TRANSPARENCY IN WIN(JS OF LEPIDOPTERA. 85 



species is about as follows : Acthria /(//crrs/^/.s, 90%; Actinia daltha, 4%; 

 Settiura tiiiiarai/dina, 6%. But theve is no difficulty in su<^gestinf^ why 

 A. IcKcasjiis became transparent if it were the first of the trio so to 

 become. The abdomen is brilliantly coloured with metallic green, and 

 the anal segment has a tuft of crimson. If the scaling that once existed 

 on the wings was not very conspicuous, nor of a warning nature, it 

 probably detracted from the highly warning colour of the abdomen, 

 but, by a nearly complete loss of scales on the wings, the abdomen has 

 been made considerably more conspicuous, and the insect is probably 

 rendered more immune and serves as a greater warning than it would 

 otherwise have done. It is, of course, beyond the scope of this paper to 

 enter into the why and wherefore of the transparency of every species, or 

 even to all the cases of grouped transparent species where the phase was 

 only developed by one, or perhaps two, species in the first instance, the re- 

 mainder having conformed to the type. Where a large group is concerned 

 one may be at fault in considering this or that species as the dominant 

 or type form, when all speculative reasons as to what gave rise to 

 transparency umst be worse than useless. The chief reasons, however, 

 that I believe to have given rise to transparency have already been 

 laid before you, and recapitulated are these: (1) Increased conspicuous- 

 ness and warning colour; (2) the reverse — greater concealment or 

 eflfacement ; (3) increased speed on the wing. With regard to (2) one 

 might inquire, if efiacement be the object of one transparent member 

 of a group, why not for all ? The answer is probably that, with the 

 aid of an already transformed member, the elfacive coloration will be 

 developed very much quicker, and that mimicry must be held to have 

 had at least a share of the influence. Moreover, if the group had 

 attained large dimensions, any fresh arrival in the neighbourhood 

 would be transformed to its ranks in less time than if natural selection 

 were working on pure efiacement alone. 



I now pass from argument to fact and come to the second 

 part of my paper — "How has transparency arisen?" From my 

 own research, and from the research of others, many means 

 have been revealed to attain this end. As to precisely how many 

 difterent methods have been employed, it is not yet known. I 

 will say at once that it was owing to Professor Poulton's valuable 

 ])aper in Trans. Linn. .S"w. Land., 1898, on "Natural Selection, 

 the Cause of Mimetic Resemblance and Common Warning Colours," 

 that I Avas fired to incjuire into the causes that bring about trans- 

 parency. In the Professor's paper a section is devoted to this 

 phenomenon, and Mrthnna cnnfasa and T/n/ridia /isidii, members of 

 the subfamily Ithoiniinac, are compared with Itiina ilione and Ittina 

 ji/niiantf, members of the subfamily Ihmainae. The microscopic 

 examination was made on the apical area of the forewings and revealed 

 a completely difterent method in the two subfamilies, and a slightly 

 difterent method even between the two Ithomiines. The latter were 

 found to have two difterently-shaped lots of scales, alternating with 

 one another, but in the transparent area the narrow scales had become 

 hairs and the l>roader scales much reduced in size. In the Danainae 

 only scales of one shape were present, and, in the transparent portion, 

 the scales were simply less numerous but not altered in shape. 



Following up his research the Professor examined the Pierine, 

 IHsinnrfihia nrisc, certain Castniids, and certain Pericopid-Hypsids. all 

 of which fall into the transparent group, of which, doubtless, Itana 



