158 Dr. H. Eltringham's Further Observations on the 



tion, but still recognisable as corresponding to the hairs 

 in the niavius brush. PL XII, fig. 3, is a section a little 

 further towards the distal end. Here the central bodies 

 occupy a reduced area. They are still surrounded by the 

 first-ring bodies, but what may now be called the normal 

 hairs occupy a greater space. It should be particularly 

 noted in this figure that the outer layer of the normal 

 hairs includes relatively many having a flattened section. 

 PI. XII, fig. 4, is a section still further outwards. The 

 central bodies have almost come to a point, and are sur- 

 rounded by a fairly thick mass of first-ring structures. The 

 normal hairs, or " second-ring," occupying a large area, 

 are those of the yellow tuft of the brush, corresponding 

 with the yellow hairs of niavius. One of these hairs is 

 shown in section at PI. XVIII, fig. 7. The central lumen 

 is small, the medullary substance presents a more or less 

 radially striate appearance, and the outer surface is longi- 

 tudinally ribbed. 



(3) The black hairs have now begun to arise from the 

 bag membrane. They are of variable diameter, and are 

 separated from the yellow hairs by those of flattened 

 section which occupied the extreme outer layer in fig. 3. 

 We may call these the third-ring bodies. In PI. XII, fig. 5, 

 the central cone has finally disappeared, the first-ring bodies 

 are coming to a point, whilst the normal hairs, yellow and 

 black, occupy nearly the whole of the area. In PI. XII, fig. 6 

 the normal hairs remain, divided, however, by the third-ring 

 bodies, which are larger and flattened towards the end of 

 the brush, and are here seen to be distinctly difterentiated. 

 In some species the hairs which thus become terminally 

 flattened eventually break up into small fragments, and 

 thus provide the " dust " which so frequently occurs in 

 connection with these organs. 



In A. egialea this breaking up does not seem to occur, 

 nor do any of my sections show that dust is produced at 

 all in this species. 



The structure of the organs in A. egialea differs from that 

 in A . niavius in the following manner. The cupped glands 

 of the wing-patch are not continuous, but have rows of 

 blind glands between them. The scales are of a different 

 shape, and do not provide a complete protection for the 

 cups. The gland-cells do not occupy the whole space