302 THE VARIATION OF ANIMALS IN NATURE 



such a condition could have arisen : either (a) the species 

 originally fed on one kind of host plant only and some 

 individuals suddenly turned to a new food, or (b) the species 

 originally fed on both plants and has since split into two, 

 each restricted to part of the old food range. It is probable 

 that either of these processes may have occurred in different 

 species. 



(a) We are to suppose that as a result of mutation a new 

 variety of a species arises which attacks a new food-plant. It 

 is only exceptionally that a vegetarian species so overeats its 

 food-supply that it is actually limited by a shortage of food ; 

 thus the new mutation would only be an obvious adaptation 

 if it happened to occur at one of the periods of famine. We 

 also have to consider whether competition with all the various 

 other dependants on the new host is more or less keen than that 

 met with on the old. It is possible that the animal was already 

 able (as far as digestion, etc., is concerned) to eat the new plant, 

 but hitherto lacked the instinct to attack it ; or, again, it may 

 require, besides the new instinct to eat, changes in its physiology 

 to ensure successful assimilation. In the latter case the new 

 instinct might at first be a positive disadvantage. The new 

 variety, even if more or less adapted to its food, could not be 

 established permanently without the help of some sort of isola- 

 tion. Such speculations can be elaborated indefinitely without 

 much profit ; all that we can be sure of is that we cannot 

 assume, in the absence of detailed knowledge, that a change 

 of food-plant would necessarily be beneficial : it appears that 

 such a change might be harmful, neutral or adaptive according 

 to circumstances. 



One point which seems to be of some importance is that 

 while an extension of the food range may at no stage be of 

 much advantage to the individuals who are actually breaking 

 new ground, yet there is probably a considerable gain to the 

 species as a whole. In the course of competition with other 

 species and in the fluctuations of conditions during geological 

 epochs, a species with a wide range of foods has a better 

 chance of survival than one more specialised. But, though 

 the species becomes in a sense better adapted, no necessary 

 advantage accrues to the various races of which it is composed. 



(b) We have more direct knowledge of the way in which a 

 polyphagous species splits into several races with a restricted 



