﻿CHEMISTRY OF INSECT COLOURS. 183 



L. quercus, M. oleracea, 0. antiqua, &c, are all classed as brown, 

 in distinction to chestnut, which not only includes — naturally — 

 the colours of Epinephele tithonus, Coenonympha pamphilus, and 

 Argynnis, but also such apparently un-chestnut colours as 

 L. phloeas and V. urticce. The behaviour of these colours, under 

 the influence of various reagents, demonstrates, nevertheless, 

 that they are essentially the same as the colour of E. tithonus, 

 which seems most accurately described as chestnut. So, too, it 

 seems best to class the pink of Smerinthus and the ambiguous 

 purplish reddish of Xanthia with the reds, since their behaviour 

 is the same. Unfortunately, however, I am compelled to class 

 together, under one name, all greens, although there are several 

 differently constituted colours here, but all apparently green. 

 Exactly a parallel case is found among the yellows ; but I had no 

 choice in such cases, since it is impossible to call a pronounced 

 yellow or green by any other name. 



By far the gravest part of this difficulty lay, however, in 

 naming the changed colours. Everyone can satisfy himself at 

 once, by a glance at his cabinet, what colour I intend when 

 referring to the natural colour ; but the only conception possible 

 to him of the colour-change produced in my experiments 

 must derive from my description, and I found it a very serious 

 difficulty to name these changes with anything like clearness, 

 distinction, accuracy, and uniformity. It will be readily under- 

 stood that when a colour, say red (not in itself a very distinct 

 name to start with), treated by a dozen different reagents, 

 undergoes several different degrees of change, — each of which 

 suggests at the moment some description, such as "reddish 

 orange," "orange-red," "faded red," "orange," "deep yellow," 

 " lemon-yellow," " pale flesh," and so on ; and then several 

 weeks later a more or less similar red is found to change in 

 more or less similar ways, — it is exceedingly difficult to pre- 

 serve anything like uniformity in the description of the colours 

 produced. In fact, without some kind of tintometer or colorimeter, 

 applicable in such work, all such names must be looked upon as 

 more or less " fluid " and approximate. Without doubt, I have on 

 one occasion described a new colour as deep flesh, and on another 

 as reddish orange, and so on, in numberless instances. Too rigid 

 interpretation of the colour-names would be, therefore, inaccurate ; 

 but, nevertheless, I hope that these are sufficiently descriptive 

 and uniform to make all my results intelligible and available. I 

 think that the difficulty will be quickly appreciated by anyone 

 who may try to name a few score of even natural colours 

 uniformly, without any standard. 



Reagents used. — This is a chronicle of disappointments. No 

 one, who had not learned by experience, would imagine how 

 stubborn and callous to powerful reagents are many — as one 



