278 Rajendralala Mitra —The Havanas of Sanskrit Writers. [No. 3, 
knowledge of those facts, and their mutual similitude is therefore natural. 
And the argument which applies to Ptolemy would apply equally to all 
the rest, whenever it can be shown, and this can be done often, that there 
are similitudes between Greek and Hindu astronomical terms and theories. 
I have already shown that the locale of the Yavana country as given in 
Sanskrit literary or quasi-historical works, is extremely uncertain. The 
same uncertainty obtains in astronomical works, in which a greater precision 
was to have been expected ; thus, according to Parasara, Yavana lies to the 
southwest of Madhyades 'a, and taking the latter to be Mathura or the 
Sauraseni country, we should have to look for it in Balochistan; but ac¬ 
cording to Varahamihira, its capital Yavanapura was 60° to the west of 
the meridian of Lanka. This would take us to the meridian of the 
Libyan Desert. “ Rome was, however,” according to Dr. Kern, “supposed 
to be 90 degrees west from the meridian of Lanka, so that the longitude of 
Yavanapura is two-thirds of that of Rome, and this, however erroneously 
the absolute longitude is given, suits approximately the situation of Alexand¬ 
ria, which accordingly may be understood by Yavanapura.”* In a footnote 
to this passage he adds : “ So far as the longitude is concerned, Constanti¬ 
nople would answer as well as Alexandria, but I am not aware that any 
astronomer drew his first meridian over Constantinople, and without that it 
would not be taken as a point of departure.” This inference is, however, 
contradicted by a passage in the Milinclappana , a Pali Euddhistical work of 
Ceylon, in which Milinda, alias Menander, who is said to be a Yona king, is 
made to state that he was born at Kalasi in Alasadda, = Alexandria, which 
was two hundred yojanas or 800 miles from Sagal where he reigned, and 
which was twelve yojanas or 48 miles to the west of Kashmir.f Both these 
places are mentioned in the same sentence by Isiodorus—“ et Sigal urbs, ubi 
regia Sacarum, propeque Alexandria urbs et non procul Alexandriapolis urbs.” 
The Mahavansa makes Alassada the capital of the Yona country. Now, 848 
miles would barely represent 12° 20 ' to the west of Kashmir, and take us 
only to the eastern side of Persia. But whether this Yavanapur be Alexand¬ 
ria or Constantinople or a town in Persia, we do not get to Greece, and 
the word Yavana must, therefore, be taken to mean either Egypt, Arabia, 
or Persia. 
I have very little to say with reference to the fourth argument. It is 
not my intention, and in fact it is impossible, to deny that apart from the 
similitude which results in the languages of the Hindus and the Grecians, 
from the circumstance of the two races having sprung from the same 
source, there has been extensive interchange of terms between them as the 
# Brihat Sauliita, Preface, p. 54. 
f D’Alwis, Introduction to Kachchayana’s Grammar, p. xlii. 
