countries differ from each other also in law and religioo, there is so much attachment to it in the hearts of those who live in them that it can never be rooted out." On Bonarca The Hindus bave some places which are venerated skylum. for reasons connected with their law and religion, e.g. Benares (Baranasi). For their anchorites wander to it and atay there for ever, as the dwellers of the Ka'ba stay for ever in Mekka. They want to live there to the end of their lives, that their reward after death should be the better for it. They say that a murderer

CHAPTER LXVI. 147

ie held responsible for his crime and punished with a punishment due to his guilt, except in case he enters the city of Benares, where he ohtains pardon. Regard- ing the cause of the holiness of this asylum they relate the following story :- "Brahman was in ehape four-headed. Now there happened some quarrel between him and Samkara, i.c. Mahadeva, and the succeeding fight had this result, that one of the heads of Brahman was torn off. At that time it was the custom that the victor took the head of the slain adversary in his hand and let it hang down from his hand as an act of ignominy to the dead and as a sign of his own bravery. Further, a bridle was put into the mouth (?). Thus the head of Brahman was dishonoured by the hand of Mahadeva, who took it always with him wherever he went and whatever he did. He never once separated himself from it when he entered the towns, till at last he came to Benares. After he had entered Benares the head dropped from his hand and disappeared." A similar place is Pukara, the story of which is this: On the holy Brahman once was occupied in offering there to the Fakara, fire, when a pig came out of the fire. Therefore they Mahtr, Taneshar,

represent his image there as thst of a pig. Outside and Multan. Kashmir,

the town, in three places, they have constructed ponds which stand in high veneration, and are places of worship. Another place of the kind is Taneshar, also called Kurukshetra, i.c. the land of Kuru, who was s peasant, a pious, holy man, who worked miracles by divine power. Therefore the country was called after him, and venerated for his sake. Besides, Taneshar is the theatre of the exploits of Vasudeva in the wars of Bharata and of the destraction of the evil-doers. It is for this reason that people visit the place. Mahura, too, is a holy place, crowded with Brahmans.

148 ALBERUNTS INDIA.

Page 276. It is venerated because Vasudeva was there born and brought up, in a place in the neighbourhood called Nandagola. Nowadays the Hindus also visit Kashmtr. Lastly, they used to visit Multan before its idol-temple was destroyed.

( 149 )

CHAPTER LXVII.

ON ALMS, AND HOW A MAN MUST SPEND WHAT HE EARNS.

IT is obligatory with them every day to give alms as mnch as possible. They do not let money become a year or even a month old, for this would be a draft on an unknown future, of which a man does not know whether he reaches it or not. With regard to that which he earna by the crops or from the cattle, he is bound first to pay to the ruler of the conntry the tax which attaches to the soil or the pasture-ground. Further, he pays bim one-sixth of the income in recognition of the protection which he affords to the subjects, their property, and their families. The same obligation rests also on the common people, but they will always lie and cheat in the declarations about their property. Further, trading businesses, too, pay a tribute for the same reason. Only the Brahmans are exempt from all these taxes. As to the wey in which the remainder of the income, after the taxes have been deducted, is to be employed, there are different opinions. Some destine one-ninth of it for alms. For they divide it into three parts. One of them is kept in reserve to guarantee the heart against anxiety. The second is spent on trade to bring profit, and one-third of the third portion (i.e. one-ninth of the whole) is spent on alms, whilst the two other thirds are spent according to the same rule. Others divide this income into four portions. One-

150 ALBERUNI'S INDIA.

fourth is destined for common expenses, the second for liberal works of a noble mind, the third for alms, and the fourth for being kept in reserve, ia not more of it than the common expenses for three years. If the quarter which is to be reserved exceeds this amount, only this amount is reserved, whilst the remainder is spent as aims. Usury or taking percentages is forbidden. The sin which a man commits thereby corresponds to the amount by which the percentages have increased the capital stock. Only to the Sudra is it allowed to take percentages, as long as his profit is not more than one- fiftieth of the capital (ic, he is not to take more than two per cent.).

( 15' )

CHAPTER LXVIII.

ON WHAT IS ALLOWED AND FORBIDDEN IN EATING AND DRINKING.

ORIGINALLY killing in genaral was forbidden to them, as it is to the Christians and Manichæans. People, however, have the desire for meat, and will always fling aside every order to the contrary. Therefore the herc-mentioned law applies in particular only to the Brahmans, because they are the guardians of the reli- gion, and because it forbids them to give way to their lusts. The same rule applies to those members of the Christian clergy who are in rank above the bishops, viz. the metropolitans, the catholici, and the patriarcbs, not to the lower grades, euch as presbyter and deacon, except in the case that a man who holds one of these degrees is at the same time a monk. As matters stand thus, it is allowed to kill animals by List of anf- means of strangulation, but only certain animals, others and unlaw. mala lawful

being excluded. The meat of such animals, the killing of eaten. ful to be

which is allowed, is forbidden in case they die a sudden death. Animals the killing of which is allowed are sheep, goats, gazelles, hares, rhinoceroses (gandha), the buffaloes, fish, water and land birds, as sparrows, ring- doves, francolins, doves, peacocks, and other animals Page 277. which are not loathsome to man nor noxions. That whioh is forbidden are cows, horses, mules, asses, camels, elephants, tame poultry, crows, parrots, nightingales, all kinds of eggs and wine. The latter is

152 ALBERUNIS INDIA.

allowed to the Sudra. He may drink it, but dare not sell it, as he is not allowed to sell mest. . Why the Some Hindus say that in the time before Bharata it mot of cowa was forbid- was allowed to eat the meat of cowa, and that there don. then existed sacrifices part of which was the killing of cows. After that time, however, it had been forbidden on account of the weakness of men, who were too weak to fulfil their duties, as also the Veda, which originally was only one, was afterwards divided into four parts, simply for the purpose of facilitating the study of it to men. This theory, however, is very little substantiated, as the prohibition of the meat of cows is not an alle- viating and less strict measure, but, on the contrary, one which is more severe and more restrictive than the former law. Other Hindus told me that the Brahmans used to suffer from the eating of cows' meat. For their country is hot, the inner parts of the bodies are cold, the natural warmth becomes feeble in them, and the power of. digestion is so weak that they must strengthen it by eating the leaves of betel after dinner, and by chewing. the betel-nnt. The hot betel inflames the heat of the body, the chalk on the betel-leaves dries up everything wet, and the betel-nnt acts as an astringent on the teeth, the gums, and the stomach. As this is the case, they forbade eating cows' meat, because it is essentially thick and cold, I, for my part, am uncertain, and hesitate in the question of the origin of this custom between two diffe- rent views. (Lacuna in the manuseript.) As for the economical reason, we must keep in mind that the cow is the animal which serves man in travel- ling by carrying his loads, in agriculture in the works of ploughing and sowing, in the household by the milk and the product made thereof. Further, man makes use of its dung, and in winter-time even of its breath.

CHAPTER LXVIII. 153 Therefore it was forbidden to eat cows' meat; as also Alhajjaj forbade it, whan people complained to him. that Babylonia became more and more desert. I have been told the following passage is from an That all Indian book: " All things are one, and whether allowed equn from a thingn are

or forbidden, equal. They differ only in weakness and onl point of philosophi-

power. The wolf has the power.to tear the sheep; view.

therefore the sheep is the wolf's food, for the former cannot oppose the latter, and is his prey." I have found in Hindu books passages to the same effect. However, such views come to the intelligent man only by knowledge, when in it hs has attained to snch a degree that a Brahman and a Candala are equal to him. If he is in this state, all other things also are equal to him, in so far as he abstains from them. It is the same if they are all allowed to him, for he can dispense with them, or if they are forbidden to him, for he does not desire them. As to those, however, who require these things, being in the yoke of ignorance, something is allowed to them, something forbidden, and thereby & wall is erected between the two kinds of things. .

( 154 )

CHAPTER LXIX.

ON MATRIMONY, THE MENSTRUAL COURSES, EMBRYOS, AND CHILDBED.

No ity of No nation can exist without a regular married life, matrtmuny. for it prevents the uproar of passions abhorred by the cultivated mind, and it removes all those causes which excite the animal to a fury always leading to harm. Considering the life of the animals by pairs, how the one member of the pair helps the other, and how the lust of other animals of the same species is kept aloof from them, you cannot help declaring matri- Page 378. mony to be a necessary institution; whilst disorderly cohabitation or harlotry on the part of man is a shame- ful proceeding, that does not even attain to the standing of the development of animals, which in every other respect stand far below him. Every nation has particular eustoms of marriage, and especially those who claim to have a religion ane law of divine origin. The Hindus marry at a very young age; therefore the parents arrange the marriage for their sons. On that occasion the Brahmans perform the rites of the sacrifices, and they as well as others receive alms. The implements of the wedding rejoic- ings are brought forward. No gift is settled between them, The man gives only a present to the wife, as he thinks fit, and a marriage gift in advance, which he has no right to claim back, but the wife may give it back to him of her own will. Hasband and wife can only be separated by death, as they have no divorce.

CHAPTER LXIX. 155

A man may marry one to four wives. He is not allowed to take more than four; but if one of his wives die, he may take another one to complete the legitimate number. However, he must not go beyond it. If a wife loses her husband by death, she cannot The widow. marry another man. She has only to chose between two things-either to remain a widow as long as she lives or to burn herself; and the latter eventuality is considered the preferable, because as a widow she is ill-treated as long as she lives. As regards the wives of the kings, they are in the habit of burning them, whether they wish it or not, by which they desire to prevent any of them by chance committing something unworthy of the illustrious husband They make an exception only for women of sdvanced years and for those who have children; for the son is the responsible protector of his mother. According to their marriage law it is better to marry Forbldden a stranger than a relative. The more distant the rela- marringe. tionehip of a woman with regard to her husband the better. It is absolutely forbidden to marry related women both of the direct descending line, viz. a grand- danghter or great-granddanghter, and of the direct ascending line, viz. a mother, grandmother, or great- grandmother. It is also forbidden to marry collateral relations, viz. a sister, a niece, a maternal or paternal aunt and their daughters, except in case the couple of relaticns who want to marry each other be removed from each other by five consecutive generations. In that case the prohibition is waived, but, notwith- standing, such a marriage is an object of dislike to them. Some Hindus think that the number of the wives Number of depends upon the caste; that, accordingly, a Brahman wives. may taka four, a Kshatriya three, a Vaisya two wives, and a Sudra one. Every man of a caste may marry a woman of his own caste or one of the castes or caste

156 ALBERUNTS INDIA.

below his; but nobody is allowed to marry a woman of a caste superior to his own. Pertus agui- The child belongs to the caste of the mother, not to that of the father. Thus, eg. if the wife of a Brahman is a Brahman, her child also is a Brahman; if she is a Śûdra, her child is a Sûdra. In our time, however, the Brahmans, although it is allowed to them, never marry any woman except one of their own caste. Duration of The longest duration of the menstrual courses which tha mon- strnal has been observed is sixteen days, bnt in reality they courtet last only during the first four days, and then the hus- band is not allowed to cohabit with his wife, nor even to come near her in the house, because during this time she is impure. After the four days have elapsed and she has washed, she is pure again, and the husband may cohabit with her, even if the blood has not yet entirely disappeared; for this blood is not considered as that of the menstrual courses, but as the same sub- stance-matter of which the embryos consist. On preg- It is the duty (of the Brahman), if he wants to co- DaDey and childbed. habit with a wife to get a child, to perform a sacrifice Pago 379 to the fire called garbhadhana; bnt he does not perform it, because it requires the presence of the woman, and therefore he feels ashamed to do co. In conseqnence he postpones the sacrifice and unites it with the next following one, which is due in the fonrth month of the pregnancy, called simamtonnayanam. After the wife has given birth to the child, a third sacrifice is per- formed between the birth and the moment when the mother begins to nourish the child. It is called jata- karman. The child receives a name after the days of the child- bed have elapsed. The sacrifice for the occasion of the name-giving is called ndmakarman. As long as the woman is in childbed, she does not tonch any vessel, and nothing is eaten in her house, nor does the Brahman light there a fire. These days are

CHAPTER LXIX. 157

eight for the Brahman, twelve for the Kshatriya, fifteen for the Vaisya, and thirty for the Sudra. For the low- caste people which are not reckoned among any caste, no term is fixed. The longest duration of the suckling of the child is three years, but there is no obligation in this matter. The sacrifice on the occasion of the first cutting of the child's hair is offered in the third, the perforation of the ear takes place in the seventh and eighth years. People think with regard to harlotry that it is allowed On the with them. Thus, when Kabul was conquered by the prostitu- causes of Muslims and the Ispahbad of Kabul adopted Islam, he tion.

stipulated that he should not be bound to eat cows' meat nor to commit sodomy (which proves that he abhorred the one as much as the other). In reality, the matter is not as people think, but it is rather this, that the Hindus are not very severe in punishing whoredom. The fault, however, in this lies with the kings, not with the nation. But for this, no Brahman or priest would suffer in their idol-temples the women who sing, dance, and play. The kings make them an attraction for their cities, a bait of pleasure for their subjects, for no other but financial reasons. By the revenues which they , derive from the business both as fines and taxes, they want to recover the expenses which their treasury has to spend on the army. In a similar way the Buyide prince 'Adud-aldanla acted, who besides also had a second aim in view, viz. that of protecting his subjects against the passions of his anmarried soldiers.

( 158 )

CHAPTER LXX.

ON LAWSUITS.

On pru- THE judge demands from the suitor a document written codura. against the accused person in a well-known writing which is thought suitable for writs of the kind, and in the document the well-established proof of the justice of his suit. In case there is no written document, the contest is settled by means of witnesses withont a written document. Number of The witnesses must not be less than four, but there may be more. Only in case the justice of the deposi- tion of a witness is perfectly established and certain before the jdge, he may admit it, and decide the qnes- tion alone on the basis of the deposition of this sole witness. However, he does not admit prying about in secret, deriving arguments from mere signs or indica- tions in pablic, conclnding by analogy from one thing which seems established abont another, and using all sorts of tricks to elicit the truth, as 'Iyas Ibn Mn'a- wiya used to do. If the suitor is not able to prove his claim, the de- fendant must ewear, bnt he may also tender the oath to the euitor by saying, "Swear thon that thy claim is true. and I will give thee what thou claimest." Different There are many kinds of the oath, in accordance with kinda ot oaths and the valne of the object of the claim, If the object is ordonls. of no great importance, and the suitor agrees that the accused person shall swear, the latter simply swears before five learned Brahmans in the following words:

CHAPTER LXX. 159

"If I lie, he shall have as recompense as much of my goods as is equal to the eightfold of the amonnt of his claim." A bigher sort of oath is this: The accused person is invited to drink the bish (visha ?) called brahmana (?). It is one of the worst kinds; hut if he speaks the truth, the drink does not do him any harm. A atill higher sort of ordeal is this: They bring the Page 28o. man to a deep and rapidly flowing river, or to a deep well with much water. Then he speaks to the water: "Since thou belongest to the pure angels, and knowest both what is secret and public, kill me if I lie, and preserve me if I speak the truth." Then five men take him between them and throw him into the water. If he has spoken the truth, he will not drown and die. A still higber sort is the following: The judge sends both claimant and defendant to the temple of the most venerated idol of the town or realm. There the defen- dant has to fast during that day. On the following day he dresses in new garments, and posts himself together with the claimant in that temple. Then the priests pour water over the idol and give it him to drink. If he, then, has not spoken the truth, be at once vomits blood. A atill higher sort is the following: The defendant is placed on the scale of a balance, and is weighed; wherenpon he is taken off the acale, and the acale is left as it is. Then he invokes as witnesses for the truth of his deposition the spiritual beings, the angels, the heavenly beings, one after the other, and all which he speaks he writes down on a piece of paper, and fastens it to his head. He is a second time placed in the scale of the balance. In case he has apoken the truth, he now weighs more than the first time. There is also a still higher sort. It is the following : They take bntter and sesame-oil in equal quantities, and

160 ALBERUNPS INDIA.

boil them in a kettle. Then they throw a leaf into it, which by getting flaccid and burned is to them a sign of the boiling of the mixture. When the boiling is at its height, they throw a piece of gold into the kettle and order the defendant to fetch it out with his hand, If he has spoken the truth, he fetches it out. The highest kind of ordeal is the following: They make a piece of iron so hot that it is near melting, and put it with a pair of tongs on the hand of the defen- dant, there being nothing between his hand and the iron save a broad leaf of some plant, and under it some few and scattered corns of rice in the husks. They order him to carry it seven paces, and then he may throw it to the grouud.

( 161 )

CHAPTER LXXI.

ON PUNISHMENTS AND EXPIATIONS.

IN this regard the manners and customs of the Hindus resemble those of the Christians, for they are, like those of the latter, based on the principles of virtue and abstinence from wickedness, such as never to kill under any circumstance whatsoever, to give to him who has stripped yon of your coat also your shirt, to offer to him who has beaten your cheek the other cheek also, to bless your enemy and to pray for him. Upon my life, this is a noble philosophy; but the people of this world are not all philosophers. Most of them are ignorant and erring, who cannot be kept on the straight road save by the sword and the whip. And, indeed, ever since Constantine the Victorious became a Chris- tian, both sword and whip have ever been employed, for withont them it would be impossible to rule. India has developed in a similar way. For the Hin- The Brah- dus relate that originally the affairs of government and nally the mans orlgi-

war were in the hands of the Brahmans, hut the country nation. , rulem of the

became disorganised, since they ruled according to the philosophic principles of their religious codes, which proved impossible when opposed to the mischievous and perverse elements of the populace. They were even near losing also the administration of their religious affairs. Therefore they humiliated themselves before the lord of their religion. Whereupon Brahman in- Poge 281. trusted them exclusively with the functions which they now have, whilst he intrusted the Kshatriyas with the VOL. 1I. L

162 ALBERUNI'S INDIA.

duties of ruling and fighting. Ever since the Brahmans live by asking and begging, and the penal code is exer- cised under the control of the kings, not under that of the scholars. Law of The law abont murder is this: If the murderer is a murder. Brahman, and the murdered person s member of another caste, he is only bound to do expiation consisting of fasting, prayers, and almsgiving. If the murdered person is a Brahman, the Brahman murderer has to answer for it in a futare life; for he is not allowed to do expiation, because expistion wipes off the sin from the sinner, whilst nothing can wipe off any of the mortal crimes from a Brahman, of which the greatest are: the murder of a Brahman, called vajra- brahmahatyd ; further, the killing of a cow, the drink- ing of wine, whoredom, especially with the wife of one's own father and teacher. However, the kings do not for any of these crimes kill a Brahman or Kshstriya, bot they confiscate his property and banish him from their conntry. If a man of a caste under those of the Brahman and Kshatriya kills a man of the same caste, he has to do expiation, but besides the kinge inflict upon him a punishment in order to establish an example. LAw oftheft. The law of theft directs that the punishment of the thief should be in accordance with the value of the stolen object, Accordingly, sometimes a punishment of extreme or of middling severity is necessary, sometimes a course of correction and imposing a payment, sometimes only exposing to pnblic shame and ridicule. If the object is very great, the kings blind a Brahman and mutilate him, cutting off his left hand and right foot, or the right hand and left foot, whilst they mutilate a Kshatriya without blinding him, and kill thieves of the other castes. Puniehment An adulteress is driven out of the house of the hus- of an adulterem. band and baniahed. I have repeatedly been told that when Hindu slaves

CHAPTER LXXI. 163

(in Muslim countries) escape and return to their country Hlndu and religion, the Hindus order that they should fast by war, how prisonor of way of expiation, then they bury them in the dung, returning to treatod after

stale, and milk of cows for a certain number of days, try. their coun- till they get into a state of fermentation. Then they drag them out of the dirt and give them similar dirt to eat, and more of the like. I have asked the Brahmans if this is true, bnt they deny it, and maintain that there is no expiation possible for such an individual, and that he is never allowed to return into those conditions of life in which he was before he was carried off as a prisoner. And how should that be posaible? If a Brahman eats in the house of a Sudra for sundry days, he is expelled from his caste and can never regain it.

( 164 )

CHAPTER LXXII,

ON INHERITANCE, AND WHAT CLAIM THE DECEASED PERSON HAS ON IT.

Law of in- THE chief rule of their law of inheritance is this, that borttanoe. the women do not inherit, except the danghter. She gets the fourth part of the share of a son, according to a passage in the book Manu. If she is not married, the money is spent on her till the time of her marriage, and her dowry is bought by means of her share. After- wards she has no more income from the house of her father. If a widow does not burn herself, bnt prefers to remain alive, the heir of her deceased husband has to provide her with nourishment and clothing as long as she lives. The debts of the deceased must be paid by his heir, either ont of his share or of the stock of his own pro- perty, no regard being had whether the deceased has left any property or not. Likewise he must bear the just-mentioned expenses for the widow in any case whatsoever. As regards the rule abont the male heirs, evidently the descendants, i.e. the son and grandson, have a nearer claim to the inheritance than the ascendants, i.e. the Page 28z. father and grandfather. Further, as regards the single relatives among the descendants as well as the ascen- dants, the nearer a man is related, the more claim he has on inheriting. Thus a son has a nearer claim than a grandson, a father than a grandfather. The collateral relations, as, e.g. the brothers, have less

CHAPTER LXXII. 165

claim, and inherit only in case there is nobody who has a better claim. Hence it is evident that the son of a danghter has more claim than the son of a siater, and that the son of a brother has more claim than either of them. If there are several claimants of the same degree of relationship, as, e.g., sona or brothers, they all get equal shares. A hermaphrodite is reckoned as a male being. If the deceased leaves no heir, the inheritance falls to the treasnry of the king, except in the case that the deceased person was a Brahman. In that case the king has no right to meddle with the inheritance, bot it is exclusively spent on almagiving. The dnty of the heir towards the deceased in the Dutios of first year consists in his giving eixteen banquets, where wards the the helr to-

every guest in addition to bis food receives alms also, viz. on the fifteenth and sixteenth days after death; further, once a month during the whole year. The ban- quet in the sixth month must be more rich and more liberal than the others. Further, on the last but one day of the year, which banquet is devoted to the deceased and his ancestors; and finally, on the last day of the year. With the end of the year the duties towards the deceased have been fulfilled. If the heir is a son, he mnat during the whole year wear mourning dress; he mnst mourn and have no intercourse with women, if he is a legitimate child and of a good stock. Besides, you mnst know that nourish- ment is forbidden to the heirs for one single day in the first part of the mourning-year. Besides the almsgiving at the just-mentioned sixteen banquets, the heirs must make, above the door of the house, something like a shelf projecting from the wall in the open air, on which they have every day to place & dish of something cooked and a vessel of water, till the end of ten daya after the death. For possibly the spirit of the deceased has not yet found its rest, but

.166 ALBERUNIS INDIA.

moves still to and fro around the bouse, hungry and thirsty. Parallel A similar view is indicated by Plato in Phaedo, where trom Plato. he speaks of the soul circling round the graves, because possibly it still retains some vestiges of the love for the body. Further he asys: " People have said regarding the soul that it is its habit to combine something cohe- rent out of the single limbs of the body, which is its dwelling in this as well as in the future world, when it leaves the body, and is by the death of the body sepa- rated from it." On the tenth of the last-mentioned days, the heir spends, in the name of the deceased, much food and cold water. After the eleventh day, the heir sends every day sufficient food for a single person and a dirham to the house of the Brahman, and continues doing this during all the days of the mourning-year without any interruption until its end.

. ( 167 )

CHAPTER LXXIII.

ABOUT WHAT IS DUE TO THE BODIES OF THE DEAD AND OF THE LIVING (i.e. ABOUT BURYING AND SUICIDE).

. IN the most ancient times the bodies of the dead were Primitive exposed to the air by being thrown on the fields without toms. burtal-cus-

any covering; also sick people were exposed on the fields and in the mountains, and were left there. If they died there, they had the fate just mentioned; but if they recovered, they returned to their dwellings. Thereupon there appeared a legislator who ordered Page 283. people to expose their dead to the wind. In conse- quence they constructed roofed buildings with walls of rails, through which the wind blew, passing over the dead, as something similar is the case in the grave- towers of the Zoroastrians. After they had practised this custom for a long time, Narayana prescribed to them to hand the dead over to the fire, and ever since they are in the habit of burn- ing them, so that nothing remains of them, and every defilement, dirt, and smell is annihilated at once, so as scarcely to leave any trace behind. Nowadays the Slavonians, too, burn their dead, whilst Oreek par- the ancient Greeks seem to have had both customs, allels.

that of burning and that of burying. Socrates speaks in the book Phaedo, after Crito had asked him in what manner he wanted to be buried: " As you wish, when you make arrangements for me. I shall not flee from you." Then he spoke to those around him: "Give to Crito regarding myself the opposite guarantee of that

168 ALBERUNTS INDIA.

which he has given to the judges regarding myself; for he guaranteed to them that I should stay, whilat you now must guarantee that I shall not atay after death, I shall go away, that the look of my body may be tolerable to Crito when it is burned or buried, that he may not be in agony, and not say: 'Socrates is carried away, or is burned or buried.' Thou, O Crito, be at ease about the burial of my body. Do as thou likest, and specially in accordance with the lawa." Galenus saya in his commentary to the epothegms of Hippocrates: "It is generally known that Asclepius was raised to the angels in a column of fire, the like of which is also related with regard to Dionysos, Heracles, and others, who laboured for the benefit of mankind. People say that God did thus with them in order to destroy the mortal and earthly part of them by the fire, and afterwards to attract to himself the immortal part of them, and to raise their souls to heaven." In these words, too, there is a reference to the burning as a Greek custom, but it seems to have been in use only for the great men among them. In a aimilar way the Hindus express themselves. There is a point in man by which he is what he is. This point becomes free when the mixed elements of the body are dissolved and scattered by combustion. Firo and the Regarding this return (of the immortal soul to God), tunbeam as the nearet the Hindus think that partly it is effected by the rays ronds to God. of the sun, the soul attaching itself to them and ascend- ing with them, partly by the flame of the fire, which raises it (to God). Some Hindu used to pray that God would make his road to himself as a straight line, be- cause this is the nearest road, and that there is no other road upwards save the fire or the ray. Similar to this is the practice of the Ghuzz Turks with reference to a drowned person; for they place the body on a hier in the river, and make a cord hang down

CHAPTER LXXIII. 169

from his foot, throwing the end of the cord into the water. By means of this cord the spirit of the deceased is to raise himself for resurrection. The belief of the Hindns on this head was confirmed by the words of Vasndeva, which he spoke regarding the sign of him who is liberated from the fetters (of bodily existence). "His death takes place during utta- rayana (i.c. the northern revolntion of the sun from the winter solstice to the summer solstice), during the white half of the month, between lighicd lamps, i.e. be- tween conjunction and opposition (new moon and full moon), in the seasons of winter and spring." A similar view is recognised in the following words Quotation of Mant: "The other religious bodies blame us because Page 284- from MAnt

we worship sun and moon, and represent them as an image. Bnt they do not know their real natures; they do not know that sun and moon are our path, the door whence we march forth into the world of our existence (into heaven), as this has been declared by Jesus." So he maintains. People relate that Buddha had ordered the bodies of the dead to be thrown into flowing water. Therefore his followers, the Shamanians, throw their dead into the rivers. According to the Hindus, the body of the dead has Hindu the claim upon his heirs that they are to wash, embalm, burial, manner of wrap it in a shroud, and then to burn it with as much sandal and other wood as they can get. Part of his burned bones are brought to the Ganges and thrown into it, that the Ganges should flow over them, as it has flowed over the barned bones of the children of Sagara, thereby forcing them from hell and bring- ing them into paradise. The remainder of the ashes is thrown into some brook of running wster. On the spot where the body has been burned they raise a monument similar to a milestone, plastered with gypsum.

170 ALBERUNI'S INDIA,

The bodies of children under three years are not burned. Those who fulfil these duties towards the dead after- wards wash themselves as well as their dresses during two days, because they have become unclean by touch- ing the dead. Those who cannot afford to burn their dead will either throw them somewhere on the open field or into running water. Nodes of Now as regards the right of the body of the living, aadeido. the Hindus would not think of burning it save in the case of a widow who chooses to follow her husband, or in the case of those who are tired of their life, who are distressed over some incurable disease of their body, some irremovable bodily defect, or old age and infirmity. This, however, no man of distinction does, but only Vaisyas and Śûdras, especially at those times which are prized as the most suitable for a man to acquire in them, for a future repetition of life, a better form and condition than that in which he happens to have been born and to live. Burning oneself is forbidden to Brahmans and Kshatriyas by a special law. Therefore these, if they want to kill themselves, do so at the time of an eclipse in some other manner, or they hire some- body to drown them in the Ganges, keeping them under water till they are dead. The trea of At the junction of the two rivers, Yamuna and Ganges, there is a great tree called Praydga, a tree of the species called vata. It is peculiar to this kind of tree that its branches send forth two species of twigs, some directed npward, as is the case with all other trees, and others directed downward like roots, but without leaves. If such a twig enters into the soil, it is like a supporting column to the branch whence it has grown. Nature has arranged this on purpose, since the branches of this tree are of an enormous extent (and require to be supported). Here the Brahmans and Kshatriyas are in

CHAPTER LXXIII. I7I

the habit of committing suicide by climbing up the tree and throwing themselves into the Ganges. Johannes Grammaticus relates that certain people Greek in ancient Greek heathandom, " whom I call the wor- paradlals.

shippers of the devil"-so he says-used to beat thair limbs with swords, and to throw themselves into the fire, without feeling any pain therefrom. As we havs related this as a view of the Hindus not to commit suicide, so also Socrates speaks: " Likewise it does not become a man to kill himself before the gods give him a cause in the shape of some compulsion or dire necessity; like that in which we now are." Further he says: " We human beings are, as it were, in a prison. It does not behove us to flee nor to free ourselves from it, because the gods take notice of us, since we, the human beings, are servants to them."

( 172 )

CHAPTER LXXIV.

ON FASTING, AND THE VARIOUS KINDS OF IT.

FASTING is with the Hindus voluntary and superero- gatory. Fasting is abstaining from food for a certain length of time, which may be different in duration and Page 285- in the manner in which it is carried out. Varlous The ordinary middle process, by which all the condi- mothods of fasting. tions of fasting are realised, is this: A man determines the day on which he will fast, and keeps in mind the name of that being whose benevolence he wishes to gain thereby and for whose sake he will fast, be it a god, or an angel, or some other being. Then he pro- ceeds, prepares (and takes) his food on the day before the fast-day at noon, cleans his teeth by rubbing, and fixes his thoughts on the fasting of the following day. From that moment he abstains from food. On the morning of the fast-day he again rubs his teeth, washes himself, and performs the duties of the day. He takes water in his hand, and sprinkles it into all four direc- tions, he prononnces with his tongue the name of the deity for whom he fasts, and remains in this condition till the day after the fast-day. After the sun has risen, he is at liberty to break the fast at that moment if he likes, or, if he prefers, he may postpone it till noon. This kind is called upavdsa, ie. the fasting; for the not-eating from one noon to the following is called eka- nakta, not fasting. Another kind, called kricchra, is this: A man takes his food on some day at noon, and on the following day

CHAPTER LXXIV. 173 in the evening. On the third day he eats nothing except what by chance is given him withont his asking for it. On the fourth day he fasts. Another kind, called pardka, is this: A man takes his food at noon on three consecutive days. . Then he transfers his eating-hour to the evening during three further consecutive days. Then he fasts nninterrupt- edly during three consecntive days without breaking fast. Another kind, called candrayana, is this: A man fasts on the day of full moon; on the following day he takes only a mouthful, on the third day he takes double this amount, on the fourth day the threefold of it, &c., &e., going on thus till the day of new moon. On that day he fasts; on the following days he again diminishes his food by one mouthful a day, till he again fasts on the day of full moon. Another kind, called masavasa (masopavdsa), is this : A man nninterruptedly fasts all the daye of a month withont ever breaking fast. The Hindus explain accurately what reward the latter Reward of fasting in every single month will bring to a man for a in the single tho fasting

new life of his after he has died. They say : month.

If a man fasts all the days of Caitra, he obtains wealth and joy over the nobility of his children. If he fasts Vaisakha, he will be a lord over his tribe and great in his army. If he fasts Jyaishtha, he will be a favourite of the women. If he fasts Ashadha, he will obtain wealth. If he fasts Śravana, he obtains wisdom. It he fasts Bhadrapada, he obtains health and valour, riches and cattle. If he fasts Âsvaynja, he will always be victorious over his enemies. If he fasts Karttika, he will be grand in the eyes of people and will obtain his wishes.

I74 ALBERUNrS INDIA.

If he fasta Margaftrsha, he will be born in the most beantiful and fertile country. If he fasts Pausha, he obtains a high reputation. If he fasts Magha, he obtains innumerable wealth. If he fasts Phalguna, he will be beloved. He, however, who fasts during all the months of the year, only twelve timee breaking the fast, will reside in paradise 10,000 years, and will thence return to life as the member of a noble, high, and respected family. The book Vishnu-Dharma relates that Maitreyl, the Page 286. wife of Yajnavalkya, asked her husband what man is to do in order to save his children from calamities and bodily defects, upon which he answered: " If a man begins on the day Duve, in the month Pansha, i.e. the second day of each of the two halves of the month, and fasts four consecntive days, washing himself on the first with water, on the second with sesame oil, on the third with galangale, and on the fourth with a mixture of various balms; if he further on each day gives alms and recites praises over the names of the angels; if he continne to do all this during each month till the end of the year, his children will in the following life be free from calamities and defects, and he will obtain what he wishes; for also Diltpa, Dushyanta, and Yayati obtained their wishes for having acted thus."

( 175 )

CHAPTER LXXV.

ON THE DETERMINATION OF THE FAST-DAYS.

THE reader must know in general that the eighth and The oighth eleventh days of the white half of every month are fast- daye of each and eleventh

days, except in the case of the leap month, for it is dis- month aro half of s regarded, being considered unlucky. last-daya.

The eleventh is specially holy to Vâsudeva, because on having takeu possession of Mabura, the inhabitants of which formerly used to worship Indra one day in each month, he induced them to tranefer this worship to the eleventh, that it should be performed in his name. As the people did so, Indra became angry and poured rains over them like deluges, in order to destroy both them and their cattle. Vasudeva, however, raised a monntain by his hand and protected them thereby. The water collected round them, bnt not above them, and the image of Indra fled. The people commemorated this event by a monument on a mountain in the neighbourhood of Mahtra. Therefore they fast on this day in the etate of the most punctilions cleanness, and they stay awake all the night, considering this as an obligatory perfor- mance, though in reality it is not obligatory. The book Vishnu-Dharma says: "When the moon is On tingle in Rohini, the fourth of her stations, on the eighth day of throughout fant-days

the black half, it ie a fast-day called Jayanti. Giving the yoar. alma on this day is an expiation for all sins." Evidently this condition of the fast-day does not in general apply to all months, but in particular only to Bhadrapada, since Vasndeva was born in this month

176 ALBERUNPS INDIA.

and on this day, whilst the moon stood in the station Rohini. The two conditions, viz the moon's standing in Rohint and that the day is the eighth of the black half, can happen only once in so and so many years, for varions reasons, e.g. the intercalation of the year, and because the civil years do not keep pace with lunar time, either getting in advance of it or falling behind, The same book says: "When the moon stands in Punarvasu, the seventh of her stations, on the eleventh day of the white half of the month, this is a fast-day, called Atj (? Attâtaja). If a man does works of piety on this day, he will be enabled to obtain whatever he wishes, as has been the case with Sagara, Kakutstha, and Dandahamar (?), who obtained royalty because they had done so. The sixth day of Caitra is & fast-day holy to the aun. In the month Ashadha, when the moon stands in Anurâdha, the seventeenth of her signs, there is a fast- day holy to Vasudeva called Devasinf (?), i.c. Deva is sleeping, because it is the beginning of the fonr months during which Vasudeva slept Others add this condi- tion, that the day must be the eleventh of the month. It is evident that auch a day does not occur in every year. The followers of Vasndeva abstain on this day from mest, fish, sweetmeats, and cohabitation with the women, and take food only once a day. They make Page 287. the earth their bed without any covering, and do not use a bedstead raised above the earth. People say that these fonr months are the night of the angels, to which must be added a month at the beginning as evening twilight, and a month st the end as morning dawn. However, the sun atands then near o° of Cancer, which is noon in the day of the angels, and I do not see in what way this moon is connected with the two Samdhis. The day of full moon in the month Sravana is a fast- day holy to Somanatha.

CHAPTER LXXV. 177 When in the month Asvayuja the moon stands in Alsharatân (the lunar station) and the snn is in Virgo, it is a fast-day. The eighth of the same month is a fast-day holy to Bhagavati. Fasting is broken when the moon rises. The fifth day of Bhadrapada is a fast-day holy to the sun, called shat. They anoint the solar rays, and in particular those rays which enter through the win- dows, with various kinds of balsamic ointments, and place npon them odoriferous plants and flowers. When in this month the moon stands in Rohiņi, it is a fast-day for the birth of Vasudeva. Others add, besides, the condition that the day mnst be the eighth of the black half. We have already pointed out that such a day does not occur in every year, but only in certain ones of a larger number of years. When in the month Karttika the moon stands in Revati, the last of her stations, it is a fast-day in com- memoration of the waking up of Vasudeva. It is called deotthint, i.e. the rising of the Deva. Others add, besides, the condition that it must be the eleventh of the white half. On that day they soil themselves with the dung of cows, and break fasting by feeding upon a mixtnre of cow's milk, urine, and dung. This day is the first of the five days which are called Bhishma panca- ratri. They fast during them in honour of Vasudeva. On the second of them the Brahmans break fasting, after them the others. On the sixth day of Pausha is a fasting in honour of the sun. On the third day of Mâgha there is a fasting for the women, not for the men. It is called Gaur-t-r (gaurt-trittyd !), and lasts the whole day and night. On the following morning they make presents to the nearest relatives of their husbands.

VOL. II.

( 178 )

CHAPTER LXXVL

ON THE FESTIVALS AND FESTIVE DAYS.

YATRA means travelling under auspicious circumstances. Therefore a fenst is called yatra. Most of the Hindu festivals are celebrated by women and children only. The sud Cai- The 2nd of the month Caitra is a festival to the people of Kashmir, called Agdus (?), and celebrated on account of a victory gained by their king, Muttai, over the Turks. According to their acconnt he ruled over the whole world. But this is exactly what they say of most of their kings. However, they are incautious enough to assign him to a time not much anterior to our time, which leads to their lie being found out. It is, of course, not impossible that a Hindu should rale (over a huge empire), as Greeks, Romans, Babylonians, and Persians have done, but all the times not much anterior to our own are well known. (It, therefore, auch had been the case, we should know it.) Perhaps the here mentioned king ruled over the whole of India, and they know of no other country but India and of no other nations bnt themselves. nIth Caltra. On the 1Ith there is & festival called Hindol- caitra, when they meet in the devagriha, or temple of Vasudeva, and swing his image to and fro, as had been done with him when he was an infant in the cradle. They perform the same in their houses during the whole day and make merry. Fall moon's On the full moon's day of Caitra there is a feast day. called Bahand (vasanta f). & festival for the women,

CHAPTER LXXVI. 179

when they put on their ornaments and demand presents from their husbands. The zznd is a festival called caitra-cashati, a day of znd Caitra. merriment boly to Bhagavati, when people use to wash and to give alms. The 3rd Vaisakha is a festival for the women called ard Vald- Gaur-t-r (gaurt-trittya f), holy to Gaurt, the danghter of Page a88. the mountain Himavant, the wife of Mahadeva They wash and dress gaily, they worship the image of Gauri and light lamps before it, they offer perfumes, abstain from eating, and play with swings. On the following day they give alms and eat. On the 1oth Vaifakha all the Brahmans whom the kings have invited proceed forth to the open fields, and there they light great fires for the sacrifices during five days till full moon. They make the firee in sixteen different apots and in four different groups. In each group a Brahman performs the sacrifice, so that there are four performing priests as there are four Vedas. On the 16th they return home. In this month occnrs the vernal equinox, called vernal equi- vasanta. They determine the day by calculation and nox. make it a festival, when people invite the Brahmans. On the Ist Jyaishtha, or new moon's day, they cele- zet Jyaish- brate a festival and throw the firstfruits of all seeds tha.

into the water in order to gain thereby a favourable prognostic. The full moon's day of this month is a festival to Pall moon's the women, called rupa-panca (?). day.

All the days of the month Ashadha are devoted to Ashadha. alms-giving. It is also called dhart. During this time the household is provided with new vessels. On the full moon'a day of Srâvana they give banquets 15th Sra- to the Brahmans. On the 8th Asvayuja, when the moon stands in the sth Afvayu- nineteenth station, Mula, begins the sucking of the eugar cane. It is a festival holy to Mahanavamt, the

180 ALBERUNTS INDIA. sister of Mahadeva, when they offer the firstfruits of sugar and all other things to her image which is called Bhagavati. They give mnch alms before it and kill kids, He who does not possess anything to offer, stands upright by the side of the idol, without ever sitting down, and will sometimes pounce upon whom- soever he meets and kill him. Isth Aéra- On the 15th, when the moon stands in the last of her atations, Revatt, there is the festival Puhdt (?), when they wrangle with each other and play with the animals. It is holy to Vasudeva, because his uncle Kamhsa had ordered bim into his presence for the pur- pose of wrangling. 16th Aám- On the 16th there is a festival, when they give alms yuja. to the Brahmans. On the 23rd is the festival Afoka, also called dhot, when the moon atands in the seventh station, Punar- vasu. It is a day of merriment and of wrangling. Bhldrapeda, In the month Bhadrapada, when the moon atands in the tenth station, Magha, they celebrate a festival which they call pitripaksha, i.c. the half of the month of the Fathers, because the moon's entering this atation falls near the time of new moon. They distribnte alms during fifteen days in the name of the Fathers. grd Bhsdra- On the 3rd Bhadrapada is the festival Harbalt (?), for prda. the women. It is their custom that a number of daya before thay sow all kinds of seeds in baskets, and they bring the baskets forward on this day after they have commenced growing. They throw roses and perfumes on them and play with each other during the whole night. On the following morning they bring them to the ponds, wash them, wash themselves, and give alms. 6th Baadra- On the 6th of this month, which is called Gaihat (?), when people give food to those who are in prison. On the 8th, when the moonlight has reached half of pad4. its development, they have a festival called dhruva-

CHAPTER LXXVI. 181

griha (1); they wash themselves and eat well growing grain-fruit that their children should be healthy. The women celebrate this festival when they are pregnant and desire to have children. The 1Ith Bhadrapada is called Parvatf (?). This is rIth Bhad- the name of a thread which the prieat makes from Pago z8g. rapadd

materials presented to him for the purpose. One part of it he dyes with crocus, the other he leaves as it is. He gives the thread the same length as the atatue of Vasudeva is high. Then he throwa it over his neck, so that it hangs down to hia feet, It is a much vene- rated festival. . The 16th, the first day of the black half, is the first 16th Bhad- of seven days which are called kardra (?), when they rapadA

adorn the children nicely and give a treat to them. They play with various animals. On the aeventh day the men adorn themaelves and celebrate a featival. And during the rest of the month they always adorn the children towards the end of the day, give alms to the Brahmans, and do worka of piety. When the moon stands in her fourth atation, Rohiņi, they call this time Gundlahtd (1), celebrating a festival during three days and makiog merry by playing with each other, from joy over the birth of Vasudeva. Jivasarman relates that the people of Kashmir cele- asth, 27th, brate a festival on the 26th and 27th of this month, d4. on acccunt of certain pieces of wood called gana (?), which the water of the river Vitasta (Jailam) carries, in those two days, through the capital, Adhishthana. People maintain that it is Mahadeva who sends them. It is peculiar to these pieces of wood, so they say, that nobody is able to seize them, however much he may desire it, that they always evade his grasp and move away. However, the people of Kashmir, with whom I have conversed on the subject, give a different statement as to the place and the time, and maintain that the thing occurs in a pond called Kedaishahr (?), to the left of the

182 ALBERUNPS INDIA.

source of the just-mentioned river (Vitasta-Jailam), in the middle of the month Vaifakha The latter version is the more likely, as about this time the watero begin to increase. The matter reminds one of the wood in the river of Jurjan, which appears at the time when the water swells in its source. The same Jivasarman relates that in the country of Svat, opposite the district of Kirt (?), there is a valley in which fifty-three streams unita It is called Tranjdi (cf. Sindhi trevanjaha). In those two days the water of this valley becomes white, in consequence of Maha- deva's washing in it, as people believe. ant Kartti- The Ist Karttika, or new moon's day, when the ka. sun marcbes in Libra, is called Dibalf. Then people bathe, dress festively, make presents to each other of betel-leaves and areca-nuts; they ride to the temples to give alms and play merrily with each other till noon. In the night they light a great number of lamps in every place so that the air is perfectly clear. The cause of this festival is that Lakshmi, the wife of Vasu- deva, ouce a year on this day liberates Bali, the son of Virocana, who is a prisoner in the seventh earth, and allows him to go out into the world. Therefore the festival is called Balirdjya, i.c. the principality of Bali. The Hindus maintain that this time was a time of luck in the Kritaynga, and they are happy because the feast-day in question resembles that time in the Kritayuga In the same month, when full moon is perfect, they give banquets and adorn their women during all the days of the black half. 3rd MArga- The 3rd Margafirsha, called Guvdna-batrtj (- tri- ttyd f), is a feast of the women, sacred to Gaurt They meet in the houses of the rich among them; they put several silver statues of the goddess on a throne, and perfume it and play with each other the whole day. On the following morning they give alma

CHAPTER LXXVI. 183

On full moon'a day of the sama month there is tsth Mirga- another festival of the women, Sirahs. On most of the days of the month Pausha they pre- Pausha. Pago 290.

pare great quantities of puhaval (?), i.e. a aweet dish which they eat. On the eighth day of the white half of Pausha, which sth Pausha is called Ashtaka, they make gatherings of the Brah- mans, present them with dishes prepared from the plant Atriplex hortensis, i.e. sarmak in Arabic (= orache), and ahow attentions to them, On the eighth day of the black half, which is called Sakartam, they eat turnips. The 3rd Magha, called Mahatrij (Magha-trittya !), is grd Migha. a feast for the women, and sacred to Gauri. They meet in the houses of the most prominent among them be- fore the image of Ganri, place before it varions sorts of costly dresses, pleasant perfumes, and nice dishes. In each meeting-place they pnt 108 jugs full of water, and after the water has become cool, they wash with it four times at the four quarters of that night. On the following day they give alms, they give banquets and receive guests. The women's washing with cold water is common to all the days of this month. On the last day of this month, i.c. the 29th, when aoth MAgha there is only & remainder of 3 day- minutes, i.e. It hour, all the Hindus enter the water and duck under in it seven times. On the full moon's day of this month, called cdmaha rsth Migha (?), they light lamps on all high places. On the 23rd, which is called mansartaku, and also zrd Haghe. mahatan, they receive guests and feed them on meat and large black peas. On the 8th Phalguna, called purdrtaku, they pre- sth PhAlgu, pare for the Brahmans various dishes from flour and butter. The full moon's day of Phalguna is & feast to the rsth PhAl- women, called Oddd (?), or also dhola (i.e. dola), when

184 ALBERUNS INDIA.

they make fire on places lower than those on which they make it on the festival camdha, and they throw the fire ont of the village. 16th Pb4l- On the following night, ie that of the 16th, called Sivardtri, they worship Mahadeva during the whole night; they remain awake, and do not lie down to sleep, and offer to him perfumes and flowers. agrd Ph4l- On the 23rd, which is called puyattan (?), they eat rice with butter and sugar. A featiral in The Hindus of Multan have a festival which is called Sambapurayatrd; they celebrate it in honour of the sun, and worship him. It is determined in this way: They first take the ahargana, according to the rules of Khandakhadyaka, and subtract 98,040 therefrom. They divide the remainder by 365, and disregard the quotient. If the division does not give a remainder, the quotient is the date of the festival in question. If there is a remainder, it represents the days which have elapsed since the festival, and by subtracting these days from 365 you find the date of the same festival in the next following year.

( 185 ) . * &

CHAPTER LXXVII.

ON DAYS WHICH ARE HELD IN SPECIAL VENERATION, ON LUCKY AND UNLUCKY TIMES, AND ON SUCH TIMES . AS ARE PARTICULARLY FAVOURABLE FOR ACQUIRING IN THEM BLISS IN HEAVEN.

THE single days enjoy different degrees of veneration according to certain qualities which they attribute to them. They distinguish, eg., the Sunday, because it is the day of the sun and the beginning of the week, as the Friday is distingnished in Islam. To the distinguished days further helong amdodsyd The days of and purnimd, i.e. the days of conjunction (new moon) and full new moon

and opposition (full moon), because they are the limits moon

of the wane and the increase of the moonlight. In accordance with the belief of the Hindus regarding Page 291. this increase and wane, the Brahmans sacrifice con- tinnally to the fire in order to earn heavenly reward. They let the portions of the angels accumulate, which are the offerings thrown into the fire at moonlight during the whole time from new moon to full moon. Then they begin distributing these portions over the angels in the time from full moon to new moon, till at the time of new moon nothing any more remains of them. We have already mentioned that new moon and full moon are noon and midnight of the nychthemeron of the Fathers. Therefore the uninterrupted almsgiving on these two days is always done in honour of the Fathers.

186 ALBBRUNPS INDIA.

The foor Four other days are held in special veneration, das on Thich tho becanse, according to the Hindus, with them the foor yugar single yugas of the present caturyuga have commenced, bvo coma- moencod. viz. :-- The 3rd Vaifikha, called kshairitd (?), on which the Kritayuga is believed to have commenced. The 9th Karttika, the beginning of the Tretâyuga The 15th Magha, the beginning of the Dvapara- yuga. The 13th of Asvayuja, the beginning of the Kali- yuga. According to my opinion, these days are festivals, sacred to the yugas, instituted for the purpose of alms- giving or for the performance of some rites and cere- monies, as, e.g., the commemoration-days in the year of the Christians. However, we must deny that the four yugas could really have commenced on the days here mentioned. Criticisms With regard to the Kritaynga, the matter is perfectly thercoo. clear, because its beginning is the beginning of the solar and lunar cycles, there being no fraction in the date, since it is, at the same time, the beginning of the caturyuga. It is the first of the month Caitra, at the same time the date of the vernal equinox, and on the same day also the other yugas commence. For, accord- ing to Brahmagupta, a caturyuga contains :- Civil days . Solar months 1,577,916,450

Leap months 51,840,000

Lunar days . 1,593,300

Ûnarâtra daye . 1,602,999,000 25,082,550 These are the elements on which the resolution of chronological dates into days, or the composition of them out of days, is based. All these numbers may be divided by 10, and the divisors are wholes without any fraction. Now the beginnings of the single yugas depend upon the beginning of the caturyuga,

CHAPTER LXXVII. 187

According to Pulisa the caturyuga contains :- Civil days Solar months 1,577,917,800

Leap months 51,840,000 . 1,593,336 Lunar days . Ûnardtra days 1,603,000,010 25,082,280 All these numbers may be divided by 4, and the divi- sors are wholly without any fraction. According to this computation, also, the beginnings of the single yugas are the aame as the beginning of the caturyuga, i.c. the first of the month Caitra and the day of the vernal equinox, However, this day falls on different week daya. Hence it is evident that their theory about the above-mentioned four days being the beginnings of the four yugas, is without any foundation at all; that they could never arrive at such a result unless by resorting to very artificial ways of interpretation. The times which are specially favourable to earn a The days heavenly reward in them are called punyakala. Bala- dabila called pua bhadra says in his commentary to the Khandakhad- yaka :- "If the yogin, i.e. the ascetic who understands the creator, who chooses the good and eschews the bad, continned his manner of life during one thousand years, his reward would not be eqnal to that of a man who givea alms on punyakala and fulfila the dnties of the day, ia washing and anointing himself, saying prayers and praises." No doubt, most of the feast-days euumerated in the preceding belong to this kind of daya, for they are Page 2g2. devoted to almsgiving and banqueting. If people did not expect to gain thereby a reward in heaven, they would not approve of the rejoicings and merrimenta which are characteristic of these daya. Notwithstanding the nature of the punyakdla is such as here explained, some of them are considered as lucky, others as unlucky days.

188 ALBERUNPS INDIA.

Those days are Incky when the planets migrate from one sign into the other, especially the sun. These times Bahkrint. are called samkranti. The most propitions of them are the days of the equinoxes and solstices, and of these the most propitious is the dey of the vernal equinox. It is called bikhut or shibd (vishuva), as the two sounds sh and kh may be exchanged for each other, and may also, by a metathesis, change their place. As, however, a planet's entering a new sign does not require more than a moment of time, and, during it, people must offer to the fire the offering sdnta (?) with oil and corn, the Hindus have given a greater extent to these times, making them begin with the moment when the eastern edge of the body of the sun tonches the first part of the sign; reckoning as their middle tho moment when the sun's centre reaches the first part of the sign, which is in astronomy considered as the time of the migration (of the planet from one sign to the other), and reckoning as the end that moment when the western edge of the sun's body tonches the first part of the sign, This process lasts, in the case of the sun, nearly two hours. For the purpose of finding the times in the week when the sun migrates from one sign to another, they have several methode, one of which was dictated to me by Samaya (?). It is this :-- Mothod for Subtract from the Sakakala 847, multiply the re- the moment mainder by 180, and divide the product by 143. The calcalattng

rdati. quotient you get represents days, minntes, and seconds. This number is the basis. If yon want to know at what time in the year iu question the sun enters any one of the twelve signs, yon look out the sign in the following table. Take the number which you find side by side with the sign in question, and add it to the basis, days to days, miontes to minutes, seconds to seconds. If the wholes amount to 7 or more, disregard them, and with the remainder

CHAPTER LXXVII. 189

count off the week-days, beginning with the beginning of Sunday. That time yon arrive at is the moment of samkrânti.

Whut murt be added to the Baris. The Zodiacal Signs. Days, Ohați Casbaka.

Aries 19 . Taurns 17 0 Gemint . 2 Cancer 43 . 6 21 Leo . 2 0 . Virgo 49 0 Libra 5 49 14 0 Scorpio Arcitenens . 3 6 30

Capricornus 34 30 5 54 Amphora. 0 . . Piscee 30 0 20

The beginning of consecutive solar years in the week On the differs by I day and the fraction at the end of the the solar length uf

year. This amount, rednced to fractions of one kind, Ing to Brah- is the multiplicator (180), used in the preceding com. Pullen, and putation in order to find the surplus of each year (ie. Aryabbate.

the amount by which its beginning wanders onward through the week). The divisor (143) is the denominator of the fraction (which is accordingly 129). Accordingly the fraction at the end of the solar year is, in this computation, reckoned as , which implies as the length of the solar year, 365 days 15' 31" 28"" 61v. To raise this fraction of a day to one whole day, 10g of a day are required. I do not know whose theory this is. If we divide the days of a caturyuga by the number of its solar years, according to the theory of Brahma- gupta, we get as the length of the solar year, 365 days 30' 22" 30" oh. In this case the multiplicator or gunakara is 4027, and the divisor or bhagahdra is 3200 Page 293. (ie. I day 30' 22" 30"" o'r are equal to 9933

190 ALBERUNI'S INDIA.

Reckoning according to the theory of Pulisa, we find as the length of the solar year 365 days 15' 31" 30" of. Accordingly, the gunakara would be 1007, the bhaga- hara 800 (Le. I day 15' 31" 30" oh are equal to 1oa'). According to Aryabhsta, the length of the solar year is 365 days I5' 31" 15". In that case the gunakdra is 725 and the bhagahara is 572 (i.e. I day 15' 31* 15" are equal to ;)- Anotber Another method for finding the moment of samkrdnti mothod for onding tbe has been dictated to me by Auliatta (7), the son of Sa- hdwt (?), and is based on the system of Pulisa It is this: Subtract from the Sakakâla 918, multiply the re- mainder by 1007, add to the product 79, and divide the sum by 800. Divide the quotient by 7. The remainder you get is the basis. Whst now must for each sign be added to the basis, as has already been mentioned (ii. 188), is indicated by the following table opposite to each sign :

What must bo What must be sdded to the udded to the The Zodiacal Sigoe. Baris, The Eodiacel Sipnr. Baste.

Dayı. ,Daya. Ghat1.

Aries . 35 Libra 6 31 Tsurus 4 33 Scorpio 1 23 Gemini 0 Arciienens. Cancer 39 2 . Capricorntis 10 Leo 4 34

Virgo . Amphora 4 . I 34 6 Pisces. 5 . 4 28

Varahamihira maintains in the Pancasiddhantika Ehadartti- that the shadafttimukha is in the same degree pro- pitious as the time of samkrdnti for acquiring in it infinite heavenly reward. This is the moment of the sun's entering :- The 18th degree of Gemini; the 14th degree of Virgo; the 26th degree of Arcitenens; and the 28th degree of Pisces. The moment of the sun's entering the fixed signs

CHAPTER LXXVII. 191

is four times as propitious as the moment of his entering the other signs. For each of these times they compute the beginning and the end by meane of the radius of the sun in the same way as they compute the minntes of the enn's or moon's entering and leaving the shadow at an eclipse. This method is well known in their canones. We, however, communi- cate here only those of their methods of calculation which we think remarkable, or which, as far as we know, have not yet been explained before Muslim ears, as Mnslims know of the methods of the Hindus only those which are found in the Sind-hind. Most propitions times are, further, the times of solar Times of and lunar eolipses. At that time, according to their celipeen,

belief, all the waters of the earth become as pure as that of the Ganges. They exaggerate the veneration of these times to such a degree that many of them commit suicide, wishing to die at such a time as promises them heavenly bliss. However, this is only done by Vaisyas and Sudras, whilst it is forbidden to Brahmans and Kshatriyas, who in consequence do not commit suicide (vide, however, ii. 170). Further, the times of Parvan are propitious, i.e. those Parran and tiies in which an eclipse may take place. And even yoga.

if there is no eclipse at such a time, it is considered quite as propitious as the time of an eclipse itself. The times of the yogas are as propitious as those of the eclipsee. We have devoted a special chapter to them (chap. Ixxix.). If it happens within the course of one civil day that Uulucky the moon revolves in the latter part of aome etation, then enters the following station, proceeds throngh the whole of it and enters a third station, so that in one single day she stande in three consecutive statione, such a day is called trihaspaka (1), and also triharkasha (1). Page 24. It is an nnlucky day, boding evil, and it is counted among the punyakala, (See ii. 187.)

192 ALBERUNI'S INDIA.

The same applies to that civil day which compre- hends a complete lunar day, whose beginning, besides, falls in the latter part of the preceding lunar day, and whose end falls in the beginning of the following lunar day. Such e day is called trahagattata (?). It is unincky, but favourable to earn in it e beavenly reward. When the days of unardtra, i.e. the days of the de- crease (see ii 25), sum up so as to form one complete day, it is unlacky and reckoned among the punya- kdla. This takes place according to Brahmagupts in 6212:911 civil days, 62.1fir solar days, 63:1 ::: Innar days. According' to Pulisa, it takes place in 62;3:11; civil days, 631:1': lunar days, 62.i.4f, solar days. The moment when a complete leap-month without any fraction is summed up, is unlucky, atd is not reckoned among the punyakdla. According to Brahma- gupta, this takes place in 990,8:11f civil days, 976.811 solar days, 1006,f7 lunar days. Times which are considered as unlucky, to which no of carth- merit whatsoever is attributed, pre, e.g., the times of earthquakes. Then the Hindus beat with the pots of their households against the earth and break them, in order to get a good omen and to banish the mishap. As times of s similar ill nature, the book Samhitd further enumerates the momenta of landalips, the fall- ing of stars, red glow in the sky, tho combustion of the earth by lightning, the appearance of comets, the occurrence of events contrary both to nature aud custom, the entering of the wild beasts into the villages, rainfall when it is not the season for it, the trees putting forth leaves when it is not the season for it, when the nature of one season of the year seems trans- ferred to another, and more of the like. The book Sradhara, attributed to Mahadeva, says the following:

CHAPTER LXXVII. 193

"The burning days, ia the unlucky ones-for thns guotation they call them-are: from the book &t- "The second days of the white and black halves of Mardrn. dáara of

the months Caitra and Pausha; "The fourth days of the two halves of the months Jyaishtha and Phalguns; "The sixth days of the two halves of the months Śråvana and Vaiśakha; "The eighth days of the two balves of the months Åshâdha and Aávayuja; "The tenth days of the two halves of the months Mårgadirsha and Bhadrapads; "The twelfth daye of the two halves of the month Kârttika."

TOL IL.

( 194 )

CHAPTER LXXVIIL

ON THE KARANAS.

Expienetka WE have already spoken of the lunar days called tithi, and have explained that each lunar day is ahorter than a civil day, because the lunar month has thirty lunar days, bnt only a little more than twanty-nine and a half civil days. As the Hindus call these tithis nychthemera, they also call the former half of a tithi day, the latter half night Each of these halves has a separate name, and they all of them (i.c. all the halves of the lunar days of the lunar month) are called karanas. Fized and Some of the names of the karanas occur only once morable in a month and are not repeated, viz four of them about the time of new moon, which are called the fixed ones, because they occur only once in the month, and because they always fall on the same day and night of the month. Others of them revolve and occur eight times in a month. They are called the movable ones, because of their revolving, and becanse each one of them may as well fall on a day as on a night. They are seven in number, and the seventh or last of them is an unlncky day, by which thoy frighten their children, the simple mention of which makes the hairs on the head of their boys stand on end. We have given an exbaustive description of the karanas in another book of ours, They are mentioned in every Indian book on astronomy and mathematics.

CHAPTER LXXVIII. 195 If yon want to know the karanas, first determine the Rute bow ta lunar days, and find out in what part of them the date toroyu. and the

in question falls, which is done in this way :-- Subtract the corrected place of the sun from the corrected place of the moon. The remainder is the distance between them, If it is less than six zodiacal signs, the date falls in the white half of the month; if it is more, it falls in the black half. Reduce this number to minntes, and divide the pro- duct by 720. The quotient represents tithis, ie. com- plete lunar days. If you get by the division a re- mainder, multiply it by 60 and divide the prodnct hy the mean bhukti. The quotient represents ghatts snd minor fractions, ie. thet portion of the current day which has already elapsed. This is the method of the canones of the Hindus. The distance between the corrected places of sun and moon must be divided by the mean bhukti. This, how- ever, is impossible for many of the days. Therefore they divide this distance by the difference between the daily revolutions of sun and moon, which they reckon for the moon as 13 degrees, for the sun as I degree. It is a favourite metbod in rules of this kind, especi- ally in Indian ones, to reckon by the mean motion of sun and moon. The mean motion of the sun is sub- tracted from the mean motion of the moon, and the remainder is divided by 732, which is the difference between their two middle bhultis. The quotient then represents days and ghafts. The word buht is of Indian origin. In the Indian Eaplana- language it is bhukti (= the daily motion of a planet). Mattt. If the corrected motion is meant, it is called bhukti sphua. If the mean motion is meant, it is called bhukti madhyama, and if the buht which renders equal is meant, it is called bhuktyantara, i.c. the difference between the two bhuktis.

ALBERUNIS INDIA.

The lonar days of the month have apocial names, ddde Which we exhibit in the following diagram. If you know the lunar day in which you are, you find, by the side of the number of the day, ita name, and opposite it the karana in which you are. If that which has elapsed of the current day is loss than half a day, the karaşa is a diurnal one; if that which has elapsed of it is more than half a day, it is a nocturnal one. This is the diagram :-

The white balt The black half. The karaya are comamoi to both balves.

Their namạs, Their namen. Their names. Their names, In daytlme In the nighh of the days. of the daya, The bumber The nnmber The number o the daya. of the days.

CHAPTER LXXVIII. Amávasyâ 0 0 O Catusbpada. Naga. o Barkhu. Kinstughna. Bava. 0 0 0 O

Biya. Navin. 17 Barkhu. 24 Attn. Balava. Kanlava. 3 I0 3 Triya. Dahîn. I8 Biya. 25 Navin. Taitila. Gara. 4

Caut. I2 Yaht. 19 Triya. 26 Dahtn. Banij. Vishți. S

Panct. I3 Duvâhî. 20 Caut. 27 Yaht. Bava. BAlAva

Sat. Trobt. 21 Panci. 28 Duváht. Kanlara Taitila. 14

Satin. 15 Caudaht. 22 Sat. 29 Troht. Gara. Banlj.

Attn. 16 f Pornima ) 23 Satin. Viahti. BaYa pancahi. O o

197 0 0 O 30 Candaht. Viahti. Šakuni.

198 ALBERUNTS INDIA.

The Hindus attribute to some of the karanas domi- with thuir nants, as is their custom. Further they give rules showing what during each karans must be done or not, rules which are similer to collections of astrological prognostics (as to lucky or unlucky days, &c.). If we give here a second dingram of the karanas, we thereby simply mean to coufirm what we have said already, and to repeat a subject which is unknown among na Thus it is rendered easy to learn the subject, because learning is the fruit of repetition.

THE FOUR FIXRD KARANAS.

Thelr Thelr they All dorninants. of the mooth

Favourable for the action of medicines. of drugs against the bite of nerpente, of in- cantations, of learning, of conncil-hold- ing, and of reciting boly texts before tho balf. Kali. idols. Śakuni In the black

Favourable for placing a king oa a throne, giving alms in the name of the Fathers, for making ase of four-footed animals in Agriculturo. Tha rodiacal Catushpada sigh Taurws.

Favourable for weddings, laying a founda- tion-stone, ozamining the state of snake- bitten pornona, for frightoaing people and seizing them. The snake. Iu the white half. Ruins all actiom and is favourable only for things connected with marriage, for the constraetion of pararols, the pioreing of the ear, and for works of pioty. The wind. Kinstughna.

CHAPTER LXXVIII. 199

THE SEVEN MOVABLE RARANAS

The pranostica of the karanas, apd for what thing cach of them it favourablo. Their Thetr they fatL dominanta of the month In whjch balf

When there is a camkrânti in this karana, it is sitting, and the fruita will, during it, suffer some mishap. It is favourable for travelling, for beginning with thing whioh are intended to last long, for cleaning oneself, for compoundiog the drugs which Bara. make the women fat, and for the sacrificea Śukra.

which the Brahmans offer to the fire.

When there is a samkranti in it, it is sitting, not good for the fruits. It is favourable for the affairs of future life, and for ac- quiring & heavenly reward. BAlava. Brahmau.

When thero is a samkranti in it, it ia stand. ing. All that is sown in it will prosper and drop with sneeulence. It is favour- able for making friendshlps with people. Mitra. Kaulava.

