254 Transactions of the 



time since, wlien speaking of the structure of the " Test scale," and 

 to which he alkides in his remarks on that object in the ' M. M. 

 Journal,' No. XXI., p. 124, but which happy idea he had not, it 

 appears, then carried out, for want of the coarse scales ; but I learn 

 from him, April 6, that he has done so since with much success.* 

 This will claim notice farther on. 



Hitherto, with the generality of observers, I had regarded the 

 scales of the Lepidoptera as composed of two distinct membranes 

 with their longitudinal ribs (often showing a very beaded look) and 

 cross-bars ; but in addition considered these parts, which have been 

 rather too vaguely defined, formed a framework of union between 

 themselves and the membranes, converting the interspace into more 

 or less minute areas, and hence falling towards Dr. Bowerbank's 

 views, though somewhat differing from them in one or two parti- 

 culars. He makes them to consist of three distinct layers, viz. " the 

 upper membrane, which contains the colouring matter ; the strias, 

 which constitute the frame ; the under membrane, which is nearly 

 colourless ; and finally, the ribs, which are in reality tubes." Where 

 we are at variance will be seen when the special relations of the 

 parts are considered. It has been only after much trouble I could 

 find a solution to the many appearances, old and new {i. e. with 

 aplanatic searcher), in conformity with such arrangement. It was 

 at last to injured, dried and charred specimens more particularly, 

 attention was directed. Without detailing the trouble in attempt- 

 ing to break up scales in particular directions — in fact, to dissect 

 them in various ways, and for convenience — the following remarks 

 will apply to scales of two common characters — though not natu- 

 rally divided, for some scales partake of both, viz. the simple and 

 the wavy scale — derived from their appearance under low powers. 



The simple scale, as alluded to by others — as Quekett, De la Kue, 

 Dujardin, Carpenter, Beck, Gosse, Slack, Mclntire, &c. — is undoubt- 

 edly composed of two or more membranes. The outer, as so beau- 

 tifully shown by Mr. Jos. Beck's simple experiment,! on Thysanu- 

 radeae scales, and which he kindly made before me, and enabled me 

 to repeat myself, is composed of an almost structureless membrane 

 (Dr. Bowerbank's under membrane?), in most scales exceedingly 

 thin and transparent, in others thicker and coloured (possibly 

 capable of being split into two layers or laminae, though I have 

 only met with a few examples, and these might have consisted of 

 three distinct membranes), occasionally marked on the inner surface, 

 with faint indications of attachments of the opposite surface of the 

 ribs and bars, or of oblique lines, " striae " of others, and sometimes 

 containing fine pigment granules (Fig. 9). 



The inner one, as made uj) of a coarser and darker material, 



* Now published, vide 'M. M. Joimial,' No. XXIX., p. 216. 

 t ' M. M. Journal,' No. XXIII., p. 253. 



