118 THE entomologist's RECORD. 



would thus appear that transparency, or semi-transparency, of the 

 scales, was here a prior development to the loss in number, and that 

 the two processes were evolved at different periods of the species' 

 existence. Although the end of the cell in the hindwing is almost 

 completely transparent, the basal portion is a long way from that con- 

 dition, but, on the underside, a greater development is to be found than 

 on the upperside, many of the scales having there become semi- 

 transparent, while in the same position on the upperside comparatively 

 few are at all advanced. 



The second special group of transparent- winged species is one 

 made up of four Ithomiines of four different genera, viz., Napeogencs 

 potaronns (described in Appendix), n.sp., LeKcothi/ris zarepha, Ithomia 

 fiorula, HeterosaiH sj/lphis : two Erycinids, Stalachtis phaedusa and 

 StalacJitis evelina : one Hypsid, Laiiron partita: and one Geometrid, 

 Hyrniina pmfecta (described in Appendix), n.sp. This group, as 

 the last, has a particular interest, in that the various members are in 

 widely different families. At the same time, the individuals all bear a 

 remarkable similarity. The mode of scale-development to bring about 

 this end is, as before, completely different in the different families, and 

 slightly different in different genera within the same family. Each 

 species of the group is detailed as follows : — 



Napeogenes potaronus. — The normal scales are short and broad, and 

 long and narrow alternately, and fairly closely packed together. In 

 the transparent area the scales have developed into short hairs, but 

 there is no reduction in the number, and on both fore- and hindwings 

 there is no area denuded of everything. 



t Leucothyris zarepha. — In this species, the normal scales are much 

 the same shape as in the last, but they are larger. Here also the 

 scales are of two shapes. Development has, however, gone on in 

 another, though similar, direction, for each broad scale has, in the 

 transparent areas, become a mere skeleton or outline, and the result is 

 a V, composed of two bristles that originally formed the posts to 

 which the membrane of the scale was attached. The long scales 

 have become long bristles. In some cases the two bristles are stalked, 

 and a Y-shaped object is the result. This may be taken, however, as 

 simply the result of the normal scales being slightly variable, and not 

 as any special subsequent development. On those portions of the 

 wing that are contiguous to the transparent areas, a certain number of 

 scales are to be seen in a transition state, having partly lost the mem- 

 brane only. 



Ithomia florula. — As in the last, the scales are of two shapes, 

 but the long scales are difficult to detect m the unmodified portion of 

 the wing. In the transparent areas, these long narrow scales have 

 become stiff bristles, and these are much longer than the V- or Y- 

 shaped hairs, and, consequently, brush oft" much more quickly. The Y- 

 shaped hairs usually remain attached to the sockets, and are, in fact, to a 

 certain extent, protected by the long single bristles, until all these 

 have been removed. In addition to the scales becoming modified 

 they also become greatly elevated, but, in all cases, the hairs and 

 bristles become slightly curved, the Y-shaped ones in particular. 



t Two other species of Leucothyris have been since detected from the same 

 locality, viz., L. fora, and L. —? n.-sp. 



