In short, therefore, it is clear that if meaningless resemblances are numerous and striking, one can accept protective resemblance and mimicry only in instances which have been fully demonstrated. And we may in the meanwhile mark as doubtful numerous cases which now pass current in zoological literature. Among these would, I believe, fall the famous leaf-like butterfly, Kallima, which Weismann has adopted as the ne plus ultra of protective resemblance, for in the lack of adequate experimental evidence even this form may prove to be a meaningless resemblance, and not the product of selection. That it may be, and probably is, of protective value at the present time can as readily follow from an accidental resemblance which happens to turn out to be valuable as from one which has been the product of
Fig. 9. Tree-hoppers whose appearance suggests birds. After Gibson in Century.
(Cut kindly loaned by Dr. Skinner.)
numberless selected variations. In fact, it is quite credible, it seems to me, that accidental favorable variation may have furnished the basis of many a useful resemblance—as some mutationists believe. And there are no peculiar "adaptive characters" in Kallima which can safely be construed as more complicated than the meaningless characters of the Taira crab. For in what way is the resemblance of a
