palæontology for evidence as to these phenomena; thus, if we are denied the right of attempting to explain them by causes acting at the present time, we shall have to abandon the whole question in despair. But, what is more important, mimicry and resemblance are only particular aspects of the principle of natural selection, and therefore if the factors of mimicry do not exist to-day, then, a posteriori, neither do those of natural selection. A single glance at nature is sufficient to justify the rejection of such a conclusion, and we must, therefore, admit that the factors of mimicry are in actual operation now; were they not, we should have no grounds for assuming that they had operated in past geological epochs. If, therefore, we find that certain cases appear difficult of exact explanation in the present state of our knowledge, we are by no means justified in disposing of the difficulty by referring them to causes operating only in the dim past, which we can neither prove nor disprove. Rather must we continue the laborious search for further evidence, not by a study of the anatomy of extinct animals, but by seeking a deeper and more intimate knowledge of the real life-histories of living organisms; for we are still profoundly ignorant of the immensely complex factors which go to make up the conditions of life of the very commonest animal upon this earth.
Nevertheless, it must be conceded as possible that there may be certain cases of mimicry or resemblance which cannot be attributed to exact causes acting at the present time; but these would be only exceptional, and would probably be due to a recent change in the enemies or the general environment of the species. I say "recent" advisedly, for we have very good grounds for believing that complicated protective characters would gradually disappear soon after the need for them ceased, whether this disappearance be attributed to pammixis or to disuse.
Later on, in the papers under consideration, we find an excellent suggestion that all examples of mimicry and resemblance should be classified under various headings, such as—Demonstrated—Suggested or Probable—Disputed or Mistaken—Purposeless—or Active. If such an arrangement could be thoroughly and carefully carried out, it would be of considerable value to students of these phenomena. Mr. Distant could, of course, only give us a mere sketch of the subject; but it is remarkable that there is not even a reference to the lengthy and important paper by Prof. Poulton, who has so thoroughly identified himself with this line of research, on "The Experimental Proof of the Protective Value of Colour and Markings in Insects in reference to their Vertebrate Enemies" (Proc. Zool. Soc. 1887, pp. 191–274), in which all the reliable experiments on British insects, up to that date,