When there is a samkrânti in it, it is sretcked on the ground. It indicates that the prices will akk, and is favourable for the kneading of aromatic unguents and the compounding of perfumes. Taitila. Aryaman.

Wheo there is a samkranti in it, it is dretched on the ground. Itindicates that the prices will be depressed, and is favour. able for sowing and laying the founda- Both in the white and the black halves.

tion-stone of a building. Gara. Parvata.

When there is a samkrnti in It, It is stand- ing. All corn will prosper (tacuna), and is favourable for commerce. Śrt. Banij.

When there is a samkrdnti in it, it ie stretched on the ground. It indicates that the prices will be insufficieot. It is oot favourable for anything save the crush- ing of tha sugar-cane. It is considered as Vishți. unlacky sod is not good for travelling. Marut.

20O ALBERUNIS INDIA.

Bole for the If you want to find the karanas by compnitation, cotapats- tion of the subtract the corrected place of the sun from that of Paga 197- the moon, reduce the remainder to minntes and divide the number of them by 360. The quotient represents complete karanas. What remains after the division is multiplied by 60, and divided by the bhuktyantara. The quotient re- presents how mnch has elapeed of the current karana. Every unit of the number is equal to half a ghaft. We now return to the complete karanas. If they are two or less, you are in the second karana. In that case yon add one to the number and count the sum off, beginning with catushpada, If the number of karanas is 59, you are in sakuni. If it is less than 59 and more than two, add one to them and divide the sum by seven, The remainder, if it is not more than seven, count off, beginning with the beginning of the cycle of the movable karanas, i.c. with bava. Thereby you will arrive at the zame of the current karana in which you happen to be. The Larațas Wishing to remind the reader of something relating se borrowod by Alkindt to the karanas which he perhaps has forgotten, we and other Arb must tell him that Alkindt and others like him have sutbor. hit upon the system of the karanas, bnt one which was not sufficiently explained. They did not comprehend the metbod of those who use the karanas. At one time they trace them back to Indian, another time to Babylonian origin, declaring all the time that they are altered on purpose and corrupted by the inadvertence of the copyists. They have invented a calculation for them which proceeds in a better order than even the original method itself. But thereby the thing has become something totally different from what it origin- ally was. Their method is this: they count half days, beginning with new moon. The first twelve hours they regard as belonging to the sun, as burning, ie. unlucky, the next twele hours as belonging to Venus, the

CHAPTER LXXVIII. 201

following twelve hours as belonging to Mercury, and so on according to the order of the planets. Whenever the order returns to the sun, they call his twelve houra the hours of Albist, i.e. vishti. However, the Hindas do not measure the karanas by civil, but by lunar days, nor do they begin with those burning hours following npon now moon, Accord- ing to the calculation of Alkindt, people begin, after new moon, with Jupiter; in that case the periods of the sun are not burning. On the other hand, if they begin, according to the method of the Hindus, after new moon with the sun, the hours of vishti belong to Mercury. Therefore, each method, that of the Hindus and that of Alkindi, must be treated sepa- rately. Because vishti recurs eight times in a month, and because the points of the compass are eight, we shall exhibit in the eight fields of the following table their åorpoxoyoupera regarding the karanas, observations the like of which are made by all astrologers regarding the shapes of the planets and regarding those stars which rise in the single third parts of the zodiacal signs. Puge 298.

DESCRIPTION OF THE BIMOLE " VISHTIA." riso. riehtis. fall. Names of tha Their natdes Tbe diractions in whiob they In what part of the conth they book Srudhava. Their Dncsbars. according to the

It haa three eyes The hair on its head is like growing sngar-cane. In one band it has an iron hook, in the other a black serpent. It is strong and violent like run- I. ning water. It has a long ... East. tithi. tonguo. Its day is only good for war, and those actions in which there is deception VadavAmnkha. and falsitication. In the night of the 5th

202 ALBERUNPS INDIA.

DESORIPTION OF THE SUGLE tAlL. Name of the The direotione in which they Thelr paaet the mobth they Their num bert. zooording to the In what part of book Brddkare.

It is green, and bas a sword in its hand Its place is in the lightning, thundering, stormuy, and cold cloud. Its time is favourable for tear- ing out fattening herbe, for ... .II. drinking medicine, for com- Blv (1). AisAna. merco, and for casting gold in a monld. In the day of the gth

It has a black face, thick lipe, busby eyebrows, long hair of the head. It is long, and rides during its day. In the hand it bas a sword, it is intent upon devonring men, it emita fire from its mouth, and eays bd bd &d Its time is only good for fighting, for III. North. Ghora. Ghora. killing misereants, for our- ing ill people, and for fetch. ing serpents ont of their boleo. In the night of the 12th tithi.

It has five faces and ten eyes. Its time is favourable for punishing rebols, for divid- ing the anny into single corpe. During it a man must not turn with his face IV.

towards the direction where VAyava. 16th tithi. Krala (!). it rises. In the day of the

It is like a smoky flame. It has three heads, in each three eyes torned apeide down. Its hair is standing on end. It sits on the hend of a human being, it screams v. ... like thunder. It is angry, West. devours men It holds in Jwala (!). one hand a knife, in the other an axe. In the night of the 1gth

CHAPTER LXXVIIL 203

DESCRIFTION OF TEE BINGLE rin. Sridkarg. Names of the Thetr namer in what part of in which they Thetr bumabera. The directione the month they Cocording to the

It is white, has three eyes, and rides on an elephant, which always remains the same. In the one hand be has a huge rock, in the other a vajra of iron, which it throwa. It destroys the cattle over which it rises. He who makes war coming from the direction whence ... VI. it rises will bo victorions. Nuirrita. A man must not tura with his face towarde it wheo tearing ont fattening herbe, digging ont treasures, and trying to satisly the wants In the day of the 23d tithi.

of life.

It has the colour of crystal. In one band it bolds a three- fold parafadha, and in the otber a rosary. It looks towards heaven, and says hd ha Ad. It rides on an ox. Ita time is favourable ... VIL tithi. for handing over the chil- South.

dren to the schools, for con- KAlarttri.

cinding peace, giving alma, and works of piety. In the night of the 26th

It is pistachio-coloured like a parrot. It looks like some- thing globular, and has three eyes In one band it has a mace with an iron hook, in the other a sharp diseus. It sita on ita throne, frightening people, and say- ... ... ing ad ad ad. lts timo is VIIL Ågneya. hot favourable for beginning anything. It is only yood for doing servico to relacions and for house-work. In the day of the 3oth tifhi.

( 2044 )

CHAPTER LXXIX.

ON TRE YOGAS.

Page s99. THESE are times which the Hindus think to be most nnlucky and during which they abstain from all action. They are numerous. We shall here mention them. There are two yogas regarding which ell Hindus sod mizi- agree, viz :- rite. (1.) The moment when sun and moon together stand on two circles, which are, as it were, scizing each other, ia each pair of cireles, the declinations of which, on one and the same side (of either solstice), are equal. This yoga is called vyatipata. (2.) The moment when sun and moon stand together on two egual circles, ie. each pair of circles, the de- clinations of which, on different sides (of either solstice), are eqnal, This is called vaidhrita. It is the signum of the former that in it the sum of the corrected places of sun and moon represents in any case the distance of six zodiacal signs from O° of Aries, while it is the signum for the latter that the same sum represents the distance of twelve signs. If yon compute the corrected places of sun and moon for a certain time and add them together, the sum is either of these signa, ic. either of these two yogas. If, however, the sum is less than the amount of the signum or larger, in that case the time of equality (i.c. the time when the sum is equal to either of the sirna) is computed hy means of the difference between this sum and the term in question, and by means of the

.CHAPTER LXXIX. 205

eum of the two bhulti of sun and moon instead of the bhuktyantara, in the same manner as in the canones the time of full moon and opposition is computed. If you know the distance of the moment from noon On middle or midnight, whether yon correct the places of sun and tme.

moon according to the one or the other, its time is called the middle one. For if the moon followed the ecliptic as accurately as the sun, this time would be that which we want to find. However, the moon deviates from the ecliptic. Therefore, she does not at that time stand on the circle of the sun or on the circle which, as far as observation goes, is equal to it For this reason the places of sun and moon and the dragon'e head and tail are computed for the middle time. According to this time they compute the declinations Method for of sun and moon. If they are equal, this is the time wyattpdta computing

which is sought for. If not, yon consider the declina- rita. and vaidh-

tion of the moon. If, in computing it, yon have added her latitude to the declination of the degree which she occupies, you subtract the latitnde of the moon from the declination of the sun. However, if, in computing it, you have subtracted her latitude from the degree which tho moon occupies, you add her latitude to the declination of the sun. The result is reduced to ares by the tables of the kardajat of declination, and these arcs are kept in memory. They are the same which are used in the canon Karaņatilaka. Further, you observe the moon at the middle time. If she stands in some of the odd quarters of the ecliptic, ie. the vernal and autumnal ones, whilst her declina- tion is less than the declination of the sun, in that case the time of the two declinations equalling each other -- and that is what we want to find-falls after the middle, ia the future one; bnt if the declination of the moon is larger than that of the sun, it falls before the middle, ia the past one.

206 ALBERUNTS INDIA.

If the moon stands in the owa quarters of the ecliptic (ie the summer and winter quarters), just the reverse takes place. Another Puliss adds together tho declinations of sun and mthed by moon in eyattodta, if they stand on different sides of the colstice, and in vaidhrita, if they stand on the same side of the solstice. Further, he takes tho difference between the declinations of aun and moon in vyattpdta, if they stand on the same side, and in vaidhrita, if they stand on different sides. This is the firat value which is kept in memory, ic. the middle time. Further, he reduces the minutes of the days to mdshas, supposing that they are leas than one-fourth of a day. Then he computes their motione by means of the bhukti of sun and moon and the dragon'a head and tail, and Le computes their places according to the amoant of middle time, which they occupy, in the past and the fnture. This is the second value which is kept in memory. By thia method he manages to find out the condition of the past and the future, and compares it with the middle time. If the time of the two declinations equalling each other for both sun and moon is paat or future, in that case the diffcrencs between the two valnes kept in memory is the portio divisionis (divisor); but if it is past for the one and future for the other, the sum of the two values kept in momory is the portio divisionis. Page 3oo. Further, ho multiplies the minutes of the days, which have been found, by the firat value kept in memory, and divides the product by the portio divisionis The quotient represents the minutes of the distance from the middle time which minntes may either be past or future. Thas the time of the two declinations equalling each other becomes known, The author of the canon Karanatilaka makes us return to the arc of the declination which has been

CHAPTER LXXIX. 207

kept in memory. If the corrected place of the moon Anothor is less than three zodiacal signs, it is that which we thesntr want; if it is between three and six aigns, he subtracts pattols. of tho Etrt- it from six signs, and if it is betwesn six and nine signs, he adds six signs thereto; if it is more than nine signs, he subtracts it from twelve eigna. Thereby he gets the second place of the moon, and this he compares with the moou's place at the time of the correction. If the second place of the moou is less than the first, the time of the two declinations equalling each other is future; if it is more than the first, the time of their equalling each other is past. Further, he multiplies the difference between the two places of the moon by the bhukti of the sun, and divides the product by the bhukti of the moon. The quotient he adds to the place of the aun at the time of the cor- rection, if the second place of the moon is larger than the first; but he subtracts it from the sun'a place, if the second place of the moon is less than the first. Thereby he finds the place of the sun for the time when the two declinations are equal to each other. For the purpose of finding it, he divides the difference between the two places of the moon by the bhukti of the moon. The quotient gives minutes of days, indi- cative of the distance. By means of them he com- putes the places of sun and moon, of the dragon'a head and tail, and of the two declinations. If the latter are equal, it ia that which we want to find, If they are not equal, the author repeats the caloulation so long till they are equal and till the correct time has been found, Thereupon he computes the measure of sun and moon. However, he disregards half of the sum of them, so that in the further calculation he uses only the one half of their measures. He multiplies it by 60 and divides the product by the bhuktyantara, The quotient represents the minutes of the falling (pâta !)

208 ALBERUNPS INDIA.

The correct time, which has been found, is marked in three different places. From the first number he subtracts the minutes of the falling, and to the last number he adds them. Then the first number is the time of the beginning of oyatipata or vaidhrita, which- ever of the two you want to compute The second number is the time of its middle, and the third number the time of its end. The author We have given & detailed account of the bases on books oo che. subject. which these methods rest in a special book of ours, called Khaydl-alkusdfain (i.c. the image of the two eclipses), and have given an accurate description of them in the canon which we have composed for Sydva- bala (7), the Kashmtrian, and to which we have givan the title The Arabic Khandakhadyaka. About the Bhattila (?) thinks the whole day of either of these torus being ualmaky. two yogas to be unlncky, whilst Varahamibira thinks only that duration of them to be unlucky which is found by the computation, He compares the unlncky portion of the day to the wound of a gazelle shot with a poisoned arrow. The disease does not go beyond the environs of the poisoned shot; if it is cut out, the injury is removed. According to what Pulisa mentions of Parafara, the Hindus assume a number of vyattpatas in the lunar etations, bnt all of them are computed by the same method which he has given. For the calculation does not increase in its kind; only the single apecimens of it become more numerous. Quotation The Brahman Bhattila (7) says in his canon :- troao Bba'y "Here there are 8 times, which have certain gange- measures. If the sum of the corrected places of sun and moon is equal to them, they are unlucky. They are: " 1. Bak-shuta (1). Its gauge-measure is 4 zodiacal signs. "2. Ganddnta, Its gange-measure is 4 signs and 13} degrees.

CHAPTER LXXIX. 209

"3. Lata (?), or the general vyatipdta. Its gange- measure is 6 signa. "4 Casa (7). Its gauge-measure is 6 signs and 63 degrees. "5. Barh (?), also called barhuyatipata Its gauge- measure is 7 aigns and 163 degrees. "6. Kaladanda. Its gauge-measure is 8 signs and 13} degrees. "7. Vydkshata (?). Its gauge-measure is 9 signs and 23} degrees. "8. Vaidhrita. Its gauge-measure is 12 signs." These yogas are well known, but they cannot all be traced back to a rule in the same way as the 3d and 8th ones Therefore they have no certain duration determined by minutes of the falling, but only by general estimates. Thus the duration of vydkshdta (1) and of bakshuta (1) is one muhurta, according to the statement of Varahamihira, the duration of Gandanta and of Barh (1) two muhurtas. The Hindus propound this subject st great length and with much detail, but to no purpose. We have given an account of it in the sbove-mentioned book. (See ii. 208.) The canon Karanatilaka mentions twenty-seven Twenty- yogas, which are computed in the following manner: teren yogas according to Add the corrected place of the sun to that of the tilata the Karana- moon, reduce the whole sum to minutes, and divide the Pago 301. number by 800. The qnotient represents complete yogas. Multiply the remainder by 60, and divide the product by the aum of the thuktis of sun and moon. The quotient represents the minutes of days and minor fractions, viz, that time which has elapsed of the cur- rent yoga. We have copied the names and qualities of the yogas from Sripala, and exhibit them in the following table :-

VOL. II. 0

210

TABLE OF TBE TWENTY-SEYEN "TOGAS."

The Whether The Whether Tbe food or bad. Dumber. Their nsiam. Whether good or Led, ber. Their namm. rood or bad. burober. Thetr barte

ALBERUNTS INDIA. Vishkambba. Good, 10 Ganda. Bad. 19 Parigha Bad.

Priti. Good. 11 Vriddhi. Good. 20 Šiva. Good.

Bad. Siddba 3 Rajakama (?) 12 Dhruva. Goed. 21 Good.

SaubhAgya. Good. 13 Vykghâta (7) Bad 22 SAdhya. Middling. 4

Sobhana. Good. 14 Harabaņa. Good. 23 Bubha. Good. 5

6 Atigaņda. Bad Vajra. Bad 24 ģukra. Good.

7 Sukarman Good. 16 Biddhi: Good. 25 Brahman. Good.

Dhriti Good. 17 K-n-n-Ata (?) Bad. 26 Indra. Good.

Bad. t8 Variyas Bad. 27 VaidhritL Bad.

( 211 )

CHAPTER LXXX.

ON THE INTRODUCTORY PRINCIPLES OF HINDU ASTRO- LOGY, WITH A SHOBT DESCRIPTION OF THEIR METHODS OF ASTROLOGIOAL CALCULATIONS.

OUR fellow-believers in these (Muslim) countries are Indian not acquainted with the Hindu methods of astrology, unknown satrology and have never had an opportunity of studying an Mubam- Indian book on the subject In conseqnence, they "dnm imagine that Hindu astrolugy is the same as theirs and relate ell sorts of things as being of Indian origin, of which we have not found a single trace with the Hindus themselves. As in the preceding part of this our book we have given something of everything, we shall also give as much of their astrological doctrine as will enable the reader to discuss questions of a similar nature with them. If we were to give an exhaustive representation of the subject, this task would detain us Pago 302. very long, even if we limited ourselves to delineate only the leading principles and avoided all details. First, the reader must know that in most of their prognostics they simply rely on means like anguring from the flight of birds and physiognomy, that they do not-as they ought to do-draw conclusions, regarding the affairs of the snblunary world, from the seconds (sic) of the stars, which are the events of the colestial sphere. Regarding the number seven as that of the planets, On the there is no difference between us and them. They call Panot. them graha. Some of them are throughout lucky, viz

212 ALBERUNTS INDIA.

Jupiter, Venus and the Moon, which are called saum- yagraha. Other three are throughout unlucky, viz. Saturn, Mars, and the San, which are called kruragraha. Among the latter, they also count the dragon's head, though in reality it is not e star. The nature of one planet is variable and depends upon the nature of that planet with which it is combined, whether it be lucky or unincky. This is Mercury. However, alone by itsolf, it is lucky. The following table represents the natures of the seven planeta and ovarything else concerning them :-

Name of the planats. Sun. Moon. Mars. Meroury. Jupiter. Venus. Saturn.

Whethor they are Unlncky. Lacky, bat de- Unlucky. Lucky, when Lucky. Lucky. Unlnoky. lucky or unlucky. fpending upon tho planet noar ber. it is alone. Yiddilng in the Else depond- Art, lueky In the socond, aud ing opon the unlucky in the nature of the Last ten days of planet near it. the moouth

What elemants Fire. Bartn Heaven. Water. Wind. CHAPTER LXXX. ... they indicate.

Whether they in- Male. Female. Male. Neither male Male. Female. Neither male dicate male or nor female. nor female. femalo beings.

Whether they in. Day. Night. Night. Day and Day. dicate day or zight to- Day. Night.

night gether. What point of the Eest. North-west. Sonth. North, North-east. Between east West. compass they and west. indicate.

What colour they Bronze- White. Light red Piatachio- Gold-coloar. Many Black. indicate. colour. green, coiours.

What time they Ritu, ic. a indicate. Ayana Muhurta. Day. Month Paksha, i.e. Year. 213 sixth part of the year. half a mozth.

Name of the plancts. Sun. Moon. Mars. Mercury. Jupiter. Venus. Saturn, 234 What season they 0 Veraha. Grtshma. Bard. Hemanta. Vasanta śiáira. indicate.

What taste they Bitter. Saltish A mixtnre of Sweet. indicata. ... all tastes, ...

What material Bronze. Gold. they indicate. Crystal Small pearls. Silver, or If ALBERUNI'S INDIA. the constella- Pearl. Iron.

tion is very strong, gold. What dress and Thick. New. Barned. Wet from Between new Whole. and shabby. Barned. clothes they water. indicate.

What angel they Nema (1). Ambn, the Agni, the Brahman, Mahadeva. Indra. ... ìndicate. water. Bre Page What carte they Kshatriyas Vailyas and Śtdraa and Brahmans indicate. Kshatriyas Brahmans commanders and general .. and minis- ... and com- princes. and minis- mandera, tera. ters. Which Veda they indicate. O Stmaveda. Atharvaņa- Rigveda. Yajurveda. veda. The months of The fourth The fifth The second The soventh The third The Arst The sixth pregnancy. month, in wbich the month, in which the mnonth, in

bones become skin appears. which the month, in which the month, in

child becomes which the month, in

bard. embyro at. talna consist. Himbs begin which the month, whon the hair

complete, and to branch off semen and the menstrua] grows.

eucy. recoives the blood become memory. mixod.

Character as hased Satya Satya. Tamas, Rajas, Satya. Rajas. Tamas. on the three primary forces.

Mitra. Friendly Mercury. Japiter, Sun, Ban, Venua Son, Moon, Saturn, Venns, planets. Jupiter, Mars, Sao, Mooa. Moon. Mars. Mercury. Moroury.

Šatru. Hostile Saturn, There is no Merenry. Moon. Venus, Sun, Moon. Mars, Bun, planets. Venus. plauet hostile Mercury. Moon, to her. CHAPTER LXXX. Vi- SIndifferent Mercary. Saturn, Jupi- Venus, Saturn. Baturn, Japiter, Satarn. Jupiter, miśra. ? planets. ter, Mars, Mars. Mars. Japiter.

Venus.

What parts of the The breath The root of the' The flesh Vaice and Intellect and Semen, Sinows, flesb, body they indicate. and the tongue and skin. the blood. and brain fat. and pain. bones.

The scale of their 1 2 6 4 25 (1) 7 magnitude.

Years of shadaya. 19 25 15 I2 15 21 20

Years of nai- 20 2 9 18 20 5o s argka.

215

216 ALBERUNI'S INDIA.

Erplass- The column of this table which indicates the order tory Doteo to the of the aize and power of the planets, serves for the following purpose :- Sometimes two planete indicate exactly the same thing, exercise the same influence, and stand in the same relation to the event in question. In this case, the preference is given to that planet which, in the column in question, is described as the larger or the more powerful of the two. The montha The column relating to the months of pregnancy is to of preg- be completed by the remark that they consider the eighth month as standing under the influence of a horoscope which causes abortion. According to them, the embryo takes, in this month, the fine substances of the food. If it takes all of them and is then born, it will remain alive; bat if it is born before that, it will die from some deficienoy in its formation. The ninth month stands under the influence of the moon, the tenth under that of the sun. They do not speak of a longer duration of pregnancy, but if it happens to last longer, they believe that, during this time, some injury is brought about by the wind. At the time of the horoscope of abortion, which they determine by tradition, not by calculation, Pago 304 they observe the conditions and influences of the planets and give their decision accordingly as this or that planet happens to preaide over the month in question, The qnestion as to the friendship and enmity of the of tho and mntty planets among each other, as well as the influence of Friandship

plansta. the dominus domds, is of great importance in their astro- logy. Sometimes it may happen that, at a particular moment of time, this dominium entirely loses its original character. Further on we shall give a rule as to the computation of the dominium and its single years. The sodinen3 There is no difference between us and the Hindus regarding the number twelve as the number of the signs of the ecliptic, nor regarding the manner in which the dominium of the planets is distribnted over them. The following table shows what qualities are peculiar to each zodiacal sign as s whole :-

The Zodiaca! Tourns. Gemini. Arci- Capri- Am- Signs. Aries. Cancer. Leo. Virgo. Libra. Scorpio. teneas. phora. Pinces.

Their doml- Mar, Vonus. Moroury. Moon. Sun. Meroury. Venus. Mart. Jopiter. Satarn, Saturn, Jupitar. Dants

(Degroms, 10 3 15 90 tudea Altitade. San. Moon. 0 Jupiter. Mercury. Saturn. Hars. 0 CHAPTER LXXX. Dominants of Mara Moon. O Suo. Moroury. Venus. Jupiter. o BatorD. the mAlatritona,

Whether male Female. Male. Female. Male. Female. Malo, Female. Male. Femalo. Malo. Femalo, or female.

Whetber incky Unlueky. Lueky. Unlucky. Lucky. Unlucky. Lucky. Unlncky. Luoky. Unlucky. Lucky. Unluoky. Lucky. or unlud y.

The colours. Reddisb. White. Green. Yollow- Gray. Many Blaok. Golden. Striped Brown. Dust inh. coloured. White oolonrod. and blaok,

The directiona. Dua B.S.B. W.8.W. N.N.W. E.N.E. Due Due Due E.S.B. S.8. W. W.N.W. N.N.B. conth. Weit north.

NIn what manner Stretched Stretched Lying on Stretched Standing Standing Standing Standing Stretched Stretched Standing | Standing they rio, ou the on the the nde. on the ereot. Oreot. creot. crect. on the ou the erect. creat, grotnd, ground. ground. ground ground. 217

218 Th Xodinon! Ara. Tewrws. Cmmini Cancer. Firgo. Libra. Scorpio. Arci- Caprt piont.

Whather tarn- Moriag. Rartiog. Movins Moving. Rasting. Moving Movine Resting. Moving Moring. tac, fxed or And and and donals-bodied. roting resting together. toguther. togother. ALBERUNTS INDIA. Whether at At At At Dariag Daring Daring Dariog At At During Duriay might, or during night. night. nicht. aight. day. day. day. day. sight. abat, day. day. day, sooording to rotae prople.

What parts of Hond. Fhor. Shoal- Breast. Belly. Hip. Under Malo The The The the body thoy dors aDd the abd loins. Knoc. indicate. baods. female ginital

Beasons. Vasanta. Grithma. Grisbtan. Varhs. Varaba. Barad. garad. manta.

Their Agures. A ram. An ox. A man Orab. Lion. A hore, A kind Two vith a with an coospios. tho with the lyro, and ear of taos of of bost ad aad or burgs. a elab in oorn in apper bis band. her band. hilt of Taere is

bare bomss

.

What kinds of Ampbi- The Arst bions. Quadru- Biped. The The Arat bedngs they sre. Quadru- Qoadra- Homaa Amphi- Biped, Watery. nad ped. biped. pod. blogs, apper balf a half a balf a biped, biped, biped

the the Jatter etber lower balf a balf balf watary, quadru+ or the pod. whole a

bung. CHAPTER LXXX. The times of At During Duriog At During During Duriog The Dariog The Darias their strougest night. night. the the night. the day, the day. the human the bamaa the influence ac- day. mamhdhi mrbdbi. part dur- camhdhi purt in cuadhL cording to the ing the daytime, didarent kinds. day, tho other at night. otber at nigbt.

219

220 ALBERUNI'S INDIA.

The heighs or altitudo of a planet is called, in the Indian language, wccastha, ita particular degree paramoo- Mtrology. castha. The depth or dejectio of a planet is called nicastha, its particular degree paramantcastha. Mela- trikona is a powerful infiuence, attributed to a planet, when it is in the gaudium in one of its two houses (cf. ii. 225). They do not refer the aspectus trigoni to the elements and the elementary natures, as it is our oustom to do, but refer them to the points of the compass in general, as has been specified in the table They call the turning zodiacal sign (rporwov) carardái, ic moving, the fixed one (arepeor) sthirardsi, i.c. the resting one, and the double-bodica one (8lawua) dvisva- bhdva, i.c. both together. As we have given a table of the zodiacal signs, we next give a table of the houses (domus), showing the qualities of each of them. The one half of them above Pupo 306. tbe earth they call chatra, ic. parasol, and the half nnder the earth they call nau, ie. ship. Further, they call the half ascending to the midst of beaven and the other half descending to the cardo of the earth, dhanu, i.a the bow. The cardines they call kendra (sivrpor), the next following houses panaphara (èraradopa), and the inclining houses apoklima (anonMua) :-

CHAPTER LXXX. 221

What thay On the aspicts,

indioste, being taked as basta. in them. The Houss. shadow of noou, JOrs of the House. in fusnoo In them, yeara of the Houre. Which sodtaon dle the grestest induenes How they aro dirided tracted from the lucky Which plauots arercico How mnuoh ia to be sub- How much is to be sab- socording to the bortson. tracted fro the unlnoky divided according to tho Into what elasses they are

Hond and Basis for the The Mer- 0 noul. caleulation, huraan sigot. eury

Jopi- ter. Ascendens.

Pace and Two stand in 0 proparty. arpeot with the ascendens. II.

The two sras and brotbers. The ascendent Jooks towards Ascending bow. it, bat it does not look to- wards tho ascendens. Ship. Heart, parenta, Two stand in The Venas 0 friends, house, and joy. aspeot with the ascendens. watory and signa Moon. IV.

Bally, child, Two stand in And clever. aepoot with the arcendens. v.

The two sidos, tha enety It looks to- o

tad riding ward the ascendens, bat animala. tha ascendens does not look VI.

towards it.

Under the Two stand in haval and aspect with ¿ of toro. then. the ascendens. them. womon. VII. Desoending bow. Return aad Tho ascendens looks towards ... 0 desth. . it, bas it does not jook to- wards the VIII. arendens. ParasoL

The two loins, Two stand in ... 0 journey and debt. Apect with the ascedeRt. IX.

222 ALBERUNPS INDIA.

What thay On the ciputs, indlests. beny takeo n in ther, The Houssa, sbedow of Boon. Inducnoe ia the. year of the Houce yom of the Houss. oxeroise the grestact the gristest induenos How they are divided Whioh plansta czereics tracted fron the luoky tooording to the borisoa. How much la to be cub- How much le to be sub- divied aooording to the traoted from the upmeky Into what claaees thay ar The two kases Two stand in The and sction. spect with the ascendens. quadra- peds. X.

The twe O calves and It looks to- wards the axcendens, but the aucendens does not look towards it. ParasoL

The two fest Two do Bot O o The Ascending bow.

and expenees. stand io espeot wbole. vith the XII.

Page 307. The hitherto mentioned details are in reality the cardinal-points of Hindu astrology, viz the planets, zodiacal signs, and houses. He who knows how to find out what each of them means or portends deserves the title of a clever adept and of a master in this art. Ơn the Next follows the division of the zodiacal aigns in diviion of a todiacs1 minor portions, firat that in nimbahras, which are called nign fn hord, ic. hour, because half a sign rises in about an hour's time. The firat half of each male sign is unlucky as standing under the influence of the sun, because he produces male beings, whilst the second half is lucky as standing under the influence of the moon, because she produces female beings. On the contrary, in the female aigns the first half is lucky, and the second unlucky. .. In dret- Further, thers are the triangles, called drekkdna. There is no use in enlarging ou them, as they are simply identical with the so-called draijandt of our system. 3- In mgi- Further, the nuhbahrat (Persian, "the nine parts"),

CHAPTER LXXX. 223 called navdmfaka. As our books of introduction to the art of astrology mention two kinds of them, we shall here explain the Hindu theory regarding them, for the information of Indophiles. You reduce the distance between O' of the sign and that minute, the nuhbahr of which you want to find, to minutes, and divide the number by 200. The quotient represents complete nuhbahras or ninth-parts, beginning with the turning sign, which is in the triangle of the sign in question; you count the number off on the consecutive signs, so that one sign corresponds to one nuhbahr. That sign which corresponds to the last of the ninth-parts which yon have is the dominant of the nuhbahr we want to find. The first nuhbahr of each turning sign, the fifth of each fixed sign, and the ninth of each double-bodied sign is called vargottama, i.e. the greatest portion. Further, the twelfth-parts, called the twelve rulers. 4. Iu For a certain place within a sign they are found in the parta. twelfth.

following manner :- Reduce the distance between O° of the sign and the place in question to minutes, and divide the nomber by 150. The quotient represents complete twelfth-parts, which you count off on the following signs, beginning with the sign in question, so that one twelfth-part corresponds to one sign. The dominant of the sign, to which the last twelfth-part corresponds, is st the same time the dominant of the twelfth-part of the place in question. Further, the degrees called trinsdmsaka, i.c. the s In 30 thirty degrees, which correspond to our limits (or opia). aa. degrees or

Their order is this: The first five degrees of each male sign belong to Mars, the next following five to Saturn, the next eight to Jupiter, the next seven to Mercury, and the last five to Venus. Just the reverse order takes place in the female signs, viz the first five degrees belong to Venus, the next seven to Mercury, the next eight to Jupiter, the next five to Saturn, and the last five to Mercury.

224 ALBERUNPS INDIA.

These are the elements on which every astrological caleulation is based. On the The nature of the aspect of every sign depends upon knds of the the nature of the ascendens which at a given moment risea above the horizon. Regarding the aspets they have the following rule. A sign does not look at, Le. does not stand in aspetu with the two signs immediately before and after it On the contrary, each pair of signs, the beginnings of which are distant from each other by one-fourth or one- third or one-half of the circle, stand in aspect with each other. If the distance between two signs is one-sixth of the cirele, the signs forming this aspet are coonted in their original order; but if the diatance is five- twelfths of the circle, the signs forming the aspect are counted in the inverse order. There are various degrees of aspedts, viz :- The aspect between one sign and the fourth or eleventh following one is a fourth-part of an aspect; The aspect between one sign and the fifth or ninth following one is half an aspect; The aspect between a sign and the sixth or tenth following one is three-quarters of an aspect; The aspect between a sign and the seventh following one is a whole aspect. The Hindus do not speak of an aspeet between two planets which stand in one and the same sign. Friendehtp With reference to the change between the friendship of cortala and enmity of single planets with regard to each other, planata fn rlation to the Hindus have the following rule :- Cecb otber. If a planet comes to stand in signs which, in relation Pare 308. to its rising, are the tenth, eleventh, twelfth, first, second, third, and fourth signs, its nature undergoes a change for the better. If it is most inimical, it becomes mo- derated; if it is moderated, it becomes friendly; if it is friendly, it becomes most friendly. If the planet comes to stand in all the other signs, its nature undergoes a

CHAPTER LXXX. 225

change for the worse. If originally it is friendly, it becomes moderate; if it is moderate, it becomes ini- mical; if it is inimical, it becomes even worse. Under such circumstances, the nature of a planet is an acci- dental one for the time being, associating itself with its original nature. After having explained these things, we now proceed The four to mention the four forces which are peculiar to each each planot. planet :- I. The habitual force, called sthanabala, which the Laghujsta- planet exercises, when it stands in its altitudo, its house, IL & kem, ch.

or the house of its friend, or in the nuhbahr of its house, or its altitudo, or its mulatrikona, ie. its gaudium in the line of the lucky planets. This force is peculiar to sun and moon when they are in the lucky signs, as it is peculiar to the other planets when they are in the un- Incky signs. Especially this force is peculiar to the moon in the first third of her lunation, when it helps every planet which stands in aspect with her to ecquire the same force. Lastly, it is peculiar to the ascendens if it is a sign representing a biped. II. The force called drishtibala, i.e. the lateral one, Lgh il II. also called drigbala, which the planet exercises when etanding in the cardo in which it is strong, and, accord- ing to some people, also when standing in the two houses immediately before and after the cardo. It is peculiar to the ascendens in the day, if it is a sign representing a biped, and in the night, if it is a four-footed sign, and in both the samdhis (periods of twilight at the beginning and end) of the other signs. This in particular refers to the astrology of nativities. In the other parts of astrology this force is peculiar, as they maintain, to the tenth aign if it represents a quadruped, to the seventh sign if it is Scorpio and Cancer, and to the fourth sign if it is Amphora and Cancer. III. The conquering force, called ceshtabala, which Lagh. il s a planet exercises, when it is in retrograde motion, VOL, II. P

226 ALBERUNPS INDIA.

when it emerges from concealment, marching as a visible atar till the end of four signs, and when in the north it meets one of the planets except Venus. For to Venus the south is the same as the north is to the other planets. If the two ( --? illegible) stand in it (the south), it is peculiar to them that they stand in the ascending half (of the sun's annual rotation), pro- ceeding towards the summer solstice, and that the moon in particular stands near the other planets-except the sun-which afford her something of this force. The force is, further, peculiar to the ascendens, if its dominant is in it, if the two stand in aspect with Jupiter and Mercury, if the ascendens is free from an aspect of the unlucky planets, and none of them-except the dominant-is in the ascendens. For if an unlucky planet is in it, this weakens the aspect of Jupiter and Mercury, so that their dwelling in this force loses its effect. Lnghnjata- IV. The fourth force is called kalubala, i.c. the tem- kam, tl 6. poral one, which the daily planets excrcise in the dsy, the nightly planets during the night It is peculiar to Mercury in the samdhi of its rotation, whilst others maintain that Mercury always has this force, becsuse he stands in the ssme relation to both day and night. Further, this force is peculiar to the lucky plsnets in the white half of the month, aud to the unlucky stars in the black half, It is always peculiar to the ascendens. Other astrologers also mention years, months, days, and hours among the conditions, under which the one or other of the four forces is peculiar to a planet. These, now, are the forces which are calculated for the planets and for the ascendens. If several planets own, each of them, several forces, Pago 309 that one is preponderaut which has the most of them. If two planets have the same number of balas or forces, that one has the preponderance the magnitude of which is the larger. This kind of magnitude is in the table of

CHAPTER LXXX. 227

ii. 215, called naisargikabala. This is the order of the Lagh, il. 7 planets in magnitnde or force. The middle years which are computed for the planets The years are of three different species, two of which are com- the single of life which

puted according to the distance from the altitudo. The bestow. planets

measures of the first and second species we exhibit in species of Threo

the table (ii. 215). these yeara.

The shadaya and naisargika are reckoned as the degree of altitudo. The first species is computed when the above-mentioned forces of the sun are prepon- derating over the forces of the moon and the ascendens separately. The second species is computed if the forces of the moon are preponderating over those of the sun and those of the ascendens. The third species is called amsaya, and is computed if the forces of the ascendens are preponderating over those of sun and moon, The computation of the years of the first species for The first each planet, if it does not stand in the degree of its species,

altitudo, is the following :- Yon take the distance of the star from the degree of LAgh. vi. I. its altitudo if this distance is more than six signs, or the difference between this distance and twelve signs, in case it is less than six signs. This number is multiplied by the number of the years, indicated by the table on page 812. Thus the signs sum up to months, the de- grees to days, the minutes to day-minutes, and these values are reduced, each sixty minutes to one day, each thirty days to one mouth, and each twelve months to oue year.' The computation of these years for the ascendens is this :- Take the distance of the degree of the star from o° of Lagh, vi. 2. Aries, one year for each sign, one month for each 2ł degrees, one day for each five minutes, one day-minute for each five seconds.

228 ALBERUNTS INDIA.

The nonond The computation of the years of the second species for the planets is the following :- Take the distancs of the star from the degree of its altituda according to the just-mentioned rule (ii. 227). This number ie multiplied by the corresponding num- ber of years which is indicated by the table, and the remainder of the computation proceeds in the same way as in the case of the first species. The compntation of this species of years for the ascendens is this :-- Take the distance of its degree from o° of Aries, a year for each nuhbahr; months and days, &c., in the same way as in the preceding computation. The number yon get is divided by 12, and the remainder being less than 12, represents the number of years of the ascendens. The third The compntation of the years of the third species is speeiea, the same for the planets as for the ascendens, and is similar to the computation of the years of the ascendens of the second apecies. It is this :- Take the distance of the star from o° of Aries, one year for each nuhbahr, multiplying the whole distance by 108. Then the signs sum up to months, the degrees to days, the minutes to day-minutes, the emaller mea- sure being reduced to the larger one. The years are divided by 12, and the remainder which you get by this division is the number of years which you want to find. Laghujt- All the years of this kind are called by the common takam, ch. vl I. name dyurddya. Before they undergo the eqnation they are called madhyamdya, and after they have passed it they are called sphutdya, i.c. the corrected ones. The yor The years of the ascendens in all three species are of life be- stowed by corrected ones, which do not require an equation by the ascen- means of two kinds of subtraction, one according to the position of the ascendens in the æther, and a

CHAPTER LXXX. 229

second according to its position in relation to the horizon. To the third kind of years is peculiar an eqnation by Varions means of an addition, which always proceeds in the tions for the computa-

same manner. It is this :- duration of Hfo. If a planet stands in its largest portion or in its honse, the drekkdna of its house or the drekkana of its altitudo, in the nuhbahr of its house or the nuhbahr of its altitudo, or, at the same time, in most of these posi- tions together, its years will be the donhle of the middle number of years. Bnt if the planet is in retrograde motion or in its altitudo, or in both together, its years Pago 3I0 are the threefold of the middle number of years. Regarding the equation by means of the subtraction (vide ii. 228) according to the first method, we observe that the years of the planet, which is in its dejectio, are reduced to two-thirds of them if they are of the first or second species, and to one-half if they belong to the third species. The standing of a planet in the house of its opponent does not impair the number of its years. The years of a planet which is concealed by the rays of the sun, and thus prevented from exercising an in- fluence, are reduced to one-half in the case of all three species of years. Oniy Venus and Saturn are excepted, for the fact of their being concealed by the rays of the sun does not in any way decrease the numbers of their years. As regards the eqnation by means of subtraction according to the second method, we have already stated in the table (ii. 221, 222) how mnch is subtracted from the unlucky and lucky stars, when they stand in the houses above the earth. If two or more planets come together in one house, yon examine which of them is the larger and stronger one. The subtraction is added to the years of the stronger planet and the remainder is left as it is. If to the years of a single planet, years of the third

230 ALBERUNPS INDIA.

species, two additions from different sides are to be made, only one addition, viz., the longer one, is taken into account. The same is the case when two subtrac- tions are to be made. However, if an addition as well as a subtraction is to be made, yon do the one first and then the other, becanse in this case the seqnence is different. By these methods the years become adjusted, and the sum of them is the duration of the life of that man who is born at the moment in question. It now remains for us to explain the method of the Tho single Hindus regarding the periods (sic). Life is divided elementa of the com- in the sbove-mentioned three species of years, and pntation of thedurstion immediately after the birth, into years of sun and of lifa. moon. Thst one is preponderating which has the most forces and balas (vide ii. 225); if they equal each other, thst one is preponderating which has the greatest portio (sic) in its place, then the next one, &c. The companion of these years is either the ascendens or that planet which stands in the cardines with many forces and portiones. The several planets come together in the cardines, their infiuence and sequence are determined by their forces and shares. After them follow those planets which stand near the cardines, then those which stand in the inclined signs, their order being determined in the same way as in the preceding case. Thus becomes known in what part of the whole human life the years of every single planet fall. However, the single parts of life are not compated exclusively in the years of the one planet, bat accord- ing to the infiuences which companion-stars exercise upon it, i.e. the planets which stand in aspect with it. For they make it partake in their rule and make it share in their division of the years. A planet which stands in the same sign with the planet ruling over the part of life in question, shares with it one-half. That which stands in the fifth and ninth signs, shares with

CHAPTER LXXX. 231

it one-third That which stands in the fourth and eighth signs, sheres with it one-fourth. That which stands in the seventh sign, shares with it one-seventh. If, therefore, several planets come together in one position, all of them have in common that share which is necessitated by the position in question. The method for the computation of the years of such a companionship (if the ruling planet stands in aspect How one with other planets) is the following :- planot is Affected by Take for the master of the years (i.e. that planet of another the nature

which rules over a certain part of the life of a man) one " ou8. as numerator and one as denominator, i.e. , one whole, because it rules over the whole. Further, take for each companion (i.e. each planet which stands in aspect with the former) only the numerator of its denominator (not the entire fraction). You multiply each denominator by all the numerators and their sum, in which operation the original planet and its fraction are disregarded. There- by all the fractions are reduced to one and the same denominator. The eqnal denominator is disregarded. Each nnmerator is multiplied by the sum of the year and the product divided by the sum of the numerators. The quotient represents the years kálambuka (kala- bhaga ?) of a planet. As regards the order of the planets, after the question as to the preponderance of their influence has been decided (? text in disorder), in so far as each of them Page 3ut. exercises its individual infinence. In the same way as has already been explained (vide ii. 230), the preponde- rating planets are those standing in the cardines, first the strongest, then the less strong, &c., then those standing near the cardines, and lastly those standing in the inclined signs. From the description given in the preceding pages, speclal the reader learns how the Hindus compute the dura- inquiry of methods of

tion of human life. He learns from the positions of astrologers. tho Hindn

the planets, which they occupy on the origin (i.e. at

.232 ALBERUNPS INDIA.

the moment of birth) and at every given momont of life in what way the years of the different planets are distributed over it. To these things Hindu astrologers join certain methods of the astrology of nativities, which other nations do not take into account. They try, e.g., to find out if, at the birth of a human being, its father was present, and conclude that he was absent, if Lartjtta the moon does not stand in aspect with the ascendens, Icar, ch. ili 3- or if the sign in which the moon stands is enclosed between the signs of Venus and Mercury, or if Saturn is in the ascendens, or if Mars stands in the seventh sign. Chap. iii. 4 (?) .- Further, they try to find ont if the child will attain full age by examining sun and moon. If sun and moon stand in the same sign, and with them an unlucky planet, or if the moon and Jupiter just quit the aspect with the ascendens, or if Jupiter just quits the aspect with the united sun and moon, the child will not live to full age. Further, they examine the station in which the sun stands, in a certain connection with the circumatances of a lamp. If the sign is a turning one, the light of the lamp, when it is transferred from one place to the other, moves. If the sign is a fixed one, the light of the lamp is motionless; and if the aign is a double-bodied one, it moves one time and is motionless another. Further, they examine in what relation the degrees of the ascendens stand to 30. Corresponding to it is the amount of the wick of the lamp which is consumed by burning. If the moon is full moon, the lamp is full of oil; at other times the decrease or increase of the oil corresponds to the wane and increase of the moonlight. Chap. iv. 5 .- From the strongest planet in the car- dines they draw a conclusion relating to the door of the house, for its direction is identical with the direction of this planet or with the direction of the sign of the ascen- dens, in case there is no planet in the cardines. Chap. iv. 6 .- Further, they consider which is the

CHAPTER LXXX. 233

light-giving body, the sun or moon. If it is the sun, the bouse will be destroyed. The moon is beneficent, Mars burning, Mercury bow-shaped, Jupiter constant, and Ssturn old. Chap. iv. 7 .- If Jnpiter stands in its altitudo in the tenth sign, the honse will consist of two wings or three. If ita indicium is strong in Arcitenens, the honse will have three wings; if it is in the other double-bodied signs, the honse will have two wings. Chap. iv. 8 .- In order to find prognostics for the throne and ita feet they examine the third eign, its squares and its length from the twelfth till the third signs. If there are nnlncky planets in it, either the foot or the side will perish in the way that the unlucky planet prognosticates. If it is Mars, it will be turned; if it is the sun, it will be broken; and if it is Saturn, it will be destroyed by old age. Chap. iv. 10 .- The nnmber of women who will be present in & house corresponde to the number of etars which are in the signs of the ascendens and of the moon. Their qualities correspond to the images of these con- etellations Those stars of these constellations which stand sbove the earth refer to those women who go away from the house, and those which stand under the earth prognosticate the women who will come to ths honse and enter it. Further, they inquire into the coming of the spirit Laghujata- of life in man from the dominant of the drekkdna of 3, + kam, ch. xti. the stronger planet of either sun or moon. If Jupiter ie the drekkâna, it comes from Devaloks; if it is Venus or the moon, the spirit comes from Pitriloka; if it is Mars or the sun, the spirit comes from Vriscikaloka; and if it is Saturn or Mercury, the spirit comes from Bhriguloka. Likewise they inquire into the departing of the soul after the death of the body, when it departs to that planet which is stronger than the dominant of the

234 ALBERUNIS INDIA.

drekkana of the sixth or eighth houses, according to a similar rule to that which has just been laid down. Page 412 However, if Jupiter stands in its altitudo, in the sirth house, or in the eighth, or in one of the cardines, or if the ascendens is Pisces, and Jupiter is the strongest of the planets, and if the constellation of the moment of death is the same as that of the moment of birth, in . that case the spirit (or soul) is liberated and no longer wanders about. I mention these things in order to show the reader the difference between the astrological methods of our people and those of the Hindus. Their theories and On comets, methods regarding aerial and cosmic phenomena are very lengthy and very subtle at the eame time. As we have limited ourselves to mentioning, in their astrology of netivities, only the theory of the determina- tion of the length of life, we ahall in this department of science limit ourselves to the apecies of the comets, according to the statements of those among them who are aupposed to know the anbject thoroughly. The analogy of the comets shall afterwards be extended to other more remote snbjects. The head of the Dragon is called rahu, the tail ketu. The Hindus seldom speak of the tail, they only use the head. In general, all comets which appear on heaven are also called ketu. Quotations Varahamihira says (chap. iji. 7-12) :- froru the Bakkitd of "The Head has thirty-three sons who are called Varthami- bira. tamasaktlaka. They are the different kinds of the comets, there being no difference whether the head extends away from them or not. Their prognostics correspond to their shapes, colours, sizes, and positions. V. 8 .- The worst are those which have the shape of a crow or the shape of a beheaded man, those which have the shape of a sword, dagger, bow and arrow. V. 9, 10 .- They are always in the neighhourhood of sun and moon, exciting the waters so that they become

CHAPTER LXXX. 235 thick, and exciting the air that it becomes glowing red. They bring the air into such an uproar that the tornadoes tear out the largest trees, that flying pebbles beat against the calves and knees of the people. They change the nature of the time, so that the seasons seem to have changed their places. When unlucky and calamitous events become numerous, such as earthquakes, land- elips, burning heat, red glow of heaven, uninterrupted howling of the wild beasts and screaming of the birds, then know that all this comes from the children of the Head. V. II .- And if these occutrences take place together with an eclipse or the effulgence of a comet, then recognise in this what thou hast predicted, and do not try to gain prognostics from other beings but the Sons of the Head. V. 12 .-- In the place of the calamity, point towards their (the comets) region, to all eight sides with relation to the body of the snn." Varahamihira says in the Samhitd (chap. xi. I-7) :-- "I have apoken of the comets not before having exhausted what is in the books of Garga, Parasara, Asita and Devala, and in the other books, however numerous they may be. "It is impossible to comprehend their computation, if the reader does not previously acquire the knowledge of their appearing and disappearing, because they are not of ona kind, but of many kinds. "Some are high and distant from the earth, appearing between the stars of the lunar stations. They are called divya. "Others have a middle distance from the earth, appearing between heaven and earth. They are called ântarikshya. "Others are near to the earth, falling down upon the earth, on the mountains, houses and trees. "Sometimea yon see a light falling down to the earth, which people think to be a fire. If it is not fire, it is keturupa, i.c. having the shape of a comet. "Those animals which, when flying in the air, look

236 ALBERUNTS INDIA.

like sparks or like fires which remain in the houses of the pisdcas, the devils, and of the demons, efflorescent substances and others do not belong to the genus of the comets. "Therefore, ere you can tell the prognostics of the comets, you must know their nature, for the prognostics Pago 313- are in agreement with it. That category of lights which is in the air, falling on the banners, weapons, houses, trees, on horses and elephants, and that category coming from a Lord which is observed among the stars of the lunar stations-if a phenomenon does not belong to either of these two categories nor to the above-men- tioned phantoms, it is a telluric ketu. V. 5 .- "Scholars differ among each other regarding the number of the cometa According to some there are 101, according to others 1000. According to Narada, the sage, they are only one, which appears in a multitnde of different forms, always divesting itself of one form and arraying itself in another. V. 7 .- "Their influence lasts for as many months as their appearance lasts days. If the appearance of a comet lasts longer than one and a half month, subtract from it forty-five days. The remainder represents the months of its inflnence. If the appearance lasts longer than two monthe, in that case state the years of its influence to be equal to the number of the months of its appearance. The number of comets does uot exceed the number 1000." We give the contents of the following table in order to facilitate the stndy of the subject, although we have not been able to fill out all the single fields of the diagram, because the manuscript tradition of the single paragraphs of the book either in the original or in the copy which we have at our disposal is corrupt. The anthor intends by his explanations to confirm the theory of the ancient scholars regarding the two numbers of comets which he mentions on their anthority, and he endeavours to complete the number 1000.

From what Their names, Their descent. How many ntura each Sum total. Thelr queiitien direetion Thair prognostiea comet has. they appoar. Pago

of Kirana. 25 Similar to pearls in rivaleta of crystal or gold-colonred. Only in It bodes the fighting of the ... The children 25 east and kings with each other. west.

The childreo of the Fire (1). 25 5o Green, or of the colour of fire or It bodes pestilence. of lac, or of blood, or of the S.E.

blossom of the tree. CHAPTER LXXX. It bodes hanger and pesti- of Death. 25 75 With orooked tails, their colour ... The children inclining to black and dark. lence.

The ohildren 22 of the Earth. 97 Round, radiant, of the colour of water or sesame oil, withont N.E. It bodes fertility and wealth.

taile,

The children It bodes ovil, in conseqnence of the Moon. 3 100 Like rosea, or white Jotus, or sil- N. ... ver, or polished iron or gold. It shines like the moon, of which the world will be turned topay-turvy.

Brahma- Son of Brab- I 101 Having three colours and three In all It bodes wickedness and de- danda man. taila directions. strnction.

The children 84 185 White, large, brilliant. N. and N.E. It bodes evil and fear. ... of Venus.

Kanaka. The children Radiant, as if they were horns. In all It bodes misfortune and of Suturn, ... ... directions death. Vikaca, The ohildren 65 Brilliant, white, withont any 8. ' It bodes destruction and mis- of Jopiter. ... fortune. 237 tnils.

238 stara eaoh Bum From what Thelr namos, Their descont. How many total. Thoir qualitles. direction Their prognostica. comet ban. they appoar.

Taskara, In all Jt bodes misfortune. ... White, thin, long. The eye is i.c. the thief. The children of Mercury. 51 dazzled by them. directions.

Kanůkuma. The children 60 It bodes the extromity of evil. .. + It bas three tails, and the colour N. of Mars. of the flamo.

TAmasa- The children 36 About the It bodes fire. of the Hend. ... Of different shapes. ALBERUNPS INDIA. ktlaka. son and

The children Of a blazing light like the flame. moon. Viśvarůpa. 120 It bodes oviL of the Fire. ... ...

Aruna. The children 77 They have no body, that you It bodes general destruction. of the Wind. ... could see a star in them. Only ...

their rays are united, so that these appear as rivulets. Their colour is reddish or greenish. ...

Gaņnka. The children 204 It bodes much evil and de- ... of Prajapati. ... Square comets, eight in appear- ance, and 304 in number. etruction.

Kańka. The children of the Water. 32 Its (?) are nnited, and it is shin- It bodes much fear and evil ... ... ing like the moon. in Pundra.

Ksbandha. The children Like the cat-off head of a man. It bodes much destruction. of the Time. ... ...

... One In appearance, nine in num- In all ber. White, Jarge. directions. It bodes pestilonce. ... ...

CHAPTER LXXX. 239

The author (VarAhamihira) had divided the comets Page 3rs. into three classes: the high ones near the stars; the Further flowing ones near the earth; the middle ones in the air, from the quotations

and he mentions each one of the high and middle classes Varthaml- > Samhitd of

of them in our table separately. He further aays (chap. xi. 42): " If the light of the middle class of comets ahines on the instruments of the kings, the banners, parasols, fans, and fiy-flaps, this bodes destruction to the rulers. If it abines on a house, or tree, or mountain, this bodes destruction to the empire. If it shines on the furni- ture of the honse, its inhabitants will perish. If it ahines on the aweepings of the house, its owner will perish." Further Varahamihira says (chap. xi. 6) :- "If a shooting-star falls down opposite to the tail of a comet, health and wellbeing cease, the rains lose their beneficial effects, and likewise the trees which are holy to Mahadeva- there is no use in ennmerating them, since their names and their essences are unknown among us Muslims-and the conditions in the realm of Cola, Sita, the Huns and Chinese are troubled." Further he says (chap. xi. 62) :- "Examine the direction of the tail of the comet, it being indifferent whether the tail hangs down or stands erect or is inclined, and examine the lunar station, the edge of which is touched by it. In that case predict destruction to the place and that its inhabitants will be attacked by armies which will devour them as the pea- cock devours the snakes. "From these comets you must except those which bode something good. " As regards the other comets, you must investigate in what lunar stations they appear, or in what station their tails lie or to what atation their tails reach. In that case you must predict destruction to the princea of those countries which are indicated by the lunar

240 ALBBRUNPS INDIA. stations in question, and other erents whioh aro in- dicated by those stationa" The Jews hold the same opinion regarding the comets as we hold regarding the stone of the Ka'ba (viz. that they all are stones which have fallen down from heaven). According to the same book of Vara- hamihira, comets are such beings as have been on account of their merits raised to heaven, whose period of dwelling in heaven has elapsed and who are then redescending to the earth. The following two tables embody the Hindu theories of the comets :-

TABLE OF COMETS OF THE GREATEST HEIGHT IN THE ÆTHER VOL II. 1 Vast. West. It is flashing and thick, and extends It bodes death and excessive wealth itself from the north. and fertility.

2 Asthi. West. Less bright than the first. It bodes hunger and pestilence.

3 Sastra. West. Similar to the firat. It bodes the fighting of the kings with each other. CHAPTER LXXX. 4 Kupâlaketu. Enst. Its tail extends till nearly the midst of heaven. It has a smoke-coloor and It bodes the abundance of rain, mnch

appears on the day of new-moon. hunger, illness and deatl.

5 Raudra. From the cast in Půrva- With a sharp edge, surronnded by mys. Bronze-coloured. It bodes the fighting of the kings with

shAdha, Pur- third of heaven. It occapies one- each other.

vabhadrapa- df, and Re- vatt.

6 Calaketu. West. During the first time of its appearance it has a tail as long as a finger towards It ruins the country from the tree

the south. Then it turns towards Prayaga till Ujjayim. It ruins the

the north, till it becomes as long as to Middle Country, whilet tho other regions fare differently. In some Q the south, the Great Bear and the Pole, then the Falling Eagle. Ris- places there is pestilence, in othors

ing higher and higher it passes round drought, in others war. It is visible

to the south and disappears there. between 10-12 mouths. 241 Page 316.

242

TABLE OF COMETS OF THE GREATEST HEIGHT IN THE #THER-Continued.

7 Švetaketn. South. It appears at the beginning of night and is visible during seven days. Its tail extends over one-third of heaven. When these two comets shine and It is green and passes from the right lighten, they bode henlth and wealth. ALBERUNTS INDIA. side to the left. If tho time of their appearance ex- ceeds seven days, two-thirds of the affeirs of men and of their lives are ruined. The sword is drawn, revolu- tions prevail, and there will be mis- 8 Ka. West. It appenrs in the first half of night, fortnne during ton years. Its flame is like scattered peas, and remuins visible during seven days.

9 Rasmiketu (?) The Pleiades. It has the colonr of smoke. It ruins all human affairs and ereates namerous revolutions.

10 Dhruvaketu(?) Appears be- It has a big body, It has many sides (!) | It bodes health and peace. tween heaven and colours, and is bright flashing. and earth wherever it likes.

TABLE OF COMETS OP MIDDLE BEIGHT IN THE SKY.

Their names, From what direction they Description, Thetr prognostics, appear, Their burn ber.

I Kumuda. Weat. Namesake of the lotus, which is com- pared with it. It remains one night, It bodes lasting fertility and wealth

and its tail is directed townrds the fur ten years.

CHAPTER LXXX. couth.

2 Maņiketu. West. It lasts only one quarter of a night Its tail is straight, white, similar to It bodes a great number of wild animals

the milk which spurts out of the and perpetnal fertility during four

breast when it is milked. and a half months.

3 Jalaketu. Weet. Flashing. Ite tail has a curve from the It bodes fertility and well-being of the west side. subjects during nine months.

4 Bhavaketu. East. | It has a tail like that of a lion towards It is visible only une night. It bodes the south. perpetual fertility and well-being dur- ing as many months as its appear- ance last muhurtas. If ita colour becomes less bright, it bodes pesti- 243 lence and death.

Pago 317-

244

TABLE OF COMETS OF MIDDLE HEIGHT IN THE SKY .- Continued.

Their names. From what! direction they Desoription. Thoir prognostics. appear. Thetr ALBERUNTS INDIA. numbar.

5 Padmaketu. South It is as white as the white lotue. It lasts one night. It bodes fertility, joy, and happiness for seven years.

6 Âvarta. Wost. It appears at midnight, bright shining It bodes wealth during as mauy months and light gray. Its tail extends from the left to the right. as its appearance lasts muhurtas.

Samvsrta West. With a tail with a sharp edge. It has the colour of smoke or bronze. It The lonar station ia which it appears becomes anlucky. It ruins as well extends over one-third of heaven, and that which it bodes, as the lunar appears daring the sandhi. etation. It bodes the ansheathing of the weapons and the destruction of the kings. Its influence iasts as many years as its appearance lasta muhdrtas.

CHAPTER LXXX. 245

This is the doctrine of the Hindus regarding the Pago 318. comets and their presages. Only few Hindus occupy themselves in the same On meteoro- way as physical scholars among the ancient Greeks logy.

did, with exact scientific researches on the comets and on the nature of the other phenomena of heaven (Ta ueTewpa), for also in these things they are not able to rid themselves of the doctrines of their theologians. Thus the Matsya-Purana says :-- "There are four rains and four mountains, and their basis is the water. The earth is placed on four elephants, standing in the four cardinal directions, which raise the water by their trunks to make the seeds grow. They sprinkle water in summer and snow in winter. The fog is the servant of the rain, raising itself up to it, and adorning the clouds with the black colonr." With regard to these four elephants the Book of the Medicine of Elephants says :- "Some male elephants excel man in cunning. There- fore it is considered a bad omen if they stand at the head of a herd of them. They are called manguniha (?). Some of them develop only one tooth, others three and four; those which belong to the race of the elephants bearing the earth, Men do not oppose them; and if they fall into a trap, they are left to their fate." The Vayu-Purana says :- "The wind and the sun's ray raise the water from the ocean to the sun. If the water were to drop down from the sun, rain would be hot. Therefore the sun hands the water over to the moon, that it should drop down from it as cold water and refresh the world." As regards the phenomena of the sky, they say, for instance, that the thunder is the roaring of Airavata, i.c., the riding-elephant of Indra the ruler, when it drinks from the pond Manasa, rutting and roaring with a hoarse voice.

246 ALBERUNTS INDIA.

The rainbow (lit bow of Kuzah) is the bow of Indra, as our common people consider it as the bow of Rustam. Conciuson. We think now that what we have related in this book will be sufficient for any one who wants to con- verse with the Hindus, and to discuss with them questione of religion, science, or literature, on the very . basis of their own civilisation. Therefore we shall finish this treatise, which has already, both by its length and breadth, wearied the reader. We ask God to pardon us for every statement of ours which is not true. We ask Him to help us that we may adhere to that which yielda Him satisfaction. We ask Him to lead us to a proper insight into the nature of that which is false and idle, that we may sift it so as to distinguish the chaff from the wheat All good comes from Him, and it is He who is clement towards His alaves. Praise be to God, the Lord of the worlds, and His blessings be upon the prophet Muhammad and his whole family !

ANNOTATIONS.

VOL. L

ANNOTATIONS.

VOL. I.

P. 1. Title .- The author proposes to investigate the reality (=hakika) of Hindn modes of thought in the entire extent of the subject. He describes the religious, literary, and scientific traditions of India, not the country and its inhabitants. However, in some chapters he gives more than the title promises; ef his notes on the roads and on the courses of the rivers. The contents of the eighty chapters of the book may be arranged under the following heads :- Chap. I. General Introduction. Chap. 2-I1. On Religious, Philosophical, and cognate subjects. Chap. 12-17. On Literature and Metrology, Strange Customs and Superstitions. Chap. 18-31. On Geography, Descriptive, Mathemati- cal, and Traditional, i.e. Pauranic, Chap. 32-62. On Chronology and Astronomy, inter- spersed with chapters of Religious Tradition, e.g. on Nârâ- yana, Vâsudeva, &c. Chap. 63-76. On Laws, Manners and Customs, Festivals and Fast Days, Chap. 77-80. On Astrological Subjects. The word makdla, translated by category, is a technical term of Arabian philosophy. It was coined by the first Arabian translators of Aristotle for the purpose of render- ing karryopia, and has since become current in the school language of Islam (cf. the Arabic title of Aristotelis Cate- goria Grace cum versione Arabica, &c., edid. J. Th. Zenker, Lipsia, 1846). The Syrian predecessors of those Arabian translators had simply transferred the Greek word just as

250 ALBERUNPS INDIA.

it is into their own language; ef. e.g. Jacob of Edessa in G. Hoffmaun'a De Hermeneuticis apud Syros Aristoteleis, Lipsia, 1869, p. 17. That & Muslim author should investigate the ideas of idolaters, and not only such as Muslims may adopt, bat also such as they must reject and condemn, that he quotes the Koran and the Gospel side by side (p. 4-5), is a proof of a broadness of view and liberality of mind more fre- quently met with in the ancient timea of Islam, in the centuries before the establishment of Mnhammadan ortho- doxy by Alghazzali (died A.D. IIII), than later. There was more field for utterances of mental individuality before the ideas of all the nations of Islam were moulded into a unity which makes it difficult to recognise the individual influences of every single nation on the general develop- ment of the Muhammadan mind, before all Islam had become one huge religious community, in which local and national differences seem to have lost most of their original importance for the spiritual life of man. The work of Alberuni is unique in Muslim literature, as an earneat attempt to study an idolatrous world of thonght, not pro- ceeding from the intention of attacking and refnting it, bnt uniformly showing the desire to be just and impartial, even when the opponent'a views are declared to be inad- missible. There can be hardly a doubt that under other circumstances, in other periods of Muslim history and other countries, the present work might have proved fatal to its author; and it shows that the religious policy of King Mahmud, the great destroyer of Hindu temples and idols, under whom Alberuni wrote, must have been so liberal as to be rarely met with in the annals of Islam (ef. pp. 268, 269).

P. 5. The master 'Aba-Sahl, de .- Al-tiflist, i.c. e native of Tiflis in the Caucasus, is not known from other sources. I snppose he was one of the high civil functionaries of the realm or court of Mahmud. The name Sahl occurs very freqnently among men of Persian descent of those times, and the title Ustdh=master, is in the Ta'rikh-i-Baibaki alwaya prefixed, if not precisely as an official title, at all events as a title expressive of profound respect on the part of the speaker, to the names of the ministers and

ANNOTATIONS. 251

highest civil officials of Mahmud and Masud, such as Bu Sahl Zauzant, Bu Sahl HamdOni, Bu Nasr Mushkau, the minister of state, whose aecretary Al-baihakt was, as well as to the name of Alberuni (Ar1, 16), but never to the namea of the great military men (cf. on titles in the Ghaz- nawi empire, A. de Biberstein Kazimirski, Menoutchehrt, Paris, 1887, p. 308). Administrative akill was a legacy left by the organisation of the Sasanian empire to the Persians of later centuries, whilst military qnalities aeem entirely to have disappeared among the descendants of Rustam. For all the generals and officers of Mahmud and Mas ud were Turks, aa Altontash, Arslan Jadhib, Ari- yarok, Bagtagin, Bilkatagin, Niyaltagin, Noshtagin, &c. The Ghazna princes apoke Persian with their civil function- aries, Turkish with their generals and soldiers (cf. Elliot, History of India, ii. 81, 102).

