116 THE entomologist's RECOKD. 



quickly than its model, and itself become the model for the time being. 

 Af/i/rta nririlia has, however, an additional transparent apical patch, 

 and, as two other members of the group also have it, we may suppose 

 that, unless the proportion of the different species is materially 

 altered, the Eiiaf/ra and the Geometrid will, in time, also develop 

 apical transparent patches. The details of all five species are as 

 follows : — 



Agyrta micilia. — The transparency is produced by the scales first being 

 slightly reduced in size and numbers, in the scales becoming transparent, 

 and finally assuming a hair-like structure and standing up from the 

 membrane. In the hindwing, the transparency is almost complete in 

 the centre of the wing, probably immediately after first flight, as, in a 

 seemingly perfect specimen, to the naked eye, the membrame is almost 

 quite bare of all scales, only a very few being discernible, and these 

 stand up at an angle of about 60'^' with the wing-surface, and are almost 

 bristle-like. On the inner edge of the outer marginal black band the 

 dark scales are less closely packed than at the middle of the band, and 

 point to the scales having become reduced in number previous to the 

 transparency and angular situation. A few stray black scales are to 

 be seen in position in their sockets, well out into the main transparent 

 area. On the nervures, a few transparent scales are to be detected 

 amongst the black scales. The most transparent portion in the hind- 

 wing is between nervures l/> and 2, where even the sockets of the scales 

 have vanished. In the forewing, transparency is not nearly so com- 

 plete, most of the scales are present, but reduced in size, and trans- 

 parent. In the apical patch, a large number of much-reduced dark 

 scales are mingled with the transparent ones, and clearly indicate that 

 this is the most recent of the transparent patches. The lowest portion 

 of this apical patch shows fewest dark scales, and may be considered 

 to have started earliest. 



EuAORA ccELESTiNA. — The process of becoming transparent is 

 practically the same in this species of Syntomid as in the last. In the 

 hindwing the transparent area is seen to be thinly covered with quite 

 small bristles or hairs, many of which are indicated only by the empty 

 scale-sockets. Doubtless these sockets had hairs in them on emer- 

 gence, which became removed in flight. In the forewing, in the central 

 portion of the transparent area, much the same is to be seen, except 

 that the hairs are distinctly stouter and larger, and deserve the term 

 bristle much more than the hairs in the hindwing. On the edges of 

 the transparent portion, the remarkable fact is revealed that the scales 

 are transparent or semitransparent, and also smaller than the normal 

 black or blue scales of the remainder of the wing, clearly indicating 

 that, primarily, the hairs on the transparent patches were modified 

 reduced scales, which later became semitransparent, and finally 

 became bristle-like and hair-like. It may be well here to remark that, 

 as in the former species, Aiiyrta micilia, the transparency of the fore- 

 wing is at present effected by the scales still being reduced and semi- 

 transparent, while, on the hindwing only, the scales have developed 

 further into bristles. It is reasonable to suppose that, in luiagra coelcs- 

 tina, the transparency is an older phenomenon, with scales developed 

 into bristles in both fore- and hindwings. This is also interesting as 

 almost conclusively proving that, however strong one member of a 

 Miillerian group may be at the present time, it by no means follows 



