82 WONFOR^ ON BUTTERFLY SCALES. 



•white (Pontia or Pieris brassicce, fig. 11, and P. rapcB, fig. 13), 

 and argued that something similar ought to be found on other 

 members of the same family. The first I tried was the green- 

 veined P. napi (fig. 14). This gave a scale differing slightly 

 from the small white, but somewhat broader and more trian- 

 gular. The orange tip (P. cardimines, fig. 12) for a long 

 time puzzled me, as my specimens were battered ; but having 

 caught insects in good condition, I found the short brush- 

 like scale differing considerably from the other whites. On 

 the Bath white {Mancipium or Pieris daplidice, fig. 15) I 

 found a scale half-way between the orange tip and small 

 white, that is, the ribbon-like form of the one and triangular 

 brush of the other. All these whites differ also in their 

 modes of attachment to the Aving, the stalk being of a 

 different construction from that of the ordinary scale or the 

 battledore of the blues. Though the arrangement of the 

 scales is in rows and at intervals, as in the battledores, they 

 are not so readily made out in situ, but from their greater 

 length present the appearance of hairs. 



In the case of the Hipparchia family, I happened while at 

 Dorking this summer toco me across plenty of the H. semele, 

 fig. 18 (grayling), and conceived, as there was a well-known 

 scale, brush-like and tapering after the manner of the large 

 white, but differing from it in the markings on the ribbon- 

 like portion, on the H.jariva, fig. 17 (meadow broAvn), that 

 there might be something on the grayling. At first I was 

 disappointed, until I discovered my specimens were all 

 females. The next morning I caught some males, when a 

 decidedly shaving-brush like scale was the result. Pursuing 

 the same plan with all the Hipparchifc I could procure, I 

 have obtained the following results : distinctive scales, differ- 

 ing from each other in H. tithonus (large heath), fig. 16; 

 H. pamjjhilus (small heath), fig. 19; H. cegeria (wood argus), 

 fig. 21; and H. magara (wall argus), fig. 20. In all these 

 cases the brush-like scales are plentifully arranged in rows, 

 and project considerably beyond the ordinary scales. I have 

 not yet had the opportunity of pursuing^ my investigations 

 among the other families ;* but as far as I have gone, I think 

 it is clear there are in the three families of Polyommatus, Pieris 

 or Pontia, and Hipparchia, forms of scales found only on the 

 males. In addition to this, the ordinary scales in males and 

 females are the same, so that these peculiar scales may be 

 taken to be characteristic of sex. What purpose, if any, they 

 serve, I cannot conceive. They seem to me to have their 



* I have since found characteristic scales on members of the Argynnidae 

 (Eritillaries). 