P. 5. The Mutazila sect .- The dogma, God has no know- ledge, is part of their doctrine on the qualities of God, maintained especially by Ma'mar Ibn 'Abbad Al-Sulami. (Cf. on this and related subjects the treatise of H. Steiner, Die Mutaziliten oder die Freidenker im Islam, Leipzig, 1865, pp. 50, 52, 59, and Al-Shahrastanf's " Book of Reli- gious and Philosophical Sects," edited by Cureton, London, 1846, p. 30, ll. 7-9). Proceeding from the study of Greek philosophy, the doctors of this achool tried to save the free will of man as against predestination. There was once in Arabic a large literature composed by them and by their opponents, most of which ia unknown, at all events not yet brought to light. Most of these books were of a polemical nature, and it is against their polemi- cal bias that the criticism of Alberuni is directed." With regard to his own work, he expressly declares (p. 7) that it is not a polemical one. The book which Abu-Sahl had before him, and which gave rise to the discussion between him and our author, was probably one like that of Abul- hasan Al-'ash'ari (died A.D. 935), the great predecessor of Alghazzali, "On the Qualities of God," in which he attacks the Mu'tazila doctrine of the negation of God's omni- science. (Cf. W. Spitta, Zur Geschichte Abulhasan Al- 'Asharts, Leipzig, 1876, p. 64) The same author has also written an extensive work against the antagonists of

252 ALBERUNIS INDIA.

the orthodox faith, against Brahmins, Christians, Jews, and Magians (v. ib. p. 68). Our information regarding the ancient literature on the history of religion and philosophy (the latter proceeding from a work of the Neoplatonist Porphyrius) is very scanty, and mostly limited to titles of books. The work of Shahrastâni (died A.D. 1153) is a late compendium or y (v. his pref., I, 8). His editor, Cureton, intended to give "Observations respecting the sources from which this author has probably derived his information " (English pref., p. iv.), but, as far as I am aware, he has not carried ont his intention. There is an excellent treatise on the history of religions in the Fihrist of Al-nadim (composed about A.D. 987) on p. rh r.l. The same author mentions, (p. \v) an older work on doctrines and religions by Alhasan Ibn Musa Alnaubakhti (mentioned by Mas'udi), who also wrote against metempsychosis. Parts of a simi- lar work of Ibn Hazm, an Arab of Spain (died A.D. 1064), are extant in the libraries of Vienna and Leyden. Mr. C. Schefer has recently pnblished in his Chrestomathie Persane, Paris, 1883, a useful little book in Persian called composed by Abul-Ma'ali Muhammad Ibn كتاب بيان الاديان 'Uķail, who wrote in Ghazna, under the king Mas'ud Ibn Ibrahim (A.D. 1089-1099), half a century after Alberuni, whose Indica he quotes in his book, He calls it al .I, i.e. " The Doctrines of the Hindus" (p. VrA). Two more treatises in Persian on the history of religions are mentioned by C. Schefer, Chrestomathie Persane, pp. 136, 137. An anthor who seems to have written on subjects con- nected with the history of religions is one Abû-Ye'kub of Sijistan, as Alberuni (i. 64-65) quotes his theory on the metempsychosis from a book of his, called Kitdb-kashf- almahjub.

Pp. 6-7. Aleranshahri and Zurkan .- Our anthor has not made any use of the Muhammadan literature on the belief of the Hindus, as far as such existed before his time; evidently he did not give it the credit of a bond fide source of historical information. Throughout his book he derives his statements exclusively either from Indian books or from what he had heard himself. He makes an exception of this rule only in favour of Aleranshahri, the

ANNOTATIONS. 253

author of a general work on the history of religions. Alberuni seems t have known this book already (A.D. 1000) when he wrote his " Chronology," for there he gives two quotations, one an Eranian, and the other an Armenian tradition, on the anthority of Aleranshahri (v. " Chrono- logy of Ancient Nations," &c., translated by Dr. C. Edward Sachau, London, 1879, pp. 208, 211). The word Eranshahr was known to the Arabs as the name of the whole Sasanian empire, from the Oxus to the Euphrates. So it is used, c.g. by Abt-Alt 'Ahmad Ibn 'Umar Ibn Dusta in his geographical work (British Museum, add. 23,378 on fol. 1206), where he describes the whole extent of it. If, however, Eranshahr here means the place where the anthor Abul'abbas was born, we must take the word in the more restricted meaning, which is mentioned by Albaladhurt. For it is also the name of a part of the Sasanian empire, viz one of the four provinces of Khurasan, the country between Nishapur, Tus, and Herat. Accordingly, we suppose that Aleranshahri means a native of this particular province. Cf. Almuķaddasi, p. rir, Yâkût, i. PA. According to another tradition, the name Eranshahr also applied to Nishapur, i.c. the name of the province was used to denote its capital. Cf. Almu- ķaddasî, p, r. Aleranshahri, a sort of freethinker according to Albe- runi, is only once quoted (i 326, a Buddhistic tradition on the destruction and renovation of the world). But as Alberuni praises his description of Judaism, Christianity, and Manichæism, we may suppose that the information of the Indica on these subjects, e.g. the quotation from the Gospel (p. 4-5), was taken from Erânshahri. Incorporated in the work of Eranshahri was a treatise on Bnddhism by an author, Zurkan, who is entirely unknown. Althongh Alberuni speaks very slightingly of this anthor, and although he does not mention him anywhere save ini the preface, he seems to have borrowed from him those notes on Buddhistic subjects which are scattered through his work (v. Inder Rerum, s.v. Bud- dhists). This sort of information is not of a very high standard, bnt other sources on Buddhism, literary or oral, do not seem to have been at the command of Alberuni. The Hindus with whom he mixed were of the Brahminical

254 ALBERUNI'S INDIA.

creed, not Buddhists. In the countries where he had lived, in Khwarizm, Jurjan, the country round Ghazna .(Zabnlistan), and the Panjab, there had been no oppor- tunity for studying Buddhism; and also among the nume- rous soldiers, officers, artisans, and other Indians in the service of Mahmud in Ghazna and other places, there do not seem to have been Bnddhists, or else Alberuni would have used such occasions for filling ont this blank in his knowledge. In the Fihrist (ed. G. Flugel, Leipzig 1871), on p. rm re there is an extensive report on India and China, which is derived from the following sources :- I. The account of Abu-Dulaf of Yanbt, who had travelled to India and China about A.D. 94I. 2. That of a Christian monk from Najran, who by order of the Nestorian Katholikos had also travelled to India and China in the years A.D. 980-987. 3. From a book dated A.D. 863, of an unknown anthor, a book which had passed through the hands of the famons Alkindt. Was this perhaps the work of Aleranshahri, and the note on Buddha on p. rrv by Zurkân? The origin of the chapter on Indian snbjects in Shah- rastâni (ed. Cureton, London, 1846), on p. Prr seq. is not known. At all events, this author has not made use of Alberuni's work.

Pp. 7-8. Greeks, Sufis, Christians .- In order to illustrate the ideas of the Hindus, and to bring them nearer to the understanding of his Muslim readers, Alberuni qnotes related ideas- I. Of the Greeks (ef. i. 24). 2. The Christians. 3. The Jews. 4. The Manichæans; and 5. The Sufis. Pantheism in Islam, the doctrine of the Sufis, is as near akin to the Neoplatonic and Neopythagorean schools of Greek philosophy as to the Vedanta school of Hindu philosophers. It was in our author's time already repre- sented by a very large literature. He qnotes some Sûfi sentences, eg. of Abu Bakr Al-shibli, and Abt Yazid Albistamt, who are known from other sources (i 87, 88),

ANNOTATIONS. 255

and a Stfi interpretation of a Koranic passage (i. 88). Cf. besides, the Inder Rerum s.v. Sufism. He gives i. 33, 34, several etymologies of the word Stfi, which he himself identifies with Zodla. The notes relating to Mani and the Manichæans (v. Index Rerum), and the qnotations from their books, are probably mostly taken from Aleranshahri (v. p. 18). However, it must be kept in mind that, at the time of- our author, the works of Mant still existed, and he him- self found the "Book of Mysteries" and others in his native country, though perhaps at a time subsequent to . the date of the composition of the Indica. Of. Chronologie Orientalischer Volker, herausgegeben von Ed. Sachau, Leipzig, 1878, Vorwort, pp. xi. and xxxvi. The following works of Mani are quoted: "Book of Mysteries," Ls NM; Thesaurus vivificationis ohadt ys, i. 39. Cf. Mani, seine Lehre und seine Schriften, by G. Flugel, Leipzig, 1862. As regards the Jews, I am not informed to what degree Jewish colonies were in those times spread over Cen- tral Asia Alberuni derived probably his knowledge of Judaism also from Aleranshahri (p. 253). That in earlier years, during his stay in Jurjan, he was acquainted with a Jewish scholar is apparent from his chronological work ("Chronology of Ancient Nations," p. 269). Alberuni's knowledge of Christianity may have been communicated by various channels besides the book of his predecessor Aleranshahrt, as during his time it was far spread in Central Asia, and even at the court of Mahmud in Ghazna (e.g. Abulkhair Alkhammar, p. 256), there lived Christians. It has not yet been investigated in detail how far Nestorian Christianity had been carried eastward across Central Asia towards and into China. Cf. Assemani's Notitia Ecclesiarum Metropolitanarum et Episcopalium que sunt Patriarcha Nestoriano Subjecta (Bibliotheca Orientalis, vol. iv. p. DCCV. seg.). Barhebræus speaks of Uiguri monks bid e Lp2 (ib. ii. 256), and from the same time date some of the Syriac inscriptions on Christian tombstones recently found in Russian Central Asia and published in Petersburg, 1886. Alberuni men- tions Christians in his native country Khwarizm (Khiva), and in Khurasan, and not only Nestorians, but also Mel-

256 ALBERUNIS INDIA.

kites, whilst he expressly states that he does not know the Jacobites. Cf. " Chronology of Ancient Nations," pp. 283. 4; 292, 12; 295, 22 ; 312, 16. Where Alberuni, learned Greek philosophy, and who introduced him to the study of Plato's Dialogues and Leges, he does not state himself. The Arabic translations which he used, and which are tolerably correct, had passed throngh Syriac versions which are now no longer extant (e.g. those of Plato). Alberuni was personally acqnainted and had literary connections with a man who was one of the first representatives of Greek learning in the Muslim world in that age, Abulkhair Alkhammar, and it was perhaps to him that Alberuni owed part of his classical edncation. Abulkhair was born a Christian in Bagdad, A.H. 942. He lived some time in Khwarizm, and migrated. thence, together with Alberuni and others, to Ghazna, A.D. 1017, after Mahmad had annexed that conntry to his empire. He died in Ghazna during Mahmud's reign, i.e. before A.D. 1030, and is said to have become a Muslim towards the end of his life. He was a famons physician, and wrote on medical snbjects and on Greek philosophy; besides he translated the works of Greek philosophers (e.g. Theophrast) from Syriac into Arabic. Of his writ- ings we may mention a " Book of Comparison of the Theory of the (Greek) Philosophers and of the Christians," "Explanation of the Theory of the Ancients (i.e. Greek philosophers) regarding the Creator and regarding Laws," "The Life of the Philosopher," " On the DAn," " On Meteo- rology," &c. His pedigree points to a Persian descent. Cf. Chronologie Orientalischer Völker, Einleitung, p. xxxii., Fihrist, p. rie, and the work of Shahrazuri don, cidt dey ehy'l (manuscript of the Royal Library of Berlin, MSS. Orient. oct. 217, fol. 144b-146a); C. Schefer, Chresto- mathie Persane, p. 141. It must be observed that Alberuni, in comparing Hindu doctrines with those of Plato, follows in the wake of Megasthenes, who says: Παραπλέκουσι δέ και μύθους, ώσπερ καί Πλάτων, περί τε άφθαρσίας ψυχής καί τών καθ' δου κρίσεων καί άλλα τοιαύτα (Schwanbeck, Bonn, 1846, p. 138).

P. 8. Sdnkhya (or Samkhya) and Patanjala-The

ANNOTATIONS. 257 former word is here written sangu ach. It may be donbtful whether the second is to be read Patañjala or Patanjali. Alberuni generally says Jost obs, which may be translated the book of (the author) Patanjali, or . the book (which is called) Patanjali or Patanjala. Only in one place, i 68 (re, 5), he says صاحب كتاب باتدجل , the author of the book of Patanjali, where apparently Jot means the title of the book, not the name of the author. The long a in the Arabic writing would rather indicate the pronunciation Patanjala than Patanjali, but in this respect the transliteration is not always uniform, as sometimes a short Indian a has been rendered by a long d in Arabic, سات لوك ,gandharva كاندهرب ,brahman براهم ,tala تال .gه madhyaloka, Jy- sutala, we: vijayanandin, 4 para, y4 vasu, sol. mathurd, Jue mahatala. Only in two places the word Jost evidently means the author, i. 70 (TP, 20), and 87 (Pr, 3). The name of the anthor seems to have been current also as meaning his book. Therefore, and be- cause in Sanskrit generally the name Patanjali is quoted, I have given the preference to the latter form of the name. Alberuni has transferred large portions of his transla- tions of the books Samkhya and Patanjali, which he had published at an earlier date, into the Indica.

Pp. 17-19-In a similar way to Alberuni, the poet Mir Khusrau discourses on classical and vernacular in his Nuh-sipihr. He mentions the word Sanskrit, whilst Albe- runi only speaks of Hindt (v. Elliot, " History of India," iii. 562, 556; also v. 570, "On the Knowledge of Sanskrit by Muhammadans"). There were Hindu dragomans in the service of Mahmud, both in the civil administration and in the army, large portions of which were Hindus under Hindu officers (Elliot, ii. 109; some fought in Karman, Khwarizm, and before Merw for their Muslim master, ib. ii. 130, 131). Part of these troops were Kannara, i.e. natives of Karna- țadeśa (here i. 173). A specimen of these interpreters is Tilak, the son of Jai Sen (i.e. Tilaka the son of Jayasena). After having pursued his studies in Kashmir, he became interpreter first to Kâdi Shirazi Bulhasan 'Ali, a high civil official under

258 ALBERUNI'S INDIA.

Mahmud and Mas'hd (Elliot, ii. 117, 123), then to Ahmad Ibn Hasan of Maimand, who was grand vizir, A.D. 1007- 1025, under Mahmud, and a second time, 1030-1033, under Mas ud, and rose afterwards to be a commanding oflicer in the army (Elliot, ii. 125-127). This class of men spoke and wrote Hindi (of course with Arabic cha- racters) and Persian (perhaps also Turkish, as this language prevailed in the army), and it is probably in these circles that we must look for the origin of Urdu or Hindustani. The first anthor who wrote in this language, the Dante of Muhammadan India, is one Mas td, who died a little more than a century after the death of King Mahmud (A.H. 525 =A.D. 1131). Cf. A. Sprenger, " Catalogue of the Arabic, Persian, and Hindustany. Manuscripts of the Libraries of the King of Oudh," Calcutta, 1854, pp. 407, 485. If we had any of the Hindi writings of those times, they would probably exhibit the same kind of Indian speech as that which is found in Alberuni's book.

P. 18 .- The bearing of the words e tcy lesa, (9, 14, 15), which I have translated " and must prononnce the case-endings either," &c., is doubtful. The word 'irab means the process or mode of Arabizing a foreign word, and refers both to consonants and vowels. An 'frab mashhur would be a generally knou Arabie mode of pronunciation of a word of Indian origin, an 'i rab ma mal such a pro- nunciation of an Indian word in Arabic as is not yet known, but invented for the purpose. E.g. the Sanskrit word dvipa appears in two different forms, as dib, -, which must be classed under the first head, and as dbip, As, which belongs to the second class. If it is this the anthor means, we must observe that the former class, i.e. the class of words which had already general currency in Arabic before he wrote his Indica, is insignificantly small in comparison with the large number of words which by Alberuni were for the first time presented to a reader of Arabic (v. preface of the edition of the Arabic original, p. xxvii.). Another meaning of the word 'i'rab is the vowcl-pronun- ciation at the end of the words, chiefly the nouns; in fact, the case-endings. Accordingly, 'irab mashhur may mean case ending (in German, vocalischer Auslaut) as it is gene-

ANNOTATIONS. 259

rally used in Hindi, e.g. La gita, Fs, revati, and 'i rab ma mul, a case-ending added to a word purposely in order to make it amenable to the rules of Arabic declension (dip- toton and triptoton), eg. old lanku =Skr. lankha, ys gauru =Skr. Gaurt, ay bindu=Skr. Vindhyd, The vocalisation of these words is liable to lead us into an error. Is ty an Arabic diptoton, or is its final vocal the termination of the noun in Hindi? If the former were the case, we ought also to have s in genitive and acensative, and we ought to read 5 a caste (varna), s Lt an impure one (mleccha), L a measure (mana), &c. But these forms do not occur in the manuscript, and therefore I hold the termination u to be the Indian nominative, developed ont of the 6 of Prakrit, and still extant in Sindhi. (Cf. E. Trumpp, Dic Stammbil- dung des Sindht, " Journal of the German Oriental Society," xvi. p. 129; his " Grammar of the Sindhi Langnage," p. 32). The Arabic manuscript is not snfficiently accurate to enable us to form an opinion to what extent names in Alberuni's Hindi terminated in w, bnt we must certainly say that this is the case in the vast majority of nouns. If we are correct in this, the term 'i'rab mamul cannot mean an artificial case-ending or one invented or added for the pur- pose, because it existed already in the Iudian dialect wheuce Alberuni took the word. الاحتيال لمبطها بتغيير النقط والعلاماب وتقييدها باعراب Of the words Jyne Ct, ate ut, the former halt refers to the writing of the consonants (and perhaps of the Lesezeichen). Accord- ingly the latter halt onght to refer to the vowels; but 'irab does not mean rowels or vocalisation; it only means the vocalisation of the final consonant of the word. There- fore I am inclined to prefer the first of the two interpreta- tions here proposed, and to translate for in order to fix the pronunciation we must change the points (i.e. the dia- critical points of the consonants, & 3 )j, &c.) and the signs (perhaps he means the Hamza, which cannot be applied to Indian sounds), and must secure its correet pro- nunciation by such a process of Arabizing as is either already in general use or is carried out (or invented) for the pur- pose. This is an example (and there are hundreds more) of the concise style of the author, so sorely franght with

260 ALBERUNPS INDIA.

ambiguity. Every aingle word is perfectly clear and cer- tain, and still the sentence may be understood in entirely different waya.

P. 19, 3 Which in our Persian grammatical system are considered as, &c .- Literally, "Which our companions call having," &c. Speaking of his fellow-Muslims in opposition to the Hindus, the anthor alwaya says our com- panions, our people, not meaning national differences, Arab, Persian, or Turk, bnt exelnsively the difference of creed. In Sanskrit a word (a ayllable) may commence with one, two, or three consonants, e.g. dvi, jyd, stri, kshveda, which is impossible in Arabic, where each syllable begins and ends with one consonant only. Alberuni's comparison cannot, therefore, refer to Arabic. In Persian, the rules for the beginning and end of the syllable are different. Whilst in the ancient forms of Eranian speech a syllable could commence with two con- sonants, as, e.g. fratama, khsapa, Neo-Persian permits only one consonant at the beginning of a syllable, fardum, shab. However, the end of a syllable may consist of two con- aecutive consonants, as in yaft wdy, baksh se, khushk alts, mard dy, &c. Alberuni seems to hint at these ex- amples, and at a doctrine of certain grammarians, who are not known, to this effect, that the first of these two conso- nants is to be considered as having not a complete or clear vowel, but an indistinet hidden one, something like a schica mobile of Hebrew grammar. There is a small number of words (or ayllables) in Neo- Persian which indeed commence with the two consonants but they were at ,واستن,خواهر,استخوان, غواب, خويش .as,e.g,خو the author's time pronounced as a single one, if we may judge from the metrical system of the Shahnama of his contemporary Firdausi, who was only a little older than himself. (Cf. aimilar remarks of the author, i 138, 139.)

P. 20. Sagara .- The story of Sagara is related in Vishnu- Purdna, translated by Wilson-Hall, vol. iii, p. 289-295. The words po'w sdet, and eT des ess might make us think that these events happened within the recollec- tion of the author; but this is not necessarily the case. The former words may be interpreted, " I recollect the story

ANNOTATIONS. 261

of a Hindu who," &c., ie. " I recollect having heard the story," &c .; and the words with which he winds up the story may mean, " I feel thankful to my fate that it was not I and my contemporaries whom he treated thus, but former generations."

P. 21. Shamaniyya-The Buddbists are in Arabic called by this name, which is derived from a Prakritic form of Sanskrit sramana (Strabo Eappavat, Hieronymus Sam- anaei), and by the word ih, i.e. the red-robed people (= raktapata), which refers to the red-brown (= kashaya) cloaks of the Buddhist monks. Cf. Kern, Der Buddhismus und seine Geschichte in Indien, tibersetzt von H. Jacobi, Leipzig, 1882, ii. 45. See another note of our author's on Buddhism in his " Chronology of Ancient Nations," pp. 188, 189. It is extremely difficult, from the utter lack of historic tradition, to check the anthor's statements as to the western extension of Buddhism, which certainly never reached Mosul. Before all, it will be necessary to examine how far Alberuni, when speaking of the ancient history and institntions of Eran, was under the influence of the poets of his time, Dakiki, Asadi, and Firdausî, who versified Eranian folklore for the edification of the states- men of the Samanian and Ghaznavi empires, all of them of Eranian descent. Hearing the songs of the heroic exploits of their ancestors consoled them to a certain degree for the only too palpable fact that their nation was no longer the ruling one, bnt subject to another; that Arabs and Turks had successively stepped into the heritage of their ancestors. It must be observed that the negotiators of the cities of Sindh, whom they sent to the Muslim conquerors when first attacked by them, were invariably sramanas (v. Albaladhuri), which seems to indicate that Sindh in those times, i.c. about A.D. 710, was Buddhistic. Cf. H. Kern, Der Buddhismus und seine Geschichte in Indien, ii. 543.

P. 21. Muhammad Ibn Alkasim .- The brilliant career of the conqueror of Sindh falls into the years A.D. 707- 714 By Albaladhuri (p. Pry), Ibn-Al'athir, and others he is called Muh. Ibn Alkasim Ibn Muhammad, not Ibn Almunabbih, as here and p. 116. When Alberuni wrote,

262 ALBERUNPS INDIA.

Islam was known in Sindh already 350 years (since A.D. 680), and was established there 320 years (since abont A.D. 710). On the history of the conquest of Sindh, ef. Albaladhuri's Kitab-alfutuh, p. tri, translated by Reinand, "Fragments," p. 182; Elliot, History of India, i. 113. Instead of Bahmanva read Bamhanvd = Brahmanavata.

P. 23. The words of Varahamihira are found in his Brihat-Samhitd, translated by Kern in the " Journal of the Royal Asiatic Society," 1870, p. 441 (ii. 15): "The Greeks, indeed, are foreigners, bnt with them this science is in a flourishing state. Hence they are honoured as thongh they were Rishis; how muoh more then a twice- born man, if he be versed in astrology." P. 25. Think of Socrates, &c .- The anthor speaks of a Socratic fate or calamity, meaning a fate like that which . befell Socrates. I do not know from what particular source Alberuni and his contemporaries derived their in- formation abont the history of Greek philosophy. There is a broad stream of literary tradition on this subject in Arabic literature, but it has not yet been investigated what was its origin, whether it proceeded from one source or from several. Those men, mostly Greek heathens from Harran or Syrian Christians, who had enjoyed the Greek edncation of the time, not only translated Greek literature into Syriac and Arabic for the benefit of their Arab mas- ters, but wrote also general works on the history of Greek learning and literature, probably translating and adopting for their purpose some one of the most current school- books on this subject, used in the schools of Alexandria, Athens, Antioch, &c. Among authors who wrote such books, some being mere compilations of the famous sen- tences of Greek sages (doxographic), others having a more historic character, are Hunain Ibn 'Ishak, his son 'Ishak Ibn Hnnain, and Kusta Ibn Luka (i.e. Constans the son of Lucas). Bnt what were the Greek works from which they took their information, and which they probably communicated to the Arabs exactly as they were? I am inclined to think that they used works of Porphyrins and Ammonius, the Greek originals of which are no longer extant.

ANNOTATIONS. 253

P. 25. Jurare in verba, &c .- The Hindus consider, e.g. the sciences of astronomy and astrology as founded upon tradition, and their anthors prodnce in their books side by side their own perhaps more advanced ideas and some silly notions of any predecessors of theirs, although they are fully aware that both are totally irreconcilable with each other. Cf. the words of Varahamihira to this effect in Brihat Samhita, ix. 7, and the note of his commentator Utpala to v. 32. Alheruni pronounces most energetically against this kind of scientific composition when speaking of Brahmagupta in chapter lix on eclipses.

P. 27. Beyond all likeness and unlikeness, an expression frequent in the description of the Deity. Literally trans- lated: things that are opposite to each other and things that are like each other. Perhaps the rhyme didd and nidd, addad and 'anddd, has contributed to the coining of this term. As for the idea, it may be compared with the term dvandvas in Hindu philosophy = pairs of opposites, as pleasure and pain, health and sickness. Bhagavad-Gita, ii 45, vii. 27; " Yoga Aphorisms of Patañjali" (edited by Rajendralala Mitra), ii. 48, p. I11.

P. 27. Who is the worshipped one? &c .- The greater part of this extract from Patafijali has been translated into Persian by Abulma'ali Muhammad Ibn 'Ubaid-Allah in his Kitab-bayan-al'adyan; v. C. Schefer, Chrestomathie Per- سوال كداسست أن معبود كه همه كان بتوفيق أو رأه: ١٣٩-١٣٨ . sane, i يابند بعبادت أو جواب أنكه همه أسيدها بدرست وهمة بيمها آلع

P. 27. Patanjali .- The book of this name used and trans- lated by the author had the form of a conversation between two persons, simply called " the asking one," and " the an- swering one," and its subject was the search for libcration and for the union of the soul with the object of its medita- tion (i. 132), the emancipation of the soul from the fetters of the body (i. 8). It was a popular book of theosophy, propounding in questions and answers the doctrine of the Yoga, a theistic philosophy developed by Patañjali out of the atheistic Samkhya philosophy of Kapila. Cf. J. Davies, " Hindu Philosophy," Sankhyd Karika of Isvara Krishna, London, 1881, p. 116. The latter is called nirisvara=

264 ALBERUNTS INDIA.

not having a lord, the former sesvara =having a lord. It mostly treats of moksha (salvation) and metempsychosis. It contained not only theory, but also tales (i 93), Hag- gadic elements by way of illustration. Alberuni's Patanjali is totally different from " The Yoga Aphorisms of Patanjali" (with the commentary of Bhoja Raja, and an English translation by Rajendralala Mitra, Calcutta, 1883), and, as far as I may jndge, the philosophic system of the former differs in many points essentially from that of the Sutras, Moreover, the extracts given in the Indica stand in no relation with the commentary of Bhoja Raja, although the commentator here and there mentions ideas which in a like or similar form occur in Alberuni's work, both works being intended to explain the principles of the same school of philosophy. Besides the text of Patanjali, a commentary also is mentioned and quoted (i. 232, 234, 236, 238, 248), h Jt ls or Jsty. It is most remarkable that the extracts from this commentary are all of them not of a philosophic, but of a plainly Pauranic character, treating of cosmographic subjects, the lokas, Mount Meru, the different spheres, &c. The name of the commentator is not mentioned. If the quotations on i 273 seg. may be considered as derived from this commentary, the author was Balabhadra. V. index i. s.v. Patanjali.

P. 29. Gitd .- The book Gitd is, according to Alberuni, a part of the book Bharata (i.e. Mahabharata, which term does not occur in the Indica 1), and a conversation between Vasndeva and Arjuna (o,> yot JU). It is largely quoted in chapters relating to religion and philosophy. We have now to examine in what relation Alberuni's Gitd stands to the well-known Bhagarad-Gita as we have it in our time. Cf. "Hindu Philosophy," "The Bhagavad-Gita, or the Sacred Lay," translated by J. Davies, London, 1882. The latter is described as a skilful union of the systems of Kapila and Patanjali with a large admixture of the prevailing Brahmanic doctrines. Although the opinions regarding its origin differ widely, it can scarcely be denied that it is not free from having been inflnenced to a certain degree by 1 Cf. Alberuni on the Mahdbharata, i. 132, 133.

ANNOTATIONS. 265 Christianity, and that it could not have been composed before the third Christian century. Chapter xi. gives the impression of having been modelled after a Christian apocalypsis. The quotations from the Gitd (or Song) may be divided into three classes :- (1.) Such as exhibit a close relationship with certain passages in the Bhagavad-Gita. Parts of sentences are here and there almost identical, but nowhere whole sen- tences ; v. i. 40, 52, 73, 74, 86, 87, 103, 104, 218 (v. note), 352; ii. 169. (2.) Snch as show a certain similarity, more in the ideas expressed than in the wording, with passages in the Bhagavad-Git; v. i. 29, 70, 71, 78, 79, 103, 104, 122. (3.) Such as cannot be compared, either in idea or in wording, with any passage in the Bhagavad-Gita; v. i. 52, 53, 54, 70, 71, 73, 74, 75, 76, 78, 79, 80, 92, 122; iL 137, 138. The single texts will be discussed in the notes to the places in question. The quotations given by Alberuni cannot have been translated from the Bhagavad-Gita in its present form. Admitting even that the translator translated as little literally and accurately as possible (and the texts of Albe- runi do not give this impression), there remains a great number of passages which on no account could be derived from the present Sanskrit text, simply because they do not exist there, Or has Alberuni translated a commentary of the Bhagavad-Gitd instead of the original? The text of the extracts, as given in the Indica, is remarkably short and precise, extremely well worded, withont any repeti- tion and verbosity, and these are qualities of style which hardly point to a commentary. Alberuni seems to have used an edition of the Bhagavad- Gitd totally different from the one which we know, and which also in India seems to be the only one known. It must have been more ancient, because the notorious Yoga elements are not found in it, and these have been recog- nised by the modern interpreters as interpolations of a later time. Secondly, it must have been more complete, because it exhibits a number of sentences which are not found in the Bhagavad-Gita.

266 ALBERUNPS INDIA.

Various generations of Hindu scholars have modelled and remodelled this book, one of the moss precions gems of their literature, and it seems astonishing that an edition of it which existed as late as the time of Alberuni should not have reached the nineteenth century. As regards the quotation on this page (29), it exhibits only in the substance a distant relationship with Bhagavad- Gitd, x. 3: "He who knows Me as unborn and without beginning, the mighty Lord of the world, he of mortals is free from delusion, he is free from all sin."

P. 30. Simkhya .- The book Samkhya, as used and translated by Alberuni, had the form of a conversation between an anchorite and a sage, and it contained a treatise on the origines and a description of all created beings (i. 8), a book on divine subjects (i. 132). It was composed by Kapila. The anthor quotes it largely on questions of religion and philosophy. The Samkhya philosophy of Kapila is the most ancient system of thought among the Hindus, the source of the Yoga doc- trine of Patanjali. Cf. Colebrooke, " Essays," i. 239-279; J. Davies, " Hindu Philosophy," &e., p. Iot seq. The relation between Alberuni's Samkhya and the so- called Samkhyapraracanam ("The Sankhya Aphorisms of Kapila," translated by Ballantyne, London, 1885) is a very distant one, and is limited to this, that there occurs a small number of passages which show a similarity of matter, not of form. The latter book (the Sutras) seems to be a late secondary prodnction; v. A. Weber, Vorle- sungen uber Indische Literaturgeschichte, p. 254, note 250. Besides, the philosophic system propounded by Alberuni under the name of Samkhya seems in various and essen- tial points to differ from that of the Sutras; it seems altogether to have had a totally different tendency. The Sutras treat of the complete cessation of pain; the first one runs thus: " Well, the complete cessation of pain, (which is) of three kinds, is the complete end of man;" whilst the Sainkhya of Alberuni teaches moksha by means of know- ledge. Next we have to compare Alberuni's Samkhya with the Samkhya Karikd of Isvara Krishna (v. Colebrooke, " Essays," i. 272; J. Davies, " Hindu Philosophy," London,

ANNOTATIONS. 267

1881). Both werks teach moksha by means of knowledge, and contain here and there the same subject-matter. It mnst be observed that of those illustrative tales which Alberuni's Samkhya gives in full length, shert indications are found in the Samkhya Karika. Its author, Îsvara Krishna, says at the end of his boek that he has written his seventy Sutras, excluding illustrative tales. This is not quite correct, as semetimes, though he has net teld them, he has at all events indicated them. His words show that he has copied from a book like the Samkhya of Albe- runi, in which the tales were not only indicated, but related at full length. Cf. A. Weher, Vorlesungen über Indische Literaturgeschichte, Berlin, 1876, p. 254, note 250. Hall considers the S. Pravacanam to be younger than the S. Karika. If, in the third place, we examine the Bhashya of Gaudapada, we find that it is net identical with Alberuni's Samkhya, but a near relative of it. Cf. the Samkhya Karika, &c., translated by Colebroeke, also the Bhashya of Gaudapada, translated by H. H. Wilson, Oxford, 1837; Celebreeke, "Essays," i. 245. Most of the quotations given by Alberuni are feund only slightly differing in Gaudapada, aud some agree literally, as I shall point out in the notes to the single passages. Almost all the illus- trative tales mentioned by Alberuni are found in Gauda- pâda, being, as a rule, more extensive in Alberuni than in Gaudapâda The latter seems to have taken his inferma- tien from a work near akin te, or identical with, that Samkhya book which was used by Alberuni. According te Colebrooke (in the preface of the work just mentioned, en p. xiii.), Gandapâda was the teacher of Samkara Acarya, who is said to have lived in the eighth Christian century. Cf. alse A. Weber, Vorlesungen, pp. 179, 254, and 260: Alberuni does not mentien Gaudapâda, as far as I can see. Or is he perhaps identical with Gauda the anchorite, whem Alberuni mentions even befere Kapila ? Cf. the passage, i. 131-132: "Besides, the Hindus have books, &c., on the process of becoming Ged and seeking liberation from the world, as, e.g. the book composed by Ganda the anchorite, which goes by his name." Kapila, the father ef the Samkhya philosophy, is men- tioned by Alberuni also as the author ef a book called

268 ALBERUNI'S INDIA.

Nydyabhasha, " on the Veda and its interpretation, also showing that it has been created, and distinguishing within the Veda between snch injunctions as are obliga- tory only in certain cases and those which are obligatory in general" (i. 132). The subject of this book is evidently not related to the Nyaya philosophy, bnt to the tenets of the Mimamsa philosophy, ie. the Pûrvamimâmsa (Colebrooke, " Essays," i. p. 319-349; J. Davies, " Hindu Philosophy," p. 2; Thibant, Arthasamgraha, Benares, 1882), a system of rules which are applied to the text of the Veda and its sacrificial prescriptions.

P. 31. The anthropomorphic doctrines, the teachings of the Jabriyya sect, &c .- The sect called Jabriyya, Jabariyya, and Mnjbara teaches that the actions of man proceed from God. They are the followers of Al-najjar. Of. Fihrist, p. 179 seq. The All-altashbil, or anthropomorphists, teach that God is similar to His creatures. Cf. Statio Quinta et Sexta et appendix libri Mevakif, edit, Th. Scrensen, Leipzig, 1858, p. 362; Kitab-i-Yamint of Al-Utbi, translated by J. Rey- nolds, London, 1858, preface, pp. xxv. xxix .; " Book of Religious and Philosophical Sects," by Alshahrastani, edited by Cureton, pp. 59, 61, and 75 seq. I understand the passage Adl (s, (1, 11, 12) as meaning the prohibition of the study (not discussion, as I have translated, which would be Ssit) of a subject, i.e. a question of a religious bearing; but I am not aware what particular event the anthor hints at by these words. At the intolerant religious policy of the Khalif Alkâdir? King Mahmud was a great Ketzerrichter. Probably a stont adherent of the theory of the harmony of throne and altar, which his contemporaries Al-'Utbi (in his preface) and Alberuni (i. 99) call twins, he tried to cover the illegitimate, revolutionary origin of his dynasty, which was still fresh in the memory of the men of the time; he maintained the most loyal relations with the spiritnal head of Islam, the Khalif of Bagdad, Alkadir (A.H. 381-422), who had clad the usurpation of his family with the mantle of legitimacy ; and in order to please him, he hnnted down the heretics in his realm in Khurasan as in Multan (ef. Reynolds, L 1., p. 438 seg.), impaling or stoning them. He tried to rid

ANNOTATIONS. 269 the Khalif of the real or suspected votaries of his oppo- nent, the Anti-Khalif in Egypt, the famous Hakim, famous by his madness and by being considered by the Druzes as the originator of their creed. The religious policy of Mah- mud may be retraced to the following principles :- (1.) Perfect toleration for the Hindus at his court and in his army. (2.) Persecution of certain Muslim sectarians in the interest of the Khalif, of the Karmatians and other sects of Shiitic tendencies. (Cf. A. von Kremer, Geschichte der herschenden Ideen des Islam, Leipzig, 1868, p. 127.) (3.) Predilection for a Muslim sectarian from Sijistan by the name of Abu-Abdillah Ibn Alkiram, by whose influence both Sunnites and Shiites had to suffer (cf. Alshahrastant, p. r.). How long the influence of this man had lasted, and how far his doctrines bad been carried into practice, does not appear from Alshahrastani's account. That, notwithstanding all this, there was a large margin for liberty of religious thought under the rule of Mahmud and his immediate successor, is sufficiently illustrated by the tenor of Alberuni's work Altogether, it must bc . kept in mind that before Alghu zali the Muslim Church was not that concentrated organisation nor that all-over- whelming force which it has been ever since and keeps up in our days. To those who only know the centuries of Muslim history after the establishment of the orthodox Church, it sounds next to incredible that the military chief of a Khalif should have been an infidel (a Zoroastrian ?) Cf. the story of Afshin, the general of the Khalif Almn'- tasim, in Menoutchetri, Poète Persan, par A. de Biberstein. Kazimirski, p. 149.

P. 33. To XavOavew .- The word bumun, which I have thus rendered, means to be hidden. Not kuowing to what school of Greek philosophers the anthor refers, I can only give the note of Reiske, " orat Jel, Philosophi qui omnes animas simul et semel creatas et reconditas in Adamo putant" (Freytag, Lexicon Arabicum, s.h.v.).

P. 33. Pailasopa, &c .- As Syrian scholars were the author's teachers in Greek philosophy, he knows the Greek word dirooodos only in its Syrian garb leom o.

270 ALBERUNI'S INDIA.

The Ahl-aspuffa were certain persons, poor refugees and honseless men, who during the first years of Muhammad's. stay in Medina passed the night in the suffa of the mosque of the Prophet in Medina, which was a covered place, an appurtenance of the mosque, roofed over with palm-sticks (Zane). Abulfath Albusti was a famous poet of the time. A native of Bust in Northern Afghanistan, he was in the service of the governor, who held the place nnder the Samani dynasty, and after the conquest of Bust by Sabuk- tagin he entered the service of this prince and of his sou Mahmud. Under Mastd he lived still in Ghazna, for Baihakî mentions that he had fallen into disgrace and had to carry water for the royal stables. By the inter- vention of Baihaki, he was restored into the good graces of the prime minister, Ahmad Ihn Hasan of Maimand. Cf. Elliot, " History of India, ii. 82, 84, iv. 161; Ethé, Radagi's Vorlaufer und Zeitgenossen, p. 55. According to Hâjt Khalifa (iii. 257, iv. 533), he died A.H. 430 (A.D. 1039). For further information see Shahrazuri, Nuzhat-al arwah, fol. 182b (MS. of the Royal Library, Berlin, MSS. Orient. octav. 217); Al-Baihaķi, Tatimmat-puwdn-alhikma, fol. 22b (MS. of the same lihrary, Petermann, ii. 737); also Mirchondi Historia Gasnevidarum Persice, by F. Wilken, Berlin, 1832, p. 144 Towards the end of his life he is said to have travelled with an embassy of the Khakan of Transoxiana to that country, and to have died there.

P. 34 Galenus .- The author quotes the following works of Galenus :- (Ι.) λόγος προτρεπτικός. (2.) A commentary to the aphorisms of Hippokrates, a book of which I do noc know the Greek original (ef. i. 35, ji. 168). - Tepi avvDe =(مام from the Syriac ) كتاب المياسر (.3) σεως φαρμάκων τών κατά τόπους. (4.) lJl ls =the book of the proof, of which I do not know the Greek original; cf. i. 97. (5.) n get = de indole animæ (repì 0a ?), of which the Greek original likewise is not known to me; ef. i. 123, 124. (6.) i1U =περί συνθέσεως φαρμάκων κατά γένη.

ANNOTATIONS. 271

Besides, the author gives some quotations from Galenus without mentioning from what particular book they were taken ; ef. i. 222, 320. Cf. on Galen's works in Arabic Dr. Klamroth, " Journal of the German Oriental Society," vol xl. 189 seq. The passage here given is found in Προτρεπτικός έπι Tas Texvas, ed. Abrah. Willet, Lugduni Bat., 1812, chap. ix. pp. 29, 30 :- ώς και τών άνθρώπων τούς αρίστους θείας αξιωθηναι τιμής, ούχ ότι καλώς έδραμον έν τοϊς άγωσιν ή δίσκον έρριψαν ή διεπάλαισαν· άλλά διά την άπό των τεχνών ευεργεσίαν. Άσκληπιός γέ τοι καί Διόνυσος είτ' άνθρωποι πρότερον ήστην είτ' άρχήθεν, τιμών αξιούνται μεγίστων, ό μεν διά την ιατρική», ο δ' δτι τήν περί τούς άμπέλους ήμας τέχνην έδίδαξεν. The two passages on p. 36 are probably taken from the Protrepticus too. With the former compare the words in chap. ix. (on p. 22 editio Kuhn, vol. i.) : Ei 8' oUK eGEXes έμοι πείθεσθαι, τόν γε θεον αϊδέσθητι τον Πύθιου. Shortly afterwards follows the second quotation, verses quoted by Galen from Herodotus, i. 65:

"Ηκεις, ώ Λυκόεργε, έμον ποτί πίονα νηόν. Δίζω ή σε θεόν μαντεύσομαι ή άνθρωπον, άλλ'έτι. ά μάλλον θεόν έλπομαι, ω Λυκόεργε.

P. 35. Plato .- The author quotes the following works of Plato :- Phædo. Timœus (cf. also Proclus).

Of the three quotations on this passage, the middle one (3.) Leges.

is found in Timceus, 41A: -Eπεl δ ούν πάντες κ.τ.λ., λέγει πρός αύτους ό τόδε το πάν γεννήσας τάδε · θεοί θεων κ. τ.λ., αθάνατοι μεν ούκ έστε ούδ' άλυτοι το πάμπαν· ούτι μέν δη λυθήσεσθέ γε ούδε τεύξεσθε θανάτου μοίρας, της έμης βουλήσεως μείζονος έτι δεσμού καί κυριωτέρου λαχόντες εκείνων οις οτ' εγίγνεσθε ξυνεδείσθε. The first and third quotations are not found in the Greek text, and Ed. Zeller, to whom I applied for help, thinks that both are taken from a commentary on Timæus by some Christian author, as e.g. Johannes Philoponus, the former having been derived from 40D (περί δέ τώων άλλων

272 ALBERUNI'S INDIA.

δαιμόνων είπειν καλ γνώναι την γένεσιν κ.τ.λ.), the latter from passages like 32B and 92B. The index of the works of Johannes Philoponus or Scho- lasticus (Steinschneider, Al-Farabi, p. 152 seq.) does not mention a commentary on Timgus, if it is not concealed under the title of one of his books, slil, ost i, ie. on czisting and perishing. As he was a literary opponent of Nestorins, he seems to have been a strict Monophysite, which would be in keeping with the third quotation, "God is in the single number," &c. Cf. the note to pp. 56, 57-

P. 36. Johannes Grammaticus (identical with J. Philo- ponus and Scholasticus) is five times quoted. There are three extracts from his Refutatio Procli, and two more, the origin of which is not mentioned, but probably taken from the same book. The passage here mentioned is found in Joannis Grammatici Philoponi Alerandrini contra Proclum de Mundi aternitate, libri xviii., Venetiis, 1551, Greek and Latin, in the 18th Aoyos, chap. ix. (there is no pagination; ef. the Latin translation, p. 95) :- μή δέ γάρ είδέναι πω εκείνους αλλό τι θεον πλήν των φαινομένων σωμάτων ήλίου καί σελήνης καί τών λοιπών, ώσπερ καί μέχρι νύν τών βαρβάρων υπολαμβάνειν τους πλείστους, ύστερον δέ φησιν είς εύνοιαν και των άλλων θεών τών άσωμάτων έλληνας έλθόντας, τώ αύτφ κάκείνους προσαγορεύσαι δνόματι. I have not succeeded in identifying the other four quotations, i. 65, 226, 231, 284- Cf. on this author, Fihrist, p. 254, and Dr. Steinschneider, Alfarabi, pp. 152, 162. P. 37. Baal .- The form of the word dy (Syriae Hsa) shows that the Arabic Bible-text which Alberuni used had been translated from Syriac.

P. 39. Mant .- Vide note to pp. 7, 8. P. 40. Gita .- Cf. with these words the Bhagavad-Gita (of J. Davies), xv. 14, 15 :- " Entering into the earth, I sustain all things by my vital force, and becoming a savoury juice, I nourish all herbs (v. 14). "I become fire, and enter into the bodies of all that breathe, &c. And I am seated in the hearts of all : from

ANNOTATIONS. 273

Me come memory, kowledge, and the power of reason," &a (v. 15). Davies snpposes the whole of verse 15 to be an interpo- lation, but this remark mnst, as it seems, be limited to the final sentence of verse 15 only, i.c. to the words: "I form the Vedanta, and I am one who knows the Vedas."

P. 40. Apollonius-A Greek book of Apollonius of Tyana of this title is not known to me, but it exists in Arabic, Jit ls (Liber de Causis), in the library of Leyden, cf. Wenrich, De Auctorum Gracorum Versionibus et Commentariis Syriacis, Arabicis, &c., p. 239.

Pp. 40-44-The Samkhya doctrine of the twenty-five tattvas is found in the commentary of Gaudapada to the Samkhya Karikd of Isvara Krishna, where also the saying of Vyasa (here i. 44 and 104) is found. Cf. the translation of H. H. Wilson, p. 79, i. 14- P. 40. Buddha, dharma, sangha .- This note on the Buddhistic trinity probably rests on the authority of Zurkan, as he was quoted in the book of Eranshahrt: ef. note to pp. 6, 7. It shows that Alberuni had no original information regarding Buddhism, and it justifies his harsh judgment on the worth of the tradition of Zurkân, v. i. 7. The name Buddhodana is nothing, and by mistake derived from Suddhodana, the name of Buddha's father. Perhaps Zurkan had read not ospey but osyesy, which would be Sauddhodani, i,e. the son of Suddhodana or Buddha.

P. 41. Vayu Purana .- Of the Puranas the author had the Aditya, Matsya, and Vayu Purdnas, i.e. only portions of them (i. 130), and probably the whole of Vishnu- Purdna. Most of his Pauranic quotations are taken from Vayu, Vishnu, and Matsya Purânas. Cf. on the Purânas, A. Weber, Vorlesungen, p. 206, aud note 206 on p. 208. P. 42 .- The five mothers are a blunder of the anthor's instead of the five measures, i.e. pancamatrani (pancatan- malrâni). The combination between the senses and the elements, as it is given here and on p. 43, also occurs in the Vaise- VOL. II.

274 ALBERUNPS INDIA.

shika-philosophy of Kanada: ef. Colebrooke, " Essays," i. 293 seq. Compare also Vishnu-Purdna, i. 2, p. 35, and Hall'a note I. There are aimilar elements in the philosophy of the Banddhas or Saugatas : v. Colebrooke, L.c. i. 416, 417.

P. 42 .- The quotation from Homer is not found in the . Greek text, nor do I know the Greek original of the second verse. Were they taken from some Neo-Pythagorean book ?

P. 43. Porphyry .- This is the only quotation from Por- phyry, from a book of his which is not extant in the Greek original. According to Wenrich, I.c. p. 287, there has once been in Syriac a translation of the fourth book of a Liber Historiarum Philosophorum, probably identical with the work here mentioned. The note on the Milky Way (i. 281) is perhapa taken from this same source.

.P. 43. Lacuna-In the Arabic text (n, 15) is missing the relation between the hcaring and the air, the comple- ment to the words hearing airy in L 14

P. 43. Plato .- As the author does not mention the source whence he took these words, I conjecture that they were derived from Timaus, 77, A, B, or from some commen- tary on this passage: ef. note to p. 35.

P. 45. Matres simplices .- Cf. note to p. 42. On the Samkhya theory regarding the nnion of soul and matter, ef. Samkhya Karika, vv. 20, 21, 42, and Gaudapada'a Bháshya.

P. 47. Dancing-girl .- This example is likewise found in Gaudapâda, p. 170 (Bhashya to v. 59 of the Samkhya Karikd); that of the blind and the lame on p. 76 (to t. 21).

P. 48. Mant .- Vide note to pp. 7, 8.

P. 48. The book of Samkhya, &c .- The theory of pre- dominance among the three primary forces (guna), v. in Gaudapâda, pp. 92, 93, to v. 25, p. 49 to v. 12; the com-

ANNOTATIONS. 275

parison of the soul with a spectator on p. 72 to v. 19 (also Bhagavad-Gitd, xiv. 23) ; the story of the innocent among the robbers on p. 74 to v. 20.

P. 49. The soul is in matter, dc .- The soul compared to a charioteer, v. in Gaudapada, p. 66 to v. 17.

Pp. 52-54. Vasudava speaks to Arjuna, &c .- Of those quotations from Gitd, compare the passage, " Eternity is common to both of us, &c., whilst they were concealed from you," with Bhagavad-Gitd, iv. 5: " Many have been in past time the births of me, and of thee also, Arjuna. All these I know, but thou knowest them not, O slayer of foes!" Of the other quotations on these two pages, I do not see how they could be compared with any passage in Bhagavad-Gita, except for the general tenor of the ideas. With the phrase, " For he loves God and God loves him," ef. Bhagavad-Gitd, xii, 14-20, " Who worships me is dear to me."

P. 54. Vishnu-Dharma .- Alberuni gives large quota- tions from this book. He speaks of it i. 132, and trans- lates the title as the religion of God. I do not know the Sanskrit original of the book, for it is totally different from the Vishnu-Smriti, or Vishņu- Sutra, or Vaishnava Dharmasastra, translated by J. Jolly ("The Institutes of Vishnu," Oxford, 1880), a law-book in a hundred chapters, similar to those of Apastamba, Yajna- valkya, Vasishtha, the Grihyasutras, &c. Our Vishnu- Dharma is a sort of Purana, full of those legends and notions characteristic of the literature of Puranas; but the author does not assign it to them. Most of the ex- tracts here given are conversations between the sage Mârkandeya and Vajra, others a conversation between the king Pariksha and the sage Satauika. The extracts treat of mythological subjects (i. 54); the twelve suns (i. 216, 217); the pole (i. 241); the planets and fixed stars (i. 287, 288); star-legends (i. 291); the story of Hiranyaksha (ii. 140); the names of the Manvantaras (i. 387); the domi- nants of the planets (ii. 121); in particular, of chrono- logical and astronomical subjects. The author has taken several series of names from the Vishnu-Dharma. He

276 ALBERUNPS INDIA.

seems to quote it sometimes withont mentioning its title. So, ag. I am inclined to attribute the traditions of Satnaka (i. 113, 126) to this book. The qnotation (ii. 398) on Vasndeva, Samkarshaņa, Pradyumna, and Aniruddha, as the names of Hari in the four Yugas, is fonnd likewise amoug the doctrines of the Vaishnava sect, the Pâncara- .. tras, or Bhagavatas : cf. Colebrooke, " Essays," i. 439, 440. Vishon is the chief god of those Hindns with whom Alberuni held relation. Were they Vaishnava sects, and was the Vishnu-Dharma a special code of theirs? On the heterodox sect of Vishnn or Vasudeva worshippers just mentioned, cf. Colebrooke, L.c. pp. 437-443- Colebrooke mentions a book, Vishnu-Dharmottara- Purdna, which is said to have comprehended the Brahma- siddhauta of Brahmagupta : ef. "Essaya," ii. 348. This work is perhaps identical with the Vishnu-Dharma used by Alberuni. As he had a copy of the Brahmasiddhanta, he had it perhaps as s portion of this larger work.

P. 54 Lakshmt, who produced the Amrita .- For the legend of Lakahmi v. Vishnu-Purdna, i. 9, where it is Dhanvantari who brings the Amrita-cnp, not Lakshmi. Apparently this goddess is meant here, and not Lakshmana, as the mannscript has it, the brother of Rama When Alberuni wrote this, he seems to have mistaken Lakshmt for a masculine being, or else we must write 4y in the text rv, 3, instead of sy'. The Arabic hand'a (=aisance, felicite) is an attempt of Alberuni'a to translate the Sanskrit amrita=smbrosia, which scarcely any one of his readers will have understood. Cf. the Arabic text, ir 6 (here i. 253).

P. 54 Daksha, who was beaten by Mahddeva .- Cf. the story of the destruction of Daksha's sacritice by order of Siva, as communicated by Hall in his edition of Wilson's Vishņu-Purâna as appendix to i. viii. p. 120 seq. (Sacrifice of Daksha, from the Vayu-Purdna).

P. 54. Vardhamihira-Of this author Alberuni quotes the following works :- (1.) Brihatsamhita. (2.) Brihajjatakam, i 158, 219, 220, ii. 118.

ANNOTATIONS. 277

(3- Laghujatakam, i. 158. (4.) Pancasiddhantika, i. 153, ii. 7, 190. Books of the eame author, which Alberuni mentione without giving extracts from them, are Shatpancasikd and y(?), both with astrological contents (i. 158). Perhapa the two books called Yogaydtra and Tikant (?)- yatra (i. 158) are also to be attributed to Varahamihira. Besides there are mentioned several commentaries, one of the Brihat-Samhitd by Utpala, from Kashmir (i. 298), and one of the Brihajjatakam by Balabhadra. One of the sources whence Alberuni has drawn most copiously is the Brihat-Samhita, or, as he calle it, the Samhitd: v. the edition of the Sanekrit original by Dr. Kern, Calentta, 1865, and his tranelation in the " Journal of the Royal Asiatic Society " for the years 1870, 1871, 1873, 1875. Alberuni praises Varabamihira as an honest man of science (i. 366), and maintains that he lived 526 years before his own time, which is A.D. 1030. Accord- ingly, the date of Varahamihira would be A.D. 504. Cf. ii. 86. In the preface to the edition, p. 61, Kern mentions the Shatpancasikd and the Yogaydtra. Both the Brihat-Sam- hitd and Laghujatakam had heen translated into Arabic by Alberuni. The passage here (p. 54) quoted is fonnd in chap. iii. 0. 13-15 ("Journal of the Royal Asiatic Society," 1870, p. 446).

P. 54. Mant .- Vide note to pp. 7, 8.

P. 55. Patanjali .- Vide note to p. 27.

Pp. 56, 57. Phædo .- The two quotations from Phædo are the following :- 700. παλαιός μεν ούν έστι λόγος, ού μεμνήμεθα, ώς είσίν ενθένδε άφικόμεναι έκει, κα πάλιν γε δεύρο άφικνούνται καλ γίγνονται έκ τώων τεθνεώτων, καί εί τουθ ούτως έχει, πάλιν γίγνεσθαι έκ τών άποθανόντων τούς ζώντας, άλλο τι ή είεν άν αί ψυχαί ήμων έκεϊ, κ.τ.λ. αρ' ούτωσί γίγνεται πάντα, ούκ άλλοθεν ή έκ των έναν- τίων τά έναντία, κ.τ.λ. The sentences which in the Arabie follow after these

278 ALBERUNI'S INDIA.

words ("Our souls lead an existence of their own," &c.) cannot be combined with the Greek text, and I auppose they were taken from some commentary. The second qnotation is fonnd 72Ε ότι ήμιν ή μάθησις ούκ άλλο τι ή ανάμνησις τύγχανει ούσα, και κατά τούτοε ανάγκη που ήμας έν προτέρω τιν χρόνω μεμαθηκέναι & νών αναμιμνησκόμθα, τούτο δε άδύνα- του, εί μή ην που ήμων ή ψυχή, πρίν έν τδε τώ άνθρω- πίνω είδει είναι, κ.τ.λ. 73D. ούκοϋν οίσθα ότι οί έρασταλ, όταν ίδωσι λύραν ή ιμάτιον ή άλλο τι, οίς τά παιδικά αύτων είωθε χρήσθαι, πάσχουσι τούτο, έγνωσάν τε τήν λύραν και έν τη διανοία έλαβον το είδος του παιδός, ού ην ή λύρα; τούτο δέ έστν ανάμνησις. In some sentences the Arabic and Greek texts agree literally; in others they differ to such an extent that this extract, too, does not seem to be taken from a simple trans- lation of the text of Phædo, but rather from a work in which text and commentary were mixed together, and the original form of a dialogue was changed into that of a aimple relation. Alberuni erroneously held this to be the original form of the book. We have arrived at a similar result in the case of Plato's Timauts. Proclus has composed a commentary on the saying of Plato that the soul is immortal, in three sections : r. Wen- rich, De Auctorum Gracorum Versionibus, &c., p. 288; and Zeller, Philosophie der Gricchen, iii. 6, 780, I. This was probably an Arabic edition of Phædo, and possibly that one which Alberuni used. Cf. note to p. 35- The qnotations from Phædo given farther on (pp. 65-67) agree more accurately with the Greek original, but in them, too, the dialogistie form has disappeared.

P. 57. Proclus is twice quoted, here and i. 86. Both extracts seem to be derived from some commentary on Timaus, which was different from that commentary known in onr time and edited by Schneider, Breslau, 1887. The words here mentioned probably refer to Timœus, 44 ABC :- κάι διά δή ταΐτα πάντα τα παθήματα νύν κατ' άρχάς τε άνους ψυχή γύγνεται τό πρώτον, όταν είς σώμα ένδεθη θνη- τόν κ.τ.λ. χωλήν τού βίου διαπορευθείς ζωήν, άτελης καί ανόητος είς Διδου πάλεν έρχεται.

ANNOTATIONS. 279

The commentary of Proclus referring to these words (pp. 842, 843, ed. Schneider) is entirely different from the Arabic words. The other quotation (i. 86) is derived from the same book, and refers to Timaus, 44D :- eis adaipocdes copa ενέδησαν, τούτο δ νύν κεφαλήν έπονομάζομεν, ο θειότατον τ' έστλ καί των έν ήμεν πάντων δεσποτοϋν, κ.τ.λ. The commentary of Proclus (ed. Schneider) breaks off a little before this passage, at the beginning of 44D. I am inclined to believe that the work, simply intro- duced by " Proclus says," is identical with that one which he calls Timæus (ef. note to page 35), a work which was -- (1.) Not a simple translation of the book, but a transla- tion and a commentary together, the one running into the other; and which (2.) Was different from the now extant commentary of Timcus by Proclus. Therefore Proclus must either have made two editions of Timœus, or he is not really the author of the book used by Alberuni. In the one place the name .أبروقلس in the other ,بروقلى is written

P. 57 .- The seat (العرض) and the throne (الكرمى) of God. By these two words Muhammad calls the throne of God in the Koran. Allah's sitting on his throne, as mentioned in the Koran, has been a subject of deep speculation among Muslim theologians. Cf. Zur Geschichte Abulhasan Al- Afart's, von W. Spitta, Leipzig, 1876, pp. 106, 107, and the note on p. 144.

P. 60. Vishnu-Purana .- The passage is found in Book II. chap. vi. (Wilson-Hall, ii. p. 216). The order in which the bells are enumerated and their names differ to some extent :- Alberuni. Sanskrit original. Raurava. Raurava Rodha. Rodba. Taptakumbba, Sūkara, Mahájvala. Tala, S. Savala. Krimiáa. 5. Taptakumbha. Taptaloha. LAlabhaksha. Mahajvala Visazana Lavana, Adbomukha. Vimoba. 10. Rudhirandba. 10. Krimibhaksha.

280 ALBERUNTS INDIA.

Alberuni. Sasstrit original.

Rudbira. Krimtóa Vaitaranl. Lalabhakaha. Kriahps Vedhaka. Asipatravans. Viśasana. 15. Vahnijvtla 15. Adhomukba. Půyavaha. .Rudhirandha. Vaitarani Krisbņa. 20. Asipatramans. Vahnijvala. Sandaméa Śvabhojana

P. 62. Mmnkhya .- I do not find anything corresponding in the Samkhya Karikd nor Gaudapada's commentary. As for the idea, cf. " Samkhya Aphorisms," iv. 32.

P. 63. Atinahika .- On the ativahika =that which is swifter than the wind in passing from body to body, ef. Smkhya Karikd, ed. Colebrooke-Wilson, p. 133. The Barzakh is mentioned in the Koran, 23, 102; 25, 55; 55, 20.

P. 63. Vishnu-Purdna-This quotation is related in substance to Book II. chap. vi. pp. 221-224: cf. the unin- terrupted thinking (samsmarana) with the remembrance of Hari, the meditation on Vasudeva, Are the words of Alberuni an extract from this passage ?

P. 64 Samkhya .- The S. Karika and Gandapada do not seem to offer anything analogous to this passage.

P. 64 .- A theosoph, &c .- The passage relating to the four degrees of metempsychosis has been translated into Persian by Abulma ali Muhammad Ibn 'Ubaid-Allah in his Bayan aladydn: v. C. Schefer, Chrestomathie Persane, i. IPA, 1. 3-8. Abd-Ya'kub and his work are not known to me from other sources.

P. 65. Johannes Grammaticus-Vide note to p. 36. Phædo .- The quotations on pp. 65-67 agree pretty accorately with the Greek text.

ANNOTATIONS. 281

The body is earthy, dc., 81 C, D :- Έμβριθες δε γε, ώ φίλε, τούτο οίεσθαι χρή είναι καί βαρύ και γεώδες καί όρατάν· & δή καί έχουσα ή ταιαύτη ψυχή βαρύνεταί τε καί έλκεται πάλιν είς τόν ύματόν τόπον φόβφ του αέιδους τε καί "Διδου, ώσπερ λέγεται, περί τά μνήματά τε καί τους τάφους κυλινδουμένη, περί & δη καί ώφθη άττα ψυχών σκιοειδή φαντάσματα, οία παρέχονται αί τοιαύται ψυχαί είδωλα αί μή καθαρώς

όρωνται. άπολυθεϊσαι, άλλά και τού όρατού μετέχουσαι, διό καί

It appears that these are not the souls, &c., 81D-82A :- Εικός μέντοι, ώ Κέβης· και ού τί γε τάς τών άγαθών ταύτας είναι, άλλά τάς των φαύλων, α περί τά τοιαύτα αναγκάζονται πλανάσθαι δίκην τίνουσαι της προτέρας τροφής κακής ούσης · καί μέχρι γε τούτου πλανώνται, έως αν τη ξυνεπακολουθούντος του σωματοειδούς επιθυμία πάλιν ένδεθώσιν είς σώμα, 'Ένδουνται δε, ώσπερ εϊκός, εις τοιαύτα ήθη όποι άττ' αν καί μεμελετηκυίαι τύχωσιν έν τω βίφ. Τά ποΐα δή ταϋτα λέγεις, & Σύκρατες; Οίον τούς μέν γαστριμαργίας τε καί ύβρεις και φιλαποσίας μεμελετηκότας καί μή διευλαβημένους είς τά τών όνων γένη καί των τοιούτων θηρίων είκός ένδνεσ- θαι · η εύκ οίει; πάνυ μέν ούν είκός λέγεις. Τούς δέ γε αδικίας τε καί τυραννίδας καί άρπαγάς προτετιμηκότας είς τά τών λύκων τε κα ίεράκων και ϊκτίνων γένη. If I did not think that I am going, &c., 63B :- εί μεν μή ώμην ήξειν πρώτον μεν παρά Θεούς άλλους σοφούς τε καί αγαθούς, έπειτα καί παρ' άνθρώπους τετε- λευτηκότας άμεινούς των ένθάδε, ήδίκουν αν ούκ άγανακτών τω θανάτφ.

