tinuous ear-piercing scream of a number of the large Pœcilopsaltria horizontalis, Karsch, which seems to make the whole air pulsate, without betraying the exact locality of a single individual. Although in many cases I have actually tracked down individuals by their cry, in order to learn the calls of the different species, yet such a method is far too laborious for ordinary collecting purposes. So experienced a collector as Dr. Percy Rendall says: "In the Transvaal I have also taken them at rest on tree-trunks, but I do not think they were taken in consequence of their song having thus localized them. At Zomba I caught a large species by actually localizing its noise, but that was the only instance of the kind that I remember" ('Zoologist,' 1897, p. 520).
It must not be supposed that I do not recognize that the Cicada's cry must, under certain circumstances, be dangerous for individuals as, indeed, are many other secondary sexual characters; but Mr. Distant appears to have overestimated the danger, and the contention that this noise invalidates their admirably protective colouration appears to be an inverted way of looking at the question. It is more reasonable to suppose that the protective resemblance of these insects is so efficacious, that they have been able to develop these extraordinary cries through the process of sexual selection (or perhaps even natural selection, supposing aesthetic appreciation on the part of the female be denied), without unduly endangering the safety of the species. On this view, the Cicada's song, far from proving that the insect's colouring is inefficient for protective purposes, would stand as a testimony of its very high efficacy. In fact, I venture to think that, in the vast majority of cases in which animals produce conspicuously loud sounds, they will be found to possess either highly protective colouration or habits, or else distasteful or other qualities which render concealment unnecessary.
In conclusion, I can only hope that sufficient has been said to show that there are good grounds for opposing the suggestion that active mimicry is of any general occurrence in the animal kingdom; and, further, that the attempt to minimise certain phenomena of ordinary protective resemblance, in order to bring them within the scope of that principle, is not justifiable upon the evidence adduced. The subject, however, is such a wide one, that it is impossible to deal adequately with all its aspects within the limits of a paper such as this.
