208 EASTERN ARCTOG^A. [CHAP. 



cation between the Ethiopian and Oriental regions at least as late 

 as the lower Pliocene epoch. With regard to where this connec- 

 tion was situated, we may note, in the first place, that Dr Wallace : 

 was of opinion that even the Pikermi fauna made its way into 

 Africa chiefly through Syria, although a brief connection of Europe 

 with Tunis is admitted. When the passage in question was 

 written, little or nothing was, however, known as to the Pliocene 

 fauna of Algeria. And although this undoubtedly indicates a 

 western connection between Europe and Africa, yet even in the 

 Pliocene the Sahara probably formed, as now, a barrier 2 across 

 which the fauna of northern Africa could not pass south. Accord- 

 ingly, even the Pikermi fauna may have come round byway of Egypt. 

 Be this as it may, it seems clear that the Siwalik fauna entered 

 Africa by way of Syria or Arabia, or possibly by both. The most 

 direct line of communication would be via the Gulfs of Oman and 

 Aden ; and some indication that such a line of connection may 

 have existed is afforded by the distribution of the goats of the 

 genus Hemitragus. As already stated, fossil species of this genus 

 occur in the Siwaliks of Perim Island and the Himalaya, while of 

 the three existing forms, one is Himalayan, a second confined to 

 the Nilgiri and certain other South Indian ranges, and the third 

 inhabits Oman. So far as it goes, the evidence of these goats is 

 strongly suggestive of the former existence of a land-bridge across 

 the mouth of the Persian Gulf, as otherwise we should expect to 

 find living species in Persia and other parts of western Asia. If 

 the existence of such a bridge be admitted, we only require another 

 across the narrow strait of Bab-el-Mandeb to give a free line of 

 communication between India and Africa in this direction. 



Whether, however, the migration from India to Africa took 

 place at the north or south end of the Red Sea, or at both ends, 

 it is certain that the connecting land must have been of consider- 

 able width, and suited to the passage of mammals of all kinds. 

 In referring to the nature of the connection, Dr Wallace 3 remarks 

 that " we may now perhaps see the reason of the singular absence 



1 Geographical D^strib^ttwn of Animals, Vol. I. p. 288. 



2 The idea that there was a Tertiary sea in the Sahara is incorrect; see 

 Blanford, Quart, Journ. GeoL Soc. Vol. XLVI. p. 90 (1890). 



3 Op. cii. p. 291. 



