1899] THE SCOPE OF NATURAL SELECTION 123 
was living in; and lastly, if we were fully acquainted with the char- 
acter of the organism and its environment it would still be difficult 
to form any adequate opinion on the value of such a variation, owing 
to the fact that this apparently simple organism would differ so widely 
from our own functional activity and life that any conclusions formed 
on comparative methods of testing its powers, etc., would be extremely 
likely to be fallacious. If, however, we keep in mind the facts that 
(1) the whole and not merely a part of the organism is selected, and 
that, therefore, each variation does not require to be of the same value 
as if selection depended on it alone; (2) specialisations are largely 
quantitative, between man at one extreme of development and a simple 
unicellular organism at the other, the difference though very great, is 
mainly due to the fact that man is a huge multicellular colony; this 
difficulty will be much simplified. To estimate the qualitative difference 
it is necessary to endeavour to determine the specialisation of an in- 
dividual cell in one of those collective specialisations or organs: the 
difference between a cell in, for instance, the cerebral cortex of man 
and the character of an amoeba is no doubt great, but the amoeba 
reacts to stimuli, though in a less specialised form just as the cortex 
cell does; in the same way the reaction to light in the mammalian eye 
is not a new development—it has its beginnings in the preference for 
light or darkness shown by many unicellular organisms. These two 
points that selection is organismal and that specialisations are as, or 
more, largely quantitative than qualitative, weaken if they do not 
abolish all those difficulties to natural selection that are founded 
on this objection, and it is further necessary to recollect that no 
specialisation has yet been found which has not a primitive counter- 
part in the earliest known forms of life. 
5. The Imperfections of the Geological Record.—This is obviously 
a much less important objection than the preceding one. The very large 
areas of the world that have yet to be examined tend very much to 
weaken any objection founded on imperfections and absence of links. 
And as with increasing research these missing links are being steadily 
filled in, it follows that this objection has become weaker and not 
stronger with advancing knowledge. 
There are, however, certain points which it is essential to recollect 
in any consideration of the imperfections arising from this cause. 
Lloyd Morgan has pointed out that, as the tendency of natural 
selection is to favour, under appropriate conditions, definiteness both 
in the soma and in the germinal structures, the geological record should 
not be expected to provide evidence that does not correspond to this 
definite line of development. 
There is also another point which does not appear to me to have 
been sufficiently emphasised. In the earlier part of this paper I drew 
attention to the fact that Darwin considered the mutual action of a 
different set of inhabitants arising from the birth of a new generation 
