^{page 527 note 1} Art. VIII. in J.R.A.S. 1888, and Art. III. in J.R.A.S. 1889.

^{page 528 note 1} Sayce, , Hibbert Lectures for 1887, pp. 142–3, 179, 187Google Scholar.

^{page 529 note 1} See hymn translated by Sayce, , Hibbert Lectures for 1887, p. 382Google Scholar.

^{page 529 note 2} This is apparently doubtful. Though Zimmer, , Altindisches Leben, p. 372Google Scholar, proves that there were only three seasons known to the ancient Greeks and Germans, and though this can also be proved with regard to the Latin races, yet, ancient as this division is, the sanctity attached to the number three is probably derived from a still earlier source It was apparently coincident with the division of the year into thirteen months, ten being, as shown in pp. 544 and 553, months of gestation, and three being those of reproduction; these being the creative months were especially sacred.

^{page 529 note 3} The dead were the fathers (Pitris) of the new believers who had died in the belief of the ancient and, to the sun-worshippers, obsolete creed.

^{page 531 note 1} Sayce, , Hibbert Lectures for 1887, p. 23Google Scholar.

^{page 531 note 2} Sayce, , Hibbert Lectures for 1887, pp. 115, 406Google Scholar.

^{page 532 note 1} DrSayce, , in his Introduction to the Science of Language, vol. ii. pp. 195–196Google Scholar, speaks of the old Turkish or Ural-Altaic year of thirteen months of twentyeight days each, and of the Accadian week of seven days. He also seems to think that the primitive Turkish-Tatar horde who spoke the parent language of the Ural-Altaic speech is connected with the Accadians, who also spoke an agglutinative language.

^{page 532 note 2} Two names of the thirteenth month are given by Zimmer, , Altindisches Leben, p. 371Google Scholar. One, Amhaspati, from the Vājasaneya Samhita 7, 30 and 22, 31. The second, Malimlucha, from the Kāthaka 28, 14 and 35, 10.

^{page 532 note 3} Aitareya Brāhmana, i. 3, 12; Haug's translation, vol. ii. p. 26. See also i. 3, 14, pp. 33–34. From these passages it is clear that Soma worship was first introduced into the ritual of the Aryan Brahmins of India from the east Another passage, Aitareya Brāhmana, ii. 2, 22, p. 122, Haug's translation, which states that Soma was made intoxicating hy being licked by an Asura woman, shows that it was originally introduced into religious worship on account of its intoxicating properties. The Soma rites must have been derived from the Baratas, or Bars, the Kolarian tribes of Eastern India, among whom the women prepared the beer to he drunk at the seasonal festivals. Mitra-Varuna, according to the Aitareya Brāhmaṇa, made it non-intoxicating by mixing it with curds.

^{page 532 note 4} The date of the rule of Menes, the first king of the dynasty, who was apparently like his successors a sun-worshipper, is fixed at about 5000 b c. by Mariette, and about 3900 b c. by Lepsius.

^{page 533 note 1} As well as those of Asia Minor, Greece, and Italy. Proofs of this assertion are given in the subsequent parts of this paper.

^{page 534 note 1} The Sphinx is a lion with a human head, and probably, as will be shown in the sequel, represented the moon-god.

^{page 534 note 2} This is tiue even if they only brought with them the original division of the year into thirteen months, and imported later the astronomical elements introduced by the system of the Nakshatras. I have adduced proof in the Appendix to this paper to show that the names of the thirteen months are earlier than those of the Nakshatras, which took their names from the months.

^{page 535 note 1} Sayce, , Hibbert Lectures for 1887, pp. 142, 143Google Scholar.

^{page 535 note 2} I have given reasons in p. 529, note 2, and also later on in pp. 564, 565, for supposing that this division was made before the year of the three seasons was worked out.

^{page 535 note 3} Sayce, , Hibbert Lectures, p. 110Google Scholar.

^{page 535 note 4} Sayce, , Hibbert Lectures, p. 193Google Scholar.

^{page 535 note 5} Sayce, , Hibbert Lectures, pp. 155, 165, 166Google Scholar.

^{page 535 note 6} Ādi (Sambhava) Parva, lxv. p. 185. All references to the Mahābhārata in this, as in previous papers, are to Pertāp Chundur Roy's translation.

