CHEMISTRY OF INSECT COLOURS. l37 



undue length on this point, but it seemed to me necessary to 

 clear up any misunderstanding that may have arisen from what I 

 now see to have been a very misleading statement. 



As to the yellow colour of G. cleopatra, concerning which I 

 had originally surmised that it would prove very immovable,* I 

 believe that I have already corrected this error by stating that on 

 experimenting with this species I found the yellow to dissolve 

 out, leaving a white wing. 



As additional points of interest that have recently turned up, 

 I may note that among some American Bombyces, sent me by 

 Mr. F. M. Jones, were several species of a kind of chestnut 

 colour, which were changed to yellow by acids ; thus, again, 

 exemplifying the close relationship of chestnut and yellow. 

 With these should be compared the reaction of Melitcea athalia.j; 

 Moreover, in two species, Samia cecropia (chestnut) and Callo- 

 somia promethea (a sort of red-brown), after this transformation 

 to yellow, a slight reversion effect to the original colours could be 

 obtained, just as in the reds. This appears to me specially 

 interesting, and to point to the fact that these species are — in a 

 coloric sense — connecting links. Furthermore, one of these same 

 American Bombyces, viz., C. promethea, was marked with such 

 an unusual looking black that I was induced to experiment upon 

 it. My suspicions were partially justified, for one reagent — nitric 

 acid — converted this black into a definite brown. This is certainly 

 a notable exception to the usual behaviour of black ; but whether 

 in this species there be really any pigment present, or whether — 

 as I strongly incline to hold — the reaction here was simply that 

 of a physical colour, it is as yet impossible to say. Anyhow, the 

 fact is worth recording. In this connection, too, I may add that 

 various black-marked species, such as Arge galatea, A. paphia, 

 &c., have recently been soaked in 30 per cent, hydrochloric acid 

 for many clays without tlie black being affected.^ 



Before proceeding to the genealogy of the colours it seems to 

 me only right to point out one or two real or apparent difficulties 

 connected with my rigid separation of physical from pigmental 

 colours. These difficulties arise from the several instances in 

 which the red (pigmental) of one species is replaced by blue 

 (physical) in a closely-allied species. 



First of all there is the case of Catocala fraxini (the " Clifton 

 nonpareil" or "blue underwing"). In the larger genus of 

 Catocala there are a great number of species with red or crimson 

 bands on the under wing ; these reds are of course pigmental. 

 There are also species with orange or yellow bands ; and these, 

 too, are clearly pigmental. But in the case of G. fraxini we have 



• Ante, p. 38. t Entom. xxiii. p. 250. 



\ I have recently found certain evidence of the presence of pigment in several 

 dull-coloured species, such as H. semele, E. hyperanthes or S. egerides (or both), 

 where, from the superficial appearance, one might have supposed there was* 

 nothing but a physical blackish. 



