But all this has been exploded by subsequent observation. What Fabre saw he knew, but in all that he inferred without seeing he was entirely mistaken. As observed by the Peckhams, a caterpillar may be either stung so slightly as to be quite lively, and yet not succeed in shaking off the wasp larva hatched on its body, or so severely as to die almost immediately, yet without detriment to the larva who feeds on its discoloured and more or less putrified body, with the same gusto, and apparent benefit, as though it were warm with life.[87] Thus the question seems not so much to be, how can such perfection of instinct as was observed by Fabre have been attained through the process of natural selection, as why it should have been attained; or perhaps we may even go further and ask if this supposed perfection exists at all, and whether Fabre did not deceive himself. A wasp having secured a caterpillar is, of course, at liberty to sting it as often as it pleases. Why, then, should one wasp behave quite like another one in this respect? Here, as elsewhere, there would be some amount—perhaps a considerable amount—of variation in individual disposition, and wasps of milder or less savage mood would sting less frequently than their fiercer fellows. There might, therefore, as it appears to me, be a large amount of fluctuation both in the number and degree of severity of the stings—if indeed there is any regulation in this respect—and also in the consequent injury to the caterpillar or other insect, without any particular scope being offered for natural selection to play a part. What room, indeed, for such a force can there be if it makes no difference to the wasp-grub whether the caterpillar which is to be its food, is stung badly or slightly, or whether it lives or dies?