P. 66. When a man dies, a daimon, &c., 107D, 1080 :- λέγεται δέ ούτως, ώς άρα τελευτήσαντα έκασταν ό έκάστου δαίμων, έσπερ ζώντα ειλήχει, ούτος άγειν έπιχειρεϊ είς δή τινα τόπον, οι δεί τούς συλλεγέντας διαδικασαμένους είς "Διδου πορεύεσθαι μετά γεμόνος εκείνου, ώ δη προστέ-

282 ALBERUNTS INDIA. τακται τος ένθένδε έκείσε πορεύσαι,τυχόντας & έκεί, ων δεϊ τυχείν, καί μείναντας όν χρή χρόνον, άλλος δεύρο πάλιν ήγεμών κομίζει έν πολλαΐς χρόνου καί μακραίς περιόδοις, έστε δέ άρα ή πορεία ούχ ώς Αίσχύλου Τήλεφος λέγει· έκεινος μέν γαρ άπλην οιμόν φησιν είς " Αιδου φέρειν, ή δούτε άπλη ούτε μία φαίνεταί μοι είναι. ούδέ γάρ αν ήγεμόνων έδει, ού γάρ πού τις άν διαμάρτοι ούδαμόσε μιας όδου ούσης, νν δέ έοικε σχίσεις δε καί περιόδους πολλάς έχειν· άπό των όσίων τε καί νομίμων τών ένθάδε τεκμαιρόμενος λέγω. ή μέν κοσμία τε' καί φρόνιμος ψυχή έπεταί τε καί ούκ άγνοεϊ τά παρόντα· ή δ' επιθυμητικώς του σώματος έχουσα, όπερ έν τφ έμπροσθεν είπον, περί εκεϊνο πολύν χρόνον έπτοημένη καί περί τόν όρατόν τόπον πολλά αντιτείνασα καί ολλά ταθύσα βία καί μόγις ύπό του προστεταγμένου δαίμονος οίχεται άγο- μένη, αφικομένην δέ όθιπερ αί άλλαι, τήν μέν ακάθαρτον καί τι πεποιηκνίαν τοιούτον, η φόνων άδίκων ήμμένην ή άλλ' άττα τοιαύτα ειργασμένην, & τούτων άδελφά τε καί άδελφών ψυχών έργα τύγχανει όντα, ταύτην μέν άπας φεύγει τε καί υπεκτρέπεται καί ούτε ξυνέμπορος ούτε ήγεμών εθέλει γίγνεσθαι, αυτή δέ πλανάται έν πάση εχομένη απορία, έως αν δή τινες χρόνοι γένωνται, ών ελθόντων ύπ' ανάγκης φέρεται είς την αύτη πρέπουσαν οίκησιν· ή δε καθαρώς τε καί μετρίως τόν βίον διεξελθούσα καί ξυνεμπόρων καί ηγεμόνων θεών τυχούσα ώκησεν τόν αύτη εκάστη τόπον προσήκοντα. Those of the dead who led a middle sort of life, &c., and Those who repented of their sins, &c., 113D-114C :- καί οι μεν αν δόξωσι μέσως βεβιωκέναι, πορευθέντες έπι τόν Άχέροντα, αναβάντες & δή αύτοίς οχήματά έστιν, έπί τούτων άφικνούνται είς την λίμνην, καί έκεϊ οικουσί τε καί καθαιρόμενοι τών τε άδικημάτων διδόντες δίκας απολύονται, εί τίς τι ήδίκηκεν, των τε εύεργεσιών τιμάς φέρονται κατά την αξίαν έκαστος, οι δ' άν δόξωσιν ανιάτως έχειν διά τά μεγέθη των άμαρτημάτων, ίεροσυλίας πολλάς καί μεγάλας ή φόνους άδίκους και παρανόμους

ANNOTATIONS. 283 πολλούς έξειργασμένοι ή άλλα όσα τοιαύτα τυγχάνει όντα, τούτους δε ή προσήκουσα μοίρα βίπτει είς τόν Τάρταρον, δθεν ούποτε έκβαίνουσιν. οι δ δν ιάσιμα μέν, μεγάλα δε δόξωσιν ήμαρτηκέναι άμαρτήματα, οίον πρός πατέρα ή μητέρα ύπ' όργης βίαιόν τι πράξαντες, καί μεταμέλον αύτοίς τον άλτν βίον βιώσιν, η άνδροφόνοι τοιούτω τωνί άλλφ τρόπφ γένωνται, τοιούτους δε έμπεσεϊν μεν είς τόν Τάρταρον ανάγκη, έμπεσόντας δε αύτούς καί ένιαυτόν έκεί γενομένους εκβάλλει τό κύμα, τούς μεν ανδροφόνους κατά τόν Κωκυτόν, τους δε πατραλοίας καί μητραλοίας κατά τόν Πυριφλεγέθοντα. έπειδαν δε φερόμενοι γένωνται κατά τήν λίμνην τήν Άχερουσιάδα, ένταύθα βοωσί τε καί καλούσιν, οί μεν ούς απέκτειναν, δι δέ ούς ύβρισαν, καλέσαντες δ' ίκετεύουσι καί δέονται έάσαι σφάς έκβηναι είς την λίμνην καί δέξασθαι, καί θαν μεν πείσωσιν, εκβαίνουσί τε καί λήγουσι των κακών, εί δε μή, φέρονται αύθις είς τόν Τάρταρον καί εκείθεν πάλιν είς τους ποταμούς, καί ταϋτα πάσχοντες ού πρότερον παύονται, πρίν άν πείσωσιν ούς ηδίκησαν· αύτη γάρ ή δίκη ύπό των δικαστών αύτοϊς έτάχθη· οι δέ δη αν δόξωσι διαφερόντως πρός τό οσίως βιώναι, οϋτοί είσιν οί τώνδε μεν των τόπων των έν τη γ0 ελευθερούμενοί τε καί απαλλαττόμενοι ώσπερ δεσμωτηρίων, άνω δε είς την καθαράν οίκησιν αφικνούμενοι καί έπι της γής οικιζόμενοι,

P. 68. Ignorance, knowledge .- Cf. Samkhya Karika, v. 44, "By knowledge is deliverance; by the reverse, bond- . age.

P. 69. These eight things, &c .- Cf. the Commentary of Bhojaraja to " The Yoga Aphorisms of Patanjali," &c., v. xlv., also Gaudapada's Bhashya to the Samkhya Karika, v. xxiiL (pp. 83, 84), where he quotes the work of Pata- jali (Pâtanjala).

P. 69. Passing through several stages .- Cf. with these fonr stages of knowledge the "seven kinds of enlightenment" in "The Yoga Aphorisms," ii. v. xxvii., and Commentary.

284 ALBERUNTS INDIA.

The fourth stage of Alberuni's Patanjali corresponds to the seventh kind of Bhojadeva.

P. 70. In the book: Gitd .- There is no passage like this in the Bhagarad-Gita. The words, "pleasures which in reality are pains" (p. 71, 6), may be compared with Bhagavad-Gitd, v. 22: " For the pleasures that are born of (these) contacts are the wombs of pain." A similar sentence recnrs in another quotation from Gita here on p. 78, 1. pen: " Pleasures of a kind which, in reality, are disguised pains."

P. 71. Socrates .- The following quotation is composed of the two passages, Phædo, 65 B-D and 67A :- όταν μέν γάρ μετά του σώματος έπιχειρη τι σκοπεΐν, δήλον ότι τότε έξαπατάται ύπ' αυτού. Άληθη λέγεις. "Αρ' ούν ούκ έν τφ λογίζεσθαι, είπερ που άλλοθι, κατάδηλον αύτη γίγνεταί τι των όντων; Ναί. λογίζεται δέ γέ που τότε κάλλιστα, όταν μηδέν τόυτων αύτην παραλυπη, μήτε άκοή μήτε όψις μήτε άλγηδών μήτε τις ήδονή, άλλ' ό τι μάλιστα αύτή καθ' αύτην γίγνηται έώσα χαίραν τό σώμα, καί καθ' όσον δύναται μή κοινωνούσα αύτω μηδ' άπτομένη δρέγηται τού όντος. Έστι ταύτα. Ούκοϋν καί ένταύθα ή τού φιλοσόφου ψυχή μάλιστα άτιμάζει τό σώμα καί φεύγει απ' αύτου, ζητεί δε αύτη καθ' αύτην γίγνεσθαι. 67A .- καί έν ο αν ζώμεν, ούτως, ώς εδικεν, εγγυτάτω εσόμεθα τού είδέναι, εαν ο τι μάλιστα μηδέν όμιλωμεν τω σώματι μηδέ κοινωνώμεν, ο τι μή πάσα ανάγκη, μηδέ άνα- πιμπλώμεθα της τούτου φύσεως, άλλά καθαρεύωμεν άπ' αύτού, έως αν ό Θεός αντος απολύση ήμας. καί ούτω μέν καθαροί απαλλαττόμενοι της του σώματος άφροσύνης, μετά τοιούτων τε έσόμεθα καί γνωσόμεθα δί ήμων παν τό ειλικρινές· τούτο δ έστίν ίσως τό αληθές, The words die dete eceas (Te, 8) are barbaric Arabic= τότε έζαπατάται ύπ' αύτού. Probably the Syriac transla- tion had a passive word with ouso= in' dvroo, and this was literally rendered in Arabic by &. The reading of the MS. & cannot be accounted for in any way.

ANNOTATIONS. 285 P. 71. From the book Gud .- The text is not found in the Bhagarad-Gita.

P.72. Kapila, for he was born knowing .- Cf. Colebrooke, " Essays," i. 242.

P. 72. Cupidity, wrath, and ignorance .- "The Yoga Aphorisms," ii. 3 seq., mention five aflictions : ignorance. egoism, desire, aversion, and ardent attachment to life. Perhaps we may also compare Samkhya Karikd, v. Lxiii., where seven modes are enumerated hy which nature binds herself: virtue, dispassionateness, power, vice, ignorance, passion, and weakness.

P. 73 .- The three primary forces are rajas, tamas, sattva.

P. 73. To stop all motions, and even the breathing .- Cf. on the stoppage of motion and the expulsion and retention of breath, "Yoga Aphorisms of Patanjali," i. xxxiv., aud the notes of Rajendralala Mitra.

P. 73. In the book Gitd .- The two quotations as given here are not found in the Bhagavad-Gita. Only the com- parison with the lamp occurs in vi. 19: " As a lamp sheltered from the wind does not flicker;" this is the wonted simile of the Yogin who is subdned in thonght," &c. Also the comparison with the waters of the rivers not increasing the ocean is found ii. 70: "He attains to peace into whom all desires enter as rivers enter into the ocean, which is ever filled, and (yet) remains within its bounds," &c.

P. 74. The following nine rules-Five of these command- ments are mentioned in "The Yoga Aphorisms," ii. xxx., and the other four seem to be identical with the five obligations mentioned in ii, xxxii.

P. 75 .- Pythagoras .- I do not know the Greek original of this saying. The idea of the body being a fetter to the soul is freqnently met with in the book of the Neopytha- gorean philosophers, as Apollonius of Tyana and others;

286 ALBERUNI'S INDIA. ef. Zeller, Philosophie der Griechen, iii. 2, p. 156. For two more sentences of Pythagoras, o. i p. 85, where Alberuni states that he has taken them from Ammonius, v. note to p. 85.

P. 75. The book Samkhya says-It is difficult to say whether the Arabic manuscript has ant or u, and not knowing a Sanskrit parallel to this saying, I am thrown upon conjecture. Preferring the latter reading, I trans- late: "Everything which man opines (ie on which he forms an opinion) is a terminus to him, for he does not go beyond it," which may mean that as long as the thinking faculty of soul has not ceased, it is not liberated, has not attained moksha. Cf. Sdmkhya Karika, v. Ixviii .: "When separation of the informed soul from its corpo- real frame at length takes place, and nature in respect of it ceases, then is absolute and final deliverance accom- plished."

Pp. 75, 76. Gild .- The three quotations from this book are not found in the Bhagavad-Gitd.

P. 76. Socrates .- The quotations given here are found in Phado, 84E-85B :- και, ως Έοικε, των κύκνων δοκώ φαυλότερος ύμϊν είναι την μαντικήν, οι έπειδάν αίσθωνται ότι δεϊ αυτούς άποθανεΐν, άδοντες καί έν τφ πρόσθεν χρόνφ, τότε δή πλείστα καλ μάλιστα αδουσι, γεγηθότες ότι μέλλουσι παρά τόν θεόν άπιέναι ούπερ είσι θεράποντες, κ.τ.λ. άλλ' άτε, οίμαι, τού Απόλλωνος όντες μαντικοί τέ είσι καί προειδότες τά έν "Άιδου άγαθά άδουσι καί τέρπονται έκείνην τήν ήμέραν διαφερόντως η έν τφ εμπροσθεν χρόνφ. έγω δε καί αύτός ήγούμαι ομόδουλός τε είναι τών κύκνων καί ερός τον αυτού θεού, καλ ού χείρον εκείνων την μαντικήν έχαν παρά του δεσπότου, ούδε δυσθυμότερον αύτων τού βίον άπαλ- λάττεσθαι. In the middle a passage has been left out by Alberuni, or by the anthor of that edition of Phædo which he used.

P. 76. In the book of Patanjali .- To the explanation of

ANNOTATIONS. 287

the four parts of the path of liberation on pp. 76-80 I do not know a parallel from a Sanskrit source.

P. 77. In the book Vishnu. Dharma .- Cf. on this the note to p. 54 The Arabio text has not Parikshit, but Pariksha, which name is mentioned by Hall in a note to Vishnu-Purdņa, iv., chap. xx. p. 154.

Pp. 78,79. The book Gitd-These three extracts are not found in the Bhagavad-Gitd. The words, " He who mor- tifies his Inst," &c., compare with Bhagarad-Gita, iv. 21, "Void of hope, self-restrained in thought, performing merely bodily work, he contracts no siu." Regarding the passage, "Pleasures of a kind which, in reality, are dis- guised pains," v. note to p. 70. The expression, the nine doors of thy body (p. 79, 8), is also found in Bhagavad-Gita, v. 13: "in the city of nine gates," ic. in the body. Cf. also Sarikhya Karika, v. xxXV.

Pp. 79, 80. The book Gitd .- These quotations cannot be compared with anything in the Bhagavad-Gita.

P. 81. Patanjali .- There is a certain resemblance be- tween these words and the last of " The Yoga Aphorisms" (iv. xxxviii.): " Isolation is the regression of the qualities devoid of the purpose of soul, or it is the abiding of the thinking power in its own nature."

Pp. 81, 82. Samkhya,-The comparison with the wheel of the potter (not the silk-weaver) is also found in Sainkhya Karika, v. Lxvii.

P. 82. In the book of Patanjali .- I have not found these two passages anywhere else. As to the faculties of the perfect Yogin, ef. " Yoga Aphorisma," iil. 42, 44, 45.

P. 83. The Suft explain the Koranic verse, &c .- Being asked about the story of Dhulkarnaini (Bicornutus, i.e. Alexander), Muhammad says, " We (i.e. Allah) have made room for him on earth; " or, as Sale translates, " We stab- lished for him on carth," which means, We have given him

288 ALBERUNPS INDIA.

a position of well-established authority or power on carth, and this snthority or power is interpreted by Shfi com- mentators in accordance with their tenets, perfectly har- monising with those of the Yoga philosophy.

Pp. 83, 84. Samkhya .- With the tale of the man tra- velling in the night with his pnpils compare a similar one in Gaudapada's Bhashya to Samkhya Kdrikd, D. xxx. (ou p. 106).

P. 85 .-- Ammonius, a philosopher of the Neoplatonic school, v. Zeller, Philosophie der Griechen, iii.c. 829 scq. A Greek book of his which contains these extracts from Pythagoras and Empedocles is not known. He has been known to the Arabs as commentator of Aristotle: v. Wen- rich, De Auctorum Gracorum Versionibus, p. 289; Fihrist, P. rr. By Heracles in the passage, " Empedocles and his suc- cessors as far as Heracles," is probably meant Heraclides Ponticus.

Pp. 85, 86. Socrates says-The first extract is identical with Phædo, 79D, the second is composed of 80B, 80A, 8I AB, the order of the Greek text having been aban, doned.

Plucedo, 790. "Οταν δί γε αυτή καθ' αύτην σκοπή, εκείσε οίχεται είς τό καθαρόν τε καί αει δν καί αθάνατον καί ώσαύ- τως έχον, καί ώς συγγενής ούσα αύτοϋ άει μετ' εκείνου τε γίγ- νεται, ότανπερ αύτή καθ' αύτην γένηται καί εξη αύτη, καί πέπαυταί τε τού πλάνου καί περί έκείνα άεί κατά ταυτά ώσαύτως έχει άτε τοιούτων εφαπτομένη· καί τούτο αύτης τό παθημα φρόνησις κέκληται.

80Β. Σκόπει δή, έφη, ώ Κέβης, εί έκ πάντων των είρη- μένων τάδε ήμεν ξυμβαίνα, τφ μέν θει καί αθανάτω καί νοτίφ καί μονοειδεί καί αδιαλύτω καί άεί ωσαύτως καί κατά ταυτά έχοντι έαυτφ όμοιότατον είναι ψυχήν, τφ δ'άνθρω- πίνφ καί θνητφ καί άνοητα καί πολυειδεί καί διαλυτώ καί

ANNOTATIONS. 289 μηδέποτε κατά ταυτά έχοντι έαυτφ όμοιότατον αύ είναι σώμα.

8OA. έπειδάν έν τφ αύτφ ώσι ψυχή καί σώμα, τφ μέν δουλεύειν καί άρχεσθαι ή φύσις προστάττει, τη δε άρχειν καί δεσπόζειν.

81 A and B. Ούκούν ούτω μέν έχουσα είς τό όμοιον αύτη, τό άειδές, απέρχεται, τό θεϊόν τε καί άθάνατον καί φρόνιμον, οι αφικομένη ύπάρχει αύτη ευδαίμονι είναι, πλάνης καί ανοίας καί φόβων καί αγρίων έρώτων καί τών άλλων κακών των άνθρωπείων απηλλαγμένη, ώσπερ δέ λέγεται κατά τών μεμνημένων, ώς άληθως τόν λοιπόν χρόνον μετά των θεων διάγουσα; ούτω φώμεν, ώ Κέβης, η άλλως; ούτω νή Δι', έφη ο Κέβης - Έαν δέ γε, οίμαι, μεμιασμένη καί ακάθαρτος του σώματος απαλλάττηται, άτε τω σώματι άει ξυνούσα καί τούτο θεραπεύουσα καί έρωσα καί γεγοη- τευμένη ύπ' αύτού, υπό τε των έπιθυμιών καί ήδονων, ώστε μηδέν άλλο δοκεΐν είναι άληθές άλλ'η τό σωματοειδές ού τις αν άψαιτο, κ.τ.λ.

Pp. 86, 87. Arjuna says .- The comparison of Brahman with an asvattha tree is found in Bhagavad-Gita, xv. 1-6, and x. 26. The doctrine of Patanjali-Ideas similar to these Suff sentences are found in Bhagavad-Gita, vi. 28-31, describ- ing the union of the soul with Brahman.

Pp. 87, 88 .-- On Abu-Bakr Ash-shibli ef. Ibn Khallikan, translated by De Slane, i. 511-513; Abulmahasin, Annales, ii. 313. He lived in Bagdad, was a pupil of Junaid, died A.H. 334=A.D. 946, in Bagdad, and was buried there. Ou Abu-Yazid Albistami cf. Ibn Khallikan, nr. 311. He died A.H. 261 =A.D. 875. Jamt has articles on these two mystics with many quotations from them in his Nafahdt- al uns (Lee's "Persian Series," the Nafahat-alons, &c., or the Lives of the Soofis, by Jami, Calcutta, 1859, pp. 201 and 62).

P. 88. The Suft explain the Koranic passage (Sura 2, 68), &c .- " And when you had killed a person aud were dis- VOL II, T

290 ALBERUNIS INDIA.

puting among yourselves (the one throwing the blame on the other), whilst God was bringing to light what yon concealed, then we spoke: Beat him (the killed per- son) with part of her (the killed cow mentioned in the preceding"). In that case the killed person will again become alive and tell who murdered him. " Thus God brings to life the dead ones," &c. Cf. A. Geiger, Was hat Mohammed aus dem Judenthume aufgenommen ! Bonn, 1833, p. 172. Muhammad has moulded this part of Stra z from elements taken directly or indirectly from Numb. xix. 2 8eg., and Deut. xxi. 2 seg. The Sufies try to show by this sentence that the body must be mortified before the heart can become alive by myatic knowledge.

P. 89. Samkhya .- For the two enumerations of created beings, v. Gandapada to S. Karika, liii. p. 16z, and xliv. p. 143- The reading of the MS. ery is certainly wrong. The author means saumya = r, but it would have been better to write py in accordance with wys = daitya. As all the other words of this ennmeration stand in the sin- gular, it is not allowable to read this word in a plural form, o'y- like o, the Rishis, oun the Pitris.

P. 90. In the book Gita .- The first quotation on the prevalence of one of the three gunas, sattra, rajas, tamas, is to be compared with Bhagarad-Gita, xvii. 3, 4, seq., and xiv. 6-8 seq. The aecond extract, " Belief and virtue," &c. I am inclined to combine with Bhagavad-Gita, xvi. 3, 4, seq.

P. 91. People say that Zoroaster, &c .- The author was aware of the identity of the Persian dev (demon) with the Indian deva (god). It is in this way that he tries to account for the discrepancy of the meaning.

P. 92. Samkhya, v. p. 89; Vasudeva, v. p. 90, or Bhaga- vad-Gita, xvii 4

P. 95. Galenus, περί συνθέσεως φαρμάκων τών κατά TOTOUs, ed. Kühn, vol xiii. p. 268 :-

ANNOTATIONS 291

Ξανθήν μέν τρέχα βάλλε μυρίπνοον ϊσοθέοιο Ο5 λύθρος Έρμείας λάμπεται έν βοτάναις. Κρόκου δε σταθμόν φρένας άνέρος, ού γάρ άδηλον, Βάλλε δέ καί δραχμήν Ναπλίου Έυβοέως, κ.τ.λ. Δραχμήν καί ρίζης ψενδωνύμου, ην άνέθρεψε Χώρος ό τόν Πίσση Ζηνα λοχευσάμενος.

The second quotation, v. on p. 271 :- άξιοϊ βάλλειν ην ψευδώνυμον είρηκε βίζαν, έπειδη στάχυς ονομάζεται νάρδου · βούλεται δ' αύτην είναι Κρητικήν, ένθα φησίν, ην άνέθρεψε χώρος ό τόν Πίσση Ζήνα λοχεύσα- μενος, έπειδή τον Δία φασίν οί μυθολόγοι κατά τό Δικ- ταϊον όρος έν Κρήτη τραφήναι, κρυπτόμενον ύπό της μητρός 'Ρ'ας, όπως μή καί αύτός υπό τού πατρός του Κρόνου κατα-

P. 96. Europe, the daughter of Phænix, &c .- In the source whence the author drew his information about Greek legends, Greek, Hebrew, and Persian traditions seem to have been mixed together. It was synchronistic like the Chronicon of Eusebius, with which it is nearly re- lated (note to p. 105), comparing the dates of Greek his- tory with those of the Biblical and Persian history. Julius Africanus and Ensebins are the fathers of this kind of literature, but I do not know by whom the book which Alberuni used had been composed. Cf. Eusebi chronicorum canonum quæ supersunt, ed. A. Schœne, ii. p. 13 (Zeus), 26 (Cecrops), 32, 34 (Asterius); also the Syriac Epitome, p. 204, 206.

P. 96. The story of Alexander is derived from the romance of Pseudo-Kallisthenes (ed. Didot), which Eastern scholars have mistaken for a historic record. "Man cannot oppose the gods" (p. 97, 1)=mpos Tavras γαρ δυνάμεθα οί βασιλεϊς, πρός δε τούς θεους ού δυνάμεθα (ed. Didot, i. 9). "When then he died," &c., "from a wound in the neck," &ο. (p. 97, 4)=πεσών δέ Νεκτανεβώς λαμβάνει φοβερόν τραθμα κατά του ίσχίου αύτοϋ (i. 14).

392 ALBERUNPS INDIA.

P. 97. Galenus .- Cf. note to p. 34- P. 97. Aratus .- The anthor quotes the Phænomena and a commentary to them, which exhibits certain relations with the scholia edited by Immanuel Bekker, but is not identical with them. As I learn from my colleague, Pro- fessor C. Robert, this commentary is to be combined with the Catasterismi of Pseudo-Eratosthenes. The first quotation from Aratus is v. I sq. 'Έx Διός άρχώμεσθα, τόν ουδέποτ' άνδρες έμεν Αρρητον · μεσταί δ Διός πάσαι μεν αγνιαί, Πάσαι δ' άνθρώπων άγοραί, μεστή δε θάλασσα Καί λιμένες · πάντη δε Διός κεχρήμεθα πάντες. Του γάρ καί γένος είμέν· ό δ ήπιος ανθρώτοισεν Δεξιά σημαίνει, λαούς δ έπι έργον έγείρει, Μιμνήσων βιότοιο· λέγει δ στε βλος αρίστη Βουσί τε καί μακέλησι· λέγει δ' ότε δεξιαί ώραι Καί φυτά γυρώσαι, καί στέρματα πάντα βαλέσθαι. Αύτος γάρ τάγε σήματ' έν ύραιφ έστήριξεν, Αστρα διακρίνας · έσκέψατο & είς ένιαυτόν Αστέρας, οι κε μάλιστα τετυγμένα σημαίνοιεν Ανδράσιν ώράων, δφρ' έμτεδα πάντα φύνται. Tο μιν δεί τρώτόν τε καί ύστατον ίλάσκονται. Χαίρε, πάτερ, μέγα θαύμα, μέγ' άνθρώποισιν άνειαρ, Αύτός καί προτάρη γενεή, χαίροιτε & Μούσαι Μειλίχιαι μάλα πόσιν, κ.τ.λ.

P. 97. Commentary on the Phanomena of Aratus-The following quotation from the Scholia Sangermanensia, p. 55, I owe to the kindness of Professor Robert: " Crates autem Jovem dictum cœlum, invocatum vero merito ærem et stherem, quod in his sint sidera, et Homerum Jovem dixisse in aliqua parte cœlum." όs δ ότι ταρφείαι ναφέλαι Διός έκποτίονται -(llias, i. 3571). The common tradition of this verse is- όs δ ότι ταρφείαι κεφάδεις Διός ίκποτίονται,

ANNOTATIONS. 293

and thus it has been rendered by Alberuni. Cf. on the Scholia Sangermanensia, C. Robert, Eratosthenis Catasteris- morum Reiquia, Berlin, 1878, p. 21.

P.99. These twins, state and religion .- Vide note to p. 79. P. 100. When Ardashtr Ibn Babak .- Cf. with these ranks of the Persian nation under the Sasanians the "Chronology of Ancient Nations," translated by Dr. Edward Sachau, London, 1878, pp. 203 and 206; Geschichte der Perser und Araber zur Zeit der Sasaniden, by Th, Noldeke, P- 437 seg. P. 101. The Vaisya who were created from .- In the Arabic text, T, 4, there is a lacuna, where originally stood the words " from the thigh (dru) of Brahman. The Sudra who were created from." mukha-bahu-uru-paj-jandm. Cf. Manu, Dharmasastra, i. 87,

P. 101. Hadt, Doma, &c .- Of these classes of outcast people, the Badhatau are not known to me. The Candala are well known, called Sanddlia by Ibn Khurdadhbih (Elliot, " History of India," i 16). The Hadis and Dom are mentioned hy Colebrooke, " Essays," ii., " Ennmeration of Indian Classes," p. 169, note 3. On the latter (ef. Rom, the name of the gipsies), v. " Memoirs on the History, Folk-lore, and Distribution of the Races," &c., by Elliot, edited by Beames, London, 1869, i. p. 84. Are the Bad- hatau identical with the Bediyas, mentioned in the note of Colebrooke just quoted ? P. 103. Vasudeva answered .- The first quotation from Gita is identical with Bhagarad-Gita, xviil. 41-45; the second is similar to ii. 31-38. P. 104 .- The saying of Vyasa .- Vide note to pp. 40-44. P. 104. Vasudeva .- This quotation from Gitd much resembles Bhagavad-Gita, ix. 32, 33. P. 105. Minos .- I cannot acquit the book on ancient history which Alberuni used of the blunder of having split the Minos of Greek traditions into two persons, a Minos and a Mianos (sic). Cf. on this source note to p. 96.

294 ALBERUNPS INDIA.

At the time of Daris, dc .- Except the synchronism of Persian history, the whole passage relating to Numa Pom- pilius may be derived from Eusebins, Chronicon, ii. 82 :- 1. Νουμάς μετά 'Ρωμύλον βασιλεύσας 'Ρώμης πρώτος νόμους 'Ρωμαίοις είσήγαγεν· [ο αυτος το Καπετώλιον έκ θεμελίων φκοδόμησεν"] ο αυτός το ένιαυτ δύο μήνας προσέθηκε, τόν το 'Ίανουάριον καί τόν Φεβρουάριον, δεκα- μηναίου τού ένιαυτού πρό τούτου χρηματίζοντος· ό αύτος

όστράκινα, καί κογγιάριον έδωκεν, άσσάρια ξύλινα καί σκύτινα καλ

P. 105. Plato .- These extracts from Plato's Leges are the remnant of an Arabic translation. We give the Greek text for the purpose of comparison :- Ι. Ι. Αθηναίος. Θεός ή τις άνθρώπων ύμν, & ξένοι, 14 είληψε την αίτίαν της τών νόμων διαθέσεως; Κλεινίας, Θεός, & ξένε, Θεός, ώς γε το δικαιότατον είπεϊν, παρά μεν ήμιν Ζεύς, παρά δέ Λακεδαιμονίοις, δθεν οδ έστίν, οίμαι φάναι τούτους ' Απόλλωνα, Ι. 6. Ώσπερ τό τε άληθές, οίμαι, καί το δίκαιον ύπέρ γε θείας διαλεγομένους λέγειν, ούχ ώς πρός άρετης τι μόριον καί ταύτα τό φαυλότατον ετίθει βλέπων, άλλά πρός πάσαν άρετήν, κ.τ.λ. I. 6. οί Κρητών νόμοι ούκ είσι μάτην διαφερόντς έν πάσιν ευδόκιμοι τοίς Έλλησιν· έχουσι γάρ όρθως, τούς αύτοίς χρωμένους ευδαίμονας αποτελούντες· άπαντα γόρ τά άγαθά πορίζουσι, Π. 1. Θεοί δε, οικτείραντες τό τών ανθρώπων έπίπονον πεφικός γένος, άναπαύλας τε αύτοίς των πόνων έτάξαντο τάς τών έορτων άμοιβάς, κα Μούσας 'Άπόλλωνά τε μουσαγέτην καί Διόνυσον ξυνεορταστάς έδοσαν, II. Ι. ήμϊν δέ ούς είπομεν θεούς ξυγχορευτάς δεδόσθαι, τούτους είναι καί τούς δεδωκότας την ενρυθμόν τε καί έναρμόνιον αίσθησεν μεθ' ήδονης, η δή κινείν τε ήμας καί χοργγείν ήμιν τούτους, φδαίς τε καί όρχήσεσιν αλλήλους ξυνείροντας, χορούς τε ώνομακέναι το παρά της χαράς έμφυτον όνομα.

ANNOTATIONS. 295 P. 107. Śantanu .- Cf. Vishnu-Purana, iv. ch. xx. p. 158, and the notes. The story of the curse of Pandu is given in the Mahabharata Adiparvan, v. 3812 seq. Vydsa .- His mother is Satyavati: v. Vishnu-Purâna, l. c. The birth of Vyasa is mentioned in Mahabharata Adiparvan, v. 3802.

P. 108. Pancahtr, better Panchtr .- The author means the alpine countries of the Hindukush between Kashmir and a line from Faizabad to Kabul, i.e. the Hazara country, Svat, Citral, and Kafiristan. It is well known that poly- andry erists among the Tibetan tribes in the Alps between Kashmir and Tibet, but I am not aware whether it is also found among the inhabitants of the more western exten- sion of the Himalaya which he mentions, e.g. among the Siyahposh. On polyandry in the Panjab v. Kirkpatriek in " Indian Antiquary," 1878, 86. The Panchir mentioned by the author is the tributary of the Kabnl-Rud. Another Pancahir (sic) is mentioned by the Arab geographer Yakut as a city in Bactriana with rich silver mines. Among the heathen Arabs .- Cf. here i 185.

P .. 109. A certain Jewish marriage .- On this custom in India and Indian tradition, cf. Elliot-Beames, " Memoirs," i. 274, s.v. Karâo.

P. 109. Barshawar the Girshah .- This seems to be a mis- take, and I propose to read, as I have done in the edition of the Arabic text, siastoy, i.e., the Shah of Padashvargir or Prince of Tabaristan (as e.g. Gilanshah=the Shâh of Gilan). Cf. P. de Lagarde, Beitrage zur Baktrischen Lexi- cographie, p. 50 ; Sachau, "Chronology of Ancient Nations," P. 47, 19, and note; Noldeke, Geschichte der Perser und Araber zur Zeit der Sasaniden, p. 462. P. 112 .- The story of Romulus is drawn from the Chro- nographia of Joannes Malalas, book vii. (Bonn edition, p. 172). P. 113. Ambarisha .- The story of this king seems to have been taken from the Vishnu-Dharma, v. note to p. 54 Probably Ambarisha, the son of Nabhaga, is meant,

296 ALBERUNP'S INDIA.

famous as a worshipper of Vishnu. Cf. Vishņu-Purana, book iv. chap. ii. p. 257, note I. P. 116. Narada .- The story of this saint, a Mosea in India, is not known to me from other sources. P. 116. Jalam Ibn Shaiban .- The pronnnciation of the former name is conjectural, the history of this Karma- tian chief nnknown. The expedition of King Mahmud against Multan took place A.D. 1006, in the ninth year of his rule, the seventh year of his nsurpation of sovereignty, in which he had left out the name of his Samant liege- lord on the coins and in the public prayer, and had received the investiture, a robe and a title, from the source of all legitimacy in the Muslim world, the Khalif Alkadir, the great enemy and persecntor of the Karmatians. Cf. on this expedition Elliot, " History of India," ii. p. 441. P. 116, 1. 21 .- There is an error in the calculation of the years. From the end of the Kritayuga np to the year 4132 of the Kaliyuga there have elapsed- Of the TretAyuga Years.

Of the Dvaparayuga 1,296,000

Of the Kaliyngs 864,000 4132

Sum 2,164,132 As Alberuni gives but 216,432 years, it seems he has omitted by inadvertence the cipher 1 (Schram). P. 117,1. 7 .- The above snpposition is confirmed by this passage; it ought to be the 132 years instead of the 432 years. One can consider 132 years as a kind of arbitrary equivalent for the anm of about 10o years, but 432 years cannot be an equivalent for about 1oo years (Schram).

(Schram). P. 117, 1. 10 .- It must be 2,164,000 instead of 216,000

P. 117. Vardhamihira says .- This extract is a transla- tion of Brihat-Samhita, chap. lviii. $S 30-48, 56-57, on the fabrication of the idols (p. 117-120); chap. Iviii. SS 4952, on the consequences of faults in the construction of idols (p. 120); chap. x. § 19, on the varions classes of prieats (p. 121); chap. Ix. $$ 4, 5, on the effects of the

ANNOTATIONS. 297

idols (p. 121). The order of the single verses is to some extent different from that of the Sanskrit text as exhibited in the edition of Kern. In the Arabic text, p. av, t, in the lacuna after والسهم are reqnired the words والسيف والعرس ("the sword and shield ").

P. 122. Gitd .- I do not know similar passages in Bha- garad-Gitd. The first quotation distantly reminds one of Bhagavad-Gita, iv. 25.

P. 123. Plato .- This qnotation shows considerable con- fusion in the rendering of the Greek text. Cf. Leges, iv. 8. πρώτον μέν, φαμέν, τιμάς τάς μετ' Όλυμπίους τε καί τους τήν πόλιν έχοντας θεούς τοίς χθονίοις αν τις θεοΐς άρτια καί δεύτερα καί άριστερά νέμων ορθότατα τού της ευσεβείας σκοπού τυγχάνοι, τοίς δε τούτων άνωθεν τά περιττά καλ αντίφωνα τοίς έμπροσθεν ρηθείσι νύν δή· μετά θεούς δε τούσδε καί τοίς δαίμοσιν δ γ' έμφρων οργιαζοι 7' άν, ήρωσι δε μετά τούτους· έπακολουθεί δ' αύτοίς ιδρύ- ματα ίδια πατρώων θεών κατά νόμον όργιαζόμενα· γονέων δε μετά ταύτα τιμαί ζώντων, ώς θέμις, οφείλοντα άποτίνειν τα πρώτά τε καί μέγιστα οφειλήματα, κ.τ.λ. The nnderlined words are the original of the Arabic qno- tation. The translator has rendered Salpoow by &ll (gods), ipoot by wltS, by which elsewhere the word Mouoal is translated, and opyafew by jdy (instead of Jt=Ml). He seems to have mistaken the meaning of the word enaxoNoufet, translating in this way: "they (the ίδρύματα=(Lω.Ι) follow in rank after the πάτρφοι θεοί," ί.ε. yon shall not put the rarpoor eeol in the first place, but worship them secundo loco. P. 123. Galenus .- Vide note to p. 34 P. 126 .- The tradition of Sannaka from Venus (so the Arabic text), i.e. Sukra, is perhaps taken from the Vishnu- Dharma: v. note to p. 54. Vishņu-Purana .- Compare this quotation with book iii. chap. ii. p. 29 (ed. Wilson-Hall). The Great Bear is called the Seven Rishis in Sanskrit.

298 ALBERUNPS INDIA.

P. 126. Vasukra .- This reading does not quite accu- rately correspond to the Arabic signs, which must be read Vasukra. I have preferred the former, because it is men- tioned in the St. Petersburg Dictionary as the name of a man who occurs in the Veda as a poet of Vaidic hymns.

P. 127. Galenus .- The quotation from Galenus must be compared with the following passage in his περί συνθέσεως φαρμάκων κατά γένη (ed. Kubn, tom, xiii. p. 995) :- ήυρέθη δε ύπό Μενεκράτους, κ.τ.λ, Ιατικόν φάρμακον. επιγέγραπται δε τό βίβλιον, κ.τ.λ. αυτοκράτωρ όλογράμ- ματος· αυτοκράτωρ μέν, έπειδή τούτφ προσπεφώνηται, λογράμματος δε διότι χωρίς χαρακτήρων όλαις ταίς συλ- λαβαίς γέγραπται β' καί γ' καί δ' καί έ καί τών άλλων άριθμων έκαστος, κ.τ.λ. τούτο δ΄ έπραξεν ο Μενεκράτης, έπειδη πολλάκις ού μόνον άκόντων αμαρτάνεσθαι συμβαίνει κατά τάς γραφάς, άλλά καί διά φθόνον εκόντων ενίων, κ.τ.λ. εικότως ούν ηυδοκίμησε τά Δαμοκράτους βίβλια των φαρμάκων είς μέτρα γραφέντα [καί έίπερ άπαντα τον τρόπον τούτον εγέγραπτο], κάλλιστον αν ήν.

by Alberuni That which I have underlined forms the text as given

P. 127 .- Vyasa had four sishya .- Cf. Vishnu-Purana, book iii. chap. iv.

P. 128. A peculiar kind of recitation .- This is a descrip- tion of the four pathas, padapatha, kramapatha, &c. Cf. Colebrooke, " Essays," i. 18.

P. 128. Kandin-The word V evidently refers to the divisions of the Yajurveda called kandild. "The text of the Yajurveda is composed of Kanri, and its name (the name of Yajurveda ? what name of it ?) is derived from it (from kanrt?), i.e. the collection (or totality) of kanrt." It does not appear which one of the names of Yajurveda is here meant by the author as having been derived from

ANNOTATIONS. 299 kdnrt. Is there a name of Yajurveda like kdndika or kandin, meaning consisting of kandikds ? In kanri=kandika the cerebral d is rendered by an Arabic r, as in s kudava, sh vyâdi, " garuda, »s dravida, u nadi, ly vinadi, Em vaidurya, &c. The termination in long € seems to be characteristic of the vernacular form of Indian speech, and is probably a sur- vival of the more ancient termination ika, ikd. Cf. R. Hornle,. " Comparative Grammar of the Gaudian Lan- guages," § 195, 203, 205.

P. 128. Yajnavalkya-Cf. Vishņu-Purana, book iii. chap. v.

P. 129. The well-known story .- It is told by Alheruni himself, i. p. 396. P. 131. Vishņu-Purdna .- This index of the Puranas occurs in book iii. chap. vi, p. 66, 67. In the Arahic text 1r, 12, read S instead of ss.

P. 131. Smriti .- The author erroneously calls it a book. It is the literature on law, and the twenty sons of Brabman here mentioned are authors of Dharmasastras. Cf. on smriti (opp. śruti), Colebrooke, "Essays," i. 337, 466; A. Weber, Vorlesungen, p. 296, note 327; Indische Studien, i. 232. Alberuni sometimes quotes the book Smriti. However, he had not the book himself, but transferred those quota- tions from the Brahmasiddhanta of Brahmagupta. In reality it is the latter author who quotes it. As, according to him, the book smriti was composed by Manu (v. here ii. 110, 11I), he means tbe Dharmasastra of Manu. This law code is only once elearly referred to by Alberuni (ii. 164), but in a manner which makes me think that it was not in his hauds. On Manu, as the anthor of the great Manasa (a work on astronomy and astrology ?), v. p. 157.

P. 132. Gauda .- On the proposed identification with Gaudapâda, v. note to p. 30. Samkhya .- Vide the same note. Patanjali .- Vide note to p. 27. Nyayabhasha .- This my transliteration of Algu will perhaps seem doubtful, as the contents of the book have

300 ALBERUNI'S INDIA.

no relation to the Nyaya philosophy or logical system of Gautama (cf. Colebrooke, " Essaya," i. 280), bnt are clearly identical with the Mimamsa philosophy of Jaimini, who is here mentioned a few linea farther on. However, I do not know another mode of reading the word. That Kapila was the author of such a work does not seem to be known. Mimamsd .- Cf. Colebrooke, "Essays," i. 319. In oppo- aition to Kapila, Jaimini teaches that the Veda ia primeval and superhuman. This theory and the discussiona throngh which it has passed are also found in the history of Isiam applied to the Koran. According to Islam, the Koran too is primeval and superhuman. Laukdyata: read Lokayata .- It is the materialiatic doc- trine of the Carvaka sect that perception alone is a meana of proof. Cf. G. A. Jacob, " Mannal of Hindu Pantheism," Vedantasara, p. 74; Colebrooke, "Essaya," i. 426 seq., 456 seg .; J. Muir, verses from the Sarva-darfana-sangraha, &c., illnstrating the tenets of the Charvakas or Indian materialists,"Journal of the Royal Asiatic Society," 1861,p. 299, and "Journal of the German Oriental Society,"xiv. 5 19. Brihaspati is the founder of this achool; his sutra is qnoted by Bhâskara-âcârya. The Barhaspatyasutram is mentioned by A. Weber, Vorlesungen, p. 263.

P. 132. Agastya .- His doctrine is not known to me. Is it identical with that of the Jainas? Cf. Colebrooke, " Essaye," ii. 173. Vishnu-Dharma .- Vide note to p. 54 P. 132. Bharata, i.e. Mahabharata, which ia repeatedly mentioned by Alberuni. Bhagavad-Gitd is a part of it (i. 132). The story of the birth of Vasudeva and of his five brothers (i 401-406) is taken from Mahabharata. I am not quite certain whether Alberuni had a copy of the work. When giving quotationa from the book, he does not mention it, which he probably would have done if he had had it in hand. P. 133 .- With the index of the chapters of Mahabharata cf. Monier Williams, " Indian Epic Poetry," p. 91 seq. The list of Alberuni exhibits some remarkable differences.

P. 135. Panini. The reading of the MS. is panriti,

ANNOTATIONS. 301 wy which I cannot explain. If ort pdnrini is the cor- rect reading, we must remember that in the sound n there is an admixtnre of the sound r. So Hornle, " Compara- tive Grammar," p. 15, says: " The cerebral e contains the eound of r, being somewhat like rn." In this way Albe- runi has transliterated the n in the word banij, which he writes d barnij. Accordingly we should expect to find ot parnini, but the author seems to have written oy4 pânrini. P.135 .- Theword wes = fishyahita, has been deciphered by Prefessor Kielhorn, Gottingen. P. 136. Satarahana .- Other forms of the name are Salavahana, Salivahana (Hemacandra, i. 211); but Albe- runi clearly notes the pronunciation Samalvdhana, which is net known to me from other sources. P. 136,-Instead of maudakam read modakam = md udakam. P. 136 .- Abulaswad, &c., is, according to the literary tradition, the originater of their grammatical science. Cf. G. Flugel, Grammatische Schulen der Araber, p. 19 seg. P. 136. Chandas .- In translating the chapter on metrics, I have derived much help from Colebrooke, " Essays," ii. p. 57 (on Sanskrit and Prakrit poetry), and from Weber's edition of the Sutras of Pingala (Indische Studien, vol. viii). Alberuni, however, seems to have used other sonrces and to have fellowed another system, which has greatly in- creased the task of the translator. P. 137. Pingala .- What are the Sanskrit forms of the names u calitu, oms gaisitu, uif auliyându? The chapter of Brahmagupta's Brahmasiddhanta, of which the author here (p. 147-150) communicates a few extracts, is chap. xxi, On the calculation of the measures of poctry and on metrics, v. i. 155. P. 138 .- Alkhalil, also mentioned i 147, is in Arabic literature the father of the science of metrics. Cf. G. Flugel, Grammatische Schulen der Araber, p. 37. Sabab .-- Cf. Freytag, Arabische Verskunst, p. 64, 65.

303 ALBERUNTS INDIA.

P. 140. Madhya-I do not know this term in Sanskrit, and the signs &- admit of different transliterations. Both the terme madhyd and madhu are used in metrical ter- minology, but with different meanings. Cf. Colebrooke, " Essaye," ii, 14I (madhu), and ii, 136, 141 (madhyd).

P. 141 .- Haribhatta f-This name is not known to me as that of an author of a lexicographical work The MS. clearly writes hariuddu, which may represent various other forms of Sanskrit names.

P. 141 .- The single letters m, y, r, &c., denoting the eingle feet, are mentioned by Colebrooke, " Essays," ii. 63.

P. 142. Place the numeral 2, dc .- The rule, as explained in ll. 4, &c., differs from that one which is followed in the example (ll. 11-14), in so far as in the former place the snbtraction of I (" and from the product (4) he sub- tracts 1") has been omitted. Bnt even if we correct the text of the rule according to the exemplification, it cannot be correct, and we agree with Alberuni that something in the manuscript must have been wrong (also in the passage below, ll. 30-34). For it can be applied not to all eight feet, but only to two, viz., to H<(2X2=4-1-3X2=6-1=5) and to |<|(2X2*4-1=3X2=6), i.c. these two feet occupy respectively the fifth and sixth places in the arrangement on p. 141 (below).

P. 143. The Grecks, too, dc .- The comparison with Greek metrics is unintelligible, as something must have been dropped in the Arabic text.

P. 143. Consonant or syllable-I suppose the anthor means eyllable. The Arabic word Jy has the same incon- venience as Sanskrit akshara of meaning both syllable and sound (mostly consonant).

P. 143. Aryd .- This reading is a conjecture of mine, as the MS. has aral, which I cannot explain. The descrip- tion given by the anthor seems to be applicable to the

ANNOTATIONS. 303 Åry& metre, which conld be known to him from his study of Brahmagupta's Brahmasiddhanta. Cf. Colebrooke, " Essays," ii. 66.

P. 144 Skandha .- A kind of Arya metre, v. Colebrooke, "Essaya," ii. 137; or skandhaka, v. Weber, Indische Studien, viii. 295. Khaftf .- This Arabic metre, represented in European fashion, is the following :- -- --

P. 145. Vritta .- On the metre of this name v. Cole- brooke, "Essays," ii. 145. However the signs wy (b-r-4) admit of various other ways of reading. The MS. has britu. P. 147. Śloka .- On the rules relating to this metre v. Colebrooke, " Essaya," ii. 107.

P. 150. I have only seen a single leaf .- This translation is to be replaced by, " I have only studied a aingle leaf." P. 151. Galenus-The quotation is found in his repl

996 :- συνθέσεως φαρμάκων κατά γένη (ed. Kuhn), tom xiil p.

άλλ' ή γε διά τών χυλών ύπό Μενεκράτους εύρεθεϊσα διά τώνδε των τριμέτρων στοιχείων ύπο Δαμοκράτους yeyparTal.

P. 153. Siddhanta .- On the literature of the Sid- dhantas v. E. Burgess, Surya Siddhanta, p. 418-422. Śrishena is written with kh instead of ah, as bhashd= bhakha. Cf. Hornle, " Comparative Grammar of the Gau- dian Languages," § 19 and 20. Vardhamihira .- Vide note to p. 54. Pp. 153, 154. Brahmagupta-His work, the Brahma- siddhanta, has been vary largely used by Alberuni. It exists in manuscript, but has not yet been completely edited or translated. Alberuni translated it into Arabic when he wrote the Indica (A.D. 1030). We do not know whether he ever finished it. Brahmagupta was only thirty years of age when he

304 ALBERUNP'S INDIA.

wrote this work. He is accused of the sin against con- science of having propagsted futilities and lies in order to please the bigoted priests and the ignorant rabble of his nation, in order to svoid those dangers in which Socrates perished. Vide chsp. lix. on eclipses, and specially ii. 111. Besides, Alberuni accuses him of undue animosity against Âryabhata (i. 376). Brahmagupta holds a remarkable place in the history of Eastern civilisation. It was he who tanght the Arabs astronomy before they became acquainted with Ptolemy ; for the famous Sindhind of Arabian literature, frequently mentioned, but not yet bronght to light, is a translation of his Brahmasiddhanta; and the only other book on Indian astronomy, called Alarkand, which they knew, was a translation of his Khandakhadyaka. The latter work (here ii. 7) is also called Karanakhan- dakhadyaka (i. 156). It was explained in s special com- mentary by Balabhadra (ii. 187). A third composition of Brahmagupta's called Uttara- khandakhadyaka, is mentioned i. 156, and quoted ii. 87, 91. Cf. on Brahmagupta Colebrooke, " Esssys," ii. 409 seg. ; Dr. Bhau Daji, "Brief Notes on the Age and Anthenticity of the Works of Aryabhata, Varahamihira, Brahmagupta, &c.," in the "Journal of the Royal Asiatic Society," 1865, vol. i. 392 seq. Notes from Varahamihira's Pancasiddhantikd have been edited by G. Thibaut in the " Journal of the Asiatic Society of Bengal," 1884, vol, liii. p. 259. Sindhind is mentiooed ii. 191, as the only source of the information of Muslims on Indian astronomy and astrology. According to ii. 90, the Indian computation of the heliacal risings of the stars and the moon is identical with that given in Sindhind. It is called the great sindhind (Sid- dhânta) ii. 18. Alberuni has written a treatise on it. See preface to the Arabio edition, p. xx.

P. 154 Pulisa .- This name and Paulisa are written Pulisa and Pauliaa in Utpala's commentary to the Samhitd of Varahamihira; but as Alberuni writes them constantly with s u, not e, I am inclined to believe that he and his Pandits pronounced Pulisa and Paulisa. Alberuni has

ANNOTATIONS. 305

drawn from the Pulisasiddhanta almost as largely as from the Brahmasiddhanta, and was occupied with translating it (v. also i. 375). The relation between Pulisa and Paulisa is this :- Paulisa is the sage who communicates his wisdom in this Siddhanta. He was a native of Saintra, i.e. Alexandria. Pulisa is the redactor or editor of the book. The one as well as the other is called ug, Greek (not sn, Byzan- tine Greek). " Pulisa says in his Siddhanta that Paulisa the Greek had mentioned somewhere," &c., i. 266. A commentator of this Siddhanta is mentioned i. 339 med., where I now prefer to translate: " The commentator of the Siddhanta of Pnlisa," &c. Pulisa quotes Parasara (ii. 208), and is himself quoted by Âryabhata jun. (i. 316). Paulisa is quoted by Brahmagupta, i. 374 (v. note). Cf. on the Pulisasiddhanta H. Kern, The Brhat Sanhita, preface, p. 48.

P. 156 .- Aryabhata senior is clearly distinguished from Aryabhata junior, who is mostly called " that one from Kusumapura,"i.e. Pâtaliputra (Patna). Alberuni knows him only through the quotations in the works of Brahmagupta. He mentions two of his works, Dasagitikd and Aryashtasata, which have been edited hy Kern, Arya-bhattyam, 1874. Cf. Dr. Bhâu Daji, " Brief Notes on the Age and Authen- ticity of the Works of Aryabhata," &c., p. 392.

P. 156. Balabhadra .- Of his works are mentioned :- (1.) A tantra. (2.) A Samhitd.

hira (p. 158). (3.) A commentary of the Brihajjatakam of Varahami-

(4) A commentary to the Khandakhadyaka of Brahma- gupta. (5.) He is supposed to be the author of the book Khan- dakhadyakatippa. Alberuni always calls him the commentator, and fre- quently quotes him without indicating from what particular book he quotes. He gives on his authority the latitude of Kanoj and Taneshar, and passes harsh judgment on him i. 244, 275. Cf. also note to p. 27. VOL IL U

306 ALBERUNS INDIA.

P. 156. Bhanurajas (also on p. 157) .- The Arabic MS. writes Bahdaarjus, which I cannot identify. A slight alteration (of uy to uyt) would give Bhanuyasas, which name was suggested to me by G. Buhler. P. 156. Kura-babayd .- As kura means rice, , babaya, must mean mountain. Is it a vernacular form for parvata? P. 156. Khanda-khadyaka-tappd .- The MS. has tappa or tippd (tuppd), of which I do not know the Sanskrit form. V changed to of would be = fippant or com- mentary. Vijayanandin-Alberuni quotes from him a method for the computation of the longitude of a place (i. 313), a note on the dominants of year, month, and hord (i 343), on the circumpolar stars (ii 90), an ahargana rule (ii, 49, 50). An astronomer of this name is mentioned by Dr. Bhau Dajt as anterior to Srishena, the anthor of Romakasid- dhanta: v. "The Age and Authenticity of the Works of Aryabhata," &c. (" Journal of the Royal Asiatic Society," I864), p. 408. P. 156. Bhadatta (! Mihdatta) .- The MS. reads wige. Bhadatta is mentioned by Kern in the preface to his Brhat Safthitd, p. 29. Alberuni quotes from the work of Vitte- fvara a note on the motion of the Great Bear (i. 392), on the mean places of the stars (ii. 60), on the diameters of sun and moon (ii. 79), the latitude of Kashmir (i 317), the era used in the book (ii. 7). It must have been trans- lated into Arabic before Alberuni wrote the Indica, because he complains that that part of the book which he had was badly translated (ii, 55). P. 157. Utpala .- Besides these two Karanas, he has composed -- (1.) A commentary to the great Mdnasa composed by Manu. (2.) The Praśnacddamani (p. 158). (3.) A commentary to the Samhitd of Varahamihira (p. 298). (4) The book Sredhava (1), whence Alberuni has taken metrological and chronological notes (p. 334, 336, 361). Cf. on Utpala Kern's preface to his Brhat Sanhitd, p. 61.

ANNOTATIONS. 307

The book-title rdhunrakarana, i.e. breaking of the Ka- ranas, seems to be corrupt. One expects the word karana in the first place, and a word for breaking in the aecond.

P. 157. On Manu as an anthority in astronomy and astrology, v. Kern, preface to Brhat Sanhita, p. 42. Cf. note to p. 131. P. 157. Puncala (?) .- The author qnotes from him a atatement relating to the precession of the equinoxes; he apeaks highly of him, and says that a theory of his was adopted by Utpala (i. pp. 366, 367). I do not know of such an Indian name. The nearest approach to it is Munjdla, that of an astronomer quoted by Colebrooke, " Essays," ii. 330, 332.

P. 157. Bhadila (?) .-- The MS. has bahattal, and I sup- pose that the correct reading is Bhattila. The name is perhaps a derivation (diminutive ?) from bhatta, as kuma- rila from kumara, pushandhila from shandhz. Alberuni quotes him, ii. 208, in the chapter on the yogas. On Parasara and Garga ef. Kern, Brhat Sanhita, preface, pp. 31, 33; on Satya, Jivasarman, p. 51; on Manittha, p. 52. Mau is probably identical with Maya: v. Weber, Vorlesungen, p. 270. P. 158. Of Vardhamihira, de .- Thia author has com- posed not only the Shatpancasikd and Horapancahotriya (?), but also the Yogayatra, Tikantyatra (1), and Vivdhapatala: v. Kern, Brhat Sanhita, preface, pp. 25, 26; his translation of the Yogaydtra in Weber's Indische Studien, x. 161. The name of the author of the book on architecture is missing in the Arabic text. If it was not likewise a work of Varahamihira's, it may have been composed by Nagnajit or Visvakarman: v. Kern, L. c. p. 51.

P. 158. Sradhara .- I do not know the corresponding Sanakrit form. It seems to be some relative of sruti. If śrutayas had currency in the meaning of traditions, I ahould identify it with srudhava. Ia it = śrotarya? The word is the title of two different books, one by Utpala from Kashmir (v. note to p. 157), and the one here mentioned, on omina and portenta, lucky and unlucky

308 ALBERUNPS INDIA.

days, &c. It probably contained the names of the twenty- four hord (i. 344); it mentioned the names of the third parts of the day (ii. 120), the names of the vishti (ii. 201), the unlncky days of the year (ii. 192), the name of Vikram- âditya (ii. 6, vide note to the place). The reading of the word Jica as Bangdla is probably not correct. Is it = punyakdla ?

P, 158. Gudhdmana (?), in the Arabic jurdman .- As the word is translated by unknown, one thinks of a derivation of the word guh=to conceal (v. gudha). The Arabic characters may also be read cuddmani. If praśna jura- man (?) really meant what Alberuni says, one would expect gudhapraśna.

P. 158. Sangahila, Ptrurdna-I do not know the San- skrit eqnivalents of these two names. The former might be a word like śrinkhala or śringald (Syncellus?). Pritha- daka is the anthor of a commentary on the Brahmasid- dhanta: v. Colebrooke, " Essays," iL 411.

P. 159. Caraka .- The encient Arabic translation of his medical work is sometimes quoted hy Alberuni, and to jndge from these quotations the translation was not free from blunders nor the manuscript-tradition free from the effects of carelessness : o. & quotation on weights, i. 162, 163; one on the origin of medicine, i, 382. Cf. Weber, Vorlesungen, pp. 284, 289.

P. 159. Pancatantra-Cf. on this book and on Ibn Almukaffa's share in its translation, Benfey's introduc- tion to his translation of the Pancatantra (Leipzig, 1859). On the translations of the book and on the influence which King Mahmud of Ghazna has had on its fate, ef. Colebrooke, " Essays," ii 148. The work of Ibn Almu- kaffa is that one edited by S. de Sacy, 1816.

P. 160. Chapter XV .- For the translation of this chapter on metrology, I have derived much help from Colebrooke, "On Indian Weights and Measures" ("Essays," i 528 seq.), and Marsden's Numismata Orientalia, new edition, Part I., " Ancieut Indian Weights," by E. Thomas, London,

ANNOTATIONS. 309 1874; A. Weber, Ueber cin Fragment der Bhagavati, II. Theil, p. 265 note. The weight of one dirham = one-seventh mithkal, dates from the time of the Khalif Omar. The weight of one dirham = seven ddnak, is peculiar to India in the anthor's time, for in general one dirham = six danak. Cf. Sanvaire, Materiauc pour servir a l'Histoire de la Numismatique et de la Metrologie Musul- manes, Paris, 1882, pp. 43, 81, 98; on the mithkal, p. 35; on the fuls, p. 108. On the ancient denars of Sindh ef. Elliot, " History of India," i. Ir (Ahû Zaid), 24 (Mas'ûdi), 35 (Ibn Haukal).

P. 162. Varahamihira .- This passage is Brihat Samhita, chap. Iviii. v. I. The following quotation on yara, andt, masha, and suvarna, I do not find in his Samhita.

P. 162. Caraka .- The Arabic translation of this book is not extant. The Indian words which occur in the extracts from this book are not so accurately written as those in Alberuni's own work, and offer more difficulties in the way of identification : v. note to p. 159.

P. 162. Jivafarman .- The words " As I have been told (by him)," may better be translated " As I have heard it from him." Alberuni does not quote from a book of his, but only says "he has told, mentioned," " I have heard from him." Accordingly, he seems to have been a contemporary and personal acqnaiotance of Alberuni's, in the same way as Sripala, Alberuni relates on his authority details regarding a festival in Kashmir and Svat, ii, 181, 182. Besides, a Jivasarman is mentioned as the author of a Jatakam, i. 157, who seems to have been a different person altogether, and lived before the time of Varahamihira : v. Kern'e Preface to Brhat Sanhita, p. 29.

P. 164. Varahamihira .- This qnotation seeme to corre- spond to Brihat Samhita, chap. xxiii. v. 2. At all events, it is the passage to which Sripala refers. Sripala .- Alberuni quotes him a second time, i. 240, where he speaks of a star, sula, as observed in Multân, which people considered as unlucky, and ii. 209, he copies

310 ALBERUNI'S INDIA.

from him the names of the twenty-seven yogas. Perhaps Sripala was a scholar living at Multan in the time of the anthor. Alberuni does not mention a book of his.

P. 165. Sisupdla .- The story of Krishna's killing Sifu- pala (Sisupalabadha) is told in the Mahabhdrata, Sabha- l'arvan, v. 1336 seg.

P. 165. Alfazdrf is one of the fathers of Arabian litera- ture, the first propagator of Indian astronomy among the Arabs. His works are, as far as I am aware, not extant. Probably this Muhammad Ibn Ibrahim Alfazart, was the son of Ibrahim Ibn Habib Alfazari, the first constructor of astrolabes among the Arabs, who as a surveyor partook in the foundation of Bagdad, Cf. Fihrid, p. Mr. Gildemeister, in his Seriptorum Arabum de rebus Indicis loci, p. IO1, gives the translation of an article of Alkifti on our Fazart. According to the quotations of Alberuni (v. index s. v. Alfazari), this scholar used the word pala in the meaning of day-minute; he reckoned the circumference of the earth in olsl, i.e. yojanas; he (together with Ya'kub Ibn Tarik) mentions a town, Tara, in a sea in Yamakoti; he gives a method for the computation of the longitude of a place from two latitndes; his book contained the cycles of the planets as derived from Hindu scholars, the members of an embassy from some part of Sindh, who called on the Khalif Almansur, A.H. 154 (= A.D. 771). Alberuni charges him with having misnnderstood the meaning of the word Arya- bhata, which he is said to have used as meaning Toou of the measures of the great Siddhanta, i.e. the Brahmasid- dhanta of Brahmagupta. Lastly, Alfazari (together with Yakub) has used the word &Ls (padamdsa?) in the sense of adhimasa (lesp-month). On the whole, Alberuni finds that the tradition of Indian astronomy by Alfazari is not very trustworthy, and that in it the names or termini technici are often corrupt and hadly written. As Alfazari and Ya'kub IbL farik are sometimes men- tioned in the same context, there must have been a close relation between these two anthors, the nature of which I have no means for examining. Have both learned from the same Hindu scholar, and have they independently of each other committed their information to writing? Or

ANNOTATIONS.

has the one made a new edition or a commentary of the work of the other ? Vide note to p. 169 (Ya'kub).

P. 165. Sibi .- The word occurs thrice, and is written - (siyt !); only in one place it seemed to be .- , but on repeated comparison of the MS. I find that originally here, too, was written . I do not know a measure of such a name. Perhaps it is the bisi, of which 16=I pants (p. 166, 1. 2 in Somanath). Cf. Colebrooke, " Essays," i. 536; sixteen bisis = one pantt.

P. 166. Khwdrizmian .- The comparison of the measures of this country, the modern Khiva, will remind the reader that it was the native country of the author.

P. 166. Varahamihira .- I have not succeeded in find- ing this quotation in his Samhita.

P. 167. Vardhamihira .- The passage here quoted is Samhitd, chap. xviii. v. 26-28.

P. 167. 'Ajvdn-Alberuni only mentions the plural form, not the singular, which would be jun or jaun, jon. I take the word to be the Arabized form of yojana. Tho change from yojana to jon was perhaps facilitated by a Prakritic pronunciation on the part of the Hindu teachers of Alfazari, according to which a j between two vowels may be dropped. Cf. gao=gaja, raadam, rajata (Vararuci, ii. 2).

P. 168 .- Archimedes fixed w as a measnre between 34 and 31º. Cf. J. Gow, " Short History of Greek Mathematics," Cambridge, 1884, p. 235.

P. 169 .- Yakub Ion Tarik seems to have been the most prominent predecessor of Alberuni in the field of astro- nomy, chronology, and mathematical geography on an Indian basis. He is frequently quoted in the Indica, much more than Alfazari. Here he gives the measnres of the circumference and the diameter of the zodiacal sphere in yojanas, in which Alberuni recognises the system of Pulisa. He speaks of a city, Tara, within a sea in Yamakoti (i. 303). He gives the measures of the radius, diameter, and circumference of

312 'ALBERUNI'S INDIA.

the earth in yojanas (i. 312), a statement on the latitnde of Ujain, and a quotation from the book Arkand on the aame aubject (i. 316). He mentions the four manas or measures of time, sauramana, candramana, &c. (i. 353)- Hia work contained tablea of the revolutions of the planets, borrowed from a Hindu who had come in an embassy from Sindh to the court of the Khalif Almansur, A.H. 154 (=A.D. 771), but Alberuni finda in these tables con- siderable deviations from those of the Hindus (i. 15). He is accused of having misunderstood the word Âryabhata, so as to take it not for the name of an author, but for a technical term meaning roog of the measures employed in the great Siddhanta (that of Brahmagupta), on ii. 18, 19. He called the leap-month Al (padamasa!) instead of adhimasa (ii. 23). He gives an incorrect method for the computation of the solar daya in the ahargana and for the reduction of yeara into daya (by the aide of a correct one) on ii. 26, 34, 38. He gives further detaila of the ahargana compntation (ii 44, 45), and a table indicating the dis- tances of the planets from the earth, borrowed from a Hindn, A.H. 161 (=A.D. 777, 778), on ii. 67, 68. Accordingly the work of Ya'kub seems to have been a complete syatem of astronomy, chronology, and mathema- tical geography. It is called Compositio Sphærarum and also al, ic. Canon. Alberuni sometimes criticises Yakub, and maintains that he had committed errors, that he mia-spelled the Indian words, and that he aimply borrowed the tables from his Hindn authority without examining them by calculation. On his relation to Alfazari, t. note to p. 165. When Alberuni wrote his Chronology, he did not possess the work of Ya'kub, fort here he gives a note on the four manas and on the word alos (padamasa ?) on the autho- rity of Yakub, but taken from the work of another anthor. Vide my tranalation, p. 15. As Ya'kub studied in the years A.H. 154 and 161 (A.D. 771, 778), he must have lived in the second half of the eighth Cbristian century (probably in Babylonia). This is nearly all we know of him. Cf. Reinand, Memoire sur l'Inde, p. 313; Steinschneider, Zeitschrift der Deutschen Morgenlandischen Gesellschaft, 24, 332, 354. The Fihrist, p. wA, has a note on him in which there

ANNOTATIONS. 313

is some confusion. The work Compositio Sphærarum is erroneously mentioned among the works of 'Utarid Ibn Muhammad, whilst it is apparently identical with the work here called Canon. It consisted of two parts, one on the sphere and one on the periods (the yugas?). Accord- ing to Fihrist, he had written two more books, one on the division of the sine in kardajat, and another on what is derived from the arc of the meridian. Regarding the embassy from Sindh, from which the Arabs are said to have got the first information on Indian astronomy, in fact, the two works of Brahmagupta, the Brahmasiddhanta (Sindhind) and the Khandakhadyaka (called Arkand), I cannot find any historical account in the Arabic annals. We do not learn anything from Ibn Wadih or Tabari of the presence of a Sindhi embassy in Babylonia in the year 154 (A.D. 771), as Alberuni has it, nor in the year 156 (A.D. 773), as Alhusain Ibn Muhammad Ibn Aladami maintains (Gildemeister, Scriptorum Arabum de rebus Indicis loci, p. 1O1), nor of the presence of Hindu scholars in Babylonia in the year 161 (A.D. 777). This only is related by Ibn Wadih, that when Abnlabbas Saffah, the first Abbaside Khalif, was dying in Anbar, there arrived at his court an embassy from Sindh, A.H. 136 (A.D. 753). At all events, at the time of the Khalif Almansûr, Sindh obeyed this prince, and Islam had spread not only in Sindh, but far beyond it into the adjacent countries, both by war and by commerce. There must have been many occasions for petty Hindu princes in Sindh to send special missions to the political centre of the Muslim realm. When Yaktb wrote, the Arkand (Khandakhadyaka) had already been translated into Arabic. By whom? By Alfazâri ? In the first fifty years of Abbaside rule there were two periods in which the Arabs learned from India, first under Mansur (A.D. 753-774), chiefly astronomy, and secondly under Harun (786-808), by the special influence of the ministerial family Barmak, who till 803 ruled the Mnslim world, specially medicine and astrology.

