1892.] V. A. Smith —On the Civilization of Ancient India. 
63 
Finally, we mast note the prominent position which the Yavanas, 
in common with the Kambojas, S'akas, Pahlavas, Balhikas, etc., take in 
the Mahabharata, as well as in the Ramayana, and which is so signi¬ 
ficant for the determination of the period of composition of these 
works. The Romakas are also mentioned there, though but rarely. 
The city Romakapura, which plays a special part in astronomical 
literature ( see above ) should not be understood to mean Rome itself, 
but Alexandria, or perhaps, Byzantium. 
The city Rauma mentioned in the Vishnu-purana ( Wilson-Hall f 
1, 130) must be understood in the same way. A Romaka-siddhanta 
appears to have been one of the earlier works used by the astronomer 
Varalia Mihira, who lived between A. D. 504 and 587. 
The well-known part which Yavana women play in the dramas of 
Kalidasa (who is supposed to have flourished in the middle of the sixth 
century A. D.) as personal attendants of the king may be in some degree 
explained by the trade in “ good-looking girls for concubinage,” which, 
according to the author of the Peri plus, was carried on between 
Alexandria and India. Samudra Gupta’s Allahabad Pillar inscription 
[Fleet, Corpus Inscr. Ill , p. 14, V. A. S ] mentions the delivery of maidens 
as tribute by the Sassanian king of Persia, who is there called the 
Shahanushahi. The superior culture and education of these foreign 
girls may be the explanation of the introduction of certain peculiarities 
in the attributes of the Indian god of love, Kmnadeva. The chief of 
these is the dolphin ( makara ) banner which he carries, like the Greek 
Eros. He is also sometimes described as the son of the goddess of 
beauty, who, like Aphrodite, rises foam-born from the waters. But 
this latter myth may be of primitive Indo-Germanic origin, and refer 
to the dawn. Sometimes he is represented as the consort of the god¬ 
dess of desire. The ancient image of Aphrodite, accompanied by Eros 
and the dolphin may be dimly made out in a relief on the temple of 
Bhu vanes vara in Orissa, which seems to date from the seventh century 
of our era ; but the form is very degraded in execution. 
It is very difficult to understand how the Kimnara, or monkeys in 
the guise of men and women, can have been turned into ‘ heavenly choris¬ 
ters,’ for even Indian taste can hardly regard the screeching of monkeys 
as melodious. Perhaps the Kivvpa used by the Greek maidens at the 
courts of the Indian princes may be at the bottom of the conception. 
[Ktrupa was a ten-stringed instrument, and /arupo's and cognate 
words mean ‘ wailing.’ The coniecture seems to me a verv far-fetched 
one. V. A. £.] 
Another bold conjecture would explain the amended reading ‘ Khe- 
rdn ’ in the Paniniya S'iksha, when it is explained to mean the form of 