^{page 536 note 1} Sayce, , Hibbert Lectures for 1887, pp. 167, 193Google Scholar.

^{page 536 note 2} Sayce, , Hibbert Lectures, p. 374Google Scholar.

^{page 536 note 3} Sayce, , Hibbert Lectures, p. 406Google Scholar.

^{page 536 note 4} Sayce, , Hibbert Lectures, p. 145Google Scholar.

^{page 537 note 1} Sayce, , Hibbert Lectures, pp. 154, 155Google Scholar.

^{page 537 note 2} 2 Kings xviii. 4; also Prof. Robertson Smith's remarks on the passage in his article on the Totem Clans in the Old Testament in the Journal of Philology, No. 17, vol. ix. 1880Google Scholar.

^{page 538 note 1} Sayce, , Hibbert Lectures, p. 186Google Scholar, note.

^{page 538 note 2} Sayce, , Hibbert Lectures, p. 504Google Scholar. The Pali-Hindi word Nagur or Nangur for city, is probably derived from this name. The city was the “Nangur” or central moon of its dependent villages.

^{page 538 note 3} Sayce, , Hibbert Lectures, p. 156Google Scholar.

^{page 538 note 4} Sayce, , Hibbert Lectures, p. 186Google Scholar, note.

^{page 538 note 5} Sayce, , Assyrian Grammar, p. 31Google Scholar.

^{page 539 note 1} As a proof that the tribes entering India from Assyria brought their own names of natural and sacred objects with them, and have retained these terms in common use down to the present day, I may note that in a Vocabulary of the Saura dialect taken down by me from one of the tribe in 1867 or 1868 in Chattisgarh in the Central Provinces, I find the Saura name for the Sun entered as Bel The Saura were, as I have shown in Part II. of this series of papers, the Saos or Suvarnas, who were the Semite-Accads who brought the solar-lunar worship to India. Under this system the Sun was worshipped as Bel, and this name they still retained. Just as the Sauras still call the Sun by its Semite-Accad name, so I contend the descendants of the moon and snake worshippers call their sacred “linga” by its old Accadian name.

^{page 539 note 2} Sayce, , Hibbert Lectures, pp. 254, 185Google Scholar.

^{page 539 note 3} Sayce, , Hibbert Lectures, p. 245Google Scholar.

^{page 539 note 4} Ādi Parva, lxxv. pp. 229, 230.

^{page 539 note 5} He was the god represented in the Asherim, the sacred pillars or linga which are so frequently spoken of in the Bible, 2 Kings x. 26; 2 Kings xviii. 4; Genesis xxviii. 19–22, and many other places.

^{page 540 note 1} Sayce, , Hibbert Lectures, p. 134Google Scholar.

^{page 540 note 2} The Egyptian evidence here dealt with is taken from the articles on Egypt, Apis, and the Sphinx, in the Encyclopædiu Britannica, ninth (i.e. the last) edition, except where other sources are specified.

^{page 541 note 1} Sayce, , Hibbert Lectures, p. 110Google Scholar.

^{page 541 note 2} Tiele, , Outlines of the History of Ancient Religions, p. 48Google Scholar.

^{page 542 note 1} Tiele, , Outlines of the History of Ancient Religions, p. 46Google Scholar, translated by Estlin Carpenter.

^{page 542 note 2} Sayce, , Hibbert Lectures, pp. 107 note, 289, 290Google Scholar.

^{page 543 note 1} Sayce, , Hibbert Lectures, p. 163Google Scholar.

^{page 543 note 2} Sayce, , Hibbert Lectures, p. 286Google Scholar, where, in an extract from a hymn, Azūgasūga, the supreme goat of Mul-lil and Merodach, i.e. Bel-Merodach, the son of Eridu, are mentioned together.

^{page 543 note 3} Aitareya Brāhmana, ii. 1. 8, and ii. 1. 3, pp. 91 and 80, Hang's translation, vol. ii. Satapatha Brāhmana, iii. 3. 4. 23. Eggeling's, translation, Sacred Books of the East, vol. xxvi. p. 83Google Scholar.

^{page 543 note 4} See also p. 555, as to the connection between the moon and the goat sacrifice in ancient Italy and Greece.

