i8o 
C. W. SOAL 
variations are conditioned by the course of events in the somatic 
cycle must be referred to this integration of activities. We have 
seen that even so far as normal ontogeny is concerned, it is not 
possible at present to give any interpretation of the mechanism of 
functional integration in either physiological or psychological terms 
that is applicable to all cases. It must simply be accepted as a 
postulate of empirical biology that describes most satisfactorily the 
manner in which environmental changes are observed to be constantly 
correlated in the development of different types of organism. In the 
same way we are not concerned with the physiological or other 
mechanism of the functional interaction between the germinal and 
somatic cycles. The organism is a unity of which they are both 
integral parts. 
We have seen however, that in the reactions between different 
individuals there are various degrees of functional co-ordination, as 
for instance in the different types of symbiosis previously discussed. 
So also in the same organism the various activities are not all 
integrated in an equal degree. To use a modification of the previous 
simile, we may compare an individual life to a twisted cord that has 
been partially unravelled leaving a central intact bundle of strands 
surrounded by detached fibres and broken fragments. The central 
strands represent the nucleus of highly-integrated activities, which 
incorporate those reactions associated with the more stable structures 
and functions; the detached fibres are marginal activities corre¬ 
sponding to subordinate physiological processes; while the broken 
fragments represent sporadic and ephemeral leactions, which, while 
forming part of the whole reaction-complex at any moment (the 
cross-section of the cord), are not co-ordinated in any definite 
sequence in the course of development. 
Variation must be regarded from this point of view as a periodic 
change in the normal sequence of events. It may originate in some 
new environmental factor, or it may not; but in any case the con¬ 
tinued stability of the organism is dependent upon the persistent 
co-ordination of the subordinate activities throughout ontogeny. 
Hence variations affecting activities that are highly integrated will 
be co-ordinated and their corresponding reactions mutually com¬ 
pensative; where the activities affected are marginal, variations will 
be more autonomous and the corresponding reaction-changes appar¬ 
ently fortuitous. Sporadic and ephemeral reactions which are quite 
uncoordinated may vary indefinitely without being accompanied by 
any compensative readjustments in either individual or racial 
