IN EVOLUTION FROM A GEOLOGICAL POINT OF VIEW. 175 
position in their groups of many of the earliest known species 
of those groups. Certainly we have nothing in any fauna 
that indicates it to be an early stage of development from 
primeval protoplasm. Even the earlier paleozoics, though 
in parts restricted, have all the characters of fairly grown up 
faunas. Put ide vertebrates, and the amount of variety im 
the Silurian and Devonian faunas does not (allowing for the 
imperfection of the Geological Record) present any very 
striking contrast to the amount of variety in the present 
age. That is to say, if the progress of evolution be taken as 
a measure of geologic time, the Devonian and Silurian 
systems would find places in the scale which would be 
approximately near to the Recent compared with the Age of 
primordial protoplasm. If, in still older formations, simple 
forms are found which are supposed to be archetypal, there 
is nothing to show that they are not themselves stationary 
or even degenerate forms, as indeed the nearly synchronous 
existence of other higher and varied forms almost pr esupposes. 
5 (8). Again the frequency of “Characteristic Fossils” 
brings into view a broad range of difficulties. By their 
characteristic fossils the same strata may be recognized in 
different localities, often at great distances apart. ‘heir 
constant occurrence points to the great geographical exten- 
sion of species in synchronous or homotaxial minor epochs, 
and also suggests the question how far Evolution can 
account, not only for change, but for identical change over 
wide areas. To take but one instance, Rhychonella procuboides 
of the Eifelian is replaced both in England and Germany by 
Rhychonella cuboides in the Cuboides Beds. Thus we have 
an apparently very sedentary species similarly selec at mn 
distant countries. This can hardly be supposed to have been 
effected by a single operation, but seems to imply a wholesale 
modification producing in distant regions the same result. 
5(9). Yet another difficulty lies in the fact that the action of 
free evolution ought to be as much centripetal as centri- 
fugal. There is no intrinsic reason why variation from type 
should be stronger than reversion to type. When a species 
has varied under the control of man and thereby produced 
definite varieties, the first thing which happens when that 
control is removed is that the varieties merge and the old 
conformity is re-established. But such reversion is hardly to 
be noted in geological ages. We rarely, if ever, find a 
species, once modified by descent into another species, 
reverting in a subsequent age back to the original species 