^{page 544 note 1} The name of Osiris, Asar, or Asiri, looks very like the Accadian Asari, the chief.

^{page 544 note 2} Rigv. vi. 51. 5.

^{page 544 note 3} Rigv. i. 131. 1. There is also the triad of Mitra, Varuna and Aryaman. These three are represented, Rigv. ii. 27. 1, as three of the six Ādityas, but they are in other hymns, such as Rigv. vii. 60, addressed as a separate triad. That they were originally lunar in their origin is shown in the Ḷatapatha Brābmana, ii. 4. i. 18, ProfEggeling's, translation, Sacred Books of the East, vol. xii. p. 380Google Scholar, where Varuna is said to be the waxing and Mitra the waning moon. Mitra is Mithra, the god celebrated in the Mihirzast of the Zendavesta as “the lord of the wide pastures,” the god of the heaven of light, and, like the Sumerian Ea, as the god of wisdom. He held the same position in the later Persian calendar as the moon occupies in that of the Hindoos, as he gave his name to the seventh month, which was especially sacred to him as Kartik the month of the Dibāli festival, and the seventh month of the Hindoo solar-lunar year is sacred to the moon. The sixteenth day of every month, the central period of each lunation, was also sacred to him (ArtMithras, , Encyclopædia Britannica, ninth edition, vol. xvi. p. 530Google Scholar.) Aryaman is, in a note by Sayana, on Rigv. 141. 9, Muir's Sanskrit Texts, vol. v. sect. v. p. 58, spoken of as the god who goes between Mitra and Varuna. He is the Ardibelest or Aryamana of the Zendavesta, , Darmesteter's, Translation, Sacred Books of the East, vol. xxiii. pp. 4 and 13Google Scholar, Sirozas I. and II., and his name is probably connected with the Iru or bull of light, which I have suggested in p. 547 as being the representative of the ancient fire-stick. The triad seems to have been Varuna, the heaven, the abyss, Aryaman, the bull of light or fire-stick, and Mitra, the earth or snake·god.

^{page 545 note 1} Sayce, , Hibbert Lectures, pp. 179, 180, also 469Google Scholar; Magical Text from Eridu, W.A.I. (Western Asian Inscriptions in British Museum), iv. 15 and 26–30.

^{page 545 note 2} Mahābhārata, Ādi (Astika) Parva, xvi. and xxxiv. pp. 77 and 91; Rigv. x. 72. 8 and 9.

^{page 545 note 3} Sayce, , Hibbert Lectures, p. 154Google Scholar.

^{page 546 note 1} Iru, the bull, is the name of the second month of the Accadian year.

^{page 546 note 2} The sign of the fire-stick in Hittite, Cypriote, and early Cuneiform, distinctly points to a cloyen-footed form like that of the bull's foot. The signs, as given in an article by MajorConder, , on the Three Hieroglyphic Systems, in the Archæological Review for April, 1889, vol. iii. No. 2, p. 110Google Scholar, are as follows:

^{page 546 note 3} Sayce, , Hibbert Lectures, p. 294Google Scholar. Mahābhārata, Ādi (Astika) Parva, xxxii. p. 108.

^{page 546 note 4} The persistent evidence as to the early existence of tree-worship in all countries deriving their religion from Assyria apparently shows that there was a large Kolarian element in the aboriginal population.

^{page 547 note 1} Aitareya Brāhmana, ii. 1. 9 and 10; Haug's translation, vol. ii. pp. 93, 96,

^{page 547 note 2} Sayce, , Hibbert Lectures, p. 284Google Scholar.

^{page 547 note 3} These were the Dravidian Sombunsi, i.e. Sun or Sinbunsi, the sons of the moon or the lunar Rajputs. The solar Eajputs, who were descended from the Sāka-Sauvīra or Ikshvakus, were also called the lion race, but the root li was used for lion in forming their name Licchavi, as I have shown in pp. 260 and 261 of J.R A.S. Vol. XXI. Art. III. It appears probable that the earliest name of the lunar Rajputs was derived directly from sin, the moon, and the name Sinīvatī given to the new moon (Aitareya Brāhmana, 7, 2, 11). Haug's transation, vol. ii. p. 458, also shows that the earliest form of the later word Soma, the moon, was Sina, which is the same as the Accadian name.

