86 Revews—Dr. van Hoepen—Stegocephalia, Senekal. 
clavicle and the interclavicle become enormously expanded into the 
great bony plates so familiar in Metoposaurus and other types. The 
upper end of the clavicle, however, retains the slenderness and lack 
of ornament which it acquired in the Lower Permian terrestrial stage. 
Beyond a reduction in size necessitated by the flattened body the 
scapulo-coracoid remains essentially unchanged throughout the series 
of changes, and the fore-limb itself is only slightly altered, being 
smaller and more feebly ossified in later types. If Dr. v. Hoepen is 
correct the number of fingers is reduced to four and the humerus 
becomes less twisted. Except a slight weakening no change seems to 
take place in the hind-leg and its girdle. 
From the foregoing account it will be seen to how large an extent 
the evolutionary changes in the temnospondylous amphibia depend on 
two causes, the gradual flattening of the animal and its gradual return 
to an aquatic life. Study of the perfect materials in the Walker 
Museum, under Professor Williston’s charge, and in the American 
Museum, have considerably strengthened my belief in the close 
relation of Ceraterpeton, Batrachiderpeton, and Diplocaulus, animals 
whose ancestors have not been connected with the temnospondyl 
stock since very remote times. If these three creatures do really form 
a morphological family, then we have direct evidence that a similar 
flattening and return to the water have produced identical changes 
quite independently in two quite distinct amphibian stocks. In 
any case, even if the three animals mentioned above be not related, 
the resemblance in structural details between Deplocaulus and the 
Stereospondyls, which is very marked, must be due to convergent 
eyolution, because whilst Dzplocaulus is Basal Permian the stereo- 
spondyl structure did not arise till the Trias. The occurrence of four 
fingers in Myriodon whilst Eryops has five is another remarkable 
convergence to other amphibian stocks. 
We have thus direct and very strong evidence that two quite 
distinct groups of amphibia, separating very far back, have inde- 
pendently pursued similar evolutionary paths at quite different rates. 
We have seen that in both cases these changes can be referred back to 
a general flattening, and then the secondary adoption of an aquatic life. 
No trace of these trends can have been visible when the two lines 
separated, for, in one case at least, their origin must have followed 
the adoption of, and necessary adaptation to, a primarily terrestrial 
existence. This fact is only a particular case of a quite general 
feature of evolution, that allied stocks tend to pursue a similar course 
of change, the same and often striking new departures being initiated 
in diverse lines long after their separation, not necessarily at the same 
time or in the same order. Students of paleozoology have long 
been familiar with this fact, which is perhaps the most vital contri- 
bution to evolutionary data made of recent years, and it is interesting 
to see that botanists are now recognizing a similar ‘‘phyletie drift” 
in the subjects of their study, although zoologists and especially those 
whose work has dealt with the experimental study of evolutionary 
factors have so far paid little or no attention to it. M. H. Bergson’s 
philosophy, so far as it concerns biology, seems to depend on a super- 
ficial and incomplete appreciation of this great fact. 