P. 170. Soerates .- I do not know the Greek form of this dictum. It must be observed that according to the common

314 ALBERUNI'S INDIA.

tradition hides of animals were first prepared for vellum at Pergamum long after Socrates.

P. 171. On the fabrication of papyrus, ef. Wilkinson, "Manners and Customs of the Ancient Egyptians," ii. p. 180. P. 172. As for the Greck alphabet, &c .- The source of this tradition on the origin of the Greek alphabet seems to be certain scholia to the Ars Grammatica of Dionyaius Thrax: v. Immanuel Bekker, Ancedota Graca, Berlin, 1816, vol. ii. p. 780 seg. The synchronistic notes point more to Joannes Malalas; perhaps these things were originally mentioned in the lacuna O 129. Asidhas seems to be a mistake for Palamedes, Agenon for Agenor.

P. 173. Bahmanwd .- Read Bamhannd. Other forms of the name are Bamtrn and Bainvdh: v. Elliot, "History of India," i. 34, 189, 369, and the papers of Haig in the " Journal of the Royal Asiatic Society, 1884, p. 281, and of Bellasis in the "Journal" of the Bombay branch, vol. v., 1857, p. 413, 467. For Kannara, v. note to pp. 17-19. Andhradesa identi- fied by Cunningham with Telingana, o. his " Ancient Geo- graphy of India," p. 527. Bhaikshukt .- Alberuni writes Baikshuka, probably that of the bhikshu or beggar-monks, i.e. the sramana or Bud- dhistic monks. Is the Audunpur mentioned by Alberuni, identical with the famous Bnddhistic monsatery Udanda- puri in Magadha (1). Cf. H. Kern, Der Buddhismus und seine Geschichte in Indien, German by H. Jacobi, Leipzig, 1882, vol. ii. p. 545. What Malvashau is I do not know (Malla-vishaya ?). P. 175. To the orders of numbers, cf. Weber, Vedische Angaben uber Zeittheilung und hohe Zahlen, in Zeitschrift der Deutschen Morg. Gesellschaft, xv. 132.

Pp. 178, 179. This table has already been published by F. Wopcke, Memoire sur la Propagation des Chiffres Indiens, p. 103 seq ; A. C. Burnell, " Elements of South Indian Palæography," ii. ed., p. 77. Compare also E. Jaquet, Mode d'Expression Symbolique des Nombres Em-

ANNOTATIONS. 315

ployé par les Indiens, les Tibetains et les Javanais, (Extrait du Journal Asiatique); Brown, "Sanskrit Prosody and Numerical Symbols," London, 1869, p. 49 seg. P. 181. Pushandhila .- The eunuch is called shandha. This seems to be a diminutive form compounded with the word pums (G. Bühler).

P. 182. They magnify the nouns of their language, dc .- This somewhat enigmatic sentence seems to have the following meaning :- An Arabie word, e.g. karsh (a sea- animal), is magnified, i.e. receives a larger form, by being changed into the diminutivo form, i.e. kuraish (a small sea-animal, as a proper noun, the name of the tribe to which Muhammad belonged). The diminutive form serves the purpose of magnifying the form of the word : ef. Kash- shaf to Koran, 106, 2, i yotadl, (not oAsl). If the Hindus magnify their nouns by giving them the feminine gender, this must be referred to some of the pieonastic suffixes, e.g. d, f, which are added to Indian noune without altering their meaning. In appearance they ere the ter- minations of the feminine gender, in reality euphonic changes of the more ancient suffixes aka and ika, eg. patd, board, by the side of pat. Cf. Hornle, " Comparative Grammar of the Gaudian Languages," § 194 seg. P. 183 .- An explanation of the Indian chess has been published by A, Van der Linde, Geschichte und Litteratur des Shachspiels. P. 189. Nagarjuna .- Cf. on him A. Weber, Vorlesungen, pp. 306, 307; H. Kern, Der Buddhismus und seine Geschichte in Indien, ii. 501; Beal, "Indian Antiquary," 1886, 353. P. 189. Vyddi .- A lexicographer of this name is men- tioned in a certain connection with Vikramaditya by Colebroke, " Essays," ii. 19. P. 190. Raktamala = rakta = red, and amala = emblica officinalis. I do not see how the word could be understood to mean oil and human blood. P. 191. Bhojadera .- Cf. on this king of Malava, Lassen, Indische Alterthumskunde, iii. p. 845 seq.

316 ALBERUNIS INDIA.

P. 192. Vallabht-On the end of this city, cf. Lassen, Indische Alterthumskunde, iii. 532 seq., and also Nicholson and Forbea ou the ruina of the place, in "Journal of the Royal Asiatic Society," vol xiii. (1852), p. 146, and vol xvii, (1860), p. 267.

P. 196. For it is not navigable .- This passage agrees almost literally with Plato's Timaus, 25D :- διό καί ων άπορον καί αδιερεύνητον γέγονε τό έκεϊ πέλαγος, πηλού κάρτα βραχέος έμποδών όντος ον ή νήσος εζομένη παρέσχετο.

P. 197. The various tribes of the Zanj-The traditions of the Arabs regarding Eastern Africa have been collected by Marcel Devic in his Le Pays des Zendjs, Paris, 1883.

P. 197. The configuration of the northern coast of the Indian Ocean seems to have been a favonrite subject of Alberuni, for he mentions it again on p. 270.

P. 199. Mahira, so written by Alberuni, is written 4, Mahura, by his elder contemporary Al-ntbi, more in keeping with the Sanskrit vowels (Mathurd). Alberuni reckons the distances in farsakh, regarding the measure of which he unfortnnately does not give accurate information. According to i. 167, I yojana = 32,000 yards =8 miles; I mile=4000 yards; and according to i. 200, I farsakh =4 miles= I kuroh ; I farsakh= 16,000 yards. Cf. also Aloys Sprenger, Die Post- und Reiserouten des Orients, Vorrede, p. xxvi., who proves that one Arabian mile=præter propter 2000 metres=2186 yards, whilst the English geographical mile = 2025 yards. If we, therefore, want to compare Alberuni'a distances with English miles, we must reckon- 1 English mile=1 1r Arabian mile. 1 Arabian mile-H11t English mile. I farakk =4 Arabian miles=3175 English miles.

P. 200. Alberuni givea sixteen itineraries which seem to have been communicated to him by the military and civil officers of King Mahmud (on some of these roads he

ANNOTATIONS. 317

had marched with large armies, e.g. to Kanoj and to Somanatha), from merchants and aailors, from Hindn and Muslim travellers. The starting-points of these itineraries are Kanoj, Mahura (now Muttra), Anhilvara (now Pattan), Dhar in Malava, and two less known places, Bari, the tem- porary capital of the realm of Kanoj, after the old capital had been taken by the Muslima, and a place called Bazana. These itineraries are-I. From Kanoj to Allahabad, and thence towards the eastern coast of India as far as Kanci (Conjeveram), and farther south. 2. From Kanoj (or Barî) to Benares, and thence to the mouth of the Ganges. 3. From Kanoj eastward as far as Kamroop, and north ward to Nepal and the Tibetan frontier. 4. From Kanoj southward as far as Banavasi on the aouthern coast. 5. From Kanoj to Bazaoa or Narayan, the then capital of Guzarat, 6. From Muttra to Dhar, the capital of Malava. 7. From Bazana to Dhar and Ujain. 8. From Dhar in Malava towards the Godavari. 9. From Dhar to Tana, on the coast of the Indian Oceao. 10. From Bazana to Somanatha, on the south coast of Kathiavar. II. From Anhilvara to Tana, on the west coast, north of Bombay. 12. From Bazana vid Bhati to Loharani, at the mouth of the Sindh river. 13. From Kanoj to Kashmir. 14. From Kanoj to Panipat, Attok, Kabul, Ghazna 15. From Babrahan to Addishtan, the capital of Kashmir. 16. From Tiz, in Makran, along the coast as far as Setubandha, opposite Ceylon. Cf. the following latitudes and longitudes, taken from the Canon Masudicus :-

Tree of Prayaga, 25° o' lat., 106° zo' long. ; Kuraha, 26° I' lat., 1o6" 40' long. ; Tiauri, 23°.o' lut., 106° 30' long. ; Kujtraha, 24° 4' lat., 106° 50' long. ; Bazana (1) or NarAyno, 24° 35' lat., 106° to' long .; the country Kannakara, 22° 20' lnt., 107° o' long. ; Sharvar, 24° 15' lnt., 107° 50' long. ; Patalipntra, 22" 30' lat., 108° 20' long. ; Mungiri, 22° o' lat., 109° 10' long. ; Dogum, 22° 40' lat., 110° 50' fong. ; Bari, 26° 30' lut., 105" so' long. ; Dudahi, 25° 40' lat., 102° to' long. ; Dahmala, 31" 10' lat., 100° 55' long. ; Shirsharaha, 38' 5o' lat., 102" 1o' long. ; Bhil- lan:Ala, 23° 50' lat., 87 45' long. ; Bamhanva, 26° 40' lat., 85° o' long. ; Loharani, 24° 40' lat., 84" 25' long. ; Daibal, 24° 10' lat, 82" 30' long. ; Bhatiya, 28° 40 lat., 96° o' long. ; Ujain, 24" o' lat., 100° 5o' long .; 'Tiz, 26° r5' lat., 3° o' long. ; Kandi, 33° 40' lat., 95° 5o' long. ; Dun- pur, 33° 45' lat., 96° 25' long. ; Tanjore (?), I5' o' Int., 115" o' long. ; Rameshar, 13° o' lat., 118° o' long. ; Jabravar, 39° 50' lnt., 96 15' long. ; dyds 31" r' lnt., 95° 55' loug. Longitnde is reckoned from the coast of the Atlantic; that of Bagdad is 70°.

318 ALBERUNI'S INDIA.

P. 200. Barhamshil = Brdkmanasaila = Brahmin's rock (?). Tree of Prayaga= Allahabad, at the confluence of Ganges snd Jomna. In line 20 after 12 farsakh (in the Arabio only 12 with- out farsakh) there is apparently & lacuna. Uwaryahdr .- One expects an indication of Orissa (Uriyadesa). The word might also be read Uriyahar. -Is Uriyadhard meant? Urdabishau perhaps = urdhva- vishaya. Jaur's possessions, i.c. the Cola empire; o. also here, i. 209, and Lassen, Indische Alterthumskunde, ii. 435, iv. 230 sg.

P. 200. Bart-Regarding the situation of this place the following statements must be taken into account :- It was situated ten farsakh or three to four daye' march distant from Kanoj towards the east, east of the Ganges, in the neighbourhood of the confluence of the rivers , and s and Sarayu. It was twenty-five farsakh distant from Oudh. The name Bart occurs also in Elliot-Beames, " Memoirs," ii. 83, as that of a snbdivision of the district Agra.

P. 201. Kamre is apparently Kamarapa and Tilvat = Tirhoot, The latter is by mistake also written Tanvat. Are we to read Tirut? The word is perhaps composed of Tart, the name of the nation who lived there, and a word like bhukti. The empire of Shilahat .- Is this to be identified with Sylhet, the province of Assam ? Bhoteshar seems to be bhautta-fsvara, lord of the bhauttas, or Tibetans.

P. 202. Kajurdha is = kharjura-bhaga. Tiaurt .- According to a well-known rule of Prakrit (Vararuci, ii. 2), the name Tiirovpa (Ptolemy, vii. i. 63) would become something like Tiaurt As there is a lacuna in the Arabic manuscript, the situation of this place cannot be accurately defined. Kannakara .- This is probably identical with Kamkar, the realm of the Balhara, according to Mas'udi: v. Elliot, " History of India," i, 25.

ANNOTATIONS. 319

P. 202. Bardna .- The reading is conjectural. For an identification v. Archæological Survey of India, ii. 242. For Sahanya (Suhaniyd) v. ibid. ii. 399. On Guzarât, the empire of the Garjjara kings, not identical with modern Guzerat, ef. Cunningham, " Ancient Geography of India," p. 312 seq .; Elliot, l. c. p. 358. Jadura-This reading is uncertain. Perhaps all the signs of the Arabic text (syyl) are the name of a place.

P. 202. Bamahur is perhaps identical with Ptolemy's Bappoyoupa (Pf. vii. i. § 63), as in some cases an h re- presenta an elder g; eg. sel, Candarâha = Candrabhaga, Pys devahar=devagriha, kulahara (Prakrit) = kulagriha.

P. 203. Namdvur, Altspur .- Are these names to be identified with Nimar and Ellichpur in Central Indie ? Cf. G. Smith, "Geography of British India," pp. 339, 347.

P. 203. Sarabha .- This digression of the author's is repeated by Muhammad 'Aufi in his story-book : v. Elliot, " History of India," ii. 202.

P. 205. Anhilvara = Analavâța = modern Pattan in Northern Baroda: v. G. Smith, l. I. p. 297; Elliot, " History of India," i. 363. Lardesh = Aapuh of Ptolemy, vii. i. 4 Bihroj = Broach = Bapúyata, G. Smith, p. 263. Rihanjur is probably identical with 'Aypwayapa (Pto- lemy, vii. i. § 63). Two consonants frequently undergo a metathesis, if one of them is a liquid. Agrinagara has become Arginagara, and the g is here represented by an h, as in Candaraha = Candrabhagd. Loharant seems to be identical with AwviBape of Ptolemy, vii i. § 2. A metathesis of the middle conso- nants has taken place, and b has become h. It is also called Lohaniyye (i. 316).

P. 205 .- Jalandhar is the KurwSpwy of Ptolemy, vii. i. § 42, G. Smith, p. 207. Ballavar- Vallapura, v. Cunningham, l. c. pp. 135, 133. Is it identical with modern Phillaur? G, Smith, p. 208.

320 ALBERUNPS INDIA.

P. 206 .- Kavital = Rapisthala - KauBiafoxos (Megas- thenes), now apoorthala, G. Smith, p. 208. Vide also Kaithal in Elliot's " History of India," ii. 337, 353- Mandahakur. Cf. Elliot, I. c. i. 530.

P. 206. Kusnart-I am inclined to identify this river with the Kunhar (G. Smith, p. 231). Is the Mahvi= Kishen-Ganga ?

P. 207 .- Ushkard is explained by Cunningham, l. e. p. 99, as Hushkapura, Huvishkapura and Baramtla as Vardha- mûla.

P. 208 .- Takeshar is perhaps to be explained as Tak- katfvara, like Bhoteshar = Bhantta-isvara, Cf. on Takka, Cunningham, L. c. p. 749. RAjavari seems to be identical with Rajaori (G. Smitb, p. 228).

P. 208. The coast of India begins with Tiz .- Cf. with this ronte along the coast that one given by Ibn Khurdadbih in Elliot, "History of India," i. 15, 16; A. Sprenger, Die Post- und Reiserouten des Orients, pp. 80-82. Munha = Skr. mukha, Prakrit muham, Hindt munh: v. Hornle, " Comparative Grammar," $ 116. Daibal .- On the identification with Karaci v. Elliot, "History of India," i. 375. Daibal-Sindh is the Diulcindi of Duarte Borbosa, translated by Stanley, p. 49 (Hakluyt Society).

Pp. 208, 209 .- Barof = Baroda, Kanbdyat = Kambay, Bihroj= Broach. Subara is identical with Skr. Surpd- raka, Ptolemy's Zourapa, and the Sufala of the Arabs. Tana = Skr. sthana, and Sandan is perhaps = samdhana. To Subara, ef. Bhagvanlal Indraji, " Antiquarian Remains of Sopara," &c., "Journal" of the Bombay branch, 1881, 1882, vol. xv. p. 273.

P. 209 .- Panjaydvar seems to be a mistake for some older form of the name Tanjore. Ramaher = Ramesvara f-On Rama and the monkeys of the Kishkindha mountains ef. the fourth book of the Ramayana.

ANNOTATIONS. 321 P. 210 .- The theory of the rising and disappearing of the Diva islands seems to have been a favourite one of the anthor'a, for he explains it in three different places; o. p. 233, and ii, 106.

P. 211 .- Shauhat is explained by Johnson as a tree whence bows are made, and mulamma' means having diffe- rent colours. What particular sort of wood this means I do not know.

P. 211 .- Indravedi must be changed into Antarvedi, "the old name of the Lower Doab, extending from about Etawah to Allahabad." Elliot-Beames, " Memoirs," iL IO; Elliot, " History of India," ii. 124. Is Bhatal identical with Ptolemy's Haraxnun ?

P. 213. We have already mentioned, viz on p. 17.

P. 214. dpas katpwal, i.c. the ancient division of day and night, each in twelve equal parte, of whatsoever length day and night happened to be. These hours were different in the different seasons of the year. On the contrary, the ipar lonuepwai, probably of scientific origin, are the twenty- fourth part of a nychthemeron, alwaye equal throughout the course of the whole year. Of. Ideler, Handbuch der Chronologie, i. 86.

P. 214 Hord .- The Persian nimbahra means half part, and in astrology one-half or fifteen degrees of a sign of the zodiac; v. il. 222.

P. 214, 1. 30 .- The distance between the sun and the degree of the ascendens divided by fifteen gives in hours the time which has passed since sunrise; the dominus of the day being at once the dominus of the first hour, the rule here given is evidently correct (Schram).

P. 215 .- For names of planets o. E. Burgess, Surya Sidd- hanta, pp. 422, 423, and A. Weber, Indische Studien, ii 261. Instead of yT read yst, dvaneya. The word bibatd is probably some form of vivasrant. The reader will notice the Greek names heli ios, dra VOL IL X

322 ALBERUNTS INDIA.

"Άρης, hemna "Έρμης, jίνα Ζεύς, asphujit Αφροδίτη, λομα Κρόνος.

Pp. 216, 217, 218. Vishnudharma-Vide note to p. 54-

P. 217. Table .- I shall here give the names of the months as the author probably pronounced them, bnt cannot be held responsible for the details of the vowel- pronunciation: cetr, beshak, jert, ashar, shraban, bhadro, ashuj, kartik, manghir, posh, mag, pagun. Perhaps most of these names terminated in short u, as manghiru. Cf. the Hindustani names in Dowson's "Grammar of the Urdt," 1887, p. 259. The vernacular names of the suns are perhaps to be pronounced : rabi, bishnu, dhâta, bidhata, arjamu, bhagu, sabita, půsha, tvashta, arku, dibakaru, anshu. The difference between vernacular and classical speech is repeatedly referred to, Vide i. 18 (v. note), 218.

P. 218. With the tradition of the Vishnudharma .- After these words must be added the following, which I have overlooked in translating: " And further he (i.e. Vâsu- deva) has spoken in the Gid, ' I am like the vasanta, i.e. the equinox, among the six parts of the year.' This too proves that the tradition as given in the first table is correct." Cf. Bhagavad-Gita, x. 35-

P. 218 .- Compare the table of the nakshatras with E. Burgess, Surya Siddhânta, p. 468.

P. 219-Varahamihira .- Vide note to p. 54- -.

P. 220 .- The Greek names kriya kpios, tâmbiru raûpos, jituma 8iduuor, partina napfevos, &c., are declared to be not generally known. Cf. A. Weber, Indische Studien, ii. 259. Instead of jitu read cetthu.

P. 222. Galenus .- I have not been able to verify this qnotation about Asclepius in the Greek works of Galenus.

P. 223. From the belief of the nations who lived in ancient times in and round Babel, &c .- That information

ANNOTATIONS. 323

to which the anthor here refers was probably derived from the books of the Manichæans.

P. 223. Plato .- This quotation is not identical with Timaus 36 B-D, but apparently derived therefrom. It runs :- ταύτην ούν την ξύστασιν πάσαν διπλήν κατά μήκος σχίσας μέσην πρός μέσην, κ.τ.λ. την δ’ έντός σχίσας έξαχη έπτά κύκλους άνίσους, κ.τ.λ. Cf. note to p. 35.

Pp. 223, 224 .-- On Brahmagupta and Pulisa, v. notes to pp. 153, 154.

P. 225. Vasishtha, Aryabhata .- The author does not take the theories of these men from their own works; he only knew them by the quotations in the works of Brah- magupta. He himself states so expressly with regard to Aryabhata. Cf. note to p. 156, and the author, i. 370.

P. 225, 227. Balabhadra .- Vide note to p. 156.

8, 24 P. 226. Aristotle. Cf. his Phys. vii. I, and Metaph. xii.

P. 226. Ptolemy .- Cf. the edition of Halma, Paris, 1813, tome i. p. 2: τό μεν της των λων πρώτης κινήσεως πρώτον αΐτιον, εί τις κατά τό άπλουν έκλαμβάνοι, θεόν αόρατον καί ακίνητον αν ήγήσαιτο, καί τό τούτου ζητητέον είδος θεολογικόν, άνω που περι τά μεγεωρότατα τού κόσμου της τοιαύτης ενεργείας νοηθείσης &ν μένον, καλ καθάπαξ κεχωρισμένης τών αίσθητών ουσίων.

P. 226. Johannes Grammaticus .- Vide note to p. 36. I have not been able to find this quotation in the Greek text.

Pp. 228, 229 .- The author repeatedly complains of the great verbosity of the Sanskrit caused by the necessities of the authors, who will only write in metre, and require

324 ALBERUNIS INDIA.

a great number of synonyms, in order that one word may fit into the metre if others will not. Cf. i. 211, 217, 299-

P. 229. For those men who, dec .- This is the only passage in which Alberuni clearly apeaks of his Pandits. Appa- rently he tried hard to learn Sanskrit, bnt could not suc- ceed on account of the difficulties of which he himself complains, and he studied Indian literature in the same manner as the first English scholars in Bengal, by the help of native Pandits.

P. 230. Table .- Cf. Vishnu-Purdna, ii. 209, where the fifth and seventh earths are called mahatala and patdla. Also the Vayu-Purana (ed. Rajendralala Mitra, Calcutta, 1880) offers somewhat different names, viz, atalam, suta- lam, vitalam, gabhastalam, mahatalam, śritalam, patalam, and krishņa-thaumam, pându, raktam, pita, sarkara, sila- mayam, saurarna (vol, i. p. 391, V. 11-14).

P. 231. The spiritual beings, &e .- This list of names is literally taken from Vdyu-Purdna, vol. i. p. 391, v. 15- 394, v. 43 (Adhyđya, so).

P. 231. Johannes Grammaticus .- I have not been able to find this qotation in the Greek text, nor the verse of Homer. Vide note to p. 36.

P. 231. Plato .- Cf. Timœus, 41A :- Θεσί θεών ων έγω δημιουργός πατήρ τε έργων, & δι' έμου γενόμενα άλυτα έμου γ' εθέλοντος· τό μεν ούν δή δεθέν πάν λυτόν, τό γε μήν καλώς άρμοσθέν καί έχον λύειν έθέλειν κακού.

P. 232. Vishnu-Purana .- The seven lokas. Vide ii. 226, 227.

P. 232. The commentator of the book of Patanjali .- Cf. note to p. 27.

P. 233. Dibajat .- This remark was already made on p. 210.

ANNOTATIONS. 325 P. 235. Fishnu-Purana .- Vide the doipas and seas, Fishnu-Purdna, ii. 109.

P. 236. Lokaloka, which means a not-gathering place. Apparently the author had not quite understood the nature of the compound loka-aloka, i.e. world and not-world.

P. 237. Vishnu-Purdna .- The first quotation seems to correspond to ii. 211-213, the second to ii, 204, and the third (on p. 238) to ii. 225-227. Seshakhya is apparently a mistake for Sesha-akhya, i.e. having the name of Sesha.

P. 240 .- The story of Vivamitra's attempt at creating a second world is taken from Ramayana, i. chaps. lvii .- Ix .; but here the king is called Trisanku.

P. 240 .- On Śripala, v. note to p. 164 The city of Multan is in various places mentioned by the author in such a remarkable manner as makes me think that he knew it, and that he had lived there for some time. When King Mahmud, A.H. 408 (A.D. 1017), had returned from Khwarizm-Khiva after the conquest of the country, and had carried along with him the princes of the con- quered house of Ma'mun, many scholars (among them Alberuni), officers, and soldiers, did he send some of these (among them Alberuni) as state prisoners to Multan, which he had conquered years before ? In this way, nine- teen years later (A.H. 427), the princes of the family of Altuntash, who had ruled Khwarizm after the Ma'munis, were treated by Mahmud's grandson, Majdud, who sent them as state prisoners to Lahore. At all events, it is perfectly certain that Alberuni cannot have been in favour with King Mahmad, or he wonld have dedicated one of his books to him. Cf. Sachan, Zur altesten Geschichte und Chronologie von Khudrizm, i. pp. 16, 28.

P. 240 .- Aljaihant is one of the fathers of Muslim litera- tnre on geography and travels in the eastern part of the Khaliphate, minister of one of the Samant kings of Central Asia towards the end of the ninth Christian century. His work is most extensively quoted, but has not yet come to

326 ALBERUNIS INDIA.

light. Cf. Aloys Sprenger, Die Post- und Reiserouten des Orients, Vorrede, p. xvii.

P. 241. When Brahman wanted, &c .- On the division of Brahman, on Dhruva, &c., cf. Vishnu-Purana, i. pp. 104, 161 seg.

P. 242. 1020 to 1030 stars .- This is the number of stars enumerated in the star-catalogue of 'Abdurrahmân Soff (ef. Schjellerup, Description des Etoiles fixes par Alşufi, St. Petersburg, 1874), which Alberuni has transferred into his Canon Masudicus. Should those men breathe and receive, &c .- I am not quite certain whether I have found out the right meaning of these words or not.

P. 243. The commentator Balabhadra, de .- Vide note to p. 156. P. 245, 1. 10,-The values here given correspond to the greatest declination of 24°. So AT=1397' is the sine of 24°, BT=298' the versed sine of 24°, and TH the difference between this latter and the radius 3438' (Schram).

P. 245, 1. 12. Kardajat .- The word kardaja seems to be derived from the Persian karda=cut, meaning a segment. The radius is equal to 3438 minutes of the periphery, which are called kardajat. Cf. i. 275, and ii. 205.

P. 246, 1 .-- Read 24° instead of 23°.

P. 246. Aryabhata of Kusumapura is repeatedly quoted by Alberuni. He mentions the orders of the numbers from ayutam to parapadma, i. 176. Here he speaks of the height of Mount Meru, on the longitude of Kurukshetra, i. 316 (where he quotes Pulisa and Prithusvamin), on the day of the Devas and that of the Pitaras, i. 330. He calls the cashaka vinadt, i. 335. From a book of his it is quoted that 1008 caturyugas are one day of Brahman; half of it is utsarpint, the other half avasarpint (Jaina terms), i. 371. Unfortunately I cannot read the title of this book; the signs may be aull, and it must remain uncertain whether it is an Arabic word with the article or an Indian one.

ANNOTATIONS. 337 Alberuni warns the reader not to confound this Aryabhata with the elder scholar of this name, to whose followers he belongs, In this place (i. 246) Alberuni does not seem to have nsed a work of Aryabhata junior himself, but to have taken these words of his from a commentary of Balabhadra. We learn here that the book had been trans- lated into Arabic, but do not learn which particnlar work of Balabhadra's. Was it his commentary on the Khanda- khadyaka of Brahmagupta ? Vide note to p. 156. That Alberuni had made a new edition of the Arabic version of the Khandakhadyaka is known (v. edition of the Arabic original, pref. p. xx.); perhaps he had also procured him- self an Arabic translation of Balabhadra's commentary. Cf. on this younger Âryabhata, Kern, Brhat Sañhita, pre- face, pp. 59, 60, and Dr. Bhau Dajt, " Brief Notes on the Age and Authenticity of the Works of Aryabhata, Vara- hamihira," &c., p. 392. Alberuni always calls him Arya- bhata of Knsumapura (Patna), to distinguish him from his elder namesake.

P. 247. Suktibam .- This seems to be some vernacular form for Suktimat. Vishnu-Purana, ii. 127. Rikshabam = Rikshavat (?).

P. 248. The Vishnu-Purdna says-I do not find this quotation in the Vishnu-Purana. Cf. V. P. ii. 117.

P. 248. The commentator of the book of Patañjali .- Vide note to p. 27. P. 249. Aleranshahrt .- Vide note to pp. 6, 7.

P. 249. Ardiyd and Girnagar (?) are apparently the same mountains which the Avesta calls hara berezaiti and taera.

P. 254. Vishnu-Purana .- The quotations from the V. P. given in this chapter are found in ii. p. 191 seq. P. 254 .- Jaunu, as here the river Yamuna is called, corresponds to the Prakrit form prescribed by Vararuci ii. 3, viz. Jaund.

P. 257. Vayu-Purana .- The names of the rivers are

328 ALBERUNIS INDIA.

found in the 45th Adhydya, vol. i. pp. 349-350. The order of enumeration of the mountains in the Sanskrit text is this: Pariyatra, Riksha, Vindhya, Sahyn, Malaya, Mahendra, Sukti. V. 97. vedasmritir vedavatî vritradhni sindhur eva ca varņasa candanâ caiva satira mahatt tathâ.

V. 98. para carmmanvatt caiva vidià vetravaty api sipra hy avanti ca tatha pariyatrasrayah smritAh.

V. 99. soņo mahânadas caiva narmmadâ sumahâdruma mandâkinî dasârņâ ca citrakûțâ tathaiva ca.

V. 100. tamasâ pipyalâ śroņi karatoyâ pišacika nilotpala vipasa ca banjula baluvahini.

V. I01. siteraja śuktimati makrnņa tridiva kramât rikshapâdât prasûtas tâ nadyo maņinibhodakâh,

V. 102. tapt payoshni nirbbandhya madra ca nishadha nadt venva vaitaranî caiva sitivahuļ kumudvati.

V. 103. toya caiva mahâgauri durga ca 'ntahsila tatha vindhyapâdapraśûtâć ca nadyah punyajalah śubhâh.

V. 104. godâvari bhîmarathî krishņâ vaiņy atha vafijula tungabhadra suprayoga kauveri ca tatha, paga dakshinapathanadyas tu sahyapadad vinihsritâh.

V. 105. kritamala tamravarna pushpajaty ntpalavati malayâbhijatas ta nadyah sarvah sitajalâh subhâh.

ANNOTATIONS. 329 V. 106. trisâma ritoktlya ca ikshula tridiva ca ya langulini vamfadhara mahendratanayah smritâh. V. 107. rishikâ suknmârî ea mandagâ mandavâhinî knpa palasini caiva suktimatprabhavâh smritah.

P. 259 .- Very similar to this enumeration of rivers is that in the Vayu-Purdna, adhyaya 45, VV. 94-108 :-

V. 94. piyante yair ima nadyo gangâ sindhusarasvatî satadrus candrabhâga ca yamuna sarayus tathâ.

V. 95. irâvatî vitastâ ca vipâsa devika kuhth gomati dhutapapa ca bahuda ca drishadvati.

V. 96. kanśiki ca tritiyâ tu niścirâ gaņdakt tathâ ikshu lohita ityeta himavatpadanihsritâh.

The following verse, already given in the note to p. 273, mentions the rivers flowing from the Pariyatra.

P. 259. Vedasint .- Write Vidasint.

P. 259. Kayabish .- The realm of Kayabish is here identified with Kabul. The signs may be read Kayabish or Kayabshi; only the consonants are certain. This reminds one forcibly of the name of the Indo-Scythian king Kadaphes. A dental sound between two vowels may in later forms be represented by a y,:as e.g. in Biyattu = Vitasta. Or is the word to be combined with P'anini's Kapisht (Capissene in Pliny) ? Cf. Panini and the geo- graphy of Afghanistan and the Panjab in " Indian Anti- quary," 1872, p. 21.

P. 259. Ghuzak .- This pass ('akaba in Arabic) is also mentioned in Elliot, "History of India," ii. 20, 449 (Ghûrak).

330 ALBERUNTS INDIA

P. 259. Below the town of Parodn .- It is mentioned in the maps at abont the distance of eight miles, as the crow flies, north of Tscharikar. The road from Anderab to Parvan has been sketched by Sprenger, Post- und Reise- routen, map nr. 5.

P. 259. The rivers Nur and Kird-Read Kirat instead of Ķira. Cf. Elliot, I. c. ii. 465.

P. 260 .- Bhatul seems to mean a sub-Himalayan coun- try between the Beas and the Satlej. It occurs only here and p. 211 (together with Antarvedi). (Elliot, " History of India," i. 22) mentions it as the name Masudi

of one of the five rivers of Panjab. The union of the seven rivers .- This tradition apparently refers to the hapta hendu of the Avasta, Vendidad i. 73.

P. 261. Matsya-Purdna .- Not having this book at hand, I give the corresponding passage from the Vdyu-Purâna, adhyaya 47, vv. 38-58 :-

V. 38. nadyâh srotas tn gangâyah pratyapadyata saptadha nalini hrâdinî caiva pâvanî caiva prâggatâ.

V. 39. sitâ cakshné ca sindhuś ca prattcim disam âśritâh saptami tv annga tasam dakshiņena bhagtratht, &c.

V. 42. npagacchanti tâh sarvâ yato varshati vâsavaļ sirindhran kuntalans cinan varvarân yavasan druhao.

V. 43. rushaņâmé ca kuņindamsca angalokavarâmé ca ye kritva dvidha sindhumarum sita 'gat pascimodadhim.

V. 44. atha cinamarums caiva nanganân sarvamûlikân sadhrâms tusharams tampakân pahlavân daradân śakân etan janapadan cakshuh sravayanti gato 'dadhim.

ANNOTATIONS. 331 V.45. daradams ca sakasmirân gandharan varapân hradân sivapauran indrahasan vadatihs ca visarjayan.

V. 46. saindhavan randhrakarakan bhramarabhirarohakân sunamukhams cordhvamanun siddhacaranasevitan.

V. 47. gandharvan kinnaran yakshan rakshovidyadharoragan kalapagramakams caiva paradan siganan khasan.

V. 48. kiratâms ca pulindams ca kurûn sabharatân api pancâlakasimatsyams ca magadhangams tathaiva ca.

V. 49. brahmottarâmé ca vangâmśca tâmaliptâms tathaiva ca etân janapadan aryyan ganga bhavayate subhan.

V. 50. tatah pratihatâ vindhye pravishtâ dakshiņodadhim tatas câ 'hladini puņya prâcinabhimukhî yayau.

V. 51. plavayanty upabhogâms ca nishadanan ca jatayah ghivaran rishakams caiva tatha nilamukban api.

V. 52. keralân ushțrakarnâms ca kirâtân api caiva hi kalodaran vivarnams ca kumaran svarņabhushitan.

V. 53. sâ mandale samudrasya tirobhûtâ 'nupurvataḥ tatas tu pâvani caiva prâcîm eva disan gatâ.

V. 54. apathân bhâvayanti 'ha indradyumnasaro pi ca tatha kharapathais caiva indrasankupathan api.

332 ALBERUNPS INDIA.

V. 55 madhyen 'dyânamaskarân kuthapravaraņân yayau indradvipasamudre tu pravishta lavanodadhim.

V. 56. tataé ca nalint câ 'gât prâcimâsam javena tu tomarân bhavayanti ha harsamargân sahthukân.

V. 57. purvan defarhs ca sevanti bhittva sa bahudha girîn karņaprâvaranârhs caiva prâpya câ 'śvamukhân api.

V. 58. sikatâparvatamarun gatva vidyadharan yayan nemimandalakoshthe tu pravishta sa mahodadhim.

P. 262. Vishnu-Purana .- This qnotation occurs V. P. ii. 192. Instead of Anutapata, Shikhi, and Karma, read Anutapta, Sikhi, and Kramu.

P. 263. Created .- This word seems to prove that Albe- runi already adhered to the dogma of orthodox Islam, that the Koran had been created by God from all eternity, and had been preserved on a table in heaven before God revealed it to mankind by the mouth of his prophet, Muhammad.

P. 264. Ibn Almukaffa (Abdallah) and 'Abdalkarim are also mentioned in the author's " Chronology of Aucient Nations," pp. 80 and 108.

P. 265. For this the astronomers requite them, &c .- When writing these criticisms, the author probahly thought of Brahmagupta Cf. the chapter on eclipses, il. IIO scg.

P. 267. Yamakoti, Lanka, &c .- Cf. the same names in Sarya-Siddhanta, xii. 38-40.

P. 268. Aryabhata, Vasishtha, Lata .- All the astrono- mers quoted in this context were not known to the author from their own works, but only through qnotations in the works of Brabmagupta Also the words of Varahamihira

ANNOTATIONS. 333 (here and p. 272) seem to be quotations of Brahmagupta (evidently p. 276), althongh they possibly might have been taken from Varâhamihira's Pancasiddhantikd. Pulisa, of course, mnst be excepted, as his Siddhanta was in the hands of Alberuni, and in course of being translated by him.

P. 271. Amaravati, Vaivasvata, &c .- Cf. on these four cities Vishnu-Purdņa, ii 240. P. 273. Apta-purana-kâra. -I do not see how the Arabic signs must be read. The translation of the term means the true ones who follow the Purdna.

P. 274, 1. 37 .- TA being the sine of 3f° is equal to 225', its square to 50,625; TB the versed sine of 3f° is 7', and HT = radius - TB = 3438' - 7 = 3431 (Schram). P. 275, 1. 3 .- The following calculation seems to have been made very negligently, for there are several faults in it. The radius 795° 27' 16" is correctly determined, for employing the ratio 7 : 22 between diameter and circum- ference, we are indeed led to this number, But already in the determination of Bc there is a fault. Alberuni seems to have converted o° 7' 42" into yojanas, instead of con- verting o° 7' 45"; for 360° being equivalent to 5000 yojanas, we get for 1° 13 yojana, 7 krosa, 444+ yards, for i' I krośa, 340721 yards, and for I" 12381 yards, and reckoning with those numbers we get o° 7' 42", and not o° 7' 45", which corresponds to 57,035 yards. Further, the rule he makes use of is completely erroneous; it is not true that the relation between the height of two observers is the same as the relation between the sines of their respective fields of vision. If this were the case, we should have sec a-I: sin a=sec B-I : sin f, or the quotient e sec a-I sin a would be a constant for every value of a, which, of course, is not the case. But even with his incorrect rule we cannot find the numbers he has found. This rule is 4 yards : sine of field of vision= 57,035 yards : 225', so one would have sine of field of vision=4 225 - bnt he finds 57035'

334 ALBERUNTS INDIA.

the sine of the field of vision equal to o° o' r" 3", which corresponds to 1000' 900'

runi seems to have reckoned 4 x 225 = 1000 instead of 900. 57033' and not to 57035 Therefore Albe-

Also the length of each degree is not quite correct; it is not 13 yojana, 7 krośa, 333 yards, but, as above stated, 13 yojana, 7 krośa, 4441 yards. Lastly, if we convert by means of this number o° o' I" 3" into yards, we find 1293 yards, so that the 2913 yards he speaks of seem to have been arrived at by an arroneous metathesis of the original ciphers (Schram).

P. 277. Prana-Cf. on this measure of time here i 334, 335-

P. 278. The inhabitants of Mount Meru, &c., till as tocsticard motion, almost identical with Surya-Siddhanta, xii. 55.

P. 281. There is a story of an ancient Greek, da-Pro- bably taken from Porphyry's book on the opinions of the most prominent philosophers abont the nature of the sphere. Vide note to p. 43.

P. 289. The Grecks determined, &e .- The author has given a description of the winds, according to the Arabian and Persian views, in his " Chronology of Ancient Nations," pp. 340, 341.

P. 291. Atri, Daksha, &c .- The legends here referred to are found in Vishnu-Purana, i. 153, ii 21 seg.

P. 294-The Rishi Bhuvana-kosa (ie. globe) is only mentioned in this place, and not known to me from other sources, His work, the title of which is not given, seems to have treated of geography.

P. 295. Samnara (?) .- Thus the manuscript seems to have it. The signs may also be read Samnad.

P. 297. Kurmacakra .- Vide on this term a note of H. Kern, Brihat Samhita, translation, to tha title (kurmavi- bhága) of chap. xiv.

ANNOTATIONS. 335

P. 298. Ulpala,a native of Kashmir .- Vide note to p. 157.

P. 298 .- Stone-tower, i.e. the Alowos Tpyos of Ptolemy, vi. 13, 2.

P. 299-Bûshang, a place near Herat, to the west. Sakilkand, also Iskilkand, is identified with Aleandria by Elliot, " History of India," i. 366, note 1. Perhaps it is identical with Συγάλ πόλις οf Stephanus Cf. Droysen, Geschichte des Hellenismus, iii. 2, 217.

P. 299 .- This extract from Vayu-Purdna is fonnd in adhydya 45, vol. i. pp. 350-353, vv. 109-136. Alberuni gives the directions in the following order: east, sonth, west, north; whilst the Sanskrit text has this order: north, east, south, west. In comparing the following text with Alberuni, the varietas lectionis given in the footnotes of the Calcutta edition can sometimes be used with advantage.

V. 109. kurupafcalâh salvas caiva sajangalâh

V. 110. śûrasena bhadrakârâ bodbâh satapatheśvaraih vatsâh kisashta kulyâs ca kuntalah kâśikośalâh.

V. III. atha pârsvê tilangas ca magadhas ca vrikaih saha.

V. 115 .- NORTIL vâhlikâ vâdhadhânâs ca âbhirâh kâlatoyakâh aparitas ca śûdras ca pahlavas carmakhaņdikâh.

V. 116. gandhara yavanas caiva sindhusauvirabhadrakâh šaka hradah kulinds ca parita hâraptrikâh.

V. 117. ramața raddhakatakâh kekaya dasamanikâh kshatriyopaniveśîá ca vaiśyaśûdrakulâni ca.

336 ALBERUNPS INDIA.

. 118 kâmboja daradii caiva varvarâh priyalaukikah pinas caiva tusharas ca pahlava vahyatodarâh.

V. 119 âtreyûé ca bharadvâjâh prasthalaá ca kaserukâļ lampâka stanapas caiva pidika juhndaih saha

V. 120. apagâé câ 'limadrâs ca kirâtānâā ca jâtayah tomâra harhsamâryâśca kâśmirâs tangaņâs tatha

V. 121. côlikâs câ hukâs caiva pûrnadarvâs tathaiva ca

V. 122 .- EAST. andhravâkaļ sujarakâ antargiri vabirgirâh tatha pravangavangeya malada mala varttinah.

V. 123. brahmottarâh pravijaya bhargava geyamarthakah prâgjyotishâs ca mundas ca videhas tâmaliptakâh mala magadhagovindah.

V. 124B-SOUTH. pâņdyâs ca keralas caiva caulyâh kulyâs tathaiva ca setuka můsbikas caiva kumana vanavasikâh mahårâshtra mâhishakâh kalingas ca.

V. 126. abhtrâh saha cai 'shtka stavyâs ca varâs ca ye pulindrâ vindhyamûlika vaidarbhâ dandakaih saha.

V. 127. pinika mannikas caiva asmaka bhogavarddhanâh nairņikâh kuntala audhra udbhida nalakâlikah.

V. 128. dâkshiņâtyas ca vaidesa aparans tân nibhodhata śorpåkårâh kolavana durgth kalitakaih saha,

ANNOTATIONS. 337

V. 129. puleyâs ca surâlâs ca rûpasas tâpasaih saha tatha turasitas caiva sarve caiva paraksharâh.

V. 130. nâsikyâ 'dyas ca ye cânye ye caiva 'ntaranarmadâh bhânukacchrâh samâheyâh sahasa śaśvatair api,

V. 131. kacchtyas ca surâshtrâs, ca anarttâs câ 'rvudaih saha.

V. 132 .- WEST. malavâs ca karushas ca mekalasco 'tkalaih saha uttamarņa dasarnas ca bhojah kishkindhakaiņ saha.

V. 133- tosalâh kosalas caiva traipurâ vaidikas tatha tumnrâs tumburas caiva shațsura nishadhaih saha

V. 134 anupas tundikeras ca vitihotra hy avantayah

V. 135- nigarhara hamsamargah kshnpanas tanganah khasah.

V. 136. kusapravaranas caiva huna darvâh sahûdakâh trigarttâ malavâś caiva kirâtâs tâmasaiķ saha.

Pp. 300-303 .- This extract from Varâhamihira's Sam- hitd is taken from chap. xiv. Cf. the text in Kern's edition, p. 87, the varietas lectionis, pp. 12-14, and his translation in " Journal of the Asiatic Society," 1870, p. 81-86. The number of discrepancies between these two traditions is very considerable. In many places Alberuni and his Pan- dit may not have read their manuscript with sufficient accuracy; in others, the Sanskrit manuscript-tradition may exhibit blunders arising from a nc. uncommon confusion of characters that are much like ech other. The Arabic manuscript-tradition is on the whole correct, but the VOL. LI. Y

338 ALBERUNI'S INDIA.

copyist of the Arabic text, too, may have contributed in some case to increase the number of errors. To some Indian names he has added explanatory glosses, eg. Sam- vira, i.c. Miltan and Jahravar. It is a pity he has done this so sparingly.

P. 303 .- Yakub and Alfazart .- Vide notes to pp. 169 and 165.

P. 304-Aba-Mashar, author of many books, chiefly on astrology, died A.H. 272=A.D. 885. He is known to the Middle Ages in Europe as Albumaser.

P. 306. Cupola of the carth .- If this expression has not been derived from the Indian, the question arises, Who introduced it among the Arabs? Was it Alfazârl ?

P. 306. Ravana the demon-The author refers to the fifth and sixth books of the Ramdyana, which he apparently did not know, or he would not have called it, as he con- stantly does, the story of Rama and Ramayana; v. pp. 307. 310, and ii. 3. I have not succeeded in deciphering the name of the fortress; the Arabic signs cannot be combined with the name Trikûta.

P. 308 .- A straight line from Laikd to Meru is also mentioned on p. 316. The first degree of lougitude, accord- ing to the Indian system, is also described in Surya-Sid- dhanta, i 62. Instead of Kurukshetra the anthor seems to have pronounced Kurukketru. At all events, he did not write a sh. Therefore the compound ksh must have under- gone the Prakritic change into kkh, as in pokkharo=push- kara (Vararuci, iii. 29).

P. 309 .- These wares are deposited, dc .- This kind of commerce with savage nations is the same as that carried on by Carthage with tribes on the west coast of Africa; v. Herodotns, iv. 196; C. Müller, Geographi Græci Minores, i p. xxvii., and Meltzer, Geschichte der Karthager, p. 232 and 506.

P. 310 .- Langabalus is identified with the Nicobar

ANNOTATIONS. 339

Ialands by A Sprenger, Post- und Reiserouten des Orients, p. 88.

P. 312. Desantara .- Vide the rule for its compntation in Surya-Siddhanta, i. 60, 61. Alarkand, Ibn Tarik .- Cf. note to p. 169.

P. 312 .- Al-arkand is identified by Alberuni with the Khandakhadyaka of Brahmagupta (ii. 7). In another place (ii, 48) the author identifies the word arkand with ahargana. Both of these identifications can hardly be justified phonetically, and therefore I prefer to suppose as the Sanakrit original of Arkand a word like Aryakhanda, whilst apparently the word harkan (title of an Arabic calendar, ii. 52) is identical with ahargana. The author complains of the Arabic tranalation of Al-arkand being a bad one, and at aome time of his life (probably after the composition of the Indica) he has pub- lished a new and amended edition of this translation. Cf. preface to the Arabic edition, p. xx. The Arabic Arkand has not yet been discovered in the libraries of Europe. The author has borrowed from this book the following notes :- (1) 1050 yojanas are the diameter of the earth (i. 312, 316). (z) The latitude of Ujain is 22° 29', and that of Almangura 24° I' (i. 316). Here the author states that also Ya'kub Ibn Tarik had quoted the book, but erroneously. (3) The straight shadow in Loharanif is 53 digits (i. 316). (4) Alberuni quotes from Alarkand a method for the computation of the era Shakh, by which the Gupta era is meant (ii. 48, 49).

P. 312 .- On the relation between yojana and mile, v. note to p. 199.

P. 312, 1, 22 .-- Using the ratio of 7: 22 between diameter and circumference, we find 3300 yojanas as the circum- ference corresponding to a diameter of 1050 yojanas. So 3300 yojanas is the circumference of the earth given in the handbook Al-arkand. This agrees with the last lines of p. 315, where it is aaid that 3200 yojanas are 100 yojanas less than the value given by Al-arkand (Schram).

340 ALBERUNI'S INDIA.

P. 313. The author of Karanatilaka, i.c. Vijayanandin. -Vide note to p. 156.

P. 313 .- Vyastatrairdsika is a technical term for a cer- tain algebraic calculation. Cf. Colebrooke, " Algebra," p. 34, § 76.

P. 314 .- Alfazari in his canon, which was a translation of the Brahmasiddhanta of Brahmagupta; v. note to pp. 153, 165.

P. 314, 1. 11 .- The calculation of the desdntara is, as Alberuni remarks, quite erroneous, as the difference of longitude is not taken into acconnt (Schram).

P. 315, 1. 25 .- The number in the lacnns must be 80, for Alberuni says at the bottom of the page," If we invert the calculation and rednce the parts of the great circle to yojsnas, according to this method we get the number 3200." But to get 3200 we must multiply s4o by 80. The rule, " Multiply the yojanas of the distance between two places by 9 and divide the product by 8o," serves to convert this distance given in yojanas into degrees. This distance, then, is considered as the hypothenuse of a right-angled triangle, one of the sides of which is the dif- ference of the latitudes, the other the unknown difference of the longitudes; this latter is found by taking the root of the difference of the squares of hypothennse and known aide. This difference of longitade is then expressed in degrees; to get it expressed in day-minutes we must further divide by 6, as there are 360° in a circle, but only 60 day-minutes in a day (Schram).

P. 316 .- The line connecting Lankd with Meru, already mentioned on p. 308.

P. 316. Yakub Ibn Tarik, Alarkand .- Vide note to p. 169, 312.

P. 317. Catlaghtagin .- Not knowing the etymology of this Turkish name, I am also ignorant of its pronnnciation. The second part of the compound seems to be tagin=

ANNOTATIONS. 341

valorous, as in Toghrultagin, i.e. valorous like a falcon. ot, julghan, means a large spear, one might think of As

reading Jughattagin, i.e. valorous with the spear, but this is very uncertain. Another name of a similar formation ja kutlughtagin, katlagh, but probably entirely different. Vide Biberstein-Kazimirski, Menoutschehri preface, p. 136; Elliot, " History of India," ii, 352, iii. 253.

P. 317 .- Karanasđra by Vitteśvara; v. note to p. 156.

P. 317 .- The fortress Lauhur, also mentioned p. 208 as Lahur, must not be confounded with Lauhavar or Lahore. Situation unknown. According to the author's Canon Masudicus, it has latitude 33° 40', longitude 98° 20'. Comparing these latitudes with those given in Hunter's Gazetteer, we do not find that they much differ :--

Ghazna Huntor. AlberunL

Kabnl 33 34 33° 35 Peahavar . 34 30' 33 47 . Jailam 34° [' 45" 34 44

Siyalkote 32 55' 26" 33° 20

Multan . 32' 31' 32° 58' 30° 12' 29° 40'

On the identity of Waihand and Attok, ef. Cunning- ham, " Ancient Geography of India," p. 54 Mandakkakor (the name is differently written) was the fortress of Lahore, according to the author's statement in his Canon Masudicus. Nandna is explained by Elliot (" History of India," ii. 450, 451) as a fort on the mountain Balnath, a conspicuous mountain overhanging the Jailam, and now generally called Tilla, Cf. also Elliot, l. e. ii. 346, note 347, 366. The places Dunpur (pronunciation perfectly uncertain) and Kandt (also read Kiri), the station of the Amtr, seem to have been on the road from Ghazoa to Peshavar. Near the latter place was fought the decisive battle between King Mashd and his blinded brother Muhammad, A.D. 1040, and there the former was murdered by the relatives of those who ten years earlier had thonght to win his favour by betraying his brother, and were killed or mal- treated in reward. Cf. Elliot, l. c. iv. 199, note 1, 138, ii. 150, 112 (Persian text, p. 274), 273, note 3.

342 ALBERUNTS INDIA.

I conjecture Dunpûr to have been identical with Jala- labad or some place near it. Latitude of Jalalabad, 34° 24'; that of Dunptr, 34° 20'. Kandi, more sonthern than Dunpur and nearer to Kabul, must have been a place like Gandamak or near it. If it is called the station (post-relai) of the Amar. We may understand by this Amir the father of King Mahmud, the Amir Sabuktagin, who first constructed the roads lead- ing to the Indian frontier, as Alberuni informs us on p. 22. On the identification of Bamhanwa or Almanstra in Sindh, v. Cunningham, I. L. p. 271 seg. The statementa of Alberuni regarding the Kabul valley and environs have been laid down in a sketch-map of Aloys Sprenger, Post- und Reiserouten des Orients, No.

No. 13. 12; the Punjab and the approaches of Kashmir, ibid.

P. 319-Muhammad Ibn, &c., is the famous Razes of the Middle Ages, who died probably A.D. 932. The anthor bas written a catalogue of his works which exists in Leyden; v. Chronologie Orientalischer Volker von Alberuni, Einlei- tung, p. xi; Wtistenfeld, Geschichte der. Arabischen Aerzte, No. 98.

P. 320-Alerander of Aphrodisias is the famons com- mentator of Aristotle, who lived in Athens abont 200 after Christ. Cf. Fibrist, p. 252, and Zeller, Geschichte der Griechischen Philosophie, 3, 419. The quotation is found in Aristotle, Phys. vii. I.

P. 320. Vardhamihira .- This quotation corresponda to Samhita, i. v. 6, 7. Instead of Kumbhaka the Sanskrit text has Kanâda.

P. 322. Timcus-This quotation seems to be derived from 42 D E :- τό δέ μετά τόν σκόρον τοϊς νέοις παρέδωκε Θεοίς σώματα πλάττειν θνητά, κ.τ,λ, καί λαβόντες άθάνατον άρχήν θνητού ζώου, κ.τ.λ.

ANNOTATIONS. 343 In the Arabic text, p. Mr, 17, read de instead of ga., and bit instead of u.

P. 324. That being who is above him, i.e. a being of the next higher order .- The opposite of the term ade o is de ow (for the being of the next lower order) on p. Wvy, 20 (translation i. 351).

P. 325. Vishnu-Purana .- The first words, Maharloka lies, dc., there is one kalpa, are found in ii. chap. vii. p. 226. The sons of Brahman are mentioned in Vishnu- Purdna, ii. 200, note. The name Sanandanada (Sananda- natha ?) is perhaps a mistake for Sanatana. Cf. Samkhya Kdrikd with the commentary of Gandapada by Colebrooke- Wilson, p. I.

P. 325. Abd-Mashar .- Vide note to p. 304

P. 325. Aleranshahrt-Vide note to pp. 6, 7.

P. 327. The country without latitude, i.e. niraksha in Sanskrit .- Vide p. 267, and Surya-Siddhânta, xii. 44, note.

P. 330. Aryabhata of Kusumapura, i.e. junior .- Cf. note to p. 246.

P. 333 .- The terms parardha and kha have been ex- plained, pp. 175, 178.

P. 334 The book Srudhava by Utpala .- Vide notes to pp. 157, 158. A system of the measures of time has also been given by Colebrooke, "Essays," i. 540 seg.

P. 336. S-M-Y .- This name is so written here and p. 337. The Arabic signs are to be read Shammt or Shamiyyu. I do not know a Sanskrit name of this form. Is it= Samaya ? The same name seems to occur a third time, ii. 188, but is there written S-M-Y. Alberuni says that S-M-Y had dictated a method for the compntation of the samkranti;

344 ALBERUNPS INDIA.

he therefore, perhaps, was a scholar of the time and a per- sonal acqhaintance (teacher ?) of Alberuni's. The title of a book of his is not mentioned.

P. 338 .- The sped muhra or white shell, an Indian blow- ing instrument, is also mentioned by Elliot, " History of India," ii. 215, note. Purshûr (sx), as the manuscript has, is probably a mis- take for „t, Purushdvar, i.c. Peshavar.

P. 338. Horæ æquinoctiales and temporales-Vide note to p. 214-

P. 339. The commentator of the Siddhanta, Pulisa .- Read instead of this, "The commentator of the Siddhanta of Pulisa," and compare note to pp. 153, 154 Who this com- mentator was is not mentioned.

P. 340 .- Abhijit means the 8th muhurta of the day. The Arabic form corresponds perhaps to Sanskrit abhijiti.

P. 340. Vydea .- This atatement points to Mahabharata, the Adi-parvan, v. 4506; but the chronological detail is not found there.

P. 340. Śisupala .- Vide note to p. 165.

P. 342 .- The names of the dominants of the muhurtas are also mentioned in the following four linea taken from Aufrecht's Catalogue of the Sanskrit manuscripts of the Bodleian Library, p. 332a :--

rudrâhimitrapitaro vasuvâriviśve vedha vidhi satama- khah puruhûtavahnî.

naktamcaraś ca varuņâryamayonayaś ca proktâ dinê daśa ca parca tatha muhûrtâh

nisamuhurta girisajapadahirbudhnyapûshaśviyamâgna- yaśca.

ANNOTATIONS. 345 vidhâtricamdrâditi jtvavishņutigmadyutitvâshțrasamî- ranas ca.

P. 343. Except the astrologers-Cf. the meaning of hord in astrology, ii, 222.

P. 343. Vijayanandin-Vide note to p. 156. The title of his book would be in Arabic wlan &e (Ghurrat- alzijt).

P. 344. Names of the hords .- I have not found these names in Sanskrit. Perhaps they are mentioned in some commentary to Surya Siddhanta, xii. 79. On Srudhava, v. note to p. 158.

P. 347. Physical scholars know, &c .- There is a similar passage on the physical effects of moonlight in the author's " Chronology of Ancient Nations," p. 163. I am afraid I have not canght the sense of the sentence, "and that she affects (?) linen clothes," &c.

P. 348. Atuh (?) .- The MS. seems to read dtvahhu. The word 4a, BRBA, is perhaps a mistake for dy, barkhu, which, according to the table, ii. 197 (ef. Trumpp, " Gram- mar of the Sindhi Language," p. 158), is the name of the first day of a paksha.

P. 348. Veda .- The author gives six quotations from the Veda: one taken from Patanjali (i. 29), one from Samkhya (i. 31), two from the Brahmasiddhanta of Brah- magupta (ii. 1IO, III), and two qnotations which wero probably communicated to him by his Pandits, as he does not mention a particular source whence he took them (i 348 and ii. 348).

P. 352. Vasudeva .- The quotation corresponds to Bha- gavad-Gita, viii. 17. The book Smriti .- Vide note to p. 131. This quotation seems to have been taken from Manu, Dharmasastra, i. 72.

346 ALBERUNI'S INDIA.

P. 353 .- The information on the four mdnas (ef. Sarya- Siddhanta, chap. xiv.), as given by Yakab, was the only one at the disposal of Alberuni at the time when he wrote his "Chronology" (o. English edition, p. 15). It was commnnicated to him by the Kitab-alghurra of Aba Muhammad Alna'ib Alamuli The four different kinds of spaces of time mentioned there are the four manas, saura, savana, candra and nakshatra.

P. 353 .-- Bhukti, in Arabic buht, is the daily motion of a planet; ef. Surya-Siddhânta, i, 27, note, and here, ii. 195. The Arabic form does not seem to have passed through an intermediate stage of a Prakritic nature, for in Prakrit it would have been bhutt (Vararuci, iii, I).

P. 355. The savana-mana is used, &c .- Cf. the similar rules in Surya-Siddhanta, xiv. 3, 13, 15, 18, 19.

P. 356. Uttarayana .- On the two ayanas ef. Surya- Siddhanta, xiv. 9.

P. 357. Ritu .- Vide the description of the six seasons in Surya-Siddhânta, xiv. 10, 16.

P. 358. Dominants of the halves of the months .- I do not know a Sanskrit list of these names. The Asana (Ashunu) perhaps means Asvin or Asvint.

P. 359-Dimas (probably pronounced dimasu) = Sanskrit divasa, is the shibboleth of the Indian vernacular dialect spoken round Alberuni, and probably by himself. I do not know which dialect this was, nor whether there are any traces of it in our days. The change between v and m is also observed in the following examples:r carmanmat = carmanvatt (Chambal), wmd himamant= himavant, al.S jagamalku - yajnavalkya, macci= vatsya, AK- sugrimu=sugriva. Some examples of the change of v to m are also given by Hornle, " Comparative Grammar," § 134

P. 359 The three sounds h, kh, and sh, &c .- On the pro-

ANNOTATIONS. 347 nunciation of ah as kh, ef. Hornle, l. c. § 19, and on the further change of kh to h, ibid. § 19. Examples of the former change are numerons in the Indica; of examples of the latter, ef. su. munha = mukha, olyr babrahan= raprakhana (?), and also lei dhari, ef. ashadha, uss kihkind = kishkindha. In Prakrit muham = mukha (Vara- ruci, ii. 27).

P. 361. Sridhara by Utpala .- Vide note to p. 157,

P. 362. I ghati = 16 kald .- Cf. with these measures of time the statements on pp. 336, 337.

P. 364, Chapter XL .- It has also been translated by Reinand, Fragments Arabes et Persans, pp. 155-160.

P. 364. Samdhi udaya and saindhi astamana .- One would expect samdhyudaya and samdhyastamana, but there is no trace of a y. The forms have a vernacular character, and must be explained according to the analogy of ws duti = dyuti, and yst antazu = antyaja. Hiranyakafipu .- The story of this king and his son Prahlada is told by the Vishnu-Purdna, ii. 34 seq.

P. 366. Samdhi .- The way it is used in astrology is shown by the table, ii, 219.

P. 366. Punjala .- Vide note to p. 157. The tradition here given is very similar to that mentioned by Colebrooke, " Essays," ii 332, 333.

P. 366, 1. 35 .- We find that the beginning of the Hindn solar year 854 Sakakala takes place A.D. 932, March 22, 6 ghatt 40' 15", which corresponds to March 22, 7 h. 40 m. civil Greenwich time, whilst the real instant of the solstice is March 15, 12 h. 15 m. civil Greenwich time, so that the aolstice precedes the calculation by 6 days and 19 hours, which agrees very well with the 6° 5o' which Puñjala men- tions (Schram).

348 ALBERUNP'S INDIA.

P. 368. Ahargana - ahar+gana-The author's erroneons explanation is repeated ii. 26. Sind-hind -siddhanta-It may be qnestioned whether the inorganic n has been introduced into the word by the Arabs, or whether it existed already in the pronunciation of the Hindus from whom they learned the word. I do not know of a rule to this effect in Prakrit or vernacular, but there are certain Indian words which apparently show a similar phonetic process. Cf. e.g. Prakrit utto (Sanskrit, ushtra), which in Eastern Hindhi has become af or unt. Hornle, " Comparative Grammar of the Gaudian Lan- guages," § 149.

P. 370. Aryabhata, sen .- Vide note to p. 156. Aryabhata of Kusumapura. Vide note to p. 246. The word I cannot decipher may be read coudt, i.c. the article and three consonants with three dots above them, something like Au.

P. 371. Utsarpint, avarsarpint, are terms employed in the Jains system. Cf. Colebrooke, " Essays," ii. 186, 194-

P. 372. The book Smriti mentions-This is Mann, Dhar- maśdstra, i, 80.

P. 375. A translation of his whole work, &c .- Cf. note to pp. 153, 154. Alberuni was translating the Pulisa- Siddhanta, which until that time had not yet been trans- lated into Arabic by Muslim scholars, because they did not like its theological tendency.

P. 376. Brahmagupta .- Vide note to pp. 153, 154-

P. 378. In writing the introdnctory sentences of chap. . xliji, the anthor seems to have had in mind Plato'e Timceus, 220: πολλαλ καί κατά πολλά φθοραί γεγόνασιν άνθρώπων καί έσονται, κ.τ.λ.

P. 379. The pedigree of Hippocrates is known from Tzetzes, chil. vii. host. 115. Cf. " The Gennine Works of Hippocrates," translated by Fr. Adams, London, 1849, vol.

ANNOTATIONS. 349 i. p. 23. The name oy t seems to be a repetition of the name Hippolochos, d> !. If it is dropped from the list, we have the fourteen generations which the anthor counts between Hippocrates and Żeus. The Arabic oitt seems to be a mistake for eykt, Machaon.

P. 380. Parasurdma .- Vide this legend in Vishnu- Purana, iv. 19 (here added from the Mahabharata).

P. 380. Buddhodana .- Vide my conjecture as to the origin of this name in note to p. 40. The Muhammira-This term has been explained in note to p. 21.

P. 382. Garga, the son of .- The name of his father is written Jashu or Jash (here and p. 397). Could this be Yasoda ?

P. 382 .- 'Alf Ibn Zain was a Christian physician in Merw; cf. Shahrazuri, MS. of the Royal Library, Berlin, MS. Or., octav. 217, fol. 144b; the same in Baihakt, ibid. No. 737, fol. 6a. According to this tradition, his son was the anthor of the famous medical book Firdaus-alhikma. Cf. also Fihrist, p. 296 and notes; Wustenfeld, Geschichte der Arabischen Aerzte, No. 55. The book Caraka .- Vide note to p. 159.

P. 383. Krisa, the son of Atreya-If this is what the author means, the Arabic signs oy must be altered to uwy. Cf. A. Weber, Vorlesungen, p. 284, note 309.