^{page 547 note 4} Sayce, , Hibbert Lectures, pp. 151, 154Google Scholar.

^{page 548 note 1} Max Müller points out that Indra is derived from the same root as Indu, ‘drop, sap,’ Lectures on the Science of Language, 2nd series, 1st edition, p. 430.

^{page 548 note 2} Sayce, , Hibhert Lectures, p. 311Google Scholar.

^{page 548 note 3} Sayce, , Hibbert Lectures, pp. 206, 374Google Scholar.

^{page 548 note 4} Ouranos is the Varuna of the Rigveda, which is interpreted by the commentators to mean the heaven of night; Muir, , Sanskrit Texts, vol. v. sect. v. p. 58Google Scholar. This is an additional proof that Kronos, the god of the heaven of night, meant the moon.

^{page 548 note 5} Sayce, , Hibbert Lectures for 1887, p. 284Google Scholar.

^{page 549 note 1} Encyclopædia Britannica, ninth edition, vol. xiii. p. 778.

^{page 549 note 2} Encyclopædia Britannica, ninth edition, pp. 637, 639.

^{page 550 note 1} Sayce, , Hibbert Lectures, p. 183Google Scholar.

^{page 550 note 2} Ādi (Sambhava) Parva, lxvi. p 189.

^{page 550 note 3} They are also the fifty daughters of Endymion by Selene, the moon, and of Danans, the father of the Argive race. The legend of Endymion was endemic in Elis and the people of Argos are called by Homer Danai. The name Danai, so very like that of Dānava, the name given in the Mahābhārata to the snake-wor-shipping tribes, which, as I have suggested in Part II. p. 265, may probably have been derived from the Accadian dam, strong. The early worship of Elis and the people of the Peloponnesus in Greece was certainly moon and Phallic worship. Hermes was worshipped at Cyllene in Elis under the form of the phallus, and Kronos, the old moon god, was guardian of the Olympian games. See articles Danans, Elis, Endymion and Olympia in the Encyclopædia Britannica, 9th edition.

^{page 551 note 1} Ādi (Sambhava) Parva, lxv. p. 185.

^{page 551 note 2} Zimmer, , Altindisohes Leben, p. 370Google Scholar, gives even more cogent proof of the division of thirteen lunar months in India into pairs than that given by the list of months taken from the Mahābhārata. The names of these months as quoted by him from the Vājasaneya Saṃhita 7, 30 and 22, 31 are as follows: (1) Madhu and (2) Madhava, (3) Ṣukra and (4) Ṣuchi, (5) Nabhas and (6; Nabhasya, (7) Ish and (8) Urj, (9) Sahas and (10) Sahasya, (11) Tapas and (12) Tapasya, (13) Amhaspati. That these pairs were originally male and female is exceedingly improbable, but that this Semitic innovation was afterwards extended to these names is clear from the Mahābhārata, where, as is shown in p. 306 of Part II. of this series of papers, Madhava becomes Mādhavī, and, as shown in p. 264, Ṣuchi becomes Ṣachī, wife of Indra (Ṣukra).

^{page 552 note 1} Alberuni's India, Sachau's translation, vol. ii. chap. Ixxvi. p. 183.

^{page 552 note 2} Sayce, , Hibbert Lectures, pp. 136, 138Google Scholar.

^{page 552 note 3} This derivation appears to me to be justified for the following reasons:— First, there is no apparent Sanskrit derivation of the name. Secondly, the delta of the river, when it was first known as Sindhu, was inhabited, as 1 have shown in my previous paper, by a Dravidian population. Thirdly, that the Accads were accustomed, as is shown in the case of Sinai, to call sacred spots in foreign countries by the name of the moon-god Sin; and, lastly, that the name appears in Accadian inscriptions which Dr. Sayce thinks date from the “very dawn of the historic period in Babylonia,” and long before the Sanskrit-speaking Aryans entered India, Sayce, , Hibbert Lectures for 1887, pp. 136 and 138Google Scholar.

^{page 552 note 4} Ādi (Sambhava) Parva, lxv. p. 186.

^{page 553 note 1} See with reference to the order of the months, p. 558, note.

^{page 554 note 1} Or perhaps as shown in the Appendix the original god was Tai, the mother.

