51 
class or order to order, but by the change of species into 
varieties, varieties into species, each transition involving no 
greater alteration than is known in such cases often actually 
to take place, the transformation of the one into the many 
could have taken place. Granted, then, that there was time 
enough for such slow development to have produced the effects 
we see ; granted that the conditions of life have varied in 
different places and times to a sufficient extent to cause 
natural selection to have been carried on in exceedingly diverse 
directions ; and there is nothing in the amount of the differ- 
ences, as distinguished from their kind, which presents any 
valid obstacle to the adequacy of Darwinism. That the 
conditions of life have thus been perpetually varying, the 
testimony of geology assures us in the plainest terms. That 
the time has been enormously long, is, according to most 
geologists, equally certain ; while those who dispute the asser- 
tion do so, not by producing positive evidence that it was 
actually short, but by rebutting their opponents* arguments, 
by showing merely that it need not have been long. Still, 
therefore, even if the position of these be admitted as well 
established, it remains an open question whether, after all, 
the time may not have been amply long enough for all to 
have occurred which Mr. Darwin* s hypothesis requires. 
One further remark only is necessary before leaving this 
part of the subject. It is by no means to be imagined that 
every difference now distinguishing species from species was 
seized hold of by that natural selection which led to their 
separation. The principle of correlation of growth, on which 
the whole science of comparative anatomy and palaeontology 
depends, tells us that a difference in any one member 
involves also differences in other and related members, so 
that from a tooth only the whole structure of an animal may 
be inferred. The particular point of variation on which 
natural selection seized, might thus be but a single element 
in the total of differences that ultimately characterized the 
species, the remainder being the result of correlation. This 
should ever be borne in mind when inquiring into the possible 
way in which particular characteristics could have been exposed 
to the influence of natural selection. They may never have 
been exposed to it at all, but be the correlated results of other 
and far less apparent differences, which were so exposed.* 
3. We pass, then, thirdly, to the consistency of the hypo- 
* From here to the end of the paper was delivered extempore, being 
written out afterwards. No attempt has been made to preserve the original 
phraseology in thus reproducing it ; the matter and arrangement have, 
however, been strictly adhered to. 
E 2 
