Variation as an Organic Function 183 
by a more generalised type in which the ova were deposited indis¬ 
criminately in the bodies of many different kinds of prey, and that 
no attempt at concealment was originally made. We may further 
presume that the process of immuring the captured larva or other 
animal developed next in order, and that the various methods of 
effecting this operation are of subsequent growth. To understand 
the origin and significance of these specific divergences, each of which 
presumably implies a progressive increase in efficiency in its own 
particular phylogeny, we must again refer to the effect of environ¬ 
mental limiting factors on organic equilibrium. In this particular 
case the limiting factors are mainly the anatomical structure of the 
insect previously evolved, and the character of the prey with which 
it deals. For a particular type, of insect and a particular kind of 
prey there are, we assume, only a restricted number of different 
types of co-ordinated muscular actions by which the operation in 
question can be efficiently performed. These different systems may 
not be equally efficient, but each is complete in itself and their 
constituent elements are not mutually interchangeable. There is 
probably, however, some major determining factor, such as whether 
the prey is pushed before or dragged after the insect, which sets the 
type in each case, and which becomes established at an early stage 
in the evolution of the instinct. In other words, once a particular 
group of bodily movements have become co-ordinated to a certain 
degree (and therefore acquired a certain phylogenetic stability), the 
line along which a further increase in the efficiency of the process is 
possible is more or less physically determined. But the actual 
increased efficiency along this line is still of course an intrinsic 
property of the organism. All the specialised specific types originated 
in a generalised type where the movements were much more diverse 
and sporadic, and the factors which determined in the first instance 
what particular group became the nucleus of the new line of develop¬ 
ment, may have been some chance difference in local conditions 1 . 
This principle of environmental limiting factors is a very important 
one and is applicable to the whole field of phylogenetic evolution. 
A particular type of physiological economy becomes more efficient 
up to a certain point as the result of a succession of co-ordinated 
germinal changes, but a condition may be reached eventually in 
which little further improvement is possible on the same lines, and 
1 We do not preclude the possibility that some of the minor details of the 
various instinctive types may have arisen by the effect of natural selection 
upon sporadic variations in characters that have not yet become stabilised. 