P. 383 .- The quotation from Aratus is Phænomena, VV. 96-134 I give the text from Imm. Bekker, Aratus cum Scholiis, Berlin, 1828 :-

Άμφοτέροισι δε ποσσίν υποσκέπτειο βοώτεω Παρθένον, ή ρ' έν χερσί φέρει Στάχυν αίγλήεντα. ειτ' ούν Άστραίου κείνη γένος, δν ρά τέ φασιν άστρων άρχαϊον πατέρ' έμμεναι, είτε τευ άλλου, εύκηλος φορέοιτο · λόγος γε μέν έντρέχει άλλος

350 ALBERUNI'S INDIA. άνθρώποις, ώς δήθεν έπιχθονίη πάρος ξεν, ήρχετο δ άνθρώτων κατεναντίη, ούδέ ποτ άνδρν ούδέ ποτ' άρχαίων ήνήνατο φύλα γυναικών, άλλ' αναμιξ εκάθητο καί αθανάτη περ έουσα. καί έ Δίκην καλέεσκον · άγειρσμένη δε γέροντας iε που είν άγορη ή ευρυχόρο έν άγυιη, δημοτέρας ξειδεν επισπέρχουσα θέμιστας. ούπω λευγαλέου τότε νείκεος ήπίσταντο, ούδε διακρίσιος περιμεμφέος ούδε κυδοιμού· αύτως δ' έζωον. χαλεπή δ' άπέκειτο θάλασσα, καί βίον ούπω νήες άπόπροθεν ήγίνεσκον· άλλά βόες καί άροτρα καί αύτή πότνια λαών μυρία πάντα παρεχε Δίκη, δώτειρα δικαίων. τόφρ' ην οφρέτι γαία γένος χρύσειον έφερβεν. αργυρέφ δ' ολίγη τε καί ούκέτι πάμπαν όμοίη μίλει, ποθέουσα παλαιών ήθεα λαών. άλλ' ίμπης έτι κεΐνο κατ' άργύρεον γένος εν. ήρχετο δ' έξ όρεων υποδείελος ήχνέντν μουνάξ · ούδε τεφ επεμίσγετο μειλιχίοισιν · άλλ' όπότ' άνθρώπων μεγάλας πλήσαιτο κολώνας, ήταλει δή έπειτα καθαπτομένη κακότητος, ούδ' έτ' έφη είσωπός ελεύσεσθαι καλέουσεν. στην χρύσειοι πατέρες γενεήν ελίποντο χειροτέρην· ύμείς δε κακώτερα τεξείεσθε. καί δή που πόλεμοι, καί δή καί άνάρσιον αμα έσσεται άνθρώποισι, κακοίς δ' έπικείσεται άλγος. ως είπουσ' όρέων έπεμαίετο, τους δ΄ άρα λαούς είς αυτήν έτι πάντας ελίμπανε παπταίνοντας, άλλ' ότε δή κάκεινοι έτέθνασαν, οί δ' εγένοντο, χαλκείη γενεή, προτέρων αλοώτεροι άνδρες, οι πρώτοι κακοεργόν έχαλκεύσαντο μάχαιραν είνοδίην, πρώτοι δε βοών έπάσαντ' αροτήρων, καί τότε μισήσασα Δίκη κείνων γένος άνδρων πταθ' υπουρανί.

P. 384 The commentator of the book of Aratus-This

ANNOTATIONS, 351 commentary is not identical with the scholia edited by Bekker. Cf. Eratosthenis Catasterismorum Religuic, rec. C. Robert, pp. 82-84-

P. 385. Plato .- This qnotation is from Leges, iii. 677; but the phrases forming the conversation have been omitted. ΑΘΗΝ. Τό πολλάς ανθρώπων φθοράς γεγονέναι κατακλυσμοίς τε καί νόσοις καί άλλοις πολλοίς, έν οΐς βραχύ τι τό τών άνθρώπων λείπεσθαι γένος, κ.τ.λ. ώς οί τότε περιφυγόντες τήν φθοράν σχεδόν όρεινοί τινες ν είεν νομείς έν κορυφαίς που, σμικρά ζώπυρα του τών άνθρώπων γένους διασεσωσμένα, κ.τ.λ. καί δη τούς τοιού- τους γε ανάγκη που τών άλλων άπείρους είναι τεχνών καί τών έν τοϊς άστεσι πρός αλλήλους μηχανών ες τε πλεο- νεξίας και φιλονεικίας καί όπόσ' άλλα κακουργήματα πρός άλλήλους έπινοούσιν.

P. 387 .- Cf. with this table Vishnu-Purdna, book iii. chap. i. and ii., and the Bombay edition, 1886. Stamasa seems to be a mistake for Tamasa. Caitraka instead of caitra seems to have been derived from an erroneous reading of the beginning of the Sanskrit caitrakimpurushadydśca. Sudivya seems to have risen from a wrong division of the words Parasu (other readings Parabhu, Parama)

dyâstasya. Divya. The Bombay edition reads prajahparamadivya-

Antata, the name of Indra in the fifth Manvantara, can hardly be combined with the Vibhu of Sanskrit tradition. Sindhu, Reva .- These words, whatever their proper pro- nunciation may be, are not found in the Sanskrit text. Puru Muru is Sanskrit Uru Puru, but Pramukha is a gross mistake, for the text has urupurusatadyumnapramu- khah, i.e. Uru, Puru, Satadyumna, and others. Nabasa and Dhrishna are mistakes for Nabhaga and Dhrishta. Virajas, Ascarvart, Nirmogha .- The Sanskrit text runs riracdscorvarivdmścanirmohadyas, which Alberuni has divided into viraja-aścorvarivamsca-nirmoha. Cf. Scor-

352 ALBERUNI'S INDIA.

vart Vamsca on p. 394. Wilson reads the second name Arvartvat. Mahdoirya, name of Indra in the ninth Manvantara, instead of Adbhuta, rests on a misinterpretation of these words: tesham indro mahaviryo bhavishyatyadbhuto dvija. Sudharmatman .- The Sanskrit text has Sarvadharma. Devata Vanupadevdsca, instead of Devavat and Upadeva, rests on a wrong division of the words devavdnupadevaśca. Vicitra-adyd, a mistake for vicitrady, i.c. Vicitra and others. Urur, Gabht (sic MS.), Budhnya-adyd, a mistake for ururgabhirabudhnyadyd, ie. Uru, Gabhira, Budhnya, and others.

P. 388. The same book relates, viz., Vishnu-Purdna, iii. p. 20. On Priyavrata, v. ibid. ii. p. 101.

P. 389. A pious woman, viz., Arundhatt, v. p. 390.

P. 390. On the Seven Rishis, or Ursa Major, ef. Cole- brooke, " Essays," ii. 310.

P. 391. The almanac or calendar from Kashmir for the Saka-year 951 (A.D. 1029) is quoted in two other places, ii. 5 and ii, 8.

P. 391. On the ancient astronomer Garga, ef. Kern, Brhat Sanhita, preface, p. 33 seq.

P. 392. Only by 525 years .- Cf. on Vardhamihira note to p. 54-

P. 392. Karanasdra by Vittesvara .- Vide note to p. 156.

P. 394-This table is taken from Vishnu-Purana, book iii. chaps. i. and ii. 2. Manvantara: Dattu Nirishabha .- A mistake for Dat- toni Rishabha. Nisvara .- Alberuni read Nirsara. Ścorvari Vamśca .- The author has wrongly divided the

ANNOTATIONS. 353

word ścorvarirdmsca (ed. Bombay svorvarivamsca). Cf. note to p. 387. 4 Manvantara: Jyoti (read Jyotis) Dhaman .- Mistake for Jyotirdhaman. Caitrogni, as the author has, is a mistake for Caitragnt. Varaka .- Ed. Bombay, Vamaka; Wilson-Hall, Vanaka. 5. Manrantara: Rurdhvabahu has risen through the wrong division of the two words vedasrirurdhvabahu. Apara has by mistake heen taken for a proper noun in the following words :- urdhvabahustathaparah. Subahu (Srabahu ?) .- The Sanskrit text has svadhaman. 6. Manvantara: Atinaman .- The Arabic text has ati- manu. Or are we to read Lst instead of guil? Carshayah (= and the Rishis) by mistake derived from the following passage :- saptdsanniticarshayah. 9. Manvantara: Havya, in the Sanskrit tradition Bhavya. Perhaps we must read instead of Medhadhriti (Wilson-Hall), medhamriti (ed. Bombay). Alberuni seems to have read Vedhddhriti, if we are not to read wolesy instead of wulese. 10. Manvantara: Satya (Wilson-Hall) .- The Arabic has something like Sattayđ. Sukshetra .- The Arabic has Sushera instead of Satyaketu. Perhaps the author has overlooked this word and copied the following one, viz., Sukshetra. II. Manvantara : Niścara, in the Arabic viscara. Agnidhra = Agnitejas. The Arabic has agnitru Fust, which is perhaps to be changed to yzst (agnitejas). Nagha .- Wilson-Hall, Anagha. 12. Manvantara: Sutaya, in the Sanskrit text sutapdśca. Perhaps the author has read sutayaśca. Dyuti and Iscanyas have by mistake been derived from the following verse- tapodhritirdyutiscanyahsaptamastutapodhanah. 13. Manvantara: Tatradarsica, mistake for Tatvadar- sin, for the Sanskrit text has tatvadarsica. Vyaya, mistake for Aryaya. The author seems to bave read dhritiman vyayaśca instead of dhritimanavyayaśca. 14. Manvantara: Agniba instead of Agnibahuh. Gnîdhra .- The ed. Bombay reads magadhognidhran- vaca. Other readings, Gridhra, Agnidhra. VOL. IL Z

354 ALBERUNI'S INDIA.

Yuktasa and Jita are taken from the following verse- yuktas-tatha-jitas-cânyo manuputran atah śriņu. P. 395 .- Valakhilyas are known as pigmy sages from the Vishnu-Purdna, but I do not find there this story of them and Satakratu.

P. 396. Bali, the son of Virocana, and his Vazir Venus, i.e. Sukra-Vide Vishnu-Purana, iii. p. 19, note. There is a Hindu festival called after him Balirajya; v. ii. 182.

ii. p. 31. P. 397. Vishnu-Purana .- This qnotation is found III.

III. iii. p. 33- P. 398 .- The second quotation from Vishnu-Purana is

Kali, the son of Jasho (?) .- Vide note to p. 382.

P. 398 .- The names of tha Vyasas of the twenty-nine Dvâpara-yugas have been taken from Vishnu-Purdna, III. iii. pp. 34-37. The anthor'a tradition differs a little from the Sanskrit text, in so far as he does not always combine the same Vyasa with the same Dvapara, particularly towards the end of the list. The names agree in both traditions, except Trivrishan, for which the Arabic has something like Trivarta or Trivritta. Besides, in the word Rinajyeshtha (in Arabic Rinajertu) the anthor has made a mistake. The Sanskrit verse runs thus- kritamjayah saptadaśe riņajyoshtádaśe smritah. Alberuni has read rinajyeshtoshtadase instead of rinajycsh- tadase, and has wrongly divided these words into rinaj- yeshto-ashtadase instead of rinajyo ashtadase. Further, he has been guided by the analogy of jyaishtha (the name of the month), which in vernacular was pronounced jertu, in changing rinajyeshta into rinajertu.

P. 398. Vishnu-Dharma-In mentioning Vasudeva, Sarhkarshana, &c., as the namea of Vishnu in the yugas, this source agrees with the teaching of the sect of the Bhagavatas or Pancaratras .- Vide Colebrooke, " Essays," i. 439, 440.

ANNOTATIONS. 355

P.401 .- The story of the birth of Vasudeva, i.c. Krishna, is related in the Vishnu-Purana, book v. chap. iii.

P. 403. The children of Kaurava, &e .- The following traditions are taken from the Mahabharata: the dice- playing from book ii, or sabhaparvan; the preparing for battie from book v., or udyogaparvan; the destruction of the five brothers by the curse of the Brahmin from book xvi., or mausalaparvan; their going to heaven from book xvii., or mahaprasthanikaparvan. The introductory sentence of this relation, y,t os literally, "The children of Kaurava ,كورو على بنى العمومة were over their cousins," is odd, and perhaps not free from a lacuna. Pandu had died, and his children grow up in Hastinapura, at the court of Kaurava, ie. Dhrita- rashtra, their nacle, the brother of Pandu. One expects a sentence like " The children of Kaurava cherished enmity against their cousins," but as the Arabic words run, one could scarcely translate them otherwise than I have done. The children of Kaurava had " the charge of their cousins," &c.

P. 407. On the akshauhint cf. H. H. Wilson, " Works," 2d edit., iv. p. 290 (on the art of war as known to the Hindas). Mankalus seems to be a mistake for Myrtilus. Cf. Eratosthenis Catasterismorum Reliquia, rec. C. Robert, p. 104. The source of Alberuni aeems to have been a book like the chronicle of Johannes Malalas, The accond tradition, taken from a commentary on Aratna' Phanomena (vide note to p. 97), is found in the same book, Eratosthenis, de., p. 100, 98. For thia informa- tion I am indebted to my colleague, Professor C. Robert.

P. 408 .- The nomber 284,323 of people who ride on chariots and elephants is a mistake for 284,310. I do not see what is the origin of thia surplus of 13 men. However, the wrong number must be kept as it is, since the author reckons with it in the following computation.

ANNOTATIONS.

VOL II.

P. 1 .- The famous chronological chapter xlix. consists of two parts of very different value. Part i., on p. 2-5, an explanation of the mythical eras of the Hindns, is taken from the Vishnu-Dharma, on which work ef. note to i. p. 54 Part ii., on p. 5-14, containing information of a historical character, has not been drawn from a literary source. If the author had learned these things from any particn- lar book or author, he would have said so. His infor- mation is partly what edncated people among Hindus believed to be historic and had told him, partly what he had himself observed during his stay among Hindus and elsewhere. That their historic tradition does not deserve much credit is matter of complaint on the part of the author (on pp. 10, 11), and that altogether the description of historic chronology, as far as he was able to give it, is by no means in all points satisfactory, is frankly admitted by the author himself (on p. 9). Whatever blame or praise, therefore, attaches to this chapter must in the first instance be laid to the charge, not of Alberuni, but of his informants. What he tells us is to be considered as the vulgata among educated Hindus in the north-west of India in his time. Although the tales which had been told Alberuni may not have been of a high standard, still it is much to be regretted that he has not chosen to incorporate them into his Indica (cf. p. II, I-6). Whether his hope (expressed on p. 8), that he might some day learn something more of this subject, was realised

358 ALBERUNI'S INDIA.

or not, I cannot make out. However, the stray notes on, Indian chronology scattered throngh his Canon Masudicus, which he wrote some years after the Indica, do not seem to betray that his Indian studies had made much progress. In all researches on Indian chronology, Alberuni's state- ments play an eminent part, specially those relating to the epochs of the Saka and Gupta eras. Cf. among others the following publications :- Fergusson, "On Indian Chronology," " Joornal of the Royal Asiatic Society," vol. iv. (1870), p. 81; and "On the Saka, Samvat, and Gopta Eras," vol. xii. (1880), p. 259- E. Thomas, "The Epoch of the Guptas," ibid, vol. xiii. (1881), p. 524- Oldenberg, "On the Dates of Ancient Indian Inscriptions and Coins," "Indian Antiquary," 1881, p. 213. Fleet, "The Epoch of the Gupta Era," ibid., 1886, p. 189. Drouin, "Chronologie et Namismatique des Rois Indo-Scythes," iu "Revue Numismatique " 1888, premier trimestre, pp. 8 seq. M. Muller, " India, What can it teach us ?" pp. 281, 286, 291. P. 2 .- As the author had to compare a number of dif- ferent eras with each other, he stood in need of a common standard to which to reduce all of them, and for this pur- pose he chose the New-Year's Day or first Caitra of the year 953 of the Saka era, which corresponds to- (1.) A.D. 1031, 25th February, a Thursday. (2.) A. Hijrae 422, 28th Safar. (3.) A. Persarum 399, 19th Ispandarmadh-Mah. The Nanroz or New-Year's Day of the Persian year 400 fell on 9th March 1031 A.D., which is the day 2,097,686 of the Julian period (Schram).

P. 2, 1. 30 .- This refers to the year of the kaliyuga 3600, as there have elapsed ro divya years or 3600 years of the present yuga. On the next page Alberuni makes the calculation for the gange-year, or the year 4132 of the kaliyuga A kalpa being a day of Brahman, 8 years, 5 months, 4 days correspond to 8x720+5x60+4x2, or 6068 kalpas, or 26,213,760,000,000 years. Of the present kalpa there have elapsed six manvantaras or 1,840,320,000 years, seven samdhis or 12,096,000 years, twenty-seven caturyugas or 116,640,000 years, the kritayuga or 1,728,000 years, the tretayuga or 1,296,000 years, the dvaparayuga or 864,000 years, and of the kaliyuga 4132 years; so altogether of the seventh manvantara 120,532,132 years,

ANNOTATIONS. 359

of the kalpa 1,972,948,132 years, and of Brahman's lifé 26,215,732,948,132 years, as stated p. 3, ll. 6-9 (Schram).

P. 3. It was I who told it to Yudhishthira, &c .- The author of Vishnu-Dharma refers in these words to the third parvan (vanaparvan) of the Mahabharata.

P. 4, 1. 29 .- From the beginning of Brahman's life to that of the present kalpa there have elapsed 6068 kalpas or 6068 x 1008 X 4,320,000 or 26,423,470,080,000 years. Six manvantaras=6x 72 x 4,320,000 or 1,866,240,000 years; twenty-seven cataryugas = 27 X 4,320,000 or 116,640,000 years; three yugas+4132 years=3 X 1,080,000+4132 or 3,244,132 years. The latter number represents the years elapsed of the caturynga; adding to it successively the other numbers of years, we find the numbers given ll. 29-31 of this page. The Arabic manuscript has 26,425,456,200,000 instead of 26,425 456,204,132 (Schram).

• P. 6, 1. 3 .- In the book Srudhava, &c., cf. note to i. p. 158. Candrabija-I first took the reading of the manuscript to be xhus, bnt now I believe I can see a pale dot above the last consonant, so that we may read yeldg -. On the shashtyabda, or sixty-years cycle, ef. chap. lxii. p. 123.

P. 6. The epoch of the era of Saka, &c .- Alberuni speaks of this era in his Canon Masudicus (composed daring the reign of Mas'ud) in the following terms: d't & tl هو كال واخهر التواريم عندهم وخامة عند سجميهم مككال أى وقت سق ويحسب من سنة هلاكه لاتّه كان متعلبا عليه والرسم فيه وفى غيرة ان , Beginning of the sixth chapter ).تذكر سنيه التامة دون الناقمة

Maseam.) book i., copied from the Codex Elliot, now in the British

Translation : " Time is called Kala in the language of the Hindus. The cra most famous among them, and in particular among their astronomers, is the Sakakala, ie. the time of Saka. This era is reckoned from the year of his destruction, hecause he was ruling (rather, tyrannising) over it (i.e. over that time). In this as well as in other

360 ALBERUNI'S INDIA.

eras it is the eustom to reckon only with complete, not with incomplete or current yearg" Then the anthor goes on to give rules for the comparison of the Saka era with the Greek, Persian, and Muslim eras. A later anthor, 'Abt-Sa'id 'Abd-alhayy Ibn Aldabhak Ibn Mahmud Gardezi (Gardez, a town east of Ghazne), has reprodnced the information of Alberuni on the Saka era in Persian. Not having the original (MS. Onseley 240, Bodleian Library, Oxford) at my disposal, I give a trans- lation made years ago :- "The Hindn era is called JU, because JU (kdla) meana time, and ule (Saka) is the name of a king whose death was made an era; he did the Hindna a great deal of harm, so they made the date of his death a festival" (Oxford manuscript, p. 352). The place Kardr is also mentioned in the Chachnama. Vide Elliot, " History of India," i. 139, 143, 207.

P. 7. Al-arkand .- Cf. note to i. 312. The book does not seem to exist in the collectiona of Arabic manuscripts in Europe.

P. 8 .- The pronnnciation of the names Kanir, Bardart, Marigala, and Nirahara (Nira-griha ?) is more or less con- jectural. Alberani identifies Marigala with Takshasila (voLl ii. 302), i.e. the Taxila of the ancients. The name Marigala seems to be preserved in that of a range of hills lying only two miles to the south of Shahdhesi (Cunningham, " Ancient Geography of India," p. 111). The place is also mentioned in the Tabakati-Naşirt. Vide Elliot, "History of India," ii. 271, 273.

P. 9 .- Durlabha, a native of Multan, is only twice men- tioned. Here the anthor quotes from him a method for the compntation of the Saka era, and p. 54 & method for the compntation of ahargana. According to him, the Indian year commenced with the month Margasirsha, bnt the astronomers of Multan commenced it with Caitra (p. I0).

P. 10. Barhatakin .- The name occurs only in this one

ANNOTATIONS. 36r place. If it were an Indian name, I should think of some- thing like Vrihatkina (or Vrihatkelu ees). If it is Turkish, it is a compound, the second part of which is tagin (as in Toghrultagin and similar names). As the author declares the dynasty to be of Tibetan origin, the question is whether the name may be explained as Tibetan.

P. 10. Var .- As the Arabic verb may be connected either with the preposition bi or with the accusative, we may read either bar or or.

P. 10, 1. 25. He began to creep out .- In the Arabic text, p. t.y, 8, read as Jel instead of aee dol.

P. 11. Kanik .- Only the three consonants KNK are certain. We may read them Kanik or Kanikku, which would be a Middle-Indian Kanikkhu for Sanskritic Kan- ishka. Thus the name Turk was prononneed by the Middle-Indian tongue as Turukkhu, and Sanscritisized as Turuskha. This Zopyrus-story was reproduced by Muhammad 'Aufi. Cf. Elliot, "History of India," ii. 170.

P. 13. Lagaturman .- The unconth formation of this name seems to point to a Non-Indian (Tibetan ?) origin. I at first thought to combine it with the name of the Tibetan king, Langtarma, who abolished Buddhism, A.D. 899 (v. Prinsep, " Useful Tables," ii. 289), as our Lagatûr- man was the last of a series of Buddhistie kings, and as the names resemble each other to some extent. However, this combination seems delusive. The name Kallar is written Kallr . Could this name be combined with Kulusha (Kalusha ?), which e.g. occurs as the name of the Brahmin minister of the Mahratta Raja Sambaji ?

P. 13, 1. 17. The Brahman kings .- The word samanta means vassal. Kamalu was a contemporary of the prince 'Amr Ibn Laith, who died A.D. 911. Cf. Elliot, " History of India," ii. 172. Is the name a hypokoristikon of one like Kamalav- ardhana ?

362 ALBERUNI'S INDIA.

Anandapdla, Bhimapala, and Trilocanapala mean hav- ing Siva as protector. If, therefore, these princes, like the Indo-Scythian kings (ef. Drouin, Revue Numismatique, 1888, 48), were Siva-worshippers, we must explain the name Jaipal perhaps as Jayapala, i.e. having Durgd (the wife of Siva) as protector. Cf. the Hindu kings of Kabul in Elliot, "History of India," ii. 403 seq. (in many points antiqnated). The name Trilocanapala (here Taracanpal) has been mnch disfigured in the Arabic writing. Vide the Paru Jaipal in Elliot, Le., ii. 47, 463, 464

P. 13,1 14. The latter was killed .- The Arabic manu- script has Js, which may be read J' (narratum est) or J3 (interfectus est). I have not been able to ascertain whether the year in question was that of the enthronisa- tion of Trilocanapala, or that of his death. I prefer, how- ever (with Reinand), to read J-3, " he was killed," becanse evidently the author stood so near to the events in ques- tion that he could have ample and trustworthy informa- tion, and that, in fact, an on dit (Jes) seems here entirely out of place.

P. 13, L 22. The slightest remnant, literally one blow- ing fire, a well-known simile for nobody. Cf. e.g. Hasan Nizami in Elliot's "History of India," iL 235, L 13.

P. 15 .- For Alfazari and Yakub Ibn Tarik, ef. note to i. 165, 169. Muhammad Ibn Ishak of Sarakhs is mentianed only here and in the tables on pp. 16 and 18, besides in Alberuni's " Chronology " (English edition, p. 29).

P. 16, 1. 6 of the table .- It is not clearly ssid in the text that the anomalistic revolution is meant, but the numbers which Alberuni quotes leave no doubt on the subject. The days of a kalpa are 1,577,916,450,000, which being divided by the number 57,265,194,142, give for one revolution 2731181108144 ITiif days, or 27 days 13 h. 18 min. 33 sec., whilst the anomalistic revolution of the moon is equivalent to 27 days 13 h. 18 min. 37 sec., an agreement so very close, that every doubt that there could be meant

ANNOTATIONS. 363

anything bnt the anomalistic revolution is completely ex- cloded. Moreover, the number of the revolutions of the apsis, 488,105,858, being angmented by 57,265,194,142, is equal to 57,753,300,000, the number of sidereal revolu- tions ; and, indeed, the revolutions of the apsis, plus the anomalistic revolutions, must be equal to the sidereal re- volutions (Schram).

P. 16 .- The note in the table " The anomalistic revolu- tion of the moon is here treated," &c., is not quite clear, and probably materially incorrect. That the term jill dole means the anomaly (avopaMia in Greek, kendra (kevrpov) in Sanskrit), was first pointed out to me by my friend and colleague, Prof, Förster; but this note, which seems to be intended as a sort of explanation of the term, does not exactly render what astronomers understand by anomaly. Literally translated it runs thus: "The Hassat-alkamar stands in the place of the apsis, because the result is its (whose ? the apsis'?) share, since it (the haşsat-alkamar) is the difference between the two motions " (o a W es Accordingly, we .(ستة ال (أو not) هى ففل ما بين المركتين must translate the term as " falling to the moon as her lot or abare," viz., movement, in Arabic yal &ln as, n. Therefore, in the Arabic text, pp. ". and. Y, 8 write Lla intead of &l.

P. 19 .- Abu-alhasan of Ahwaz is mentioned only in this place. He seems to have been a contemporary of Alfazart and Ya'kub Ibn Țarik.

P. 20 .- Annus procrastinationis .- Vide the author's " Chronology" (English edition), p. 73. Malamasa, iu Hindustani malmas. Vide Dowson, " Hindustani Gram- mar," p. 258.

P. 21, 1. 24 .- A caturyuga or 4,320,000 solar years con- sists of 53,433,300 lunar months or 1,602,999,000 lunar days; so one solar year has 371-86 lunar days, and the difference between the solar and lunar days of a year is 1126. The proportion 360 :lunar days: 11-6 days =x lunar days: 30 days gives for x the number of

364 ALBERUNI'S INDIA.

97611, which is equivalent to 9761779T. 1. 23 (Schram). Vide p. 24,

P. 22, 1. 17 .- Read 22" instead of 23" (Schram).

P. 23. Padamdsa .- This seems to be an old mistake which has crept into the Arabic manuscripts of the works of Alfazari and Ya'kub. Cf. the anthor's "Chronology" (English edition), p. 15.

P. 27 .- The rule given in the first fifteen lines of this page is completely erroneous, and consequently the ex- ample calculated after this rule is so too. The right method would be the following :- " The complete years are multiplied by 12; to the product are added the months which have elapsed of the cnrrent year. The sum repre- sents the partial solar months. You write down the number in two places; in the one place you multiply it by 5311, i.c. the number which represents the universal adhimasa months. The product yon divide by 172,800, i.c. the number which represents the universal solar months. The qnotient you get, as far as it contains com- plete months, is added to the number in the second place, and the sum so obtained is multiplied by 30; to the pro- dnct are added the days which have elapsed of the current month. The sum represents the candrahargana, ie. the sum of the partial lunar days" These two proceedings would be identical, if we were not to omit fractions; but as an adhimâsa month is only intercalated when it is com- plete, we must first determine the number of adhimasa months, and, omitting the fractions, change them to days; whilst when we multiply beforehand by 30, the fractions of the adhimasa months are also multiplied, which is not correct. This is at once seen in the example which he works out after this rule, and we wonder that Albe- runi himself did not see it. He is calculating the ahar- ganas for the beginning of a year, conseqnently also for the beginning of a month, and, notwithstanding, he is not at all surprised to find (p. 30) 28 days and 51 minutes of the month already passed. The adhimasa days are nothing else than adhimasa months converted into days. As the number of the adhi-

ANNOTATIONS. 365 masa months must be a whole, so the number of the adhimasa days must be divisible by 30. Accordingly, the nnmber quoted, p. 29, 1, 30, not being divisible by 30, is at once recognised as erroneous, and it is astonishing when he says in the followving lines, " lf, in multiplying and dividing, we had used the months, we should have found the adhimasa months and multiplied by 30, they would be equal to the here-mentioned number of adhimâsa days." In this case certainly the number ought to be divisible by 30. Perhaps he would have found the fault, if not, by a strange coincidence, the difference between the true value and the false one had been exactly 28 days or four complete weeks, so that though the number con- sidered is an erroneous one, yet he 'finds, p. 30, l. 9, the right week-day. Alberuni finds, p. 29, l. 2, as the sum of days from 'the beginning of the kalpa to the seventh manvantara 676,610,573.760. Further, he finds, l. 7, that from the beginniug of the seventh manvantara till the beginning of the present caturyuga there have elapsed 42,603.744,150 days, and, l. 12, that till the beginning of the kaliyuga tlere have elapsed 1,420,124,805 days of the present catur- yuga Adding these numbers, we find that the sum of days elapsed from the beginning of the kalpa to that of the catnrynga is 720,634,442,715; but as he finds, p. 30, 1. 5, that from the same epoch to the gauge-date there have elapsed 720,635,951,963 days, so the gauge-date would be 1,509,248 days after the beginning of the kaliyuga. Now we know that the gauge-date is 25th February 1031 (see p. 2, l. 17, and note), or the day 2,097,686 of the Julian period, whilst the first day of the kaliyuga, as is generally known, coincides with the 18th February 3102 before Christ or with the day 588,466 of the Julian period, so that the difference of the two dates is 1,509,220, and not 1,509,248 days. To this result we shall also come when working out Alberuni's example after the method stated in the begin- ning of this note. Instead of p. 29, L 16, we should then have: the years which have elapsed of the kalpa up to that year are 1,972,948,132. Multiplying them by 12, we get as the number of their months 23,675,377,584 In the date which we have adopted as gauge-year there is

366 ALBERUNI'S INDIA.

no month, but only complete years; therefore we have nothing to add to this number. It represents the par- tial solar months. We multiply it by 5311 and divide the product by 172,800; the quotient 727,661,6333888 represents the adhimass months Omitting the frac- tions, we add 727,661,633 to the partial solar months 23,675,377,584, and get 24,403,039,217 as the partial lunar months. By multiplying this number by 30 we get days, viz., 732,091,176,510. As there are no days in the normal date, we have no days to add to this number. Multiplying it by 55,739 and dividing the pro- duct by 3,562,220, we get the partial finarâtra days, viz., 11,455,224,575338888. This sum of days without the fraction is subtracted from the partial lunar days, and the remainder, 720,635,951,935, represents the number of the civil days of our gauge-date. Dividing it by 7, we get as remainder 4, which means that the last of these days is a Wednesday. Therefore the Indian year commences with a Thursday. The difference between 720,635,951,935 and the beginning of the kaliyuga 720,634,442,715 is, as it onght to be, 1,509,220 days (Schram). In the beginning of chap. lii., in the Arabic text, nl1, 8, it seems necessary to write yse and yett instead of et and .اليّام P. 29, 1. I0. Thursday .- The Arabic manuscript has Tuesday.

P. 30, 1. 10-17 .- This onght to run as follows :- We have found above 727,661,6333118 for the adhimasa months; the wholes represent the nnmber of the adhimasas which have elapsed, viz., 727,661,633, whilst the fraction is the time which has already elapsed of the current adhimasa month. By multiplying this fraction by 30 we get it expressed in days, viz., 3443 days, or 28 days 51 minntes 30 seconds, so that the current adhimass month wants only I day 8 minutes 30 seconds more to become a complete month (Schram). P. 31, L 19 .- The number 1,203,783,270 is fonnd by adding the 30x 1,196,525 or 35.895,750 adhimasa days to the 1,167,887,520 solar days (Schram).

ANNOTATIONS. 367

P. 31, 1. 24-The number of days from the beginning of the caturyega to the gauge-date is here found by Pulisa'a method to be 1,184,947,570, whilst p. 33, 1. 16, the number of days from the beginning of the caturyuga to that of the kaliyuga is found to be 1,183.438,350. The difference between both numbers is (as it ought to be) 1,509,220 days (Schram).

P. 33, 1. 24-The method of Âryabhata is the same as that given before, only the numbers by which we are to multiply and to divide,are different according to his system, which aupposes a different number of revolutions in a kalpa. According to Âryabhata the elder, a caturynga has 1,577,917,500 days (see vol. i. p. 370, 1. 28). As to the revolutions of sun and moon, they seem to be the same as given by Pulisa. The tables, pages 16 and 17, are not quite correct in this, as they give, for instance, for the revolutions of the moon's node and apsis the 1oooth part of their revolutions in a kalpa, whilst in vol. i: p. 370, 1. 16, it is said that, according to Pulisa and Âryabhata, the kalpa has 1008 catnryugas. But p. 19, 1. 15, the numbers 4,320,000 for the sun and 57.753,336 for the moon are given as possibly belonging to the theory of Aryabhata. The same numbers are cited by Beutley in his " Historical View of the Hindu Astronomy," London, 1825, p. 179, as belonging to the system of the so-called spurions Arya Siddhanta, It is doubtless the same system, for if we compare the number of days between the beginning of the kalpa and that of the kaliyuga, which Bentley atates in the above-cited book, p. 181, to be 725,447,570,625, witl the aame sum quoted by Alberuni, p. 33, 1. 29, there can scarcely be a doubt as to the identity of both systems, especially as this number 725,447,570,625 is a curious one, giving Thursday for the first day of the kalpa, whilst the other systema give Sunday for this date. Of this book Bentley says, p. 183: " It would be needless to waste any more time in going over its contents; what has been shown must be perfectly sufficient to convince any man of common sense of its being a downright modern for- gery ;" and p. 190, "The apurions Brahma Siddhanta, together with the spurious Arya Siddhanta, are doubtless the productions of the last century at farthest." Perhaps

368 ALBERUNI'S INDIA.

he would have chosen more reserved expressions, if he had known that this "production of the last century " was already cited by Alberuni. When we adopt these numbers for a caturyuga, ic. 1,577,917,500 civil days, 4,320,000 revolutions of the sun and 57,753,336 revolutions of the moon, and consequently 53.433,336 lunar months, we fiud the numbers belonging to a yuga by dividing the above numbers by four, as in this system the four yugas are of equal length Thus we get for a yuga 394.479,375 civil days, 1,080,000 solar years, and consequently 12,960,000 solar months, and 388,800,000solardays, 13,358,334lunar months,400,750,020 Innar days, 398,334 adhimasa months, and 6,270,645 fnaratra days. To find the number 725.449,079,845 men- tioned, p. 33, 1. 31, as the sum of days between the be- ginning of the kalpa and the gange-date, we are to proceed as follows :From the beginning of the kaliyuga to our gauge-date there have elapsed 4132 years, which multi- plied by 12 give 49,584 as the partial solar months. This number multiplied by the universal adhimasa months 398,334, and divided by the universal solar months 12,960,000, gives 152311821 as the number of adhimasa months. This number, without the fraction added to the solar months 49,584, gives 51,107 as the number of the partial lunar months, which multiplied by 30 gives 1,533,210 as the number of the partial lunar days. This number multiplied by the universal unaratra days6,270,645 and divided by the universal lunar days 400,750,020 gives 23,9905111114 as the sum of the partial unarâtra days; and 23,990 subtracted from the partial Innar daye 1,533,210 gives 1,509,220 as the civil days elapsed of the kaliyuga till the gauge-date, identical with the number found in note to p. 27. These 1,509,220 days added to the 725,447,570,625 days which separate the beginning of the kalpa and the kaliyuga, give the number of 725.449,079,845 days cited p. 33, L 31. Finally, the number of days elapsed of Brahman's life before the present kalpa, is got by multi- plying the number of days in a kalpa, i.e. 1,590,540,840,000 (see page 370, vol, i.) by 6068, the number of the kalpas elapsed before the present one (Schram).

P. 34, 1. 32 .- There is here the same fault as that which

ANNOTATIONS. 369. led Alberuni to a false result, p. 27. The multiplication by 30 must be made after dropping the fraction of the adhimasa months, not before (Schram).

P. 36, 1. 1 .- The lacuna must have contained a phrase like this :- " In three different places; they multiply the number in the lowest place by 77, and divide the product by 69,120." This follows clearly from the explanation which he gives in the following page (Schram).

P. 36, 1. 9 .- Read lunar instead of solar, in the Arabic .الشمسية instead of القمرية ,( last word ,7,٢٢٣)

P. 36, 1. 10 .- The expression is a very concise one, so that it is not quite clear what is meant (L 14) by the "middle number."-It is to be understood in the following manner: "This number of the partial lunar days is written down in two different places, one under the other. The one of these is "in the uppermost place" (1. 17); they multiply the lower number by II, and write the pro- duct under it. Then they divide it, ie. the product, by 403,963, and add the quotient to the middle number, i.e. to the product of eleven times the partial lunar daye (Schram).

P. 36, 1. 26 .- A certain number of months A is to be divided by 551895J. 1158 If we wish to get the same result by dividing only by 65, we must subtract from A a cer- tain number .Y which is to be determined by the equation A A-X 6515933 This equation gives for I the value 65 X = A 1168 or, reduced, I=A(1038800), or at last 1155 X=A(awTso). The equation X=A( 65 1165 can also be written in the form 657883s : Tro33-A: X, that is, as Alberuni states it (L 30), "the whole divisor stands in the same relation to its fractions as the divided number to the subtracted portion" (Schram).

P. 36,1. 33 .- Alberuni has not made the calculation given VOL. IL. 2 A

370 ALBERUNI'S INDIA.

above in a general way, but he has made it ouly for a special case, for the gauge-date. He finds the fraction wsreo, which he would find for every other date, as this fraction is independent of the number A (Schram).

P. 37, 1. 26 .- Here again a certain nomber of hnarâtra days A is to be divided by 6319958. If we wish to get the same result by dividing only by 63}9, or, which is the same, by T03 Tos, we must add to A a certain number X, which is determined by the equation

4+X orA +X= 703 635040 II x 6312443 or X=A(703-IIx 6350851 II x 6318948 703-7021491 or X=A 97 70251443 39184420

or at last, dividing numerator and denominator by 97, we find X= A 40396317 The are neglected (see p. 38, 1 9) (Schram).

P. 38,1. 25 .- The Arabic manuscript has 77,139, inatead of 7739, as Dr. Schram demands; v. p. 39, L 7, and p. 40, 1. 8.

P. 39, 1. 20 .- Here he grants that the 28 daye which we get over 727,661,633 months are to be reckoned after the beginning of the month Caitra, so that the result found, p. 29, l. 30, agrees with the 28th, not with the first Caitra (Sehram).

P. 39, 1. 24-The middle number was multiplied by 4181; a solar year has 3653581 days (1. 36), or 52 weeks I day and #181 of a day. By adding the product of the number of years multiplied by $$8; to this number itself, we get the sum of days by which these years exceed whole weeks. The rest of the calculation is sufficiently explained by Alberni himself (Schram).

P. 41, 1. 19-This is the same case as p. 36, only the numbers are a little different. If A is the number of months to be divided by 3238554, and we wish to sub- tract a number from A so as to get the same result by

ANNOTATIONS. 371

dividing the difference by 32 only, we have the equation A-X 3211111 32

which gives for X the value

or X= 35552 2160000 or X=2 500 Alberuni has again made the calculation for a special case, the gange-date, and found the same fraction (Schram).

P. 41, l. 20 .- "This number of days," viz., the number of solar days corresponding to the given date (Schram).

P. 41, 1. 33 .- The MS. has 974 instead of 976.

P. 42, 1. 3 .- The number of solar days, 1,555,222,000, is here taken as divisor instead of the number of adhimasa months, 1,593,336. The fraction onght to be 976-104064 =97688385, the common divisor 24 (Schram).

P. 42, 1. 6 .- Alberuni does not seem to have understood Pulisa's calculation which is correct, although there seems to be a lacuna in its explanation. According to Pulisa's theory, there are in a caturyuga 1,555,200,000 solar days and 1,593,336 adhimasa months. Dividing the first num- ber by the second, we get as the time within which an adhimasa month sums up 976 104044 393530 days. So one would get the number of adhimasa months by dividing the given number of solar days by the number 976-94994; Pulisa prefers not to reckon with the fraction, so he diminishes the number of given days by a certain amount and divides only by 976. The number which is to be subtracted from the given days is ensily found by the following equation :- Let D be the number of given solar days; we then have D D-X 104044 104046 976 or X= D 97610406 or X = D 108500000 or X=D(- 104064

Now 384 is a commou divisor to 104,064 and the divisor 1,555,200,000. So we get I = Drosuooo, just as Pulisa finds it (Schram).

372 ALBERUNI'S INDIA.

P. 42, 1. 22 .- Not only is it not " quite impossible that this Dumber shonld, in this part of the calculation, be used as a divisor," but it needs must be used as a divisor. This we see at once when, instead of working out the cal- culation with apecial numbers, we make it algebraically. Let S be the number of solar days in a caturyuga, and A the number of adhimasa months in a caturyuga. Then the number of days within which one adhimasa month sums up, will be found by dividing S by 4. By this division we ahall get wholes and a fraction; let the wholes be represented by Q and the numerator of the fraction hy R. We then have = Q+7 or S=AQ+R. Now if, the S R

given number of solar days being D, we have to divide D by Q+ to get the number of adhimasa months, but as we wish to divide by Q alone, we must subtract from D a number X, which will be found by the eqnation R or X = D R Q+

As 4Q +R is equal to S, we have X=D where S is the

number of solar days in a caturyuga, which must necessarily be a divisor in this part of the calculation (Schram). P. 42, 1. 31 .- As one finaratra day sums up in 6351188 lunar days (see p. 37, 1. 17), we have again the equation L-X or X=L( or X=L (nf)

where L represents the number of the given Innar days.

P. 44, 1. I .- The nnmber 720,635,951,963 is not correct, as we have seen in note to p. 27. It is too great by 28 days. But the number of adhimasa days, 21,829,849,018 (L 10), is also 28 daya too great. So the difference is again correct. There is the same fault as at p. 27. The calculation ought to run as follows :- The par- tial civil days which have elapsed up to our gauge-date are 720,635,951,935. This number is given, and what we

ANNOTATIONS. 373

want to find is how many Indian years and months are equal to this sum of days. First we multiply the num- ber by 55,739 and divide the product by 3,506,481; the quotient is 11,455,224,575118128f tnaratra days. Weadd I1,455,224,575 to the civil days; the sum is 732,091,176,510 lunar days. Dividing this number by 30, we get as quo- tient 24,403,039,217 lunar months (and no fraction; so we see that the date in question consists of a nnmber of months only, or, what is the aame, that the date corre- sponds to the beginning of a month). Multiplying the lunar months by 5311 and dividing the product by 178,111, we get 727,661,63311:21 adhimasa months; 727.661,633 adhimasa months subtracted from the 24,403,039,217 lunar months give 23,675.377,584 solar months, which divided by 12 give 1,972,948,132 years and no fraction. So we find the given date corresponding not only to the beginning of a month, but also to that of a year. We find the same number of years of which the gange-date consists (see p. 29, l. 17) (Schram).

P. 45, 1. 12 .- Thia rule must indeed be based on some complete misunderstanding, for it is absolutely ( roneous, as Alberuni rightly remarks (Schram).

P. 46, 1. I .- If we calculate from the beginning of the kalpa or the caturyuga, there are in the epoch neither fractions of the adhimasa months nor of unaratra days; but as the great number of days embraced by such long periods makes the calculation wearisome, the methods set forth in this chapter start neither from the beginning of the kalpa nor from that of the caturyuga, but from dates chosen arbitrarily and nearer to the time for which they are to be employed. As such epochs are not free from fractions of the adhimasa months and finaratra days, these fractions must be taken into account (Schram).

P. 46, 1. 27 .- The nnmbers employed here do not belong to Brahmagupta's, but to Pulisa's system. The year taken as epoch is the year 587 Sakakala. As we have seen, p. 31. IL. 8-I0, that in the moment of the beginniog of our gauge-date or of the year Sakakala 953, there have elapsed 3.244,132 years of the caturyuga, there must have elapsed

374 ALBERUNTS INDIA.

3.243,766 years of the caturyuga till the beginning of the year 587 Sakakala. We must now first calculste the adhimasa months and fnaratra days for this epoch. After Pulisa's method (p. 41, 1. 29), we have: 3,243,766 years are equal to 38,925,192 solar months or 1,167,755,760 solar days. This number multiplied by 271 and divided by 4,050,000 gives 78,1381815. As here the nearest num- ber is to be taken, we get 78,139, which, subtracted from 1,167,755,760, gives 1,167,677,621. This latter number divided by 976 gives as the number of sdhimasa months 1,196,39Iris. Now 1,196,391 adhimasa months are equal to 35,891,730 adhimasa days, which, added to 1,167,755.760 solar days, give 1,203,647,490 lunar days. According to Pulisa's theory (see p. 26, L 9), there are in a caturyuga 1,603,000,080 lunar and 25,082,280 unaratra days; so one fnaratra day sums np in 6313318 lnnar days. Therefore we shonld have to divide the given namber of lunar days Z by 63:911;, bnt we prefer to snbtract from L a certain number X, and to divide the rest by 6310 or T05. The nnmber X will be given by the equation L L-X 703 II L-IIA This equation gives for X the valne X- 703 39073 Lor X= 439 48980558 Z or X I 11 L. 111573789 or nearly I1 X= Now I being equal to 1,203,647,490 lunar days, I1 I will 111573*

be equal to 13,240,122,390 lunar days; this number di- vided by 111,573 gives 118,6671T1f7s. Taking the nearest number, we subtract 118,668 from 13,240,122,390 and get . 13,240,003,722, which divided by 703 gives 18,833,57518I as the number of fnaratra days. This added to the 1,203,647,490 lunar days gives for the date of our epoch the number of civil days 1,184,813,915. This number divided by 7 gives 5 as remainder. Now the last day before the present catnrynga was a Mon- day (see p. 33, 1. 11), therefore the last day before our epoch is a Saturday, and any number of days elapsed since that epoch if divided by 7 will indicate by the remainder, the week-day counted from Snnday as t, as it is said, p. 47, 1. 19. Now the whole method is easily recognised

ANNOTATIONS. 375 as thoroughly correct. Instead of multiplying the partial solar daya by zuttlou, we multiply them by rztzr, which is aufficiently correct, as Tuidivo is equal to I

Aa besides the whole adhimasa months there is yet a I4944379

fraction of wis adhimasa months in our epoch, we add 5 before dividing by 976. The calculation of the unarâtra days has already been explained; but as in our epoch besides the whole unaratra daya there is still a fraction of ##f unarâtra days, we must add 497 before dividing by 703. The whole proceeding is thus explained (Schram).

P. 48, 1. 1I .- The calculation has been made for the complete years elapsed before our gange-date. So we get the week-day of the last day before the first Caitra of the gauge-date, and if this is a Wednesday, the first Caitra itself is a Thursday ; ef. p. 30, 1. 9. The first day of this epoch corresponds to the day 1,964,031 of the Julian period. Adding 133,655 to 1,964,031, we have for the first Caitra 953 the day 2,097,686 of the Julian period, as it ought to be (Schram).

P. 48, 1. 21 .- The 18th Isfandarmadh of Yazdajird 399 corresponds in fact to Wednesday, 24th February 1031, the day before the first Caitra 953 Sakakâla (see note to p. 2, L 17) (Schram).

P. 49, 1. 22. By six years-The Arabic manuscript has seven instead of six.

P. 50, 1. I .- The method here employed is based on Pulisa'a theory. According to this theory, the solar days must be divided by 976 ta1 to get the adhimasa months. Now 9766855U 386 with aufficient accuracy is equal to 9762 of -s0 If S represents the number of solar months, the solar days or 30 S are to be divided by 41382 same, 900 S must be divided by 29282. 3o , or, what is the

To get the fnaratra days, the lunar days must be divided by 63:1819 (see note to p. 46, 1. 27). Now 638337 is equal to II t, or with sufficient accuracy 103300, II

376 ALBERUNIS INDIA.

or at least equal to 210002 oo . So the multiplications and divisions of this method are explained. The constant numbers which are to be added, are in- herent to the epoch. The year 888 Sakakala corresponds to the year 3,244,067 of the caturyuga; 3,244,067 years are equal to 38,928,804 so.ar months, or 1,167,864,120 solar days. These solar months multiplied by 66,389 and divided by 2,160,000 give 1,196,50218688, adhimasa months, or 35,895,060 adhimasa days. This added to the 1,167,864,120 solar days gives 1,203,759,180 lunar days. Eleven times this number is equal to 13,241,350,980; this latter number divided by 111,573 gives 118678 80444 or the nearest number 118,679. Subtracting this from 13,241,350,980, the remainder is 13,241,232,301, which being divided by 703, gives 18,835,323783 tnsratra days; these days subtracted from the lunar days give for the number of civil days 1,184923,857. Dividing this last number by 7, we get the remainder 5; and as the last day before the present caturyuga was s Monday (see p. 33, L II), the last day before the epoch here adopted is a Saturday, so that any number of days elapsed since that epoch, if divided by 7, will indicate by the remainder the week-day counted from Sunday as I. The first day of this epoch corresponds to the day 2,073,973 of the Julian period. We have found in our epoch the fraction of adhimasa month ritiio, which is equal to or very nearly Tprs ir adhimasa month, so we must add 29282

661 before dividing by 29282. The fraction of unaratra days #3 is equal to 69,600,41

or nearly to $10T0s motel. Therefore we must add 69,601 before 210903

dividing by 210,902. Alberuni has, instead of this num ber 69,601, the number 64,106, 4 instead of 9, and the last three numbers reversed (Schram).

P. 50, 1. 35 .- We had 780 months; adding thereto the 23 adhimasa months, we have 803 months, which being multiplied by 30 give 24090, and not 24060 daya All the following faults are the consequences of this one (Schram).

P. 51, 1. 2 .- It ought to be "adding thereto 69,601, we

ANNOTATIONS. 377 get the snm 79,566,601. By dividing it by 210,902, we get the qnotient 377, ic. finaratra days, and a remainder of Trosos, Le, the avamas" (In the Arabic text, p. Wy, 17. the reading of the MS, onght not to have been altered.) The correct result is 23,713 civil days. If we divide this number by 7, we find the remainder 4, which shows again that the last day before our gange - date is a Wednesday. By adding 23,713 to 2,073,973, we get for the first Caitra 953 the day 2,097,686 of the Julian period, as it ought to be (Schram).

P. 51, 1. 4 .- Read 377, instead of 307.

P. 51, 1. 9-This method works with numbers much less accurate than the preceding ones. It is assumed that one adhimasa month sums up in 32; solar months. So the solar months are divided by 32+ or by 228, or, what is the same, they are multiplied by Is. For the time within which an naratra day sums up, there is simply taken 631°, and the lunar days are divided by 6319 or To3, or, what is the same, multiplied by 7. The epoch corre- sponds to the year 427 Sakakala, or the year 3,243,606 of the caturynga. This number of years is equal to $8,923,272 solar months, which, multiplied by 66,389 and divided by 2,160,000, give 1,196,33138388 adhimasa months. The anthor has taken 1,196,332 adhimasa months and neglected the little fraction 38885, so that he has no fractions of adhimasa months. These 1,196,332 adhimasa months added to the 38,923,272 solar months give 40,119,604 lunar months or 1,203,588,120 lunar days. Multiplying by I1, we have 13,239,469,320, which divided by 111,573 gives 118,661191491 or 118,662. Subtracting this from 13,239,469,320, we have 13,239,350,658, which divided by 703 gives 18,832,646589 for the number of Anarâtra days. So the fraction of unarâtra days is 420 very near to that adopted by the author of the method, viz., fif. By subtracting the unaratra days from the lunar days we get as the number of civil days 1,184,755,474, which is divisible by 7. So, as the last day before the caturyuga was Monday, the last day before this epoch is also Monday, and the number of days elapsed since this epoch if divided by 7, will give a remainder which indicates the week-day,

378 ALBERUNPS INDIA.

counting Tuesday as 1. The first day of this epoch corre- aponds to the day 1,905,590 of the Julian period (Schram).

P. 51, 1. 24-It is easily understood why this method is called that of the Siddhanta of the Greeks. It is assumed that an adhimasa month sums up in 32 or 2 $e solar months. Now 22e solar months are equal to 4 solar years, There- fore this method is apparently an application of the cycle of nineteen years of the Greeks (Schram).

P. 52, I. 2 .- 32 months 17 days 8 ghati and 34 cashaka are only another expression for 32} months (Schram).

P. 52, 1. 10 .- The number of civil days is 192096; dividing by 7, we have as remainder 2. As in this method (see note to p. 51, 1. 9) Tnesday is to be reckoned as 1, this gives for the last day before our gauge date Wednes- day. Adding 192,096 to 1,905,590, we get as the first Caitra 953 the day 2,097,686 of the Julian period, as it ought to be (Schram).

P. 52, 1. 20. Al-harkan .- This book is mentioned only in this passage. The anthor calls it a canon, etf, i.c. a collection of astronomical, chronological, and astrological. tables and calculations. Whether it was an original com- position in Arabic or translated from Sanskrit, and from what original, we do not learn from him. The word seems to be an Arabio rendering of ahargana. Alberuni quotes irom this book the computation of an era the epoch of which falis 40,081 days later than that of the Persian era, and compares it with the gange-date (p. 53).

P. 52, I. 22 .- If the epoch should fall 40,081 days after that of the era Yazdajird, it would fall on the first Caitra of the year 664 Sakakala; but this is not the case. The first of Sha'ban of the year 197 coincides with the begin- ning of Vaisakha 735. As there are 72 years to be sub- tracted, we should come to Vaisakha 663, and to begin with the beginning of a year, the epoch must be postponed to Caitra 664. But this is of no importance, as we shall see that Alberuni altogether misunderstood the method here given (Schram).

ANNOTATIONS. 379 P. 52, 1. 24 .- These two dates do not agree te a day. The first Ferwerdinmah Yazdajird coincides with 16th June 632; 40,081 days later was Monday, 12th March 742, whilst the 21st Daimah of the year 11o of Yazdaiird corresponds to Sunday, 1Ith March 742. Bnt as the date itself is erroneous, this is of no importance (Schram).

P. 52, I. 27 .- As the numbers which form multiplica- tions and divisions in this method are identical with those of the Panca Siddhantika (p. 51), we can reckon the con- stants by the directions there given. The epoch of the method of Al-harkan is the beginning of Sha ban of the year 197. But this date corresponds to the beginning of Vaisakha 735 Sakakala. So we snould have for this date the following calculation : Snbtracting 427 from 735 years and I month, we get 308 years I montb, or 3697 months; 3697 mnltiplied by 7 and divided by 228 gives for the number of adhimasa months 113115; the 113 adhimasa months added to the 3697 solar months give 3810 lunar months or 114,300 lunar days. This number multiplied by II is 1,257,300; we add 514, which gives us 1,257,814; this divided by 703 gives for the number of fnaratra days 1789183- So we should have all the numbers wanted for our epoch if, in fact, this epoch were the true epoch. Fit we have io add 864 months to the interval. Therefore these 864 months, which must always be added, must first be subtracted from the epoch, so that this latter is thrown back by 72 years. Now 72 years or 864 solar months multiplied by 7 and divided by 228 give the number of 26119 adhimasa months. These together with the 864 solar months are 890 lunar months or 26,700 lunar days, which multiplied by 11 and divided by 703 give 41716% unaratra days. So we have to subtract from the numbers first fonnd 26138 adhimasa months and 417f69 Onaratra days. The number of adhimasa months inherent to our true epoch will then be 113225-26159= 86138, or with suffcient accuracy 87 without a fraction, and the number of unaratra days 1789183-417589= 1371983. Therefore no fraction is to be added to the adhimasa months, whilst to the naratra days there mnst be added $83, or nearly 11x28. Therefore we must add 28 (not 38) before multiplying by w. The 114,300 lunar

380 ALBERUNI'S INDIA.

days of the first epoch diminished by the 26,700 lunar days of the 72 years, give 87,600 lunar days. Snbtract- ing therefrom 1371 fnaratra days, we have 86,229 civil days, which being divided by 7 give as remainder 3. So the last day before this epoch is Thursday, and the number of days elapsed since the epoch of this method, if divided by 7, will give a remainder indicating the week-day, count- ing Friday as I. The first day of this epoch corresponds to the day 1,991,819 of the Julian period (Schram).

(Schram). P. 53, 1. 1 .- It must be 28, not 38 (see preceding note)

P. 53, 1.6 .- We must add 1, if we wish to have the week- day of the date itself, not that of the last day before it.

P. 53, 1. 8 .- Here Friday is considered as the first day of the week, not, as in the Indian books, Sunday. This ought to have been remarked (Schram).

P. 53, 1. 9 .- Alberuni's notes to this method of Al- harkan are perhaps the weakest part of his work. His very first remark ehows a complete misunderstanding of the whole calculation. The method ia correct, for the months of the seventy-two years with which it begins are solar. If, as Alberuni would have them, they were lunar, and the rest of the months, as he understands it, were lunar too, then the calculation would simply be nonsense; for finding adhimasa months is nothing else than finding the nnmber which we must add to convert solar months into lunar ones. But when the months are already lunar, how can one add anything to them to make them once more lunar? (Schram).

P. 53, 1. 15 .- The example he works out is es erroneous as the remarks on the method itself. It must be clear to anybody who examines the method given on p. 52, that by the words (l. 29), " Add thereto the months which have elapsed between the first of Sha'ban of the year 197 and the first of the month in which you happen to be," there can only be meant solar months. The author fixed the initial epoch in his calendar by saying " I Sha'ban 197," instead of fixing it in the Indian calendar by saying

ANNOTATIONS. 381

"first Vaisakhs 735." This accidental circumstance, which is of no consequence, induced Alberuni to think that he was to take the interval in lunar months, as the Arabie calendar bas only lunar months, and he did not notice that lunar months in this part of the caleulation would be absolntely impossible. He takes, in fact, in the example, the interval in lunar months, for there are 2695 Innar months between the first Sha ban 197 and first Rabi' I. 422, and to these 2695 lunar months he adds the 864 months which he knows to be solar. Then he changes all these mingled months, of which the greatest part are already lunar, to lunar ones, as if they all were solar, and at last he wonders that the resnlt is nonsense, and tries to amend the method. The only fault in the matter is that he did not nuderstand the method. If we wish to exemplify the method of the canou Al- harkan in the case of our gauge-date, i.c. the first Caitra 953 Sakakal, we must proceed as follows :- Subtracting from 953 years 735 years I month, we get as interval 217 years II months or 2615 solar months; adding thereto 864 solar months, we have 3479 solar months. This multiplied by 7 and divided by 228 gives for the number of adhimasa months 10618; adding the 106 adhimasa months to the 3479 solar months, we get 3585 lunar months, or 197,550 Innar days. We add 28, and multi- plying 107,578 by II, we have 1,183,358, which number divided by 703 gives the number 1683788 for the fnarâtra days. Subtracting the 1683 fnarâtra days from the 107,550 lunar days, we have 105,867 civil days. We add I iu order to get the week-dy of the first Caitra 953, and dividing by 7, we get as remainder 7. And as here Friday is cousidered as 1, so 7 corresponds to Thursday, and the first Caitra 953 is found to be Thnrsday. By adding 105,867 to 1,991,819 we have for the first Caitra of the year 953 the day 2,097,686 of the Julian period, as it ought to be (Schram).

P. 53, 1. 33 .- The emendation is as erroneons as the example was. The 25,958 days are counted from the epoch falling 40,081 days after that of Yazdajird to the first Shaban 197. But 25,958 days are equal to 879 Arahic months, or 73 years and 3 months. Further, he

38z ALBERUNTS INDIA.

takes again the interval in lunar months, so that now in the amended method he has nothing but lunar months, which he changes to lunar months as if they were solar. So he gets a number which is, of course, absolntely errone- ous, bnt he thinks it to be correct, for in the last instance he commits a new fault by subtracting I instead of adding it. And so by an accidental combination of different faults he finds by chance a week-day which agrees with that of the day before our gange-date (Schram). P. 54, 1. 12 .- As the multiplications and divisions of this method have already been explained in the note to pp. 36 and 37, we have here to account for the constant numbers ouly which are inherent to the epoch. The epoch is 854 Sakakala, which corresponds to the year 1,972,948,033 of the kalpa. Multiplying 1,972,948,033 by 12, we find 23,675,376,396 solar months, which mul- tiplied by 1,593,300,000, the adhimasa months of a kalpa, and divided by 51,840,000,000, the solar months of a kalpa, give the quotient 727,661,597, 4463 7rios as the number of adhimasa months. Adding the 727,661,597 adhimasa months to the 23,675,376,396 solar months, we have 24,403,037,993 lunar months or 732,091,139,790 lunar days. This latter number multiplied by 25,082,550,000, the ûnarâtra days of a kalpa, and divided by 1,602,999,000,000, the lunar days of a kalpa, gives for the number of foarâtra days 11,455,224,000g11s1. Subtractingthe 11,455,224,000 fnarâtra days from the 732,091,139,790 lunar days, we find as the nomber of civil days elapsed from the begin- ning of the kalpa to this epoch 720,635,915,790, a number which divided by 7 gives as remainder o. So, as the last day preceding the kalpa was a Saturday (see p. 28, L. 31). the last day before this epoch is also a Saturday, and any number of days elapsed since this epoch, if divided by 7, shows by its remainder the week-day counted from Sunday as I. The fraction of the adhimasa months in- herent to the epoch has been found to be 463 Now 2469 mis, is equal to 2911100 before dividing by 65. The fraction of the unarâtra days 65 ", or very nearly f; so we add 29

ig 317481. Now again 91319} is equal to 685267075

nearly f8s; so we add 686 before dividing by 703. 703 =, or

ANNOTATIONS. 383

The first day of this epoch coincides with the day 2,061,541 of the Julian period (Schram).

P. 55, L 5 .- This method consists in finding first the difference of the mean longitude of sun and moon. The numbers are Pulisa's. There are in a caturyuga 4,320,000 revolutions of the sun, and 57,753,336 revolntions of the moon. The difference, 53,433,336, is the number of lunar months. In every lunar month the moon gains one revolu- tion or 360 degrees over the sun. Dividing 53,433,336 by the solar years 4,320,000, we find as the number of lunar months belonging to one solar year 12380088. So in every solar year the moon gains over the sun 12138718 revolu- tions. Omitting the whole revolutions which have no interest, the moon gains over the sun 333178 revolutions, or, what is the same, 1321' 6 degrees. Now 188% degrees are equal to 4618 v% or to 4628 minutes. So the moon gains over the sun in every solar year 132 degrees 46, minntes. By mltiplying the number of years by 132 degrees 461t minutes, we find the nnmber of degrees which the moon has gained in the given interval over the sun. Now if in the beginning of this epoch sun and moon had been together, this would be the difference of the mean longitude of sun and moon. But as this was only in the beginning of the caturynga, but not at the moment of our epoch, there is an initial differ- ence between the longitudes of sun and moon which must be added, Our epoch, or the year 821 Sakakâla, correspouds to the year 3,244,000 of the caturyuga. Multiplying 3,244,000 by the uumber of lunar months 53,433,336, aud dividing by the number of solar years 4,320,000, we find that in these 3,244,000 years the moon gained over the sun 40,124.477118 revolutions. Dropping again the whole revolutions, we see that the moon was in advance of the sun at the moment of our epoch by j1a revolutions, or 112 degrees. Therefore these I12 degrees must be added, and all the numbers of this method find in this their explanation. The result for our gauge-date, 358° 41' 46", is the number of degrees, minutes, and seconds by which the moon is in advance of the sun at the moment of the beginning of the solar year 821, that

384 ALBERUNPS INDIA.

is, in the moment when the sun enters Aries. As in the beginning of the luni-solar year sun and moon must have been in conjanction, the beginning of the luni-solar year has preceded that of the solar year by an interval which was just sufficient for the moon to make 358° 41' 46" in advance of the eun. Now as the moon gains 360 degrees in a lnnar month or 30 lunar days, so she gains 12° in every Innar day. Therefore dividing 358° 41' 46" by 12, we get the number of lunar days and fractions by which the luni- solar year's beginning preceded that of the solar year. The fractions of the lunar days are changed to ghatis and casha- kas. Thereby we get 29 days 53 ghatis 29 cashakas as the time by which the beginning of the Inni-solar year pre- ceded the sun's entering Aries, in agreement with the frac- tion of the adhimasa month found on p. 31, 1. 17. For $18ss adhimasa months are also equal to 29 days 53 ghatis 29 cashakas. The number 27 days 23 ghatis 29 cashakas which he gives, p. 55, 1. 25, is obtained by dividing 328° 4I' 46', and not 358° 41' 46", by 12 (Schram).

P. 55, 1. 17 .- The Arabic manuscript has 328 instead of 358.

P. 55, 1. 33 .- The number is 132° 461, and not 132° 46' 34" (as the Arabic manuscript has). Therefore the portio anni is not 1I° 3' 52" 500, but 11 days 3 ghatis 53 cashakas 24"; and the portio mensis not o° 55' 19" 24i1 10", bnt o days 55 ghatts 19 cashakas 27 !! The reason of this calculation is the following :- In a year or 12 solar months the moon gains over the sun 132° 4616. As she gains 12 degrees in every lunar day, the twelfth part of these degrees will represent the sum of lunar days and their fractions which the solar year con- tains over 360, that is to say, the sum of adhimasa days and their fractions. One solar month containing o adhimâsa days 55 ghatis 19 cashakas 27', the nnmber of solar months within which one adhimasa month or 30 lunar days sum up, will be found by dividing 30 days by o days 55 ghatis 19 cashakas 27". This gives 2 years 8 months 16 days 3 ghati 55 cashaka.

P. 56, 1. 1 .- There must be a great lacuna, for the first

ANNOTATIONS. 385

" lines of this page are absolutely withont meaning. I am inclined to attribute this lacuna to the source whence the author drew this information, i.c. the Arabic translation of Karaņasâra.

