CHEMISTRY OF INSECT COLOURS. 89 



I might, perhaps, add that many of these Geometrine yellows 

 are on the very border-land of yellow, and might with almost 

 equal justice be described as chestnut, with which, too, their 

 reaction betokens a strong similarity. They aiford an additional 

 example to those presently to be adduced, of the close relations 

 in constitution between chestnut and yellow. 



There remain, then, only the remarkably impervious T. pro- 

 nuba and its miniature analogue, Heliaca tenebrata. I suppose 

 that, a priori, every one would have anticipated, as I did, that 

 the magnificent orange of this species, — T. pronuba, — would be 

 rapidly affected by reagents ; but, as a matter of fact, I cannot 

 touch it. After all the preceding remarks in this section on the 

 character of yellows, it may probably be considered very safe to 

 refer this species to the subclass of those possessing a stable 

 yellow pigment ; but, as a matter of fact, I think we shall exercise 

 a wise caution in withholding any such judgment until further 

 evidence be forthcoming. The case is peculiar, since, unlike the 

 apparently similar instance of villica, &c., we have here no col- 

 lateral evidence. I can see no reason why there may not be a 

 " physical absorption " yellow as well as pigment yellows ; and, 

 granting this, we have absolutely no proof that pronuba-yeWow is 

 not such. The doubt might be cleared in two ways : in the first 

 place, the existence of a red species (or even a red variety) in the 

 genus would settle the matter at once — as in the Arctiidae — in 

 favour of pigments ; or, secondly, if other yellow species of Tri- 

 phcena were found to be affected, as is Catocala cerogama, that also 

 would be tolerably conclusive. At present I have examined no 

 other species of Triphcena, but my impression is rather that the 

 wings of the less brilliant species are more suggestive — in super- 

 ficial appearance — of physical colours than of pigments. Any- 

 how, for the present we must leave this question an open one. 



And finally, now to complete the consideration of red and yellow, 

 we have yet to take into account the intensely interesting subject 

 of reversible effects, which, so far as I at present know, are displayed 

 by this colour — red — alone. Perhaps the simplest manner of ex- 

 plaining this will be to proceed historically, — following the order 

 of my discoveries. We will start from the fact that all reds are 

 turned yellow, — by acids especially. Wishing to preserve a few 

 permanent preparations of these artificial varieties, I soaked the 

 wings of one or two perfect insects, Euchelia jacobcece and 

 Zygcena Jilipendulce, with hydrochloric acid, washed them with 

 water to remove the acid, then with alcohol (perhaps — but I am 

 not sure now) to assist the drying, and when dry put them away 

 in a store-box. On opening this some time afterwards I found, 

 to my utter disgust, that my beautiful yellow varieties had re- 

 verted to their original red. Of course, it is scarcely necessary to 

 say, that I saw at once the great importance of this unlooked for 

 phenomenon, and the consequent necessity for a thorough and 