^{page 554 note 2} Tiele, , Outlines of the History of Ancient Religions, p. 50Google Scholar.

^{page 554 note 3} DrMommsen's, History of Rome, popular edition, translated by DrDickson, , vol. i. pp. 216–218Google Scholar.

^{page 556 note 1} Encyclopædia Britannica, ninth edition, vol. xv. p. 96, s.v. Lupercalia.

^{page 558 note 1} Perhaps this may not be the correct explanation of the order of the months, and the real order may have been as follows: 1 Januarius, 2 Februarius, 3 Mercedonius, 4 Martius, 5 Apvilis, 6 Maius, 7 Junius, 8 Quintilis, 9 Sextilis, 10 September, 11 October, 12 November, 13 December. The numerals in this list correctly represent the places of the months in the year of gestation, which began in March, while the first three represent the three months of generation. The whole thirteen months thus represent a complete “annus,” or ring, composed of two smaller ones. This would account for the subsequent intercalation of Mercedonius, which, as Dr. Mommsen shows, used to take place in February when an additional month was required to correct the solar-lunar year. This would agree with the arrangement of the order of the gods in the Memphite and Th. ban Egyptian lists, where the ten gods ruling the months of gestation follow the names of the gods of the triad, and it would not be irreconcilable with the order of the Hindoo lunar months, as the thirteenth month, Kadru, is said to be the mother of the snakes, that is, of the whole body of moon-worshippers, while Vinata, the tenth month, is only the mother of Aruna and Gadura, the fire-stick and the bull of light. Böhtlingk-Both do not connect Vinati with Vinsati, twenty, but derive it from a root meaning to bow down, to bear. But for the order of the Hindoo months see Appendix.

^{page 558 note 2} Another mark of a common origin of the religion of the nations of Greece, Italy and India is to be found in the religious festivals common to the three countries, all of which can be traced back to the very earliest periods, when, as I have tried to prove, the people were all moon and snake-worshippers, and can be shown to have taken place everywhere at the same time of the year. Thus the Saturnalia in Rome, the Lenæa in Athens, and the Pongol festival in India, all took place in January, the first month of the old lunar year. The Dionusya in Attica, lasting three days, and the Lupercalia of the Latin races, are the exact counterpart of the Huli festival in Phalgun (February) in India. The Mounuchia to Artemis, the moon-goddess, the Palilia in Italy, and the sowing festival of the Hindoos and of the Kolarian tribes, all occur in April–May; but the chief festival month in all these systems is that of September–October, when the Ludi Magni at Rome were celebrated in honour of the sacred triad Jupiter, Juno, and Minerva. The counterparts of this festival in Greece and India are the Eleusinia of Athens and the Durgapuja, the great festival of Eastern India. While the simultaneity of the other festivals may be ascribed to a common adherence to the natural order of the seasons, which may have originated independently in different countries, this explanation will not account for the occurrence of the great festival of the dead in the September and October month, yet we find in that month (Boëdromïon) in Athens the great festival of the Nekusia, and in the corresponding month in India the Pitris (fathers) are worshipped for fifteen days (Alberuni's, India, Sachau's translation, vol. ii. chap. Ixxvi. p. 180Google Scholar).

^{page 559 note 1} Tiele, , Outlines of the History of the Ancient. Religions, p. 49Google Scholar.

^{page 559 note 2} Alberuni's, India, Sachau's translation, chap. lxi. vol. ii. p. 118Google Scholar, and chap. xxxvii. vol. i. p. 357.

^{page 559 note 3} See note 3, p. 550.

^{page 560 note 1} Mr. P. le Page Renouf has, in an article in the Transactions of the Society of Biblical Archæology, vol. iii. part ii. 1874, shown that the Egyptian astronomical observations were based on the culmination, not as Brugsch supposes on the rising, of the stars.

^{page 560 note 2} See Appendix, where the whole question is more fully discussed.

^{page 561 note 1} See the head of Seti I. fig. 192, in Maspero's Egyptian Archæology, translated by Miss Edwards, p. 224.

^{page 562 note 1} Sayce, , Hibbert Lectures, p. 238Google Scholar.

^{page 563 note 1} Sayce, , Hibbert Lectures, pp. 66–68Google Scholar.

^{page 564 note 1} Maspero, , Egyptian Archæology, translated by MissEdwards, , p. 61Google Scholar.

