October 7, 1920] 



NATURE 



181 



The Iridescent Colours of Insects. 

 By H. Onslow. 



II. — Diffraction Colours. 



THE structure of the scales of a number of 

 iridescent butterflies was described and 

 illustrated in the first article. The colours of 

 many of these insects are undoubtedly produced 

 by thin plates, either of chitin or of chitin and air. 

 In a few instances, however, the structure {^ave no 

 indication whatever as to how the colours were 

 • voked. 



Some definite modification of the structure gives 

 rise to the principal colour, but all the minor 

 details, such as the exact shade and the quality 

 of the surface, which are so characteristic of any 

 particular species or variety, are determined by 



Kjo. I.— Tha Oreen Haimtrcak ( THtttm ruhi). Showing the iridcKcnl green 

 under-wings, which are the k«me colour at the leaves. (Natural sire.) 



the shape and position of the scales, the amount 



and form of the surface modelling, and the colour 



• nd l(x;alisation of any accompanyinjj pigment. 



nsider, for instance, the shimmering appearance 



■ i the familiar Green Hairstreak (T)iecla rubi), 



iig. I, which makes this insect so difficult to find, 



when it sits with its wings folded high overhead, 



looking like a green leaf dancing in the breeze. 



The appearance of the green scales is well known, 



' (I'ig. 2), and a discussion on their colour, and on 



!• cause of their characteristic reticulation, was 



irried on in the Eutomnlo/risl's Record, vol. vi.. 



When observed by reflected light under the 

 microscope, this shimmer is seen to be due to the 



1 Continaed from p. ts>. 



NO. 2658, VOL. 106] 



I spangled appearance of the scale, which is 

 divided into many small, irregular areas, o, which 

 reflect a green glitter of varying intensity, like 



i so many sequins. These areas are divided from 

 one another by pale lines, which form a reticula- 

 tion. By transmitted light the reticulation shows 

 as a transparent line, and, moreover, the brown 

 colour of the polygonal areas is seen to vary 

 considerably, some being dark brown and very 

 opaque, and others pale yellow and transparent. 

 Now the intensity of the green light varies in 

 exactly the same way as this brown colour, the 

 darkest and most opaque areas reflecting the 

 brightest green. The iridescent colour is probably 

 caused by a periodic structure not unlike that 

 described in Papilio ulysses, the normal brown 



Fio. 3. 



/, Green under-scale of Tktcta rubi, showing reticulation. 

 0, dark volygonat arca<t. 



I, Section through scale of Hy^mtcta tyHamotus, var. 

 liurHUntut, showing titrati6cBtion. c, lurface cuticle. 



J, Scale of the xame weevil, in silu. r, root. 



4, Plan of scale from the !uime weevil. 



5, CruM- section of tireen scale of '/*. ruhi, showing places h, where 

 the scale is looped up. corresponding to the reticulation in /. 

 s, strije. 



6, Tangential section through the wing-case o{ Htttttfirrhinit 

 tltgmns, made through the plan* ah of section 7, showing the tops of 

 the doubly refractive rod*. 



7, Cross-section through the same wing<aac, showing layer of 

 rods r\ and a deep layer r|. c, the surface cuticle. 



AH these sections, with the exception o( j, were drawn to the 

 scale fi= I mm. with Zeiss 3 mm. apochromat, N.A. 1*4, atid Comp. 

 Oc. The scale / has a magnification of about too diams. 



colour of the scale concealing any trace of this 

 structure that might otherwise have been visible. 

 The explanation of the reticulation and the 

 spangled appearance it produces becomes at once 

 evident on cutting a transver.se section, 5 (Fig. 2), 

 through the scale. It can be seen that the network 

 corresponds to the thin pl.ices on the scale, h, 



