there were no control over the range of the intolerance, the smaller bird would have no chance in competition with the larger, and it is doubtful whether the larger would gain an advantage commensurate with the energy it would expend in ridding its area of the smaller. I have described battles in which the opponents were only distantly related; for instance, the Moor-Hen will attack almost any bird—Partridge, Lapwing, or Starling—that approaches its territory even temporarily. Nevertheless the antagonism between kindred forms is more prevalent, and, as a rule, characterised by more persistent effort; and thus it seems as if the susceptibility of the fighting instinct has its limitations, the degree of the responsiveness being dependent upon the affinity of the opponents.
Suppose now that we take an area inhabited by a number of different species requiring like conditions of existence, divide it into three sections, and imagine that in one they were all sociable, that in another they were all hostile, and that in a third those which were closely related were intolerant of one another. Let us suppose further that each one of them was represented by the full number of individuals that the law of territory would allow. In the first section an individual would establish itself, and, becoming intolerant of its own kind, would exercise dominion over an area roughly sufficient, providing conditions were normal, to insure an adequate supply of food for