^{page 567 note 1} The calendar of the fifteenth century b.c. is shown by MrRenouf, , in the Transactions of the Society of Biblical Archæology, vol. iii. part ii. 1874Google Scholar, to deal entirely with the motions of the stars.

^{page 567 note 2} Sayce, , Hibbert Lectures for 1887, p. 115Google Scholar.

^{page 568 note 1} See Lydia and Phrygia, Encyclopædia Britannica, ninth edition.

^{page 568 note 2} The union of Zeus with Here on Mount Ida is apparently an old tradition used by the author of the Iliad, bk. xiv. p. 152 seqq, for the purposes of the story he is telling. The tradition seems to point to the amalgamation of the old moon-worship with the new sun-worship. The sanctity attached to both the mountains called Ida, one near Troy and the other in Crete, also prove that the name was one associated with religious reverence, and therefore it is exceedingly likely that the name was that of the Accadian Ira, the bull of light, which was, as I have suggested, first a name of the moon-god, and afterwards of his successor the sungod.

^{page 569 note 1} 1 These facts are taken from notes I made of the review in the Times of the reports of the Italian Survey published, as far I recollect, in April, 1888. I am sorry I have not been able to verify them by a reference to the originals.

^{page 570 note 1} They, like the women of the Amazonian race described by Herod, iv. 110 (Encyclopædia Britannica, ninth edition, vol. i. p. 655, Art. Amazons), paid annual visits to the neighbouring tribes to increase their stock of children, just as now the girls of a Dravidian village visit and dance with the young men of another in the neighbourhood. These visits now result in marriages or permanent unions, but anciently they were only temporary. Herodotus's account of the banishment and crippling of the males was doubtless gathered from sun-worshipping informants who looked on the old Dravidian moon-worshippers and their tribal customs, which were so totally opposed to their idea of family life, in the most unfavourable light. Herodotus's further account of the Sarmatians, iv. 110–117, the great riding race of Central Asia, shows that among them the women occupied a position similar to that of the Dravidian women. They rode, hunted and took part in battle with the men. They are probably the horse-owning tribes of India called Gandhāra, who are shown in Part II. to have entered with, or shortly after, the Dravidian tribes called Haihayas.

^{page 572 note 1} The Dravidian customs bear a strong resemblance to those of the Dorian Spartans and Cretans. The Sussitia, or common-messes, of the Spartans and Cretans are distinctly Dravidian customs. The same may be said for the council of the elders (Gerousia), the Ephors, and the important place assigned to women in Sparta.

^{page 575 note 1} This arrangement represents the year as beginning when the sun was in the constellation of the Pleiades (Krittika). It was perhaps made in solar-lunar times. At all events it does not represent the old lunar order of the months.

^{page 577 note 1} This is clearly the god Pushan, who is spoken of in the Rig-Veda as “drawn by goats,” Muir, , Sanskrit Texts, v. 12, p. 171Google Scholar; Rgv. vi. 55.4; Rgv. vi 57.3. He was, according to Sayana, Aitareya Brāhmana, i.2.9, note 16, Haug's translation, vol. ii. pp. 20–21, a Sudra, or Non-Aryan deity. He and Soma (the moon) are the gods who give wealth and cattle, he is the tutelary god of travellers, and the clearer of paths. Like the Greek Hermes, whom he resembles, he was a snake-god. In the Ṣatapatha Brāhmana, ii. 5, 4, 18, Eggeling's, Translation, Sacred Books of the East, vol. xii. p. 418Google Scholar, he is represented as the earth, who gave up Vritra the snake-race to be slain by Indra. He is, in the passage, clearly the chief snake-god.

^{page 578 note 1} Prinsep's Indian Antiquities, vol. ii. p. 150.

^{page 579 note 1} Sayce, , Hibbert Lectures for 1887, p. 284Google Scholar.

^{page 580 note 1} This was evidently connected with snake and linga worship. The watering-pot was the sacred water used to bathe the linga, which was represented in the Zodiac by the fishes, but which was originally the fish-god Ea, symbolized by the snake.

^{page 582 note 1} Journal of the American Oriental Society, vol. vi. art. iii. Translation of the Surya Siddhanta, by Rev. E. Burgess, p. 341.