P. 59, l. 23 .- The calculation should be made in the following manner :- The sum of daya of the kaliyuga is multiplied by the star-cycles of a kalpa and divided by the civil days of a kalpa, viz., 1,577,916,450,000. So we get the revolutions and part of a revolution which the planet has made during the time elapsed since the beginning of the kaliyuga, But in the beginning of the kaliyuga all planets have not been in conjunction; this was only the case in the beginning of the kalpa. Therefore to the fractions of revolutions which the planet made since the beginning of the kaliyuga, we must add its place at this begin- ning itself, i.c. the fraction of a revolution which every planet had at the beginning of the kaliyuge, the whole revolutions being of no interest. But Brahmagupta adds these numbers before dividing by the civil days of the kalpa, and this is quite natural, both fractions having by this proceeding the same divisor. Therefore what he calls the basis, ought to be the fraction of every planet at the beginning of the kaliyuga multiplied by the civil days of the kalpa; but he has made a great mistake. Instead of multiplying the fractions by the civil days of a kalpa, viz., 1,577,916,450,000, he bas mnltiplied them by the years of a kalpa, viz., 4,320,000,000. Therefore all num- bers given on p. 60 as the bases are eutirely erroneous. To find the fractions for each planet and the bases we have the following caleulation :- From the beginning of the kalpa to that of the kaliyuga there have elapsed 1,972,944,000 years; so to get the places of the planets at the beginning of the kaliynga we ought to multiply the revolntions of each planet by 1,972,944,000, and to divide them by the years of a kalpa, 4,320,000,000. As these two num- bers have the common divisor 432,000, we multiply the revolutions of each planet by 4567 and divide them by 10,000. This will give us the place of the planet at the beginning of the kaliyuga We have thus for the single planets :- For Mars, 2,296,828,522 revolutions multiplied by 4567 VOL II. 2 B

386 ALBERUNTS INDIA.

and divided by 10,000 give 1,048,961,5851016 revolu- tions; so the place of Mars at the beginning of the kali- yuga is 4 rosss of a revolution. For Mercury, 17,936,998,984 revolutions multiplied by 4567 and divided by 10,000 give 8,191,827.435-8848 lutions; so the place of Mercury is rots revolutions 00os revo-

For Jupiter, 364,226,455 revolutions multiplied by 4567 and divided by 10,000 give 166,342,221-506s revolutions; so his place is roros revolutions. For Venus, 7,022,389,492 revolutions multiplied by 4567 and divided by 10,000 give 3,207,125,28010008 ; 8O her place is rouds revolutions. For Saturn, 146,567,298 revolutions multiplied by 4567 and divided by 10,000 give 66,937,284r0 and his place ig 8od0 uuts revolutions. oou revolutions;

For the sun's apsis, 480 revolutions multiplied by 4567 and divided by 10,000 give 219106e revolutions; and its place is Tooos revolutions. For the moon's apsis, 488,105,858 revolutions multiplied by 4567 and divided by 10,000 give 222,917,94518000 revolutions; and its place is , revolutions. For the moon's node, 232,311,168 revolutions multiplied by 4567 and divided by 10,000 give 106,096,51010008 revolutions; and its place is 1000o revolutions. Multiplying now the place of every planet by 1,577, 916,450,000, we get the following bases for the single planets :- For Mars, 1,573,813,867,230. Mercury, 1,566,555,451,560. Jupiter, 1,575,549.575-325- Venus, 1,572,235,950,780. Baturn, 1,572,551-534/070. the sun's apsis, 340,829.953,200. the moon's opsis, 550,061,674.470. the ascending node, 671,561,241,120 (Sckram).

P. 67, 1. 14. A.H. 161 .-- According to p. 15, the year was A.I 154. Cf. note to i. 169. P. 71 .- With the orbits of the planets ef. Strya-Sid- dhanta, xii. 90, note. Pp. 74 seq .- As for the Arabic terminology of these pages, it deserves to be noticed that-

ANNOTATIONS. 387 (I.) Janht jal means the true distance-Sanskrit man- dakarņa. (2.) That h yat means the true distance of the shadow's end; and (3.) Sinus totus, Jsh =Sanskrit trijfva or trijya, means the sinus of three zodiacal signs or go degrees, i.c. the radins.

P. 74, ll. 17, 18 .- Inatead of TC= the Arabic manu- script has KC= , which has been corrected by Dr. Schram.

P. 75, 1. 34 .- The lacuns must be something like the following :- " For KC must be divided by the divisor kept in memory" (Schram).

P. 78, 1. 27 .- This and the two following passages are not clear. Alberuni does not seem to have understood the subject, for the shadow is neither the greatest nor the mean, but the true shadow; and the shadow from which one is to subtract, i.c. 1581, is nothing elae than the earth'a diameter, which also is neither the mean nor the greatest, but always the same (Schram).

P. 79-Alkhudrizmt is mentioned here and ii, 114 (on the various colours of eclipses). According to Fihrist, p. Mr, he composed an epitome of the Sindhind (Brahma- Siddhanta). He is famous as the anthor of a work on algebra, edited by Rosen, London, 1831. Cf. also. L. Rodet, L'Algebre d' Alkhwdriemt et les Methodes Indienne et Grecque (" Journal Asiatique," 101 (1878), pp. 5 seg.).

P. 82. Tio suns, two moons, dc .- This theory, as well as the expression fish (a name for the polar star ?), seem to be of Jaina origin, Cf. Colebrooke, "Essaya," ii. 201.

P. 84 .- Cf. with this table of the Nakshatras a paper of Thibaut, "The Number of the Stars constituting the several Nakshatras according to Brahmagupta, &c.," the "Indian Antiqnary," 1885, p. 43; also Colebrooke, " Essays," ii. 284, and Surya-Siddhanta, p. 321.

ALBERUNIS INDIA.

P. 89, 1. 32 .- In the Arabic text, p. I4, 15, read An instead of elt. The number of years is 1800, not 2800.

P. 90. Kaldmsaka .- This term (also kdldméa) is ex- plained in Sdrya-Siddhanta, note to ix. 5. The work Ghurrat-alatjdt, only once mentioned, is per- hapa identical with the Kitdb alghurra, which Alberuni quotes in his "Chronology" (my translation, p. 15 et passim). Its author was Aba-Muhammad Alna'ib Alamuli, who has used the work of Yakub Ibn Tarik. Cf. note to i. 169.

P. g0, L 21 .- Emendation of the khandakhddyaka (also on p. 91), ie. Uttarakhandakhâdyaka. On Vijayanandin (1. 26), the anthor of Karanatilaka, ef. note to i. p. 156.

P. 101 .- The enumeration of mountains, here taken from the Matsya-Purdna, may be checked by the help of Vishnu-Purdna, ii. 141, note 2, and ii. 191 aeg. The last name is written bahdshtr in the Arabic, which I cannot identify with an Indian name. Perhaps it is a blunder for mahdsht., which might represent mahdsaila. Vide Vishnu-Purdna, II. iv. p. 197.

P. IOI .- On the Aurva legend, ef. Vishnu-Purdna, III. viii. p. 81, note.

P. 102 .- The story of Soma, the husband of the danghters of Prajapati (the innar atations), occurs in its elements already in the Vedic period. Cf. H. Zimmer, Altindisches Leben, pp. 355, 375.

P. 104-On the Hindu theory of ebb and fow, ef. Vishnu-Purdna, ii. 203, 204 The two names, of which I have not found the Indian equivalents, are written baharn and ouhar in the Arabic.

P. 105. The Vishnu-Purdna says-The anthor seems to refer to Vishnu-Purdna, IL iv. p. 204: "The rise and fall of the waters of the different seas is five hundred and ten (not 1500) inches" (or finger-breadths).

ANNOTATIONS. 389 P. 106 .- The author'a theory of the origin of the Diba- jat has already been mentioned, vol. i. 233.

P. 11O .- As to the strictures of the anthor on the sin- cerity of Brahmagupta, ef. note to p. 25 (hore ii. p. 263). The passages which excited the indignation of Alberuni do not express the view of Brabmagupta, but were simply taken by him from older books-in fact, written purva- sastranusdrena. Cf. Kern, translation of Brihat-Samhita, note to chap. iii. v. 4 (p. 445)-

P. 114, 1. 12. Kinds of eclipses .- Read instead of this, colours of the eclipses. On Alkhwârizmt, cf. note to ii. 79. What the anthor here mentions as a view of the Hindus, agrees literally with Surya-Siddhanta, vi. 23.

P. 116 .- On the Khandakhadyaka, the Sanskrit original of the Arabic Sindhind, cf. note to i. 153, 154.

P. 118 .- On the Brihajjatakam of Varahamihira, ef. note to i. 219

P. 119-Rules for finding the dominants or regents of the day, month, and year are given in the Surya-Sid- dhânta, i. 51, 52; xii. 78, 79.

P. 120 .- On the srudhava (?) of Mahadeva, not to be confounded with the book of the same title by Utpala, cf. note to i. 157.

P. 120. Table of the serpents .- The names of this table must be compared with the names in Vishņu-Purdna, ii. 74, 285. The words Suku and Cabrahasta seem to be mis- takes of the Arabic copyist for Vasuki and Cakrahasta.

P. 121 .- The names of the dominants of the planets are not known to me from a Sanskrit source. Therefore the pronunciation of some of them remains uncertain.

Pp. 121, 122 .- The names of the dominants of the Nakshatras are given by A. Weber, Veber den Vedakalen- der Namens Jyotisham, p. 94. Cf. also Surya-Siddhanta,

390 ALBERUNI'S INDIA.

viii. 9, pp. 327 xg., and Visinu-Purdna, II. viii., notes on pp. 276, 277. Instead of Mitra, the deity presiding over AnurAdha, it would perhaps be better to write Maitra, and in the Arabic ar (Vishnu-Purdna, ii. p. 277). The latter part of this list in the Arabic text is not free from confusion, The regent of Uttarabhadrapada is placed side by side with Purvabhadrapada, whilst the latter station is left without its regent, which is aja ckapdt (Sarya-Siddhanta, P. 343). A part of this word seems to be extant in the square for afvint, which has ,LS t. Perhape this is to be read asvin ajaikapad, osl yt, in which case the Arabic copyist has made two blunders, dropping part of the word ajaikapad and placing it in the wrong square,

P. 123 .- On the sirty-years cycle ef. Sorya-Siddhanta, i. 55, and xiv. 17; Varâhamihira, Brihat-Samhit, viii. 20-53.

P. 125 .- For the names Samvatsara, Parivatsara, &c., ef. Brihat-Samhitd, viii 24; Surya-Siddhanta, xiv. 17, note; Weber, Veber den Vedakalender genannt Jyotisham, p. 34-36.

Pp. 127, 128 .- The dominants of the single lustra are given in Brihat-Samhitd, chap. viii. 23. The names of the single years exhibit some differences from the Sanakrit text (Brihat-Samhitd, viii. 27-52). No. 8, uwl instead of bhdta, has risen from a wrong division of the words of the text- śrimukhabhdvasdhsau, i.e. śrimukha-bhava-sahvau. No. 9, instead of y=yuvan, is perhaps a mistake of the copyist of the Arabic text. No. 15, t, visha (in Kern's edition vrisha), is not a mistake, but a different reading. The word in brackets (Vrishabha) is to bo cancelled. No. 18, us, natu, cannot be combined with pdrthira. It corresponds to natam. Cf. Kern'a various readings to chap. viii 35.

ANNOTATIONS. 391 No. 30,F+. The name of the thirtieth year is durmukha. Perhaps the reading has risen from a wrong division of these words (viii. 38)- manmatho 'sya parataśca durmukhah, so as to represent the elements -ca dur -. No. 34, y (sarva), seems to be a mistake for sarvari or sarvarin. No. 40, pardrasu is the reading of some manuscripts for parabhava. Cf. Kern, various readings to viii. 41. No. 48. This year is called dnanda by Kern, but the reading of Alberuni, vikrama, occurs also in Sanskrit manu- acripts. Cf. various readings to viii. 45. No. 56. The ow of the text seems to be a blunder of the copyist for dundubhi (viii, 50). No. 57, amgdra or amgdri, the reading of certain manu- scripts instead of udgri (viii. 50). No. 58 and 60. The words sus (instead of sts,) and 3=raktaksha and kshaya, seem to be examples of a pho- netic change between sh and r. The same list of names is given in Surya-Siddhanta, i. 55, note. P. 130 .- With this chapter on the four parts of the life . of a Brahmau cf. Vishnu-Purana, book IIL chsp. ix.

P. 131 .- The complete verse of Bashshar is this- " The earth is dark, but the fire is bright, And the fire is worshipped, since there is fire."

This is the sayiog of a man whose parents had come as prisoners of war from Tukharistan on the Upper Oxus, bnt he was born in Basra, and lived in Bagdad under the Khalif Almahdt As he stood under the accusation of being a heretic (Zoroastrian or Manichæan), or, accordiog to another version, because he had composed satirical verses ou the Khalif, he was, notwithstanding his great age, sentenced to be beaten, and died in conseqnence, A.H. 167=A.D. 784- Cf. Ibn Khallikan, Vita, No. 112. P. 134, 1. 1 .- The south, as the direction foreboding evil, has already once been mentioned in connection with the islands Lanka and Vadavâmukha, vide i. 307, 308.

392 ALBERUNTS INDIA.

Pp. 134, 135 .- With this description of Aryavarta ef. Manu, ii, 17 seg .; VAsishtha, i. 12; and Baudhayana, i I, 9-12 ("Sacred Laws of the Aryas," translated by G. Bühler, Oxford, 1879-82).

P. 135 .- On the vegetables which must not be eater, ef. Manu, v. 5, and Vasishtha, xiv. 33. Nalt seems to be = Sanskrit ndlikd.

P. 136 .- The contents of this chapter are nearly related to Vishnu-Purdna, book IIL chap. viii.

P. 137 .- The story of King Rama, the Brahmin, and the Candala, taken from the Ramdyana, vide in Wilkins' " Hindu Mythology " (Calcntta, 1882), p. 319

Pp. 137, 138 .- The two qnotations of Alberuni from the Bhagavadgita can hardly be compared with any pas- sage in the book in its present form. Cf. note to i. 29.

P. 139-On the asvamedha or horse-sacrifice, ef. Cole- brooke, " Essays," i. 55, 56.

Pp. 140, 141 .- This legend, as given on the enthority of the Vishnu-Dharma, is not known to me from a Sanskrit source.

P. 142 .- As the original of this quotation from the Purdnas is not known to me, the pronunciation of some of the proper nouns remains uncertain.

P. 143 .- The story of Sagara, Bhagiratba, and the Ganges, is related by H. H. Wilson, " Works," vol, ii. p. 168. Cf. also Wilkins' " Hindu Mythology," p. 385. The source of this legend is the first book of Ramdyana.

P. 145 .- I do not know the original of this quotation from Varahamihira's Samhitd.

Pp. 145, 146 .- The words here attribnted to Saunaka are probably taken from the Vishnu-Dharma. Cf. note to i. 54-

ANNOTATIONS. 393 P. 147 .- The story of the head of Brehman is part of the legend of Siva's fight with the Asura Jalandhara. Cf. Kennedy's " Researches," p. 456.

P. 149 .- This and the following chapters treat of subjects which are discussed more or less in every Indian law-book, as in those of Manu, Apastamba, Gautama, and others. Alberuni, however, does not seem to have drawn direetly from any of these books, but rather from his own experi- ence, from what his Pandits had told him, and what he himself had observed during his stay in India.

P. 153 .- Alhajjaj was governor of Babylonia during twenty yeara under the Omayyade Kalif 'Abdulmalik (684-704) and his son Alwalid (704-714).

P. 153. That a Brahmin and a Candala are egual to him .- Cf. the saying of Vyasa, the son of Parasara, here vol. i. p. 44.

Manu, iii. 5. P. 155 .- On the forbidden degrees of marriage, cf.

P. 156 .- On garbadhana, simamtonnayanam, &c., cf. the Dharmasastra of Gautama, viii. 14; also the Grihyasutras of Aśvalâyans, i. 13, 14-

P. 157. Thus, when Kabul was conquered, &c .- The sen- tence added in brackets to indicate the meaning of the author's words, as I nnderstand them, ought to run thus: "(which proves that he abhorred the eating of cows' meat and sodomy, but that he did not consider harlotry as auything baneful or unlawful)." The detail in the history of Kabul here alluded to is not known from other sources, e.g. Baladhurt. During the Omayya Kaliphate of Damascus, both Kabul and Sijis- tan bravely fought against the Muslims. During certain years they were subdued and bad to pay tribute, but Kabul always remained under the away of its Hindu (Brahmin) kings of the Pala dynasty. It was incorporated into the Khalif's empire nnder the Abbaside Ma'mun; it had to receive a Muslim governor, but retained at his side

394 ALBERUNI'S INDIA.

the Hindu Shah. The same donble rule existed in Khwâ- rizm. About A.D. 950-975 the city of Kabul was already Muslim, whilst the suburb was inhabited by the Hindus (and by Jews). Kabul was the coronation-city for the Pala dynasty, as Konigsberg in Prussia for the Hohenzollerns. Even when they ceased to reside in Kabul, they had to be crowned there. By the Ispahbad, mentioned by Alberuni, I under- stand the Hindu governor who ruled over the city for the Pala king. Our anthor applies a title of the Sasanian empire to the official of a Hindu empire. In what year the negotiation referred to by Alberani took place is not known. Perhaps under Ma'mun, when the city was definitely ceded to the Muslim conquerors. It seems to have been the public opinion among Mus- lims that Hindus considered fornication as lawful, as Ibn Khurdadhbih expresses it (Elliot "History of India," i. 13), whilst, according to Alberuni, they considered it in- deed as unlawful, but were lax in punishing it.

P. 157 .- The Buyide prince 'Adud-aldanla, who held Persia under his sway, died A.h. 372=A.D. 982. Not long before Alberani wrote, the last of their dominions had been annexed to the empire of Mahmud of Ghazna,

P. 158 .- 'Iyas Ibn Mn'awiya was judge in Başra under the Omayya Khalif Omar Ibn 'Abdala'ziz, and died there, A.H. 122=A.D. 740.

P. 159-With the anthor's description of the ordeals, ef. Manu, viii. 114 seg., and a translation of the chapter on ordeala from the Vydvahara Mayukha by G. Buhler, in "Journal of the Asiatic Society of Bengal," 1867, vol. xXv. pp. 14 seg .; Stenzler, Die Indischen Gottesurtheile, in Zeitschrift der Deutschen Morgenlandischen Gesellschaft, ir. p. 661. The last-mentioned kind of ordeal (p. 160) is also described in Elliot's " History of India," i. 329 (the Sindian ordeal of fire).

P. 164 According to a passage in the book Manu-Cf. Manu, ir. 118.

ANNOTATIONS. 395 P. 166 .- For the first quotation from Phædo, 81D, ef. note to i. p.65. The second quotation can hardly be iden- tified with any passage in Phædo. Perhaps it is derived from a commentary on the following words, 8IC :- άλλά διειλημμένην γε, οΐμαι, υπό του σωματοειδούς, ο αύτη ή ομιλία τε καί συνουσία του σώματος διά τό άει ξυνείναι καί διά την πολλήν μελέτην ένεποίησε ξύμφυτον.

P. 167 .- The quotation from Phædo is found 115C- I16A :- Θάπτωμεν δέσε τίνα τρόπον; όπως άν, έφη, βούλησθε, έάνπερ γε λάβητέ με καί μή εκφύγω ύμας, κ.τ.λ. εγγυήσασθε ούν με τρός Κρίτωνα, έφη, τήν έναντίαν έγγύην ή ήν ούτος πρός δικαστάς ήγγυατο, ούτος μέν γάρ η μήν παραμενεΐν, ύμεϊς δέ ή μήν μή παραμενείν έγγυή- σασθε, έπειδάν άποθάνω, άλλά οιχήσεσθαι άπιόντα, ίνα Κρίτων ράον φέρη, και μή όρων μου τό σώμα ή καιόμενον η κατορυττόμενον άγανακτη ύπέρ έμου ώς δεινά πάσχοντος μηδέ λέγη έν τη ταφύ, ώς η προτίθεται Σωκράτη ή εκφέρει 5 κατορύττει, κ.τ.λ. άλλά θαρρείν τε χρή καί φάναι τουμόν σώμα θάπτειν καί θάπτειν ούτως, όπως αν σοι φίλον η καί μάλιστα ήγη νόμιμον είναι.

P. 168. Galenus, &e .- I do not know the Greek original of this quotation. Cf. note to i. p. 35

P. 69 .- The words of Vâsudeva are a quotation from Bhagavad-Gita, viii. 24 P. 171. Johannes Grammaticus .- Cf. note to i. 36.

P. 171 .- The two quotations from Phædo are found in 620 :- ίσως τοίνον ταύτη ούκ άλογον μή πρότερον αύτον άπο- κτιννύναι δεϊν, πρίν ανάγκην τενά θεός έπιπέμψη, ώσπερ καί την ων ήμιν παρούσαν. And 62B :- ώς εν τiν φρουρά έσμεν οί άνθρωποι καί ού δεί δή

396 ALBERUNTS INDIA.

έαυτόν έκ ταύτης λύειν ούδ άποδιδράσκειν, κ.τ.λ, τό Θεούς είναι ήμων τούς έπιμελουμένους καί ήμας τούς άν- θρώπους 2ν των κτημάτων τοίς θεοίς είναι,

P. 174 .- For the Vishnu-Purdna, vide note to i. 54. The reading Duve is not certain, as the Arabic text has درى only Tho names Dilipa, Dushyanta, and Yaydti have been verified by means of the index to Vishnu-Purdna,

P. 175, last line .- On the festival of the birth of Vasu- deva-Krishna (Krishnajanmashtamt), cf. Weber, " Indian Antiquary," 1874, p. 21; 1877, p. 161; Zeitschrift der Deutschen Morgenlandischen Gesellschaft, vi. p. 92.

P. 176, L. 11 .- The Arabic manuscript has ei, ic. dtaj. For the word attataja, ef. H. H. Wilson, " Essays and Lectures," ii. 232.

P. 176, 1. 19. Devasint-The latter half of this word is apparently a derivation from the root svap=to slesp. In Prakrit sleep =sivino (Sanskrit svapna). Vide Vararuci, i. 3.

P. 177, L 20 .- Deotthini, also called deotthdn and ditthwan. Cf. H. H. Wilson, " Glossary of Technical Terms," pp. 133, 134, 143, and " Memoirs on the History, Folklore, and Distribution of the Races of the North- Western Provinces of India" by H. Elliot, edited by J. Beames, i. 245.

P. 177 .- The here-mentioned bhishma-panca-ratri seems to be identical with the bhishma-pancakam mentioned by Wilson, " Essays and Lectures," ii. 203-

P. 177 .- The name Gaur-t-r, , occurs also ii, 179, and is apparently a vernacular form for gaurt-trittya. Cf. Wilson, 4 L p. 185.

P. 178 .- With this calendar of festivals are to be com- pared the treatise of H. H. Wilson, "The Religious Festivals of the Hindus," in his " Essays and Lectures," ii. p. 151 seg., and Garcin de Tassy, Notice sur les fetes popu-

ANNOTATIONS. . 397 laires des Hindous, Paris, 1834. This chapter, as well as the preceding one, would perhaps receive much light from the Jyotirvidhabharanam, chap. xxi. Cf. Weber, " Journal of the German Oriental Society," vol xxii. p. 719, and xxiv. p. 399. This chapter has been translated into Persian by Abu- Sa'id Gardezi (manuscript of the Bodleian Library in Oxford, Ouseley 240). Cf. note to ii. 6.

P. 178. Agdus-The Arabic has only ot, which might be aomething like ajya-divasa. Muttai .- This pronunciation is given by the mannscript. The name, not to be confounded with the Arabic name Matta (Matthæus), is perhaps identical with the name of a prince of Siwistan mentioned by Elliot, " History of India," i. 145-153. Hindolf-caitra-Cf. Dola-yatrd or Holt of Wilson, p. 223. Bahand .- Vide Wilson, l. c., and vasanta, here ii. 179.

P. 179. Gaur-t-r .- Cf. note to ii, 177.

P. 180. Gaihat (?), &c .- In the Arabic text the word b must be added before it. In the following line there is a lacuna, which in my translation I have filled up by the help of the Persian translation of Gardezi which runs thos :- کامهت بود (sic) وأين روز خشم بود كه اندر أين روز زندأليان رأ طعام we. In another place Gardezi writes eol.

P. 181 .- On Jivasarman, cf. note to i. 164

P. 182. Ktrt (?) .- This is perhaps only a misspelling of the Arabic copyist for cas, Kandt (Gandt Ribat-ala'mir). Cf. note to i. 317, and Elliot, " History of India," ii, 112, 150; iv. 138; Baihakt, ed. Morley, p. 274. It is the place where King Mas'ud was murdered.

Pp. 182. Dibalt=dtpavali (row of lamps) .- Cf. Wil- son, "Glossary of Technical Terms," p. 114. Gardezt ha3 Jys, dirdlt

398 ALBERUNI'S INDIA.

P. 183. SAgdrtam= fakdshtamt .- Cf. Wilson, "Esmys." ii 208.

P. 183 .- Camdha seems to be = caturdast mdgha, mansartagu = mansáshtaka, půrdrtaku = purdshtaka, and mahdtan = maghashtamt. Cf. Wilson, " Essays," ii. 183, 184, 181.

P. 183 .- The festival dhola seems to be identical with holi, holikd or dol-jatrd. Cf. Wilson, p. 147, 210. Instead of dhola the Persian translation of Gardezi has 3y, holf.

P. 184. Śivardtri .- Cf. Wilson, p. 210.

P. 184-Puyattanu is perhaps = pupashtamt. CL papåshtaka.

P. 186 .- On the 15th Magha, as the beginning of kaliynga, ef. Wilson, "Essays and Lectures," ii. p. 208. Alberuni seems to have taken his information regarding the yugadyd or beginning of a yuga from Vishnu-Purdna, IIL chap. xiv. p. 168.

P. 187, 1. 5 .- The number of lunar days, 1,603,000,010 (sic MS.), must, according to Dr. Schram, be altered to 1,603,000,080.

P. 188. Fishuva-On the use of this term in astronomy, cf. Surya-Siddhanta, iii. 6, note.

P. 188 .- On Samaya (1), ef. note to i. 336.

P. 189, 1. 17, after the table .- The solar year is 365 days 15' 30" 22" 30', not 365 days 30' 22" 30" ol. Accord- ingly the last line must run thus : "(ie. I day 15' 30" 22" 30" are equal to $9;1) " (Schram).

P. 190, 1. 7 .- The bhagahara is not 572, as the manu- script has, but 576, and the fraction I1& (Schram).

P. 190-Auliatta (?). The name is written t e ad, !. A more literal rendering is this: "And that which A. the

ANNOTATIONS. 399

son of S. has dictated of the same (subjeot), is based on the theory of Pulisa." This author seems to have been contemporaneons with Alberuni, as also Samaya (ii. 188).

P. 190. Vardhamihira .- Cf. note to i. 54 The term shadasttimukha is explained in Surya-Sid- dhanta, xiv. 6, note.

P. 191 .- On the Parvan, ef. chap. 1x.

P. 192. Samhitd,-The anthor qnotes here the Brihat- Samhita, chap. xxxii. 24-26.

P. 192 .- On the book Srudhava, ef. note to i. 157 and ii. 120. Is the word = sarvadhara ?

P. 194-With the theory of the karanas, ci. Surya- Siddhanta, ii. 67-69.

P. 195 .- For an explanation of the term bhukti, ef. Surya-Siddhânta, i. 27, note.

P. 197 .- The names of the common karanas are found in Sûrya-Siddhânta, ii. 69, note. The other names are Indian numerals of a vernacular stamp. The corresponding Sindht forms are barkhu (?), bio, triô, cothõ, panjo, chahô, satõ, atho, nãő, daho, yárhō, bárhő, terho, codho. Cf. Trumpp, "Sindhi Grammar," pp. 158, 174 The form pancdhf (=the 15th) has, as far as I can see, no analogy in the vernacular dialects.

P. 199 .- Samkranti means the sun's entrance into a sign of the zodiac. Cf. Surya-Siddhanta, xiv. I0, note.

P. 200. Alkindt .- The way in which this scholar has transformed the Hindu theory of the karanas is instruc- tive, as ehowing how Indian subjects were handled by the Arabs before Alberuni, even by the most learned and enlightened among them. The first knowledge of these things was probably communicated to the Arabs by the translation of the Brahma-Siddhanta (Sindhind) and Khan- dakhadyaka (Arkand) of Brahmagupta. On Alkindi, of.

400 ALBRRUNIS INDIA.

G. Flügel, Alkindt, genannt der Philosoph der Araber, Leipzig, 1857 (in vol, i, of the Abhandlungen fur die Kunde des Morgenlandes).

P. 201 .- The names of the vishtis, as taken from the Sradhava (of Mahadeva ?- ef. note to ii. 120), are not known to me from a Sanskrit source. However, vadavd- mukha, ghora, and kalardtri seem to be certain. The words s and Jlys might be plava and jodla, bat Ji? The other series of names of the vishtis, according to Alkindt, which by a mistake bave been omitted in the Arabic text, may be transliterated in this way :- (1.) Shûlpi (sulapadi !). (2.) Jamadfd (ydmyodadhi !). (3-) Ghora, (4) Nastarinish. (5.) Daruni (dhárint?). (6.) Kayâli. (7.) Bahayamani, (8.) Bikata (vyakta ?). P. 204 On the yogas .- The contents of this chapter are near akin to those of chap. xi. of the Surya-Siddhanta. Compare also in the same book ii. 65, 66. The technical term pata, which literally means fall (for its astrological meaning, ef. l. c. xi. 5, note), has in Arabic been rendered by the word bi, i.c. falling (page t", 11, 24), here ii. 207, 208, 209. In the Arabic text on p. m, 7, read Ja inatead of du, and to the word uyda, 1. 16, it must be added that the manuscript has ohw.

P. 205 .- On the Karanatilaka of Vijayanandin, ef. note to i 156.

P. 207 .- The bhuktyantara has been explained, ii. 195.

P. 208 .- Sydvabala (?) seems to have been e Hindu from Kashmir who had become a Muslim, and wanted, by means of an Arabic book, to be informed on certain chap- ters of Hindu astrology. The pronunciation Sydvabala is not certain. The Arabic manuscript has siydwpal.

ANNOTATIONS. 401 P. 208 .- On the Brahmin Bhattila, ef. note to i. 157. The names of the yogas which he mentions are not known to me from other sources. The names gandanta, kala- danda, and vaidhrita are certain, and barh is probably varsha.

P. 209-On Śripâla, ef. note to i. 164

P. 210 .- With the names of this table ef. Surya-Sid- dhanta, ii. 65, note (also p. 432). The of the Arabic seems to be a mistake for Ss vishkambha; No. 15, A45, & mistake for as, ganda. Instead of dyushmant (name of the third yoga), the Arabic has Sy (rajakama ?); instead of ryatipata it has wls (gatipata ?).

P. 211 .- The contents of this astrological chapter are principally taken from the Zaghujatakam (i.e. the smaller book of nativity) by Varahamihira, of which the chapters i. ii. have been translated A. Weber (Indische Studien, 2, 277 seq.), whilst the re ander has been translated by H. Jacobi (De Astrologia Indica hora appellate origini- bus .. Accedunt Laghujataki capita inedita iii .- xii., Bonn, 1872). Alberuni does not always adhere to the order of the paragraphs which we bave in the Sanskrit text, and for certain parts he seems to have drawn from some com- mentary. The exact meaning of the term seconds of the stars (the same page, ll. 23, 24), rr Jy, is not known to me.

Pp. 213-215 .- The table of planets is taken from chapters ii. iii, iv. of the Laghujatakam. ' For the reading of the terms naisargika, vimiśra, and shadaya (p. 215), I am indebted to Prof. H. Jacobi, Kiel. The number 25, &, in the colomn with the heading The scale of their magnitude, seems to be a mistake for 3, c.

Pp. 217-219-This table of the zodiacal signs has been taken from Laghujatakam, chap. i

Pp. 221, 222 .- This table of the Houscs has been taken from Laghujatakam, chap. i. 15. VOL II. 2 0

403 ALBRRUNTS INDIA.

P. 234 .- The notes on comets and other meteorological subjecta, with which the anthor coneludes his book, have been taken from the Brihat-Samhitd of Varahamihira

Pp. 237-238 .- This table of comets is taken from Brihat-Samhitd, chap, xi. 10-28. The children of the fire are called hutasasutdh in Sans- krit, in Arabic out aft, which I cannot explain.

Pp. 241-244-This table of comets is taken from Brihat-Samhitd, chap, xi. 29-51. The reading uSs4, instead of padmaketu, seems to be a mistake of the copyist for < f.

P. 245. Book of the medicine of elephants-On this and similar literature, ef. A. Weber, Vorlesungen jiber Indische Literaturgachichte, p. 289.

TNDEX I.

&=aditya, i. 215 abda, ii. 118 Agnijibva, i. 231

abdhi, i. 178 Agnimukha, L 231

AbhApurt, L 200 Agnitya, i. 302

Abhtstala, i. 230 Agnivesa, i. 159

Abhi, i 303 agokiru, i. 220

abhijit, i, 340, 341, 342; ii. 66, ahan, i. 368 ; ii. 26

85, 87, 122 abankara, i. 41

Abbfra, i. 800, 301, 302 abargana, i. 355, 368; ii. 26, 27

abhra, i. 178 ceg., 84, 46 seg., 48, 60, 116, 184

Abia (1), L. 299 åhari, ii. 179 Abirbudhnya, i. 342; ii, 66, 122 AcArya, i, 155 Acad (1), ii. 143 Ahot, i. 180

Adatf L 802 aborâtra, i. 359

Adhaka, i. 162, 163, 164 aindra, i. 135

adhas, i. 290 Airâvata, ii. 245

adhimts, ii. 20 scg., 23 ; univergal aiśAna, i. 290, 297 ; ii, 202

or partial, it 28 aisdoya, i. 303

AdhishthAna, i, 207 ; ii. 181 Aja, L 342, 858.

adbomukha, L 61 Aja ekapad, ii, 122

adi, i. 178; ii, 23 A jodaba (Ayodbya), i. 200 Akara, i. 301 Adi-purina, i. 180 Aditi, ii, 121 akaá, i. 178

Adittahaur, L 206 akabara, L 172

Adita, i. 116, 179, 215, 216, 291 akahaubint, i, 179, 403, 407, 408

Adityavara, i. 213 akshi, i. 178

Aditya-pnrina, L 130, 168, 217, Alika, i. 300

229, 280, 232, 248, 868 Allepûr, L 203

sdityaputra, i, 215 Amarâvatt, i. 271 AmarAvatipura, i. 271 adri, L 178 amâvasya, i 348 ; iL 185, 197 Ag, i. 178 agutys, i 182; ii. 66, 91, 92, 94 ambara, i. 178, 803

Agastyamata, i. 132 ambaratala, i. 280

Agdûs (1), ii. 178 Ambarieha, i. 113

Agnoya, i. 290, 297, 801; ii. 203 Ambashtha, t. 301 amrita, i 54, 253, 262, 844 ; ii, 107 Agneya, L 858 Agoi, i. 181, 178, 242, 342, 357, amáaka, I. 140, 144

858, 894 ; li. 121, 125 aruáaya, ii. 227 arhéu, i. 217, 230 Agniba (1), i, 894 Agnidhra, i. 394 arhsumant, i. 217 anala, iL 128 agnibotrin, i. 102 Anaudapdla, i. 135 ; ii 13

404 ALBERUNTS INDIA.

Ananta, i. 237, 247, 298 Antr, i. 205 trdrt, L 218; ii. 66, 84, 86, 121

Anartts, L 800, 302 arzha, IL 95

Andbra, L 299, 300, 801 Arhant, i. 119, 121

Andhradesa, i. 173 Arbata (1), ii. 142 Ari, L 800 Andbri, i. 178 andi, i. 161, 162, 168 Arjuna, i. 29, 52, 352, 408 ; iL 138 Arjandyana, i. 302 aAga, L 178, 501 arka, i. 179, 215, 217; iL 125 atrtra, ji. 128 Angiraa, i. 131, 215, 291, 390; fi. Lrki, i. 215 Arkutirtha, i. 200 127 Aror, i. 205, 280 Ligula, L 166 Anhilvira, i, 153, 205 ; 1L, 6, 7 artha, i 178 Arthaykshava (1), L 299 antkinl, i. 407 Arunk, i. 259 Anila L 542 Antla, 1. 248 Arona, L 253 ; iL 149, 238 Arundhat, i. 890 Abiruddba, 1. 398 Aruni, L. 894 Abjan, i. 300 Arvasndhana (?), i. 003 AAE, L 174 Ar74, i. 148 ant, L. 220 Antaka, i 342 Aryabhata, L 156, 168, 225, 227, 244, 246, 206, 267, 268, 275, 277, 280, antara, i. 178 ; li. 195 370, 378, 376, 877, 886; iL 16, 17, Antardvtpe, i. 302 Antarikaha, L 898 18, 19, 53, 111, 190 Aryabbata (of Kummapara), i. 176, Aatarikabya, fi. 235 Antarvedi, i. 211 (notm). 240, 016, 330, 336, 370

Ant hl i. 267 Aryaka, i. 254 Aryaman, L 217, 242, 842; iL 121, antya, L 175 antyaja, L 101 Arytahtaiata, L 157, 856 s, i. 837 Aiblads i 231 Arylvarta, i. 178; iL 6

anmrAdht, i. 218, 391, 898 ; L 85, ML L 179

86, 122, 176 Mala (1), i. 230 Acns (1), i. 358 Anru, i. 253 Aoutapt4, L. 202 Antrii, i. 209 Labati (1), i. 215 aanvatsara, iL 125 AścArvari, i. 387 Aruvifr, L 803 tab&dbs, L 211, 217, 218, 357, 358,

AnhmArti, L. 394 408; IL 96, 99, 100, 178, 176, 179, 193

Apara (1), i. 894 sata, L. 178

Apurtota (D, i, 300 asbtaka, li. 185 MbE, i. 179 Apariataka, i 802 A L 202 Apes, L 343 ; iL 123 Apeucasabe, L 131 wipattravana, i. 61 mita, L 215 ; ii. 235 Apratidbrishya, L S72 Ailcha, i. 218, 291 ; ii. 84, 121 aprm, i, 247, 148 AprAr, i. 903 Ammaka, i. 800, 302

Aota-purkan-kAr, i 537 Lapbujit, i. 215 Lramavke, i. 133 Moka, ti. 180 graaya, i. 183 Astagiri, L 302

abuda, i. 178 uthi, f 241

Arbodam, i. 177 or, L 90, 347, 825, 831

ardhastearl, i. 178 aávamodba, i. 138 ; Li. 2, 139

Ardie, L 202 Afraradans, i 301 Aśramukhe, i. 262

INDEX. 405

Afratara L 281, 247 ymttha, i 86 ; iL 140, 141 Balabbadra, i. 156, 157, 158, 225,

Afratthiman, i. 138, 594, 898, 405, 227, 286, 238,246 (1), 241, 243,244, 246, 273, 274, 275, 279, 281, 282, 406 Lámayuja, i. 217, 218, 358, 403 ; ii. 817, 401, 403 ; iL 70,.75, 187

98, 173, 177, 179, 186, 193 Balabhid, iL. 127

sfvin, L 159, 178; iL 122 128 Baladora, i. 118

sáviot, l. 213, 242, 369 ; ii. 84, 86, Baladevapattana, i 301 bAlAgra, L. 162 122, 128 Balabaka, ii. 101 atala, L 230 Atavya, L 300 dlava, iL 197, 199

Atharvanaveda, i. 127, 129 Bali, i. 117, 129, 231 387, 396; ii.

atidbriti, i. 179 3, 11, 145, 182

atto, ii. 197 Balirajya, ji. 182 Ballavar, i. 205 AtinAman, i. 394 Balfika (9), i. 257 ativthika, i. 63 BaluvAbint, L 257 Atman, i. 351 BAmabûr, i. 202 Atmapuruaha, i 321 Atroya, L. 163, 300, 888 Bamhanv& (not Bahmanva), i. 21,

Atri, i. 131, 291, 301, 390, 394 173, 205, 316 Bandrasi (Benares), i. 200 attstaja, iL 176 atvh (i), L 348 Banavas, L 202 Baågala, i 158 atyashti, i. 170 banij, ii. 197, 199 Audumbara, i 300 Auliatta (1), ii. 190 BanjulA, i 257 bdra i. 213 Anrva, iL 101 Antata (1), i. 387 Baramtla, i. 207 Barbara L 261 302; ii. 129 anttami, i, 387 arama ii. 38, 43, 47, 48, 50, 51, 53, Bardart, iL 8 bardt (1), i. 405 54 barh, i. 359 Araneya, L 215 Barhamsil, i. 200 Avanti, i. 298, 301 Ararta, ii. 244 B4rt, L 200, 201, 261 harkhu, i. $59, 348 (!) ; ii. 197 amaarpini, i. 871 Barod4, i. 403 avaTAs, i. 339 Bardf, i. 208 ; ii. 105 avyakts, i. 40 barah, i. 859 ayana, i. 356, 866 ; iL 118 Amans (1), i. 257 BarshAvar (Peshavar), i. 211 BArvanoat, i. 261 Ayurdaya, li. 228 BakArna (1), i. 300 ayata, L 175 bava, il. 197, 199 syutam, L 175, 177 Bav&rlj, i. 208

b=Budhi, i. 215 Bazana (!), i. 202, 205, 800 Benares, L 22, 156, 178; fi 146, Babrahtn, i. 208 147 BAdara, i. 802 badhatau (1), L 101, 103 Bhadatta (?), L 156 Bhadils (?), L 157 Baga, i 208 Bhadrakâra (!), i. 299 Baba, L. 261 Bhadra, i. 300, 301 behanu, ii, 178 Bablmarvara, i. 261 bhAdrapada, L 217, 216, 340, 858,

Bahrdj (v. Bibroj), L 205, 261 403 ; ii. 8, 98, 178, 175, 177, 180,

Behudses (1), i. 250 193

babudbinya, ii. 127 Bhadraiva, i. 249

bekahata (1), iL 208 Bhaga, L. 217, 356 ; ii. 121, 128

bala, ii. 226, 230 bhAgahara, ii. 30, 189, 190

Balabandbu, L 387 Bhagsvat, i. 255 BhAgarata, i. 121, 131

406 ALBERUNPS INDIA.

Bhagaratt, L 118, 120; ti. 177, 179, | bhoktyantara, IL 195 180 BLACOTS (1), L 342 bhaml (n), L 387

Bhaetratha, iL 148, 144 Bhomibara, L 203 bhps, L 179 Bbail-an, L 202 bhori, i. 175 bheikbukt, i 173 Bballa, L 303 Bbtrishona, L. 387

Bhanarajas (1), t. 156 bhorjs, L 171

bhanu, L 179, 215, 217 bhorloka, 1. 45, 232, 215, 238

Bhanuysáus (1), (cf. Bhtaunjs), L bhuta, i. 90, 92, 93, 178 BhAtapar, L 803

Bhtaakacchra (!), L 500 157 Bburanakof, i. 294

bbåra, L 165 bhuvarloka, L 45, 282, 238

bharz, L 130 bibats (n, i. 215 Bibat, L 201

Bbaradvajt, L 394, 898 Bihroj, L 209

bharant, L 218 ; il. 84, 123 btat (1), L 165, 166

Bbarsis, L 262, 294 Bitra, L 262

Hhrata, i. 29, 117, 182, 184 ; iL 1, Bitor, i. 259

147, 162 biya, ii. 197

Bharatavaraba, L 249, 294, 295, Biyaha, i. 259, 260

396, 297 Biyatta, i. 206, 259, 260

Bbarzara, i. 132, 215, 872, 398 Blv (? plava), iL 202

Bharms (n), ii. 120 Bodba, i. 299 bodbana, t. 215 bbarna (1), ii. 104 Bbarokaccha, L. 301 Brahmadanda, ii. 287

Bhatal, L 211 brahmadi (!), ii. 116 Brahmagupta, i. 147, 150, 153, 154, Bbatt, L 205 150, 168, 223, 224, 241, 248, 267, Bhatiya, i. 178 272, 276, 277, 279, 280, 282, 283, Bhattila, ii. 208 Bhsiul, L 260 812, 314, 835, 888, 369, 870, 872, 373, 374, 376, 377, 386 ; il. 4, 7, bhanmys, i. 215 bbantya, i. 387 15, 16, 17, 18, 19, 24, 28, 31, 46,

bbara, ii. 127 69, 71, 78, 74, 75, 76, 77, 78, 82,

bbarakotu, LL 245 90, 110, 111, 112, 186, 189, 192

Bbario (1) L 254 brabmahorktra, i. 831

Bharihya, L 181 brahmaloka, i. 233 brahman, i. 28, 64, 72, 77, 89, 92, bharishya-porans L 130 BhillamAl, L 138, 267 94, 100, 116, 118, 125, 129 ; hi SODa, L 181, 134, 153, 155, 157,

Bbtmapdla, ii. 18 159, 178, 241, 256, 268, 821, 322

Bhtmarathi, L 237 eeg., 831, 332, 842, 350, 352, 860,

Bhimsarns, i. 405 361, 365, 869, 380, 386; ii. 2;

Bhishma, L 185 ile of, ii. 28, 03, 99, 116, 116,

bhighmapaho uitrt, iL 177 118, 120, 145, 147, 199, 237

Bhoraprastha, i. 802 Brabman, era of, ii. 1

Bhorarardhana, i. 800 brahmana, i. 100, 102, 104, 121

Bhofs, i 300 brabmana (1), ii. 159

Bhotesbar, L 201, 200 brahminds, i 131, 221 xeq., 237 brahmanda-purtaa, i. 130 bhrumara, il, 92 Bbrigu, i. 77, 215, 291 Brabmanl, i, 119 Brahmapura, i. 308 bhriguputra, i. 216 brahma-purina, i. 180 bhrigaloks, il. 233 Brahmaputri, i. 887 Bhujara L 312 bbukti L 353 ; iL 80, 80, 195, 200, brahmarshi, i. 93, 247

205, 206, 207 Brahmardpa, i. 256 Brabmaskvarni, i. 387

INDEX. 407

brahmasiddhtnta, L 138, 153; table of contenta, i. 154, 228, 224, 267, Candrapura, i. 300

276, 852; ii. 110, 112 candrayaņa, ii 178

Brahmavaivarta, i. 1$1 cAntima (1), i. 344

Brahmin, ii. 95, 96, 98, 100, 109, Caraka i. 159, 162, 382

110, 111, 139 seq., 149, 151, 155, Carmadvipa, i. 201

179, 180, 181, 183, 185, 191 Carmakhandila, i. 800

Brahmottara, i. 262 Carmanvati, i. 257, 259; ii. 134

Brihaspati, i. 182, 898 Carmaranga, i. 302

brihaspativara, i. 213 Carshayab (!), i. 894

budha, i. 215 cashakr, i 334 scg., 337 ; ii. 52, 56,

budhavåra, L 213 Buddha, i. 40, 119, 121, 158, 243; caturyuga, i. 825, 354, 359, 368 seq., 189

ii. 169 872 scg., 386, 398 ; ii. 1, 2, 17, 18,

Buddhodana, 1. 40, 380 28, 57 seg., 186, 189

Budhnya, i. 387 catushpada, ii. 197, 198, 200

burla (?), i. 204 Cauiya, i. 299 candabi, ii. 197

c=candra, i. 215 caut, ii. 197

Cabrahasta (1), il. 120 ceshtabala, ii. 225

cadur (1), ii. 127 chandas, i. 186

caitra, i. 212, 217, 218, 358, 869, chidra, i. 178

394, 403 ; ii. 8, 10, 39, 48, 128, cikitas, i. 355

173, 176; fostirals, ii. 178, 186, Cina, i. 261, 803 ; iL 239

187, 193 Cipitanasika, i. 302

caitra-cashati (1), fi. 179 Ciranivasana, i. 308

Caitraka (1), i. 887 citrâ, i. 218, 342; ii 85, 121, 127 citrabhauu, ii. 127 cakbaka, i. 334 cakra, i. 114, 117, 118, 841 ; iL. 101, Citrakûta, i. 301 Citrângada, ii. 120

cakraavâmin, i 117 ; ii. 103 107 Citrapala, i. 257

Cakahabhadra (1), ii. 120 Citrakuta, L 257 Citrafala, i. 255 Cakabu, i. 261 Citrasena, i. 387 cikıbukha, i. 887 Cakabus, i, 281 C-n-d-sara (?), ii. 143 Cola, i. 301 ; ii. 239 cakahuaha, i. 387 calaketu, ii. 241 Co'ika (7), i. 301

Calitu (?), i. 137 Cyavana, i. 231

camtha, ii. 188, 184 DADHT, L. 178, 235 cAmara, i. 140 CAmunda, i. 120 dadhimanda, i. 235 dadhissgara, i. 156, 235 CDO, i 407 Dahala, i. 202 cana (?), i. 168 Cancûk i. 302 dabana, i. 178

candala, i. 101, 239, 344, 381 ; ii. dAbariya (1), i 344

137, 188, 153 dahia, ii. 197 Dahmala, i. 205 Candana, i. 259 Candartha, i. 260 Daibal, i. 208

candra, i. 178, 215, 218 ; ii, 21, 101 Daihak, i. 189

candra, i. 135, 215 daitya, i. 91, 231, 237, 247, 248,

Candrabhaga, L 259 287, 272, 279, 280, 364 ; iL 140

Candrabtja (?), ii 6 daityantara, i. 266

Candrâha, i. 206, 259 Dakcha, i. 54, 131, 291, 837

candrihargiņa, ii. 27 daksbakula, i. 357

candramana, i. 358, 354 dakshaputra, i. 387

candraparvata, ii. 143 dakshina, i. 290 Dakahiņatya, i. 300

408 ALBERUNIS INDIA.

dakabinayana, i. 856, 857 DAmara, i. 309 Dovikt, L 259 Dbaman, i. 894 damariya (1), i. 344 Damin, i. 254 Dhanahjaya, i. 281, 398

Damodara, i. 403 dhanihtha i. 218, 291 ; ii. 85

dAnadbarma, L 138 dhanishtha, ii. 122, 124 dhanu, i. 166, 220 Danak, i, 208 danava, i. 91, 281, 237, 248, 256, Dhanushman (!), i, 302

272, 830, 331 Dhanya, i. 254 Dhar, i. 202, 208 dånavaguru, i. 215 Danda, i. 308 ; il. 97 Dhara, i. 191

Dandshamar (f), ii. 176 dharani, i. 178

Dandaka, i. 800 dharma, i. 40, 132, 242, 291

Dandakâvana, i. 801 Dharmaranya, i. 800

dantin, i. 178 dharmasaverni, i. 387 Dhatri, i. 217, 238, 342 ; iL 127 Dantura, i, 301 dht, i. 178 Darada, 1. 261 Daraur, i. 200 Dhivara, i. 262

darbha, ii. 130, 131 dhôle, if. 183

Dardura, i. 301 Dhrishna, i. 387

Darra, i. 303 Dhritaketu, i. 837

Darvad, i. 209 Dhritarashtra, i. 108, 403

dafagicika, i. 157, 386 dhriti, i. 179

daśalaksha, i. 176 Dhritimat, i. 894 dhruva, i. 239, 241 dabam, i. 175 dhruvagriha (1), ii, 180 Dsameya, i. 803 Darra, i. 300 Dholika (1), i 261

Daśapura, i. 301 dhruvaketu (?), ii. 242

Dafaratha, i. 117, 806, 872 dhur& (1), ii 21

Dastrna, i. 301 dhurachadha, ii 21

Dafarna, i. 257 Dhutapâp&, i. 259

daśssahasra, i. 176 dhyAnagrahadbyaya, i. 155

Dasera (!), i. 302 asbalt, ii 182

Daseruka (!), i. 300 dikahita, i. 102

daara, i. 178, 342 Dilipa, ii. 174 dimasu, i. 359 dasta, i. 166 Diptp&, i. 262 Datta, i. 894 deotthint, ii. 177 Diptimat, i. 394

destntara, i. 312, 814, 815 Dirghagriva, i. 302

dova, i 90, 91, 82, 95, 159, 176, Dirghakeśa, i. 302

247, 248, 252, 256, 266, 272, 330, Dirghamukha, L 302

831 ; ii. 63, 66, 96, 89, 130, 140, Dirvart (Dravidt), i. 173

141, 177, 279, 280, 857 Dirvarideis, i. 178

devagriba ii. 178 diá, t. 176, 179

devaka, i. 330, 352, 369, 872 Divakanbar, i. 210

Devakirti, 5 158 Divakara, i. 158, 215, 217

Devala, i. 132 ; iL 235 Diva-küdha, i, 210

devaloka, ii. 233 divasa, i. 859

devamantrin, i. 215 DivArka (!). i. 301

DevAniga, i. 387 Divaspati, i. 387

devapitA, i. 215 divya, i. 42, 374 ; ii. 235

devapurohita, i. 215 divyahoratra, i. 829

devasini, iL 176 Divyatattva, i. 157

Devaárenhta, i, 887 divyavarsha, i, 359, 368; iL 2

Devata (1), i. 387 Diyâmau, i. 205

davojya, L 215 Dkinh (!), ii. 140 domba, i. 101, 102

INDEX. 409

Dramida, i. 302 draakshana, i. 161 Gaisitn (1), i. 137

Dravidadeśs, L 173 gaja, i. 178, 800

Dravina, ii. 101 Gajakarna, i. 281

drekkana, ii. 229, 233 Galava, i. 394

drigbala, ii. 225 gana, i. 407

Drihala (?), i. 300 gana (?), ii. 181

Drishadvati, i. 259 Ganaka, ii. 238

drishtibala, ii. 225 Ganapati (?), ii. 121

Drona, i 183, 162, 163, 164, 254, Gaparajya, i, 301

894, 398, 405, 406; ii. 101 ganda, L 203

Drûta, L 259 Gandakt, i. 259

Dodahi, i. 202 gandaota, it. 208

Dugum, i. 201 gaudba, i. 42

Dûgumpůr, i. 200 Gandhamâdana, i. 248, 249

duudubbi, ii. 128 Gandh&ra, i. 21, 259, 261, 300, 303

Dunpur, i. 206, 211, 317 gandharva, i. 89,91, 238, 247, 262, 303

Durga, i. 300 gandharva, i. 296

Durgt, i. 257 Gandharvi, ii. 142

Durgavivritti, i. 135 Gangh, 1, 200 seg., 203, 253, 254,

Durlabba, ii, 9, 10, 54 259, 261 ; ii. 14

durmati, ii. 128 Gang&dvara, i. 199

durtama (!), i. 371 Gangasâgara, i. 261

Durvisan, i. 404 Gangtaâyara, i. 201

Duryodhana, i. 138 Gangeya, i. 202

Duahyanta, ii. 174 gara, il. 197, 199

duvabl ii. 197 Garbha, i. 238

duve, ii. 174 garbhadhâna, ii. 156

Dvaipâyaua, i. 398 Garga, L 157, 342, 382, 390, 391 ;

drâpara, i. 872, 397, 398 ii 96, 110, 235

dvapara-yuga, i. 126, 378; descrip- garuda, i 114, 130 181, 198, 194,

tion, i 380; ii. 5; its begioning, 231, 253, 344 Gauda, the auchorite, i. 132 ii. 186 Dvår, i. 207 Gaudaka, i 301

dvijeśvara, i. 216 gaura, i 161

dvipa, i. 168, 233, 234, 285, 236, Ganra (1), ii. 143

243, 251 seq., 266, 295, 801, 888; Gauragrtra, i. 300

iL 144 Gaurt (Gaudi), L 173

dviavabhâva, ii. 220 Gaurl, i. 119; ii. 121, 179, 182, 183

Dyuti, i. 394 Gaur-t-r (gaurl-tritiyt), ii. 177, 179

Dyutimat, i. 394 Gautama, i. 131, 894, 398 gåyatrt, i. 147

EKAOARANA, i. 303 ghana, i. 140, 144, 146

ekam, i. 175 ghatt, i 334 seq., 837, 838, 349, 362, 368 ; iL 48, 52, 56, 189, 190, 195, ekanakta, ii. 172 200 Ekapada, i. 301 Ekavilocana, i. 302 gbatika, i. 279, 282, 286

Elapatra, ii. 120 Ghora, ii. 202 Ghorwand, i. 259 Ghosha, i. 300, 303 OA, 1. 140 ghritamanda, i. 235 gabhastala, i 230 Gabhastimat, i, 230, 296 Ghozak, i, 259

Gabhira, i. 387 Giri, i. 302

gada, i. 133 Girnagara, i. 301 git4 quoted, i. 29, 30, 40, 52-54, gagana, i. 178 70-72, 73-74, 75, 76, 78, 79, 80, gAihat (?), ii. 180 86, 90, 103-104, 122

410 ALBERUNTS INDIA.

Goibra (1), L 394 Homekotya, i. 801 go, i. 178 GodAvar, i. 203 hemanta, i. 357 Hcmattla, i 302 Godavari, i. 257 gokarna, i. 167 hemna, i. 215 Himagiri, L 249 Gomati, i. 259 himagu, i. 215 gomeda, i. 236 himamayôkhı, i. 215 gomedadvip, i. 235, 255 himara6mi, i. 215 Gomukhe, L 231 Himavin, L 302 Gonarda, i. 301 Himavant, i. 119, 246, 247, 248, Gorinds, L 299, 403 258, 261, 294, 295, 308 ; ii, 179 graba, i. 140 Hindho, ii. 129 grtha, i. 204 hindoli-caitra, i. 178 griahma, i. 357 Hiranmayı, L 249 Guda i. 300 Hirańyakadipu, i. 364 godhkmana (1), i. 158 Hiranyakahs, i. 231 ; ii. 140 gthantys, i. 344 Hiraayaroman, i. 394 gulma, i 407 homa, i 128 ; ii. 96, 183 gunakara, ii. 30, 189, 190 hort, i, 343 ; their namet, 344 gonAlahid (?), ii. 181 hork-pabca-hotriya (1), i. 158 Gupt, ii, 5, 7, 49 horadipati, i. 343 GoptakAla, ii, 7, 9, 49 guru, i. 138, 140, 145, 146, 215, Hridint, i. 281, 262 Hrihtkefa, i. 403 342; ii. 121 Hudvuda (1), i. 300 Guruht, i. 302 Hühaka (I), i. 800 guvkoa-hatrij, li. 182 Hona, i. 300, 302 Gozarât, i. 202 HutAás, ii. 127 Gralior, i. 202 buttiana i. 178

HÂDI (1), L 101, 102 HAhu (!), i. 257 IDAVATBARA, il. 126

Haibaya, i. 302 ikabu, i. 235

Hanssm&rga, i. 262 Ikshult, i. 257 ikshurasoda, i 235 Hamsapura, L 298 Hirabaura, i, 293 ikbânu (1), i. 178

Haramakot, i. 207 ikahvaku, i. 387

harbali (1), ii. 180 ilA (1), i. 230

Hari, i. 254, 342, 362, 898 liâvrita, i. 248

Haribhatta (1), L 141 Indra, L 89, 92, 98, 113, 119, 159, 217, 231, 252, 271, 292, 842, 857, Haripurusha i. 252 Hartta, i. 181 361, 366, 387, 893, 396, 398 ; jt.

harivamáa-parvan, i. 138 101, 102, 115, 127, 128, 175, 246

Harivarabs, i. 249 Indradvipa, i. 262

Harnba, ii. 5, 7 Indradyumna, i. 282

Haryâtman, i. 398 Indradyumnasaras, i. 262

hasta, i. 218 ; ii. 85, 121 Indragut, i. 842, 358 ; ii, 121

hastin, i. 140, 141, 148 İndramaru, i. 261 Indrant, L 120 hattha, i. 166 hauhava (?), ì. 408 Indravedi (v. Antarvodi), i. 211

Havisbmat, i. 394 indriys, i. 178

Harya, i. 394 Indriytņi, i. 43

Hayagrira, i, 231 Indu, i. 153, 178, 215; ii. 121

helt, i. 215 Irtri, i. 206, 260

Hemagiri, i. 802 Irkvatt, i. 259

hetnalarbe, ii. 128 Jictoyas (!), i. 894

homakâta, i. 247, 249 Ichtka, i, 300 ishtin, i. 102

INDEX. 4I1

iabu, i. 178 tivara, i. 31, 179, 861, 362, 363; il. jba, i. 215

127 Jringa, i. 302 jtgs, i. 220

JADORA ($), i. 202 Jodart, i. 211

Jagara, i. 230, 800 Jrala (1), ii. 202

Jabrivar, i. 260, 300, 302 jralana, i. 140, 141, 143, 145, 146,

Jailam, i. 206, 207, 259, 317 jyaishtha, L 217, 218, 840, 358, 408; 178

Jaimini, i. 127, 132 ii. 173; feativalo, ii. 179, 193 Jajahoti, i. 202 Jajjamau, i. 200 jyeabtha, i. 218; ii. 85, 86, 122 Jyotia, i. 394 Jajjanir, i. 206 jalaketu, ii. 243 Jyotisha, i. 300

Jalandhar, L 205 Jyotinhmat, i. 394

jalapradānika, i, 133 jalaaya, i. 178 KA, ii. 242 Kabandha (1), i. 231; ii. 238 jalatantn, i. 204 Ksbul, L 206, 259, 817 ; ii. 157 Jamadagni, i. 394 jambu, i. 235; ii. 129 Ksca (1), i. 261

jambudvipa, i. 235, 243, 251, 258 Kacoh, i. 208, 260 Kacchara, L 803 jans (?), i 163 Kacchlya, i. 800 janaloka, i. 232 JanArdana, i. 254 kadamba, i. 272 Kadara, ii. 129 Janarta (?), i. 231 Jandri, i. 202 Kadrů, L 252 Kaikaya, i. 802 Jangala, i. 299 Kaildea, i. 248, 302; ii. 142, 143 JAngala, i. 300 Janpa, i. 200 Kailavata, i. 302 Kaj, i. 260 Janujangha, i. 387 Jarmapattana (?), L 301 Kajūrāba, i. 202

Jaáu (7), i. 382, 397 Kakutstha, ii. 178 kalt, i. 160, 335, 337, 362 jataka, i. 100, 157 jatakarman, ii. 156 kâlabala, ii. 226 kalabhAga (?), ii. 231 Jatasura, i. 303 Jatadhara, i. 301 KalAjing, L 801 Kalaka, i, 302 Jajhara, i. 301 Jatt, i, 401 kal&oiaka, ii. 90

Jattaraur, i. 202 Kalanemi, i. 231

Jann (Yamuna), i. 199, 200 seg., KAlanjar, i. 202

206, 254, 259, 261 KalApagrama, i. 262

Jaur, Hindu king, i. 200, 209 kalarâtri, i 344; ii. 203

jaya, ii, 127 kalasi, i 166

Jayanta (?), i. 231 Kslatoyaka, i. 800 Kalavrinta, ii. 129 Jayantt, ii. 175 Jayapâla, i. 135; ii. 18 Kalayavana, ii. 5

Jimor, i. 209 kâlayukta, ii. 128 kali, i. 140, 382, 397 ; ii. 1, 198 Jimùta, ii 101 Jina, i. 119, 243 Kalidara, i. 262 Kalika (1), i. 261 jinaloka, i 238 Jishnu, i. 153 kalikala, ii. 1, 5

Jita, i. 894 Kalinga, i. 231, 298, 299, 301

jltu, i, 220 (? cettham) Kaliya, i. 231

jituma, i. 220 kaliyuga, i. 325, 373; description, i. 380, 897, 399 : ii. 1, 4, 17, jlva, i. 215, 358 jivabaranl, i. 844 18, 28, 59, 60; ita beginning, ii.

Jivadarman, i. 157, 184; iL 181, 182 186 Kalkoti, i. 300

412 ALBERUNI'S INDIA.

Kallar, ii. 13 Kalmiaha (1), ii. 121 karkra (1), ii. 181

kalpa, i. 54, 175, 279, 825, 833, 850, Karaekarı, i. 800

352, 354, 360, 362, 368 ccq., 886 1 KaratoyL, i. 259

li. 1, 15 seq., 17, 18, 28, 28, 57 .. g., kark (=khadga), i. 204 karkAdi, i. 356

kalpahargana, i. 368 ; ii. 118 118 karkadannu (khadgadanta!), i. 204 karkata, i, 220 kalpana, L. 868 KalyAnavarman, i. 158 Karkota, ii. 120

Kamalt, ii. 18 Karkotaka, i 247; iL 120

kAma, i. 140, 141, 146, 146 Karli, town, i. 317 Karma, i 262, v. Kramy kamandaln, i. 113 Kambala, i. 231, 247 karman, i. 321

KAmboja, i. 802 KArmaneyaka, i. 801

KAmrů, L 201 KarmAra, i. 231

Karaa, i. 840, 401, 403; ti. 180 karmendriyani, i. 44

KAmyakavana, ii. 8 Karna, i, 133

Kanaka, i. 302; ii, 237 Karnaprârarana, i. 262, 300, 802

Kanaahtharsiya (!), i. 803 KarnAta, L 173, 301 ; iL 185

Kanbarat, i. 208 Karn&tadesa, i. 178

KAnct, i. 301 karba i. 163

Kand, i. 203 kArttika, i. 217, 218, 858, 408 ; ii.

