1899] THE SCOPE OF NATURAL SELECTION 125 
of one variation happen to be present at one and the same time. His 
difficulty therefore does not appear to me to be answered by Wallace. 
Weismann,’ admitting the objection of Spencer’s as having a real 
existence, attempts to answer it by the tendency of natural selection 
itself to induce definite variability. This answer does not seem to 
me to be much more satisfactory than Wallace’s, for the point of the 
argument is, that as the accessory variations are necessary to the proper 
working of the primary they must be present from the first selection, and 
as determinate selection can only appear after selection has been con- 
tinued for some generations it must be unable to explain this occurrence 
of co-ordinated parts which occurs prior to the action of selection. 
Mr. Lloyd Morgan in the December number of Natural Science 
deals with this difficulty in a manner which appears to me to be much 
more satisfactory. We have seen in the brief summary of his views 
that he draws an important distinction between somatic response to 
environment and the selection of germinal variations, that under 
altered conditions of environment he considers somatic plasticity to 
be one of the principal determining causes of selective preservation, 
and as he admits the action of use-modification on the somatic 
structures, those organisms whose somatic structures are sufficiently 
plastic to allow of this newer co-adjustment to the newer conditions 
will survive on account of their plasticity, and this will continue to 
happen over one or more generations until chance variations happen 
to make their appearance in the same direction as the environment, 
then the offspring of this organism or these organisms will start life 
with a slight favourable predisposition to their environment, which in 
addition to somatic plasticity will give them a slightly better chance 
than those without this predisposition, hence by the fostering power 
of body response a co-ordinate structure might be formed through 
cumulative co-incident variability. This objection therefore does not 
apply to the theory of Natural Selection modified as above. 
Keeping in view this theory of co-incident variability, there is 
another consideration which will also tend to weaken this objection. ° 
As selection must be from the first organismal, and as adaptation to 
climatic conditions must be absolute, as far as it is capable of exercis- 
ing a selective action, a certain common tendency will be present in 
all more or less similar organisms living under these more or 
less similar physical- conditions, This primitive climatic basis will 
give a certain direction to the subsequent inter-organismal selection, 
and we have seen that with progressive evolution the necessary 
specialisation entails an increasingly definite tendency in the organism 
as a whole, owing to the increasing dependence of one part on another : 
hence it will follow that all variations will tend to become increasingly 
co-ordinated as they become increasingly specialised, and they will also 
become increasingly so as we pass from the lower to the higher forms. 
1 ** Germinal Selection.” 
