Variation as an Organic Function 179 
possible and extinction will follow. There are abundant paleonto¬ 
logical examples of this phenomenon, the general hiatus in many 
phyla at the end of the Cretaceous period being the most marked 1 . 
It may sometimes happen that a trend of variation conditioned 
by some persistent environmental change results in an effect mor¬ 
phologically identical with that produced by functional adaptation 
in the same circumstances, as for instance in some of the ecological 
adaptations of desert plants. But in general the two processes must 
be regarded as analogous rather than interdependent or identical^ 
In any case, it is only where functional changes affect the major 
factors in somatic equilibrium that they can produce results morpho¬ 
logically identical with those due to the more fundamental germinal 
variations. The Lamarckian theory on the other hand is generally 
used indiscriminately to account for the origin of even such superficial 
characters as callosities on the skin, i.e. precisely those where 
sporadic mutation is the most prevalent 2 . 
As however we have seen that the higher types of functional 
adaptation involve compensative readjustments which are apparently 
independent of environmental changes, so the increase in efficiency 
in phylogenetic evolution must be regarded as an intrinsic property 
of the organism, and the environment as only a limiting factor. 
V. We must now consider if it is possible to discover any general 
biological interpretation of these facts. It is suggested that they 
can be satisfactorily and consistently explained in terms of the 
activity-theory of the organism discussed at the beginning of this 
paper. From this point of view the succession of generations con¬ 
stitutes a periodic sequence of events, in which there are recognisable 
recurrent phases. The sequence is not however a simple one, and can 
be resolved into a number of associated subordinate series that we 
have termed activities. The germinal and somatic cycles are two 
primary associated activity-systems. 
It is, as we have seen, characteristic of the biological individual 
that its constituent activities are functionally integrated. This 
phenomenon of integration applies to the relation between the 
germinal and somatic cycles, no less than to the various activities 
which take part in somatic development. The way in which germinal 
1 A change in average conditions for any particular species may of course 
also result from the development of a particular character in some other 
species. 
2 The indirect evidence in favour of functional inheritance, such as the 
phenomena of Recapitulation, is equally consistent with the view discussed 
here. 
12—2 
