194 



NATURE 



\jfune 26, 1879 



the region where the palaeozoic and mesozoic marine 

 formations appear to the south of the Himdlayan axis, 

 the former in the Punjab and Afghanistan, the latter 

 both there and further south, in Cutch, Sind, and Bilu- 

 chistdn. 



The geography of India in the early eocene period may 

 be represented as something like the following : — The 

 whole of the peninsular area was then, as it now is, land, 

 with the exception of some portions of the present coast 

 tract north of Bombay. In like manner the coast of 

 Kaliwdr and the greater part of Cutch were submerged, 

 and a deep sea extended up the r resent Indus valley, and 

 over Bdluchistdn, also over the Punjab and parts of 

 Kdshmir, sending an arm up the present Ganges valley, 

 which was in part estuarine, and bounded on the north 

 by a tract of land composed of the then uncontorted rocks 

 of the lower Himalaya. To the east of Bengal the present 

 plain of Silhet and a part of the Khasi Hills were 

 probably covered by a shallow sea, and this extended to 

 the southward over Arakan and at least a large portion of 

 Burmah. 



It was in this state of things that the first great disturb- 

 ance took place, which, repeated again and again during 

 middle and later tertiary times, resulted in the present 

 chain of the Himdlaya. The river valleys which, after 

 the first great upheaval, were carved out in the then 

 youthful chain, discharged their stream-borne debris as 

 now oh the Gangetic plain ; and the accumulated con- 

 glomerates, sands and clays, which formed around ■ the 

 mouths of the valleys, again and again suffered crushing 

 and upheaval during the subsequent compression of the 

 mountain mass, were added to the hills, and in their 

 turn underwent erosion. But the valleys originally 

 marked out have preserved their general course and 

 function ; and the Sutlej, the Bias, the Tous, the Jumna, 

 and the Ganges, still flow out from the mountains along 

 essentially the same hnes of drainage which their then 

 nameless representatives followed in miocene times. 

 Such, at least, briefly stated, is the history which Mr. 

 Medlicott and his colleagues have evolved during many 

 years' study of these interesting rocks, first made famous 

 through the classic labours of Cantley and Falconer, and 

 by them named " Siwalik." 



Not the least interesting chapter of the Manual before 

 us is that in which Mr. Blanford deals with the rich and 

 varied vertebrate fauna of these tertiary rocks. The 

 original collections of Cantley and Falconer have been 

 largely added to in later years by various members of 

 the Survey ; and a comparison of the forms obtained 

 from different horizons in the Sub-Himdlaya, the Punjab, 

 Sind, Perim Island, and certain river valleys in the 

 Indian peninsula and Burmah, has led to some modifica- 

 tion of the opinions originally held of the geological age 

 of the Siwalik rocks and their contained fauna. 



As Mr. Medlicott has shown, the Siwalik rocks are an 

 ancient alluvial formation, hke the river fans described by 

 Mr. Drew in Kashmir, and like the Bhdbar deposits still 

 forming along the foot of the Himalaya at the present 

 day. In what may be termed the typical area, between 

 the Sutlej and the Ganges, disturbances of some magni- 

 tude which took place after a portion of these deposits 

 had been laid down, necessitate a subdivision of the scries 

 into three groups, upper, middle, and lower, the last of 



which is termed the Ndhun group. It is from the two 

 upper groups that (with perhaps some doubtful exceptions) 

 all the fossils of this special region have been obtained, 

 the Niihun group having remained unproductive to re- 

 peated search. An elaborate analysis of the homotaxis of 

 the 45 mammalian genera (consisting of 84 species) which 

 compose this fauna shows that the proportion of living to 

 extinct genera is greater than in most miocene deposits. 

 The presence of four extinct genera not known to range 

 above the miocene period is contrasted with the occur- 

 rence of sixteen genera, not found elsewhere at a lower 

 horizon than pliocene or post-tertiary ; and there is a close 

 approximation between some of the mammals and the 

 living species of the same genera, the most remarkable of 

 all being the connection of the fossil buffalo, Bos palceindi- 

 cus, of the uppermost Siwalik strata, that of the post- 

 pliocene Jumna and Nerbudda beds, and the Common 

 Indian arnee still existing. Of six species of reptiles 

 sufficiently well known to be comparable, three are com- 

 mon forms now inhabiting the same area, and the fresh- 

 water mollusca also all apparently belong to common 

 existing species. Putting the whole palseontological evi- 

 dence together Mr. Blanford concludes that a balance is 

 in favour of a pliocene age. This conclusion is strength- 

 ened by stratigraphical evidence. At the top of the enor- 

 mous succession of tertiary deposits of Sind, which have 

 a total maximum thickness of some 30,000 feet, occurs a 

 group termed the Manchar group, about 10,000 feet thick, 

 which is of fresh-water origin and represents the Siwaliks 

 of Northern India. The lower beds of this group pass 

 downwards into the Gaj group (1,000 to 1,500 feet thick), 

 which is of marine origin, and contains a characteristic mio- 

 cene fauna, "more probably upper than lower miocene." 

 The Lower Manchar beds have yielded a considerable num- 

 ber of mammals, and this fauna, although containing several 

 species in common with the Siwaliks, is altogether older 

 in aspect ; the majority of the forms hitherto recognised, 

 belonging to the peculiar types of even-toed ungulates 

 allied to Nierycopotamus and Ajithracotherium inter- 

 mediate in character between pigs and ruminants, and 

 peculiarly characteristic of the miocene epoch. In these 

 lower Manchar beds is also found a form of Dinothcrium, 

 a type unknown in the Siwaliks. Remarking that " there 

 can be no reasonable doubt that the Manchar beds of 

 Sind, as a whole, correspond with the Siwalik formation 

 of Northern India, the two being portions of a continuous 

 band of tertiary rocks," it is concluded that the fossiliferous 

 lower beds of the Manchar group correspond to the un- 

 fossiliferous Ndhuns, and the almost unfossiliferous 

 Upper Manchar beds to the ossiferous strata of the 

 Siwaliks. Mr. Blanford then remarks on the probable 

 chmatic causes which have preserved in the Indian 

 pliocene an unusually large number of forms elsewhere 

 characteristically miocene ; and compares the case with 

 that of the Pikermi beds in Attica, which are of unques- 

 tionable pliocene age. He considers that the general 

 cooling of the north temperate zone at the end of the 

 miocene period caused a migration of many of the cha- 

 racteristic mammals into Southern Asia, the Himdlayan 

 chain at that epoch not presenting so impassable a 

 barrier as at the present day, and that such was the case 

 seems to be confirmed by the occurrence of rhinoceros 

 and elephant remains in the tertiary deposits of Hundes 