Kandakasthala, i. 301 98, 173, 177 ; fentival, ii. 182,

Kandhar (Gandbara), i. 206 186, 193

Kandi, i. 317 ; ii. 182. K&rttikeya, i. 54 Karûr, il. 6 ktndin, ii 128 Kardaba, i. 300 Kanik, ii. 11 ceq. Kanik-caitya, ii. 11 Karvata, i. 300 Kaśerumat, 1. 206 Kantr, li. 8 Kuihmtr, i. 21, 22, 108, 126, 185, Kanit, i. 200, 209 173, 174, 205, 206 neq_ 811, 258, Kaka, iL, 101, 238 903, 817, 391, 898; 1i 8, 9, 104, Kankata i, 301 148, 178, 181 Kannakara, i. 202 kAshth, i. 336 cg., 862 Kannars L 173 Kanoj, i. 21, 165, 173, 193, 199, KAG, L 299, 800

200 cco., 281, 817 ; iL 5, 8, 11, Kaśyapa, i. 216, 242, 252, 291, 894 ; ii. 96, 100 120 Kanthedhina, i. 302 Kasyapapura, L 298 Katantra, i. 135 kany4, i. 219, 220 KAtyayana, i. 131 kapklaketu, ii. 241 Kapil, i 72, 132, 255, 302, 321, katt, L 206 kaulavı, ii. 197, 199 325,397 KaumAri, i. 120 Kapishthala, i. 800 Kauninda, i. 309 karabhi, i, 167 Kaunkums, 1i. 238 karala, i. 344 Karamoda (D, i. 257 Kaurava, t. 408

karana, i. 155, 156, 157, 354 ; i. 194 kaurba, L 220 Kanmłaka, i. 801 #9., 197, 198, 200 Kaushaka (1), i 262 karanacūdmani, i, 157 Kauśiki, i, 259 karanakhandakhsdyaka, i. 156 karaņaparatilaka, i. 157 Kaustuba, i, 261 KAuverys, i 301 karaaapta, i. 157 karanzatra, i. 166, 817, 392; it. 7, KAvand, i. 259

54, 60, 79, 80 Kavara, i. 261

karanatilaka, i. 156, 313, 343 ; ii. 7, Karktadhana (!), L 302

50, 60, 80, 205, 206 KAvert, i. 257 Karint, i 261

INDEX. 413

Kavital, i. 206 KAvya, i. 894 koti, i. 92, 175, 178, 177, 236, 248,

Káyabish, i. 259 284, 308, 804

Kerala, i. 299 kotipadma, i. 176

Keralaka, j. 301 Krala (1), i. 300; ii. 202

Kesadbara, i. 302 Kramu, fi. 282 (v. noten)

Kemmri i. 231 Krathanaka, i. 231

Keśava, i 218, 361, 502, 403 Kratu, i. 890

Keśvara, i. 342 ; il 121 krauñca, i. 235, 302

ketu, ii 234, 236 Kravye, L. 802

KetumAla, i. 249 kricobra, ii. 172

ketarůpa, ii. 235 krimtia, i. 60

kha, 1. 178 333, 350 Kripa, i, 394

Khajara, i. 302 Krira-samudra, i. 301

khadira, ii. 99 krishna, i. 61, 231, 255, 257, 398

khandn, L. 158, 295, 302 kriebnabhtmi, i. 230

khandakhadyaka, i. 156, 312 ; 1i. 7, krisbnapakaha, 1. 859

46, 49, 60, 79, 83, 80, 37, 90, 91, Krishnavaidurya, i. 301

118, 119, 184, 187 krita, l. 178, 372

khandskhådyakatippa (!), i. 166 Kritamale, i. 257

khara, ii. 127 Kritarbjaya, i. 398

Kharapatha, i. 282 Kraubcadvipa, i. 235, 254, 301

khArt, i, 164 kritayuga, i. 118, 378; description,

kharva, i. 175, 176, 177 i. 379; ii. 182; ita baginning, i.

Khasa (1), i. 262 396, 397, 398 ; ii. 186

Khasha, L. 301, 303 kriti, i. 179; ii. 129

Khastha, i. 302 krittikA, i, 140-145, 218, 291, 344 ;

khendu, i. 179 ii. 84, 121

Kumbbakarna, ii. 3 Krisa (1), L. 383

Khyati, i, 387 krige, i. 220

Kihkind, I. 209 kroda, i. 344

Kikara, i. 282 krodba, if. 128

kilaka, il. 126 krodbin, ii. 128

Kimnara, i. 262 krośa, i. 168, 167, 275

Kimpurusha, 1. 249, 251, 262; ii. Krûra (1), i. 261 krūrākshi (1), i. 215 142 kinnara, i. 91 kshatriyn, i. 101, 104, 125, 247.

kinstughna, ii. 197, 198 388; il. 95, 98, 136, 155, 157,

Kira i. 259 161, 162, 170, 191

Kira, 1. 303 kahaya, ii, 128 Kshemadhorta (1), i. 303 Kirana, i. 287 Kirata, i. 262, 300, 302, 308 Kahetrapala, i. 120

Kirl (v. Kandi), ii. 182 kahtra, i. 235, 284

Kirpa (!), i. 257 koblrodaka, i. 235 Kshudramina, i. 302 Kirva (1), i. 257 Kisadya, i. 299 kebairitA (1), ii 186

Kishkindha, i. 300, 301 kshana, i. 335, 337 kah&ra, i. 235 kishku, i. 167 Kubata, i. 261 Kodara (?), i. 300 Kubera, i. 119; iL 115 Kokala, i. 803 Kucika, i. 303 Kolavana, i. 300 Kodaishahr (!), ii. 181 Kollagiri, 1. 801 Konkana, 1. 301 kudava, i, 162, 185, 184, 185 Kuht, i. 259 kona, i. 215 kuja, i. 215 Kops, i. 300 Kosala, i. 299, 300, 301 Kukura, i. 300 kola, L 356

414 ALBERUNPS INDIA.

Kaltrjak, i. 207 Lanbart, i. 259; ii. 8 Kulats L 201 Liarolint, L 257 Kulika, i. 844, 345 Laakt, L 209, 967, 268, 301, 803, Kulinda, i. 298, 300 kaltra, i. 220 306 acq., 316, 370 LArtn, i 209 Kulota, L 303 LArdesh, i. 205 Kalatalabada, L 802 LArt, L 173 Kulya, L 299 Lta, L 153, 268, 269, 280 KamArt, i. 257 Latadofs, i. 173 kumbha, i. 220 Lanh&vur (Lahore), i. 200, 203 kumbhakarna ii, 3 Laubtr, caatle, i. $17 Kumbhaka i. 821 Laukáyata, i 182 Kumuda, i. 255; ii. 243 lankikakals, li. 9, 54 Kumudvatt i. 257 lava, i. 336, 337, 862 Kunaths i $03 Kunjaradart, i, 801 Invana, i. 235

Kank, L 200 lavaņamushti, i. 156

Kunkan (Konkan), L. 203 lavanassmudra, i. 235 Likhita, i. 181 Kuntala, i. 299, 800 likhy4, 1. 162 Kupatha (1), i. 262 ligi, i 117, 131, 181 ; 1i, 102, 103 ktra-babaya (?), L 156 Litta (1), i 300 Kuraha, L 200 Kurava, L. 802 liyaya, L 220 locant, i. 178 kArma, i. 131 Lobavar, iL 8 kormacakrı i. 297 lokakAla, ii. 8 kurma-purina, i. 180 lokAnanda, i. 157 kuroh, ii, 66 Kuru, i. 182, 249, 262, 292, 299,380 Lont, iL 6 Lohartnt, i. 205, 208, 260, 316 Kuruksbetra, i. 308, 816 ; iL. 147 Lohitt, L 269 Kurura, L 254 Lohita, 1. 231 ; ii. 143 kuśa i. 235, 897 Lohitanadi, ii 148 Juásdvtpa, i. 236, 254, 325 Lohitya, I. 801 KunhikAns, i. 262 loka, i. 59, 282, 238 KuisprArarana, i. 262 lokAloka, i. 236, 237, 249, 284, 286 Kushtrt, i. 206 lokapAla, L 247 kuruma, i 140, 146 kusumakara, i 357 LApa (?), i. 257

Kusnmanagı, i 801 Kummapura, i. 310, 380, 835, 870 MADDHTANDA (1), iL 142 MAdhava, L 403 kutkra, L 120 Madhra (1), i. 300 kuthAra, i. 181 Madbu, i. 394 Kutt, i. 205 Madhushdana, i. 403 kuttaka, 1. 155 madhya (1), L 140, 141, 143, 144, 145, 146, 175 LA, i. 140 Ladds (1), 1. 205 madbyadefa, L 178, 198, 251, 290

laghu, i 138 madbyalaka, i. 59

Lagatürmân, ii. 13 madhyama, ii, 195

laghn, i. 145, 146 madbyamâya, ii. 228 Madra, L 302 Labore L 259 Madraka, L 303 Lahor, i. 208 lakrba, i. 175, 256, 284 madri (1), L 161 Madura, 1. 298 Laksbmt, 1. 54; ji. 182 1AlAbhakaba, L. 61 madya (1), L 252

Lamghn, i. 259, 817 ; ii, 8 Maga, i. 21, 121

Lampaka, i. 300 Magadba, i. 299 Magadha, i. 262, 298, 301 .

INDEX. 415

MAgadba, i. 255, 394 InAgha, 1. 211, 217, 218, 403 ; if. Malava, i. 173, 191, 202, 219, 299

177; festivala, 183, 186 800, 809, 308

magha, i. 218, 890, 891 ; li. 84, 121, MAlavartika, i. 299

124,180 Malaya, i. 200, 247, 257, 801

mahthhūta, i. 41, 42, 321, 882 Malayaparvata, L 248

Mahkoin, i. 207 MAlindya, L 301

Mahadeva, i. 54, 92, 93, 94, 117, Malia, i. 800

118, 119, 120, 121, 130, 131, 136, Malvart, i. 173

158, 176, 179 181, 292, 342, 861, Malvashau, i 173

862 ; iL 6, 102, 103, 120, 125, MAlyavant, i. 248

140, 148, 144, 147, 179, 180, 181, mAna, i, 166, 353, 355

182, 184, 192, 239 MAnabala, L 803

MahAgauri, i. 257 manaa, i. 44

Mahagriva, i. 301 minasa, i. 157, 247, 255, 256, 366 ; ii. 143, 245 Mabajambha, i. 231 MAnasottama, i. 256 mahajvala, i. 60 manda, i. 215; ii. 142 Mabak&la, i, 202 Mandaga, i. 255 mahAkaipa, i. 332 Mandagir, i. 208 mahakhya (?), i. 230 Mandabukor, i. 206 Mahamegha, i. 231 Mahanada, i. 257 Mandakint, i 257 ; ii. 142 Mandakkakor, i. 317 Mahankra, i. 259 MandavAhint, i. 257 mahsuavamt, ii, 179 mahâpadma, i. 175, 170, 247; ii. Mandavya, i. 157, 300, 802, 309 Mandeha, L 254 120 Maharasbtra, i. 299 mangala, i. 178, 216, 261

maharloka, i. 232, 238, 825 mangalab&ra, i. 213

MahArnava, i 802 manguniha (?), ii. 245

Mahtaaila, ii. 101 maniketu, iL 243 Maniman, i. 302 mahAśańku, i. 176 Manittha, i 157 mahAtala, i. 230 manmatha, ii. 127 mthatan, ii. 183 Manojava, i 387 Hahatavi, i. 301 mahatrij, ii. 183 mAnaartagu, ii. 183

Mahavika (1), i. 257 Manu, i 181, 132, 157, 179, 241, 886 ; his children, 387, 399 ; ii. Mabavirya, i. 386 Mahendra, L 242, 247, 257, 801 110, 111, 118, 127, 162 manushy&horâtra, i. 328 mAheya, i. 215, 300 mabidbara, L 178 manuahyaloka, i. 59

Mahiaha, i. 254, 299, 325 manvantara, i. 179, 241, 291, 359,

Mahoshntsha, i. 231 301, 367, 869, 872 seq., 386 seq .; their names, 387, 393, 898; ii. 1, Mabrattadeshu, L 203 MAhura, L 199, 202; ii. 147, 175 2, 17, 118, 119 Mara, L 261 Mahyt, i. 206 Maraka, i, 302 Mainaka, ii. 101 margana, i. 178 maitra, i. 358 Maitreya, i 63, 388, 397 margaśiraha, i. 217, 218, 358, 402, 403; ii, 10, 174; festivals, 182, Maitreyi, ii. 174 193 MaivAr, i, 202 makara, i. 204, 219, 220; ii. 93 marici, i. 163, 242, 890 Marikala, i. 302 makaradi, i, 356 M&rtgala, ii. 8 maia, li. 20 MAla, L 299 Markandeya, i. 54, 131, 241, 321,

Malada (1), L 300 840, 360, 372, 386 ; iL 2, 3, 64, 66

malamtas, ii. 20 markandoya-purana, i. 130 Maru, i. 261, 300

416 ALBERUNIS INDIA.

Marucipattana, i. 301 Marukucca, i. 302 Mucukunda, i. 281

Marut ii. 199 Mudrakaraka (?), L 290

mmtas, i. 179, 359 Mubran (Sindh), L 204

mtatrdham, i. 178 muburts, L 239, 287, 337, 838

mtcha, i. 180, 161, 162, 16$ 164; 341 ; their names, 842, 366; 118, 119, 243, 244 ii. 206 Kaahaka (!), i. 299 Mukta, i. 301 intla, L 218, 298 ; iL 86, 122, 179 masopavâe, ii. 173 Mathara, i. 302 Molarthans, L 298

Mathurk, i. 800, 308, 401, 408 ; Li. 5 molatrikona, ii. 225 Molika (1),'t. 800 matri, i 139, 140 mataya, i. 131, 300 MoltAn (mūlastina), i. 21, 116, 158,

MAtaya, i. 262 205, 211, 240, 260, 300, 802, 308, 817; ii. 6, 8, 9, 54 145, 148, 184 mataya-purana, L 130, 168, 285, 236, 247, 248, 251, 252, 254, 255, Mundla (1), i. 299

258, 261, 271, 284, 285, 286, 825 ; Mungiri, i. 200

ii. 62, 65, 101, 102, 142, 245 Munba i. 208

Man, i. 157 muni, i. 98, 178, 288 Munja, i. 281 mauaala, i. 133 Muru, i. 887 IDAyA, L. 344 Meda, i. 300 Musbika, i. 299 Muttai, iL, 178 Medbadhriti, i. 394 Mogha, i. 231 NABASA (I), i 387 Megbavin, i. 302 Mekala, I. 800, 301 Nabhaga, i. 394

Meru, i. 243 seg., 257, 265, 271, 274, nadt, i. 335

302, 303, 808, 818; aocording to naga, I. 178

the Buddhists 326, 827, 829; i. nâga, i. 91, 178, 247, 267, 344; ii.

82, 96, 129, 142 120, 197, 198 Nagadvipa, t. 296 menha, i. 220 mesbadi, i. 357 uAgaloka, i. 59

Mibran, i. 260 nagara, i. 173

mnlmamet, i. 182 Nagarapura, i. 156

mlua, i. 220 Nagarasamvritta, i. 257, 296

Mithilt, L NAgarjuna, i. 189

mithuna, i. 219, 220 Nagarkot, i. 260; ii. 11

Mirat (Meerut), i. 205 Nagha, i. 894

Mitra, i. 217, 242, 842; il. 122, 199 nagna, i. 121

MitrAkhya, iL 115 Nagnaparna, i. 301

mleccba, i. 19, 802; iL 187 Nabusha, L 93

modaka, i. 138 Naipal, i. 201

mokaha, i. 70, 80; ii. 188 nairita, L 290, 207, 301

moksbadharma, i. 138 bairriti, ii. 203 paisargika, ii. 215, 227 mora, i. 166 MrArarta, i. 249 naisargikabala, ii, 227

Mriga, i. 255 Naitika (1), i. 300 nakha, i, 179 mrigalanchana, i. 187; ii. 102 nakahatra, ii. 64 mrigafiras, ii. 86 mrigadiraha, i. 218, 842; ii. 84, 121 nakahatramăna, L $53, 854 nakshatranatha, L 216 mrigavyadha, ii. 91 Nakula, i. 403 mritazamjivan, L 254 Nalaka, i. 300 mrittAla, i, 230 nalt, ii. 135 mrityastra, i. 844 Mrityu, i. 898 NAlikera, i. 301

Mrūna, i. 261 Nalint, i. 261, 262 nalva, 1. 166

INDEX. 417.

nkmakarman, fl. 156 niahkabada (?), i 281 Nam&rur, i 203 Namiyya, L. 203 Nishprakamps, i 894

Namuci, i 231 Niśvara, i. 394 nitals, i. 230 nanda, i. 178, 231, 401; ii. 120 nandagoia, i. 401 ; ii. 148 nivra, i. 140

Nandant, i. 257 ulyutam, i. 176, 177

nandana, ii. 127 npipa, 1. 179

Nandanavana, i. 244 ; LL 96 Nrisimbavana, L 802

nanda-purtna, i. 180 Nor, i. 259

Nandavishthe, i. 303 nyagrodha, i. 256

Nandikeśvara, i. 93 nyarbuda, i. 175, 176

Nandna, i. 317 ny&yabhasha, i. 132

Nara, 1. 387 Narada, i. 116, 131, 237, 857 ; ii. ox, i. 173 odad (?), il 188 98, 101, 238 Naraka, i. 238 Odra, i. 301

haraloka, i 59 Narasirnha, i. 365, 366 PADA, il. 23 pâda, L 143, 144, 145, 147, 148, oarasimba-porkna, i. 130 NAriyana, L. 94,'106, 118, 129, 132, 150, 160

176, 193, 202, 216, 241, 242, 342, padamisa, il. 29

363, 395 xg., 398, 408; ii. 127, Padha, i. 300 padms, i. 114, 131, 175, 178 ; ii, 120 145, 167 NArtmukha, i. 302 padmaketo, ii. 244

Narmada, 1. 257, 259 Padmanabbi, L 403

Nasikya, i. 300, 301 Padma-Tulya (?), i. 300

nAtha, ii: 103 Padoâr, i. 209

Nanmand (?), u. 129 Pablava, L. 300 Paila, i. 127 navakanda, l. 297 navakhandaprathams, i, 294, 296 paitAmaha, i. 153 Pajaya (?), i. 257 DAVAD, 1. 178 paksha, i. 140, 143, 145, 146, 178, navin, ii. 197 359; ii. 118 netra, i. 178 ninagha i. 357 pala, i. 162, 183, 164, 105 paitán, ii. 181 nibavnan, i. 339 PalA4iot, i. 257 nikiarva, i. 175, 176 Niia, i. 231, 247, 249; iL 142 Palhava, i. 281

Niiamakha, i. 262 pail, l. 161

oimesha, i. 335 scq., 337, 362 Palola, i. 303

Nirabara, ii, 8 Paneshasta, i. 387

niraksha, 1. 267 panctht, ii. 197. Pancaia, i. 133, 282, 298, 299 Niramaya, L 387 NirbindbyL, i. 257 panca mataraa, i. 42

Niriahabha (!), i. 394 Pañcanada, i. 260, 302 panca-siddhantika, i. 153 ; ii. 7, 51 Nirmogba, i. 387 190 Nirmoha, i. 394 nirriti, 1. 358; li. 122 Pancadikha, i. 325

Niruteuka, i. 894 pancatantra, i. 159

Nisakara, i. 342 Panchir, L 108, 259

Nicara, i. 894 pauct, it. 197 Pandava, i. 178 Niselr i 259 Paadava-kAla, ii. 1, 5 nigeáa, i. 216 Pandu, i. 107, 132, 133, 199, 300, Nishaba (?), i. 262 Niahadha, i. 247, 248, 249, 257, 801 ; 880, 403 Pâņdya, i. 299 ii. 142 Panini, L 135 VOL IL 2 D

418 ALBERUNIS INDIA.

Penipat, i 205 patys, L 235 Payoshat, i. 257

Panjayamr (), L 209 Phalgulo, L 802

pantt, i. 166 phAlguna, i. 217, 218, 858, 403 ; il. 174 ; festivala, 183, 193 ppagrahs, i 218 Part, i. 257, 259 PhanikAra, i. 801

partka, ii. 173 Phonagiri, L 302

paramapada, iL 2 pllumant (?), ii. 129

parapadma, i. 176 pinda, li. 104

partrdba, i. 174, 178, 353 Pindtraka, iL 120

partrdhakalps, i. 338 Pingala, i. 187 ; ii. 128

Partiara, i, 44, 05, 107, 131, 157. Pingalaka, L 300

869, 388, 894, 897 ; ii. 96, 208. Pinjaur, 1. 205 Pipyala, L 257 235 PAraśava, i. 802 PlruvAna (1, L 158

Paraśurkma, i 380 Pisbika (!), L 257

peraávadha, ii, 203 piíaca, i. 89, 90, 92, 247 ; li. 256

PArata, L 802 Pita, L 255

partiasu, il, 128 pttabhtmi, L 230

Paroivara, i. 158 pitfmaha, i. 178, 361

paridhAvin, ii. 128 pitands (i), ii. 142

Partkaba, i. 77, 118 pitaras, L. 89, 93, 282, 248, 830, 857; ii. 121, 128, 188 parivatsara, ii. 125 Ptriyttra, L 247, 257, 259, 300 pitri, i 842

Parjanya, i. 217, 894 pitriloka, L 293, 236, 293 ; ii. 233

ParnAs L 257,259 pitrinAmahorktre, i $28

parthivs, i. 42 ; ii, 127 pitripaksha, iL, 180

partina, L 220 pitrya, i. 358

parvan, i. 132; il. 115 ecg., 119, 191 Pivara, i. 894

Parvin, L 259 plaksha, i. 235

parvata, i 140, 141, 143, 145, 140, plava, ii. 128

178 ; ii. 101, 192 plavang, ii. 128

Parratamaru, i. 262' Pojjihana (1), i. 300

parrati (n), li 181 prabhava, ii. 127

paścima, i. 290 Pradyumna, i. 118, 158, 898

ptsbinabhomi, i. 230 Pragjyotiaha, L 299, 801

Paáupala, i. 303 prabara, i. 337 Prablada, i. 366 psta, ii. 207 pAtala, i. 59, 230, 397; ii. 140 praj&pati, i, 89, 92, 94, 159, 291, 857,

PAțaliputra, i. 200 398; ii. 102, 121, 125, 127, 238

patahjala, i. 8 prakriti, i 41

Pataniali, L 27, 55-56, 68-70, 76, pramâna, i. 853

80, 81, 82, 87, 93, 132, 189, 282, pramAdin, ii. 128

234, 235, 236, 238, 248 pramsthin, ii. 127

Patheivar, L 299 pramoda, iL 127 Pramukha (!), i. 387 patti, i. 407 prAna, L 277, 834 xg., 387, 358, 361, pattrin, 1. 178 Paulisa, i. 159, 206 394

Paundra, i. 801 Prafastadri, L 802

Paurava, L 303 praina-gâdbAmana (1), i. 158

puaha, i, 217, 218, 858, 403 ; ii. praatha, i. 162, 163, 164, 165

174, 177; festival, 183, 198 prasthina, i. 133

pAraka, i. 178 prathama, i. 205 Prathangs (1), L 299 parana, i. 178 Pivani, i. 201, 262 Pratimaujas, i. 394

pavitra, ii. 180 Pratragira (1), i, 299 Praytga, i 200; ii. 170 241

INDEX. 419

priyafoitta, i. 355 prayuts, i. 175, 176, 177 Parvadeśa, i. 173 porrapbalguni, i. 218, 201 ; il. 85, prota, i. 90 Prishaka 1. 262 121, 128 půrvishadhs, i. 218, 291; fi. 85, pritant, i. 407 122. pritbivi, i. 288 Frithu, L 292, 894 půshan, i. 217, 342, 358; ii. 122

Prithodakasvamin, i. 158 pushandila (!), i. 181

Prithusvamin, i. $16 Pushkala, i. 254

Priyavrata, i. 241, 887 Pushkalavati, i. 302

Prosbtbapada, ii. 127 pushkara, i. 235, 254, 261 ; ii. 120

puhAi (1), ii. 180 pushkaradvipa, i 235, 255, 256, 284, 286 pohaval (?), il. 183 Publinga (!), i. 299 Pushpajati, i. 257

Pulaha, i. 990 pushya, i. 218, 291 ; ii. 66, 84, 121

Pulastya, i. 390 puthi, i 171

Pulindra, i. 300 puyattanu (1) ii. 184

Pulisa, L 153, 154, 168, 169, 224, 266, 275, 276, 278, 812, 313, 816, RADA (?), i. 231

$85, 889, 370, 374, 875, 876, 8775 Rabab, i. 261

ii. 4, 18, 19, 24, 31, 41, 42, 58, 87, râhu, i. 293 ; ii. 234

69, 70, 72, 74, 91, 187, 190, 192, râbucakra, i. 292 r&huurAkarana (1), i 157

Pulina-siddhAnta, i. 153, 177, 275, 208 ral, ii. 11

838 ; ji. 31 raibhya (1), i. 387

Pakala, i. 802 raivata, l. 387

Pakara, ii. 147 Raivataka, i. 302

Puleya, i. 300 raja, i. 162

Pulinde, i. 262 râjadbarma, i. 133

punarvam, i. 218; ii, 66, 84, 121, Rajagiri, i. 205, 208

176, 180 Rajanya, i. 302

Pufcala (1), L 157, 366, 367 rajarshi, i. 93

Punjadri, i. 303 rajas, i. 40, 399 Rajaurt, i. 202 puņyakala, ii. 187, 191, 192 RAjAvarl, i. 208 puraņas, i. 92; ii. 136 puraņa, i. 130, 233, 238, 264, 273, rakshasa, i. 89, 90, 91, 92, 231, 247,

283 ; ii, 110, 113 248, 262, ; ii. 3, 128

Purandara, L 887, 897 rakta, i. 215 raktabhumi, i. 230 pūrartaku, ii. 183 Purika, i. 301 raktâksha (f), ii. 128 raktâmala, i. 190 Pora, i. 262 pūrņimA, i. 348; il. 185, 197 RAma, L 117, 121, 166, 209, 258, . 306, 807, 310, 872, 380, 397 ; ii. purohita ii. 132 Purshavar(Penhavar), 1, 206, 259,317 8, 187

Pursbur (Peshavar ?), i. 338 RAmadt (1), i. 257 Ramayana, i. 307, 310; ii. 3 Puru, L 387 purusha, i. 31, 40, 321 Rameshar (?), i. 209

puruaha, i. 324, 332, 333, 850, 351, RAmsher (?), i. 209

860, 880 ; ii. 118 Ramyaka, i. 249

Purushada, i. 300 randhra, 1. 178 Rańka, i. 192 purushboratra, i. 392 Puruabaparvata, L 248 raaa, i, 42, 178, 188

PuruabAvar (v. Puraharar), ii. 11 rasttala, i. 230 raasyana, 1. 80, 188, 191, 193 parva, i. 290 pûrvabbådrapadA, i 218, 240; ii. rasayana-tantra, i. 156

85, 122 Rashtra, i. 301, 303 raÁmi, i. 178

420 ALBERUNI'S INDIA.

mémiketn (1), 5. 242 mbht-parvan, L 188 ratha, i. 407, 408 Sadkira, i. 361, 362, 363 rAtri, i. 359 Seddtnt (1), i. 257 raucya, i. 387 ntdbtraas, il. 128 raodra, i. 844, 858; ii 128, 241 stear f 178 raurava, i. 60 Ravana, L 306, 307, 380; fi S 8agur i. 20; ii. 143, 178, 169 rartam, ii. 183

ravi, i, 216, 216, 217, 842 ezbades, i. 408 Sahanyt, i. 202 ravicandra, i. 178 Rebha (n), i. 387 sahasram, i. 175, 177 cabasrarhón, i. 179 ropu, i. 102 Sabtwi (?), ii. 190 Ravanta, i. 119 rovati, i, 218, 291, 842, 369 ; ii. 66, Sahishnn, i. 394

85, 86, 122, 177, 180 Sahya, i. 247, 257

ric i. 128 Śailoda, li. 143 Rigveda, i. 127, 128 Ribanjar, i. 205 saiodbava, i 178, 261 Saintra, i. 153 Rikaba, i. 257 Rinajyesbtha (2), i. 898 Sairikirna (1), i. 301

Richabha, L. 801 ; ii. 101 Szirindha, i. 308

rishi, i. 93, 106, 180, 287, 289, 241, Saka, L 800, 802; IL 5, 8, 8

404 ; il. 96, 108 śaka, i. 285 dakadripn, i. 235, 252, 258 Riahtka, i. 257 Risbiks, i. 801 śakakkla, L 866, 390, 891, 392 ; li.

Risbyamaka, i. 301 6, 7. 9, 28, 128, 129, 188, 190

Rishyaśringa, i. 894 4akata, i 185

Ritedhaman, i. 387 éAkarayana, i. 185'

ritu (1), i. 178, 357, 359 ; ii, 118 Kakra, i. 358 ; 11. 122

RitukalyL, i. 257 Sakranala, ii. 128 sakti, i. 118, 119, 68 rodba, i. 60 éakuni, ii. 197, 198, 200 rodbakrit, ii. 128 śakvara, i. 241 rodhini, i, 218 rohint, 1 218 844, 401; ii. 66, 84, 96, galaka, i, 239

97, 99, 100, 102, 121, 175, 176, 177 Salila, i. 261

Robttaka, i. 308, 316 BAlkot, i. 317

Romaka, i. 287, 303 ialmali, i. 235

Romaka-alddbanta, L 153 éAlmalidvtpa, i. 235, 254

rudhim, i. 61 SAlv, i. 299

rudhirtndhs, i. 61 SAlvani (D, L. 300

Rudra, i. 94, 179, 342, 862, 363 ; ii. Saly, i 133

120, 121, 140 sam, L 371

rudraputra, i. 887 Samalvahana, i, 186

rukmiksha (1), ii. 129 atman, L 129

Râm, i. 268, 272 Simanta, ii. 13

Rumana (1), i. 299 Samatata, i. 801

Romimandala, i. 262 stmaveda, i. 127, 129, 896

rapa, i 42, 140, 178 Samays (?), i. $38, 357 ; ii. 133

rape-patica (1), ii. 179 SAmba, L 118

Ropaka, i 300 eAmba-purApa, i. 130

Rorasa (1), i. 261 S4mbbapura, i. 298

Rardhwabahn (!), i. $94 SAmbbapuruyatrt, if. 184

ruvu, i. 161, 162 sundhi, 1, 128, 864, 366, 369, 372; iL. 2, 17, 110, 183, 225, 226, 244

Baatst (1), i. 261 camdbi-astamans, i. 864

éabda, i 42 mrbdhi udaya, i. 864 mrdby Lrhás, i. 372, 573

INDEX. 42E

sarhitt, i. 157, 167, 298, 209, 820, 889, 391 ; iL 66, 88, 88, 92, 107, Sarayuśati (1), ii. 148

110, 111, 115, 123, 126, 145, 192, sarkara, i. 280 Barpa, ii. 129 235 Kamt, ii. 141 sårpa, i. 358

Samkara, ii. 147 Sarp&s, ii. 121

Samkarahana, i. 898 sarpia, i. 235

Samkhya, i 8; quoted, L 30, 48, 62, Saranti, i. 257, 281, 405; ii. 105,

64, 75, 81, 83, 89, 92, 132 142

samkranti, i, 344 ; ii. 188, 189, 190, Sarva, L 259, 261

199 sarvadharin, ij. 127

samnara (!), i. 295 sarvajit, ii. 127

eamndra, i. 175, 178 Barvart ($), ii. 128

Samuhuka, i. 262 Sarvatraga, i. 387 Śaryati, i. 887 Samvarta, i, 181; ii, 244 éaśalakabena, ii. 102 samvartaka, iL 101 Samvataara, i. 242; ii. 8, 9, 123, Śasideva, i. 135

125, 129 śaśidevavritti, i, 185

Samyamanipura, i. 271 śaśin, i. 178; ii. 115 śastra, ii. 241 danaiścara, L 215 fanaiścarabara, i. 213 sat, ii. 197 śatabhishaj, i. 218 ; ii. 85, 122 Sanaka, i. 325 Sananda, i. 825 satadyumna, i, 387 Śataka, i. 303 Sanandanatha, i. 325 Satakratu, i. 896 aandarnsaka, i. 61 śatam, i. 175 Sandan, L 209 SatAnika, i. 77 Sandt (?), ii. 142 Sangabila (épinkhala 1), i 158 Satarudra, i. 259 Śataśtrsha, i. 231 Sangavanta (!), i, 261 Sat&tapa, i. 131 angha, i. 40 Sankara, i. 94 Satavahana, i. 136

śakha, i. 114, 131, 801, 338; ii. Batin, iL 197

120 sattra, i. 344

ŚankhAksha, i. 231 satva, i 40

śańku, i. 188, 175, 176 Satya, i, 157, 394, 399

Sankukarna, i. 231 Satyaka, i. 885

Sankupaths, i. 262 satyaloka, i. 232, 233, 238

sânta (!), ii. 188 Saulika, i. 301

Santahaya, i, 387 Saumya, i. 89, 215, 296, 344, 358 ; ii, 128 SAntanu, L 107 Saunaka, i. 77, 113, 126, 830; ii. 145 santi, i 133, 887, Šantika, i. 302 sauptika, i. 133

sânta, i 858 saura, i. 215 saurábargana, ii. 27 Bapten, i. 178 sauramâna, i. 853, 354 saptarabayas, i 889 Sauvtra, i. 298, 300, 302 bara, i. 178 Sâva, i. 259 sAra, i. 113 áarabba i. 203 savala, i 60

Sarad, i. 857 ; ii, 93 SavaDL, L 894

śarada, L 117, 303 s&vana, L 328; ii. 21

Śaradhêna, i 802 sâvanthargana, ii. 27

éartáltimukba, ii. 190 sâvanamana, L 353

Sarasvata, i. 158, 300, 398 Savañjula, i. 257 Savara (?), i. 300, 301 Sarasvati, ii. 99, 142 sarAvall, i. 158 BAvarni, i. 387

Sarayo, i. 259; 11. 143 aavits, ii, 121 savitri, i. 216, 217, 398 ; ii, 121

422 ALBERUNTS INDIA.

cipakı i, 178 Boorvart (1), L 394 Attarhn, i. 178, 215, 216 Atamayůkhamálin, ii 125 sonAmukha, i, 407 fltarakmi, i. 218 Scaha, i. 231 Sahtkhys, i. 287 Siva, i. 181, 342, 862, 865 ; li. 128

Sotubandha, i. 209, 307 Sivapaura, i 261

Sotuka, i. 299 Gvaratri, ii. 154 BkanSe, i 118, 181 ; ii. 140 shadays, ii, 215, 227 Shakron (1), i. 257 skanda-purdn, L 130 akandhs, & metro, i. 144 Shamtlan, i 206 strt, i. 183 Sharvar, i. 200 śloka, i. 127, 132, 137, 147 Sharvat i. 259 shanhtrabda, ii. 5, 6, 123, 124, 129 Śmaśrudhara, i. 301

chat, i. 178 ; ii. 177 amriti, i, 131, 352, 872, 878, 374,

Shataldar (Satloj), i, 259, 260 386 ; ii. 110, 111

shatpanckáik, L 158 Sneh, i. 254

Shattamana (1), L 500 śokakrit, ii. 128

nbidda (1, ii. 89 Boma, i. 215, 216, 262, 253, 342 ; iL

Shilahat, i. 201 108, 128

Sbirabaraha, i. 205 somabara, L 218

Bhmthina (1), i. 259 Somadatta, i. 239

mibl (1), L 165 somagraha, L 216

Sibika, i. 801 Somamantra, ii 97

Šibira, L 301 Somanatha, I. 117, 161, 165, 189,

siddha, i. 93, 192, 238, 247 205, 209, 261, 857, 405; ii. 9,

siddhamatrika, 1. 175 103, 104, 105, 178

siddhanta, L 158, 155 ; of Puliss, soma-purana, i. 130

224, 266, 339, 874 ; iL 18 Somasoghma, i. 398

Siddhaptra, i. 267, 288, 805, 804 Sona, L 257 soabint, L. 344 siddh&rtha, ii. 128 Śikhi, L 262, 887 sparáa, L. 42

silatala, i. 230 spbuts, ii. 198

elmaratonnayanam, ii. 156 ephuttya, ii. 228

simha, i. 220 áravana, L 218; if 85, 90, 122 śrAvans, i. 211, 217, 218, 858, 403 ; Simbala, i. 301 ii. 98, 173, 176; fe tivals, 179, Simbaladvipa, L 253 Simbika (1), ii, 111 193

Sindh, i 173, 198, 206, 259, 281, Śrl, i. 118, 119 ; iL, 6, 199

270, 298, 300, 302, 310, 387 ; ii. Sriahara, i. 408

6, 8, 15, 48, 104, 129, 182 Śri Harba, ii. 5

Sipdhustgara, L 260 drimukha, ii. 127

Singaldib, i. 209 Sringtdri, i. 249

Sini, L 257 Sringarant, L 248 Śripala, i. 164, 240 Siprt, L 259 Sirva (1), L 257 Srtparvats, i. 248

Kiahys, i 127 Srishena, i. 153, 266, 376 ; ji. 111

dishyahitavritti, i. 135 Sront, L 257

4iAir, L 357 ardhara (1), i. 158, 334, 836, 344,

Sisumara, i. 231, 241, 242 861 ; iL 8, 120, 192, 201-208

Lisuplla, L 165, 840, 841 at4masa (?), 1. 387 Stambhs, L 894 eita, i. 215 ; ii. 239 sthanabala, ii. 225 Sita, L 249 Sitt, L 261 Strtrkiya, i. 802

itt, L 178 Sabthn, i. 394

fitadidhiti, i. 215 Sab&ra, L 209

śttakals, L 857 fubba, i. 344 éubbakrit, ii. 128

INDEX. 423

subhtnn, ii. 127 Soci, L 387, 394 rutala, i 230

Buddbodana, L 380' Sutapas, i 894 . Sudharmatman (1), L 387 Sutaya, i. 394

Sudivya (1, L 887 sttra, i. 158

fadra, I. 101, 125, 247, 802; iL 8, suvarņa, i. 160, 161, 182, 168, 164

95, 98, 136, 150, 152, 155, 157, Suvarnabhūmi, i. 303

163, 170, 191 Suvarnadvlpa, i. 210 ; iL 108

Sogriva, i, 156 IUVArDAvarņa, i 280

Subms, i. 300; ii 101 avådodaka i. 235

Soka (1), iL 120 Svamukha, i. 302 Śvâpada, i. 231 Sukht L 271 Suthapura, i. 271 Svargabhomi, i. 262

bukla, HL 127 svargarohaņa, 1. 133

śuklabbumi, i. 230 svarloka, i. 45, 232, 233, 397

éuklapaksba, i. 359 svArocisha, i. 387

Sukra, i. 132, 215, 358, 394; ii. svArociya, i. 387 Svastikajaya, i. 231 121, 199 fukrabara, i 218 Srat, ii 182 avlli, i. 218, 391 ; ii. 85, 99, 100, 121 Sukrita, i. 262 Sukriti, L 394 Svayambhû, I. 398

Sukahetra, i, 387, 894 tvayambhava, i. 241, 337

Sukti, L 257 Bvata, L 248 ; ii. 142

Suktibam (1), i, 247 évatakatu, ii 242

Saktimatt, i. 257 Šyamaka, i, 303

Sukorda, i. 261 Syâvabala (I), iL 208

étla, i 119, 240 Śladanta, i. 231 TAITILA, iL 197, 199 Takeshar, i. 208 ; iL 8 Solika, i. 300, 302 Sumali, i. 231 Takabaka, i. 231, 247 ; ii. 120

Sumanas, i. 255 Takshaśila, i. 302

Sumantu, i. 127 tala, i. 290

Sumedhas, i. 394 tâla, i. 167, 230

Suunkm, i 206 Talahala, L 302

iuoya, i. 178 talaka, i. 188

Suprayoga, L 257 Tarakruti (1), L 302

aura, i. 2 Talakona (1), i. 300

Suraaa, i. 257 Talikata, i. 301 Tâmalipta, i. 262 aurakahas, i. 231 Stsena, i. 299, 800, 302 Tamaliptika, i. 301 Tamara, i. 262, 800 Suraahtra, i 300 Surejya, ii. 127 tamas, i. 40, 237, 399 Tamasa, i. 257 suronu, i. 251 tâmasa, L 800 sari, L 217 tAmasakilaka, ii. 234, 238 forpa, i. 163 Borpakarna, i. 300 tâmbiru, L 220

ŚorpakAraka, i. 800 Tamra, i, 259 Tamraliptika, i. 299 otrya, i. 179,215 Sorykdrl, I. 301 TAmraparņa, i. 301

stryaputra, L 215 Tâmravarnâ, 1 257, 296

Surya-siddbAnts, i, 163 Tana, i. 203, 205, 209, 298

Susambhavya, i. 387 tanduâ, i, 204 Tâneshar, L 117, 199, 206, 300, 808, Sultnti, 1. 387 Sushmin, i. 254 816, 317 ; il. 103, 146, 147

sůtaka, L 355 Tangana, i. 808

sutala, i. 280 Tankana, i. 301 tantra, i. 155, 156

• 424 ALBERUNI'S INDIA.

Tanvat (P), L 201 Trilocanapdla, f. 18, 14 tapane, L 178 Tipudframs, i. 301 Trinetra, i, 308 tridmhiaka, ii, 228 Tapasrin, L 894 Tipl, L 257 Tripava (D, i. 257

Tapodhriti, i. 394 Triporantika, i. 248 Tripuri, i. 801 tapoloka, L 232 TristgL, L 257 Tapomorti, i. 894 Trikira, L 231 Taporati, L, 894 Trivikrama, f. 403 taptakumbba, i. 80 Tira (1), i. 803 ; iL 84 Trivrisha, !. 898 triya, ii. 197 Tirakákaba, L 231 truti, i, 335 xq., 897, 862, 363 tarana, ii. 84 Tukbara, i. 261, 302 tArana, ii. 127 tarL i. 171 tul, i, 165, 219, 220

tArkahya-purtna, i. 130 taladi, L 357

Tarojanapkla, ii. 18, 14 Tumbavana, i. 301 Tumbura, i 300 Tard, L 201 - Tartpans (n), i. 300 Tungabhadrt, i. 257

Taakira, il. 288 Turagaona, i. 802 Târan, i 208 tattva, i. 44, 179 Tvashtri, L. 217, 342, 358 ; ii. 117, Tattradaritea (!), i. $94 121, 127 tankshiks, L 220 Tavalleshar, i. 208 Tharpura (1), i 800 UDAKA, i. 136

thohar (Sindb!), L 192 Udayagiri, i. 301

Tiaurt, I. 202 Udbhira, L 300

tikanl- (1)-yAtrt, L 158 Uddebika, i. 800

Tilltta (1), i. 300 udruvaga, i. 220 Udunpůr, L 178 Tilvat, i. 201 Timingilt4ana (1), L 301 udvataara ii, 125

tiryagloka, i. 59 Udyânamarúra, i. 262

Tishya, i. 254, 372, 380 udyogs, i, 133

tithi, i. 179 ; ii, 194, 195, 201-203 Ugrabbûti, i. 135 Ujain, i 189, 202, 259, 298, 301, Tobd, L 257 tola, i. 160, 162 804, 808, 811, 318, 816

trahagattata (1), il. 192 Ujjayiot, ii. 241

traht, trobt, ii. 197 Uiyânda (1), i. 187

Traipura, i. 300 Umadevi, i. 54

tranjai, ii. 182 Ummalnara, i. 209

trisantya, i. 344 ûna, ii. 21 tnarâtra, I. 864 ; ii. 21 ; univeraal or trayam, i. 178 partial, 23, 25, 34, 37, 186, 187, Trayytruna, i. 898 192 trott, i. 372 tretâyuga, i. 253, 373, 897, 898 ; ii. Unjara (#), i. 231

188 UpakAna, i. 282

TridhAman, L 898 upari, i. 290 Upavanga, L. 801 Tridiva, i. 257, 262 Trigarta, i, 300, 802 upavasa, ii. 172

triguna (1), i. 178 raga, i. 262

tribarkacha (!), ii. 191 Urdbabtshau, i. 200

tribaspaka (1), ii. 191 Ordhvakarna, i. 301 Urdvakujn, L 231 trijagat, i. 178 trikAla, i, 178 Urja, i. 394

trikațu, i. 178 Urur, i. 887

Trikota, i, 248 urvarL, i. 178 uéanas, i, 77, 131, 398

INDEX. 425

ushnakAla i. 357 Vaivarvata, i. 271, 887

Uahtrakarna, L-263 Vajasravas, L 898

Uakala, 1. 801 vajra, i. 119, 236, 241, 321, 360,

at&masa (}), i. 387 886; ii. 2, 3, 65, 203

Utkala, i. 300 vajrabrabmabatya, Li 162

utkriti, i. 179 Vaka, i. 299

Utpala, i. 157, 158, 298, 884, 336, vakra, i, 215 ; ii, 101

887, 861, 367 Valikhilsa, i. 395

Utpalavinl (1), i. 257 Vallabha, i. 192, 193, 209 ; ii. 5, 6 Vallabbi, i. 192; ii. 6 ntaarpini, i 871 Uttama, i. 898 Valmtki, i. 398; ii 3

Uttamsajas, i 387 VAmana, i. 129, 181, 396, 408

Uttamarna, i. 300 vamana-purâna, i. 130

UttAnapada, i. 241, 242 Vamdavara, i. 257

uttara, i. 290 Varháca (L), i. 394

uttarabhadrapada, i. 218, 342 ; ii. vAna, i. 178, 300

85, 86, 122, 127 Vanarajya, i. 803

uttarakbandakhadyaka, i, 156; ii. Vanangha, i. 302

87, 90, 91 VAnavast, L 301

uttaraktla, i. 357 Vanavâsika, i. 299

Uttarakurava, i. 302 Vanga, i. 801

nttaramanasa, ii. 142 Vangeya, i. 299

Uttaranarmada, L 300 Vaoupadevaś-ca, i. 887

uttaraphalgunl, i. 218 ; ii. 84, 121 Vaprivan, i. 398

uttardshadha, i. 218 : iL 85, 122 Vapushmat, i. 394

uttardyaņa, i, 356, 357 ; ii, 169 var (1), ii. 10

Uvaryahår (!), i. 200 vâra, i 355 Varâba, i. 131

VADAVĀMUKHA, L 266, 267, 269, 272, Vardhamihira, i. 23, 54, 117-121,

278, 279, 302, 807, 327 ; il 201 153, 157, 158, 162, 164, 166, 167,

Vndayanala, ii, 104 219, 220, 266, 268, 272, 276, 297,

Vadha, i. 800 299, 300 xg., 820, 848, 389, 891,

Vadbra, ii. 101 392; ii. 7, 51, 66, 70, 86, 87, 88, 89, 92, 95, 103, 107 aeg., 113, 115, vahini, i. 407 Vabirgira, i. 299 116, 118, 123, 145, 190, 208, 235,

Vahltka (!), i. 300 239, 240

vabnijvala, i. 61 varâba-purâņa, i. 130

Vaidarbha, i, 300 Vardbi, i. 120

Vaideda, i. 300 Varaka, i. 394

vaidbrita, ii. 204, 206, 208 VardhamAna, i. 801

Vaidika, i. 300 varga, i. 297, 298 Vâricara, i. 301 Vaidrya, i 301 Vaihand, i 206, 259, 817 varna, i. 100

Vainy&, i. 257 varsha, i. 359

Vairahma (1), i. 344 varsbakâla, i. 211, 857 ; iL 94

vaisakha, i. 217, 218, 358, 403 ; ii. Varuna, i. 217, 242, 271, 292, 342,

123, 173 ; festivala, 179, 182, 186, 858, 372 ; ii. 92, 115, 122

193 varunamantra, ii, 97 Varvara, i. 261 vaishnava, i. 857 Vaishnavi, i. 120 Vasâ, ii. 241 våsara, ii. 118 vaisvânara, i 178 Vaiśarpayana, i. 127 vasanta, i, 357 ; ii. 179

vaiáya, i. 101, 125, 247, 302 ; ii. 95, Vasâti, i. 802

98, 136, 165, 157, 170, 191 Vasavas, ii. 122

vaitarant, i. 61, 257 Vasishtha, i. 115, 131, 225, 239, 288, 280, 340, 390, 894, 398 ; ii. 66, 96

426 ALBERUNTS INDIA.

vaishtha-addbinta, L 155 vam, L 178, 291, 842, 894 Vinatt, i. 252, 258

Vasndera, i. 29, 82, 104, 107, 122, Vindyaka, i. 120, 184

185, 165, 199, 218, 251, 840, 841, Viodhya, i. 247, 248, 257, 262, 501 ;

352, 362, 897, 898, 400 ccg., 401 iL 92

seg., 403 ; il. 106, 137, 138, 147, Vindbyamoli, i. 300

148, 175, 176, 177, 178, 180, 181, Vipafoit, i. 387 Virj, i. 241

Varoki, i. 231, 247 ; si. 120 182 Virajan, 1. 887, 894

Vasnkra, L 128 Viraneana, i. 381, 862

Varuman, i. 302 Virata, i. 180 Viribor, i. 342 vata, il. 170 Vatan, i. 299, 800, 301 Virocso, i. 117, 231, 806 ; iL 11

vAyara, i. 290, 297, 802 ; ii. 202 virodhin, it 127 vilakha, i. 218, 251, 291, 891 ; il. Vayavamantra, iL 97 VAyu, i. 292 ; ii. 66, 121 86, 121 vitla i. 280, 844 vayo-purana, i. 41, 130, 168, 194, 280, 281, 232, 284, 289, 241, 248, VilAlA, i. 269

251, 257, 258, 271, 287, 295, 298, VilAlyakarana, i. 254

299 acq., 387 ; ii. 62, 63, 65, 142, vidseana, i. 61 vinba, il. 159 245 veda, i, 29, 81, 104, 125, 181, 182, Vinhnu, i. 94, 118, 130, 181, 218,

178, 848, 893, 896, 898 ; iL 21, 217, 231, 242, 253, 255, 858, 865,

22, 82, 95, 96, 110, 111, 131, 136, 382, 888, 394, 397, 808, 403 ; it

189, 140, 152, 179 107, 120, 121, 122 Vihnucandra, i. 153, 286, 876 ; il. Vedabthn, i. 394 Vedasmriti, i. 257, 259 viabnu-dbarms, i. 54, 118-115 (n), 111

Vedaśrt, i. 384 126 (1), 182, 216, 217, 218, 241, Vedavatt, i, 257 242, 287, 288, 291, 821, 329, 831, vega, i. 844 882, 844, 858, 354, 858, 860, 872, Vant, i. 301 379, 380, 881, 886, 387, 398 ; li. VOnavyAmn, i. 898 VenumatL, L 259, 302 2, 3, 21, 64, 65, 102, 121, 140,

Venva, i. 257 145 (1, 174, 175

Vibht, i. 271 Visbņupada, ii, 142

vibbava, ii. 127 : viahu-purana, i. 47, 60, 61, 68, 77, 128, 130, 131, 230, 232, 235, 297, VibhAraripura, i. 271 238, 248, 254, 255, 256, 262, 825, Vicitra, i. 387 387, 388, 398; iL 62, 105, 181, Vidarbha, i. 801 Vidtsint, i. 259 182 Vishpnputra, i. 387 VidhAtri, i. 217, 288 VidiśL i 267, 259 vinhți, ii, 197, 199

vidyadhara, i. 91, 262; li. 92 vishuva, it. 188 viśra, i. 179, 842 vidyut, i. 42 Viśvakarman, iL 121 Vidyuifihva, 1. 231 vigbati, i. 854 Viśvamitra, 1. 239, 822, 894

vijaya, ii. 127 Visvartipa, ii. 238

Vijayanandio, i. 156, 348 ; ii, 49, 90 viśvkvaan, ii. 128 vikvedevkh, i. 857, 358 ; ii, 122 Vikaca, ii. 237 vitala, i. 280 viktrin, iL 128 Vitast, ii. 181 vikrama, li. 127, 128 Vikramáditya, i. 189 ; ii. 5, 6, 7, 129 vitasti, i. 167

vikrita, il. 127 vitta (1), i. 215

vilambio, ii. 128 Vitteśvara, i. 166, 392

Vimalaboddhi, i. 158 vivabapatala, i. 158

vintal, i. 837 Vivarna, i. 262 Vivaavant, i. 217

INDEX. 437

Virirhia, i. 254 yafurveda, L 127, 128 viyat, L 178 Viyattha, i. 259 yakaha, i, 89, 91, 92, 247, 262 Yams i. 181, 178, 271, 291, 292, Vodha, i. 325 803, 342 ; 1. 115, 122 VokkAna i. 302 Yamakoti, i. 287, 268, 272, 803 Vribaspati, i. 181, 215 Vrika, L 299 yamala, i. 178 Yamunt, i. 308, 318 Vrikavaktra, i. 281 YAmuna, i. 800, 302 vriścika, i. 220 vriścikaloka, iL 238 y&mys, l. 858

Vrisha, i. 301, 387 YAmyodadhi, i. 301 Yasoda (1), L 882, 397, 401 vrisbabha (1), ii. 127 Vrishabadhvaja, i. 300 Yadovati, i. 302 yatra, ii 178 vrinhan, i, 220 Yaudheya, i. 303 vrinhnt, L 844 Vritraghnt (?), i 257 yava, i. 160, 162 Yavana, i. 153, 158, 300, 302; il. 5 vritta, i. 145 YAvana-koti, i. 306 vuhara (1, il. 104 Vyadi, i. 189-191 Yavasa (1), i. 261

VyAgramukha, i. 300 Yayati, li. 174 yoga, ii. 191, 204 seg. vylkarana, L 185 ryakta, 1. 41 yogayAtra, i. 159

Vyålagriva, i. 301 yojana, i, 153, 167, 188, 189, 224, 234, 236, 244 seg., 265, 311 ; ü. Vytna (1), fi. 121 65,67 Vjas, i. 44, 104, 107, 108, 126, 127, 131, 132, 184, 171, 288, 340, 341, yojanas of beaven, ii. 72, 74, 79

852, 369, 894, 397, 808 Yima, i. 119 Yudhishthira, i. 340, 841, 890, 891, vykaamandala, i. 238 408; iL 3 rytstatrairtdika, i. 313 vyatipata, ii. 204, 206, 208 suga, i. 298, 387, 372 scg., 897 ; ii.

Vyayn, i. 394: ii. 127 1, 2 124 ; their begionings, 188, 137

YADAVA, L 133, 404, 405 rak4, L 162 Yuktann, i. 894 raht, ii. 197 Yajna, 1. 242 Juvan, ii. 127

Yajosvalkya, i, 128, 131, 192; i'. 174 ZABAJ, L 210

yajnopavtta, i, 131 ; ii. 130, 138 Zanba (1), Hi, 142 Zindutunda (1), i. 261

INDEX II.

AIDALKARIM Ibn'Abt Alanjs', i.264 "Abdalith 1bn Almukafff, i, 159 Aphrodisian, L 407 Apollonius, de causis rerum, i 40 'Abd- Jman'im v. Abd-8thl, L 5 Arabian astronomy (innar stations), Aba-Abmad 1bn Catlaghtagin, i ii. 81, 90 817 Arabian metric, L 138, 142, 144 Abo-alabbas Alerdnshabrt (r. Aler- Arabian traditiona, i. 170, 185 anshahrt), i. 6 Arabio literature, transiation of Ca- Abo-al'aswad Al-du'alt, i. 138 raka, i. 159; Kalila and Dimna, Abo-alfath Albuatt, 1. 34 tranalation from the Indian cor. Abo-alhasan of Ahvaz, ii. 19 Abo-Bakr Al-ahibit, i. 37 rupt, L 162

Abt-Ma'shar, i. 304, 325 Araba, i 802; different forms of

Abu-Sahl 'Abd-almna'im Ibn 'Alf matrimony with them, i. 108; their idol, i. 128 1bn Nůh Al-tifltst, i. 5, 7 (also Aratus, i. 97, 383; scholia on the under 'Abd-alman'im) Phanomena, i. 97, 384 Abu-Ya'kub of Sijistan, his book Archimedes, i. 168 Kashf-almatjab,i. 64 Ardasbtr lon Babak, i. 100, 109 Abo-Yastd Albistamt, i. 88 Ardiya, Eranian, i. 249 'Adud-aldaula, ii,'157 Aristotle, letter to Alexander, i. 124, Afghans, i. 208 225, 226, 232 ; quauh depbaais, Afriaitb, i. 804 i 820 Al-erknsbahri v. Abů-alabbas, i. 6, Arjabhar, ii. 19 249, 326 Alexander, story of bis birth, i. 96 Arkand, i. 312, 310 ; ii. 7, 48, 49

Alexander of Aphrodisias, i. 820 Asclepius, i, 222

Alexandria, i. 153 Asvira, i. 207

AlfazArl, i. 165, 303, 314, 816 ; ii. BABTLONIA, iL. 153 15, 16, 17, 18, 23 Bagdad, ii. 15, 67 Al-bajjaj, ii. 153 Balkh, i, 21, 260, 304 'Alt Itn Zaio of Tabaristan, i. 882 Aljshiz, L 204 Barbatagin, iL 10 Bartdish, Eranian, i. 260 Al jaihant, book of routes, i. 240 Barmecides, L. 159 Alkhalil lbo 'Abmad, i. 138, 147 Al-khwarizmt, ii. 79, 114 Barabawår, L 109 Barzakh, i, 63 Alkindt, ii. 200, 201 Barzbya, i. 159 Alma'mura, i. 21 Bashahar 1bn Burd, ii, 131 almanac from Kashmtr, i. 391 Almanstr, Khalif. Bhatta-Sbab, i. 207

Al-manştra, i. 21, 173, 193, 205, Bhattevaryân, i. 207

260, 816 ; ii. 6 bist (=vishti), ii. 201

Ammonius, i. 85 Bolor mountaios, i. 117, 207 Bolor-Shâh, i. 208

INDEX. 429

Buddhista, 1. 7, 21, 40, 91, 121, 156; their writing, 175; their conoo- Ghurrat-alatjat, fi. 90

graphio viown, 249, 826; iL, 160 Ghuz (Tarka), ii. 168

Bushang, L 299 Gilgit, I 207 Girnagar, Eranian, i. 250

CALENDAR of Kashmir, iL 5, 8 Girbth, i. 109

Ceylon, i. 209 ; pearla, i. 211 Goepel, quoted, i. 4

cheso, i. 183-185 Greek legends, 1. 06 Greek philocophy, i, 7, 24, 88 China, it 104 Chinese, ii. 239 Greck traditions, L 105, 112, 143 ;

Chinese paper, i. 171 origin of the alphabet, 1. 172; on

Christianity, i. 6, 8 the astrolabe, 1. 215, 219, 220,

Christians, their use of the words 222; on the Milky Way, i. 261,

Father and Son, L 38 289; on the frat meridian, i. 304;

Christian viewn, i, 69 on the chariot of war, i 407

Christians, i. 94; ii. 186 Christian traditiona, i. 151, 161 HARKAN, 1L. 52 Hobrew, i. 86, 37, 38 clepsydræ, i. 337 Commodus, Emperor, i. 128 Herbadh, i. 109

Constantine, Emparor, il. 161 Hindws, their language, i. 17; classical and vernacular, i. 18;

DAIBAL, i 208 shortoomings of manuscript tra- dition, i. 18; the metrical form Daizan, i. 109 Dosk, Peraian, i 163 of composition, i. 19 ; their

Denara, 1, 809 aersion to strangers, i 20;

Dibajat (Maledives, Lacoadives), i. their systems of matrimony, i. 107; the balance they use, i. 233; iL 106 Dirhama, i. 160, 188, 164 184; relation between authors

diz (Persian), i. 304 (writers) ond the netion at large, i 265; thoir architecturo, ii. 144 Hippocrates, bis pedigree, i. 379 EMPEDOOLES, i. 85 era of the realm of Siodh, ii. 48, 49 Homer, i, 42, 98

era of Yazdajird, il 48, 49 Huns, ii. 239

Eranian traditions, 1. 249 ErAnshahr, i. 54 1RN ALMUKATTA', i. 264

Erichtbouiua, i. 407 Imptla, name of the rhinoceroe with the Negroes, i. 204 lodis rainfall, 1. 211, 212 FARTAZA, i. 299 faraakh,; Persian, i. 167, 811; if. lsfandiy&d, i. 193

87,68 lelam, sectarian viowa, i. 81, 268, 264 Ispahbad (of Kabul), ii. 157 Fulûs, 1. 160 'lyds 1bn Mu'awiya, ii, 156 Fomaoj, i. 299

GALENUS, i. 222, 820 ; de indole ani. JABRITYA, & Muslim sect, i. 81

mæ, i. 123; book of specches, i. 95 ; Jolam Ibu Shaiban, i. 116 Jam, i. 804 book of deduction, i. 97; com- Jewish tradition on the tetragram- mentary to the Apothegma of Hippocratea, il. 186; Protred- maton, i. 178

ticus, i. 84; commeutary ou the Jows, i. 6, 109 ; il. 240

Aphorisms of Hippocraten, i 35, Jobannes Grammaticus, refutatiot

36; Kard yérn, i. 127, 151 of Proclus, i. 36, 65, 226, 231 ; ii.

Gauge-year, ii. 2, 7, 25, 31, 39, 44, 47, 48, 50, 53 Jon, Arabised form of yojana, i. 167 171

Ghaso, i. 117, 206, 317 Jorjâo, i. 258, 305 ; ti. 182

Ghazntu, ii. 103 Jūzajân, i. 808

• gkdr, measuro in Khwarzim, i, 166 KABUL, i. 22 ; ita history, ii. 10, 157

430 ALBERUNTS INDIA.

Mu'lwiya, Khalil, i. 124 KAf mountain, L 193, 249 Mubammad Iba Alktsim, the con- KaikA'os, i. 304 quaror of Sindh, i. 21, 116 Kaikhusran, L 304 Kaltla and Dimns, L 159 Mubammad Tbo lsbak, of Serakbs,

Kaodt (, ii 182 ij. 15, 16, 18

Kanrdis, L, 804 Hubammad Ibn Z kaariyst Al-rtat,

Kans-al'ibyt, title ef a book of the L 819

Manichpana, quoted, L 39 Hubammira (Baddhists), L 380 Mukl, Arabic, a tree, L 208 kardajat L 245, 275 ; il. 205 Mulamma', Arabic, kind of wood, i. Karmatianı, L 116, 117. KA'a, L 103 211 Multan, i, 121 Kasbmtr, L 117 katt-birds, L 195 Mu'tasila, i. 6

Khandakbadyaka, Arabio, Li. 208 Myrtilus (1), i. 407

khôm, Eranian, L 249 NARD, & play, L 182 Khoten, 1. 206 Khaykl-alkusfaini (by Alberuul), Nauroz, ii. 2 NikAh-almakt, i. 109 LL 208 Nile, sources, L 270 Khurtato, i 21 Kbwhrism, ms of, L. 258 utmbabr, Peraian, L, 843

Kbwtrismian messurer, L. 106 ntmbahra, Persian, i. 214

kirths (papyrus), L 170 Nimroz, i. 198

Kitab-Limansharkt (by Ptolamy), ii. NishApur, L 305 nuhbahr, ii. 225, 228, 220

Kitab-tibb-alfirala, ii. 245 69

Koran,' L 4; Soft interpratation, i. ORDEALS, iL. 159, 160

88, 80 ; quoted, i. 170, 222 Oxus, i. 260

sectarian interprotations, L. 283 quoted, L 264; ii. 111, 118 PAPEB, L 171

Kalzum, L 270 papyrua, 1. 171

Kumair islands, i. 210 Persian, i 40; vazidaj=gurtda, i.

kurta, Arabio piece of drem, i. 158, 218, 214 ; musmdr, i. 241

180, 239 Persian grammar, technica torm, i.

LACCADIVES, i. 210, 233 Persian metric, i. 188 19

Langa (dove-country), L 309 Persian traditions, i. 21, 08, 100, 109, 193, 304 LangabAloa, i. 241, 310 lavang (=olove), L 309 Plato, i. 43, 65, 67; Leges, i. 105,

Lobaniyys, i. 816 123; 379, 385 ; Timcut, i, 85,

lunar stations (of the Hindun), L 223, 231, 822 ; Pkædo, L 66, 57, 65-67, 71, 76, 85, 86; ii. 166, 297 167, 171

MARMOD (Yamin-aldauls), i 22, Pontus Euxious, L 258

117 ; ii, 2, 13, 108 Porphyry, quoted, i. 48 Proclus, L 57, 86 Makrtn, L 208 Molodives, i. 210, 233 Ptolemy, Almajest, i. 226, 269;

Mand, Arabic, i. 163, 164, 168 geography, 298, 890; ii. 69

Mant, L 48, 54, 55; his Book of Pythagoras, i 65, 75, 85

Mvatorien, i, 54, 264, 881 ; ii, 105,169 RANM, island, 1. 210

Manicheant, i. 7, 39, 111, 123, 159 rati, Arabic, i. 163

Miftth-'ilm-alhai's (by Alboruni), Rome, L 806 Romulus aud Remur, L 112 L 277 Rustam, ii. 246 miyis, Arabic, i. 166 mnith, i 160, 161, 163, 164 SABUKTAGIS (NAsir-aldaula), L 22

INDEX. 431

Sakaktla, ii 46, 47, 49, 50, 51, 54, | Syria, 1. 270

Bskilkand, L 299 55 Syriac, paildsopd, L 88

Samarkand, paper of, L 171 TABRİB-AL'ATLAR (v. Ya'kub), i. Barakba, il. 15 Satti, ii. 155 316, 858; ii. 87

Seven Rishis, i. 394 Tartarus, L 67

Shakh (=śaka t), li. 48, 49 Tasbkand, i. 298

Shamaniyys (iramana), i. 21 Tansar, i. 109 Tibet, L 201, 206 Shaporkan, i. 304, 308 Shtsh, i, 298 Tibetans, ii. 10

shauhat, Arabic, kind of wood, i. Tirmidh, i. 260, 302 Tiz, i 208 211 Shiltts i. 207 Turan, i. 208

ShughnAn-Shah, i. 208 Tarks, 1. 22, 206, 252, 802 ; ii. 10,

Sioily, 1. 124 135, 178

Sidar, Arabie, piece of drems, i. 180 Tûz, Persian, name of & tree, i. 171

Sijistan, L. 198 C'NANO, i. 207 Simouidea, i. 172 Sindh, Muhammadan conquest, i. Ugain (ujain), i. 308

21, 22, 165; Eranian, L 260; mimsion from Sindh to Bagdad, VAKHÂN-SHÂH, i, 206 vellum, i. 171

Sindhind, 1 153, 832, 368 ; it. 90, ii. 15 WaKwaK, island, i. 210

Siavoniane, ii. 187 191

Slavoniana, sea of the, i. 258 YAKOB Ibn Tarik, bis Tarktb-al'a-

amalipox (a wind biowing from Ndk, i. 189, 303, 312, 318, 353 ; ii. 15, 18, 23, 26, 34, 38, 44, 45, Lanks), L $09 87.68 Socrates, i. 25, 85, 170 ; ii. 171 Sogdiana, i. 249 Yazdajird, bis era, ii. 48

spod-muhra, Persian, i. 828 Yemeu (distinguished from Arabia),

Stos, i. 98 i. 270

Sufala, i. 204, 211, 270; ii. 104 Suf, explanatioo of the word, i. 33 ZABAJ, L 210; i. 108

Sufta, i. 351 Zanj, the nationa of Eastern Africa,

Safism, i. 8, 57, 62, 89, 78, 88, 87, i. 252, 270 ; ii. 104 Zarkâo, i. 7

sokbkh, measure in Khwtrizm, i. 88 Ziodik, i. 264 -.

168 Zoroastery i. 21, 91, 96 : Zoruastriane (in Soydiana);,i. 249, suamar, Persian, i. 241 260j their dakhmas, ii. 167

FRINTAD BY CALLANTYNE, HANSON AND CO. PDINAURGH AND LONDON.