Geological Society of London. 377 



The Pleistocene mammalia may again be divided into three 

 groups, those which came from Northern and Central Asia, those 

 from Africa, and those which were living in the same area in the 

 Pliocene age. Had not the animals which lived in Europe, during 

 the Pliocene age, been insulated from those which invaded Europe, 

 from Asia, by some impassable barrier, the latter would occur in 

 our Pliocene strata as well as the former. Such a barrier is offered 

 by the northern extension of the Caspian up the valley of the Obi, 

 to the Arctic Sea. The animals of Northern and Central Asia could 

 not pass westwards until the barrier was removed by the elevation 

 of the sea-bottom between the Caspian and the Urals. 



The same argument holds good as to the African mammalia, 

 which could not have passed into Sicily, Spain, or Britain without a 

 northward extension of the African mainland. 



The relation of the Pleistocene to the Pliocene fauna is a ques- 

 tion of great difficulty. If the Pliocene fauna be compared with 

 that of the Forest-bed, it will be seen that the difference between 

 them is very great. The Pliocene Mastodon and Tapir, and most of 

 the Cervidge, are replaced by forms such as the Eoe and Eed-deer, 

 unknown until then ; but many of the Pliocene animals were able 

 to hold their ground against the Pleistocene invaders, although they 

 were ultimately beaten in the struggle for existence by the new 

 comers. The fauna which the author adopted as typically Pliocene 

 is that furnished by the lacustrine strata of Auvergne, the marine 

 sands of Montpelier, and the older fluviatile strata of the Val d'Amo. 



Discussion. — Mr. Prestwicti was hardly prepared to accept the proposed division 

 of the Pleistocene mammalia into three groups ; at all events so far as Britain was 

 concerned. Neither could he draw that distinction between the beds at Erith and 

 Grays and those higher up the Thames, which found favour with the author. The 

 barrier offered by the river itself might to some extent account for the absence of 

 Eeindeer ; and though there was a difference in the fauna in the two cases, it seemed 

 hardly enough to mark any great distinction in time. As to the Hippopotamus, 

 which occurred over the whole of Northern Europe, associated with the Musk Ox 

 and large boulders, he could not see how the conclusion was to be escaped of its 

 having been able to withstand greater cold than its present representative. Though 

 the winters might have been colder, there was evidence in favour of the summers 

 having been warmer ; and the fiora seems to have been much like that of the present 

 day. The probable migrations of the different animal groups had already been 

 pointed out by M. Lartet, though Mr. Dawkins had carried his investigation of the 

 subject further. He called attention to the fact of the Mammoth having been found 

 in Italy. 



Mr. Charlesworth regretted that the author had not included within his province 

 any of the marine Crag-deposits, some of which had been regarded as Pleistocene. 

 In these beds the fish had been regarded by M. Agassiz as tropical in character, 

 while M. Deshayes considered the raolluscan remains as arctic. A similar discre- 

 pancy had been observed in other deposits of the same series, and he considered, 

 therefore, that it was unsafe to generalize from any one series of remains, as, unless 

 the whole fauna was taken into consideration, it was probable that erroneous con- 

 clusions would be arrived at. 



Mr. Flower considered that the ossiferous caves and the river-deposits were se- 

 parable and ought to be separated. 



Mr. Evans observed that in generalizations of this kind not only the whole of the 

 palseontological evidence should be taken into account, but the stratigraphical also. 

 With regard to the author's middle division of the mammalia, he thought that even- 

 tually this would have to be modified. If it were to be maintained, there would be a 

 great diflaculty in accounting for the presence of the high beds at Shacklewcll and High- 



