136 SECTIONAL ADDRESSES 
edge of the Asiatic continent. It therefore, seems possible to date them 
not only by the associated mammalian remains, but also by the fossil 
shells in the marine formations, both earlier and later, which now form 
part of the island of Java. The successive marine deposits, as might be 
expected, are marked by an increasing percentage of existing species of 
shells among the fossils. As determined by Dr. L. J. C. van Es, the river 
deposit at Trinil rests unconformably on a marine formation which is 
shown by its fossils to be of Middle Pliocene age. The gap in the geo- 
logical series indicated by this unconformity is filled in other districts by 
marine deposits which contain 66 to 70 per cent. of existing species of 
shells, and may, therefore, be regarded as Upper Pliocene. Hence the 
conclusion that if a marine formation equivalent in age to the Trinil 
river deposit were found, it would contain more than 70 per cent. of 
existing species of shells and might thus be referred to the Pleistocene. 
The percentage of existing species of freshwater shells in the Trinil 
deposit supports this conclusion, and as the associated land mammals 
(Stegodon, Hippopotamus, etc.) much resemble those in the Lower 
Pleistocene Narbada river deposits of India, Pithecanthropus evidently 
dates back to the beginning of the Pleistocene period. 
The skulls and lower jaws of Sinanthropus were met with not in a river 
deposit, but in a cave which had evidently been occupied for a long period 
by man. They were associated with rude stone and bone implements, 
and even with remains of fires. The deposits in which they occurred 
are proved by the Chinese geologists to be older than the widespread loess 
of China, which in places contains the remains of the woolly rhinoceros 
(R. tichorhinus), and in other places the mammoth (Elephas primigentus). 
‘These, it will be remembered, are two of the characteristic fossils of the 
‘Middle Terrace’ of the Thames valley, which dates back to the latter half 
of the Pleistocene period. With Sinanthropus are found remains of a 
large extinct beaver, Trogontherium, and a rhinoceros very like R. merckt, 
which are specially characteristic of the ‘ High Terrace ’ of the Thames 
already mentioned as the probable equivalent of the Piltdown gravel. 
If, therefore, the widely distributed mammals just enumerated were 
living at the extreme eastern and western limits of their range in the Old 
World at one and the same time, as seems most probable, Sinanthropus 
dates back to the early part of the Pleistocene period and must have been 
a contemporary of Foanthropus. 
These facts, I think, are enough to show that in the beginning the human 
skull was much more varied than it is at the present day. ‘There were, 
indeed, several distinct approaches to modern man before his type 
became fixed and persistent ; just as there were parallel lines of evolution, 
effective and non-effective, in the ancestry of other modern mammals. 
That the four known examples of the earliest men were all closely related, 
is proved by the skull of Sinanthropus, which exhibits a remarkable com- 
bination of the special features of the other three. In the contour of the 
top of the head, with the great depressed bony brow-ridges, it is so like 
the skull of Pithecanthropus, that some anatomists have actually referred 
it (though without good reason) to the latter genus, In the fine spongy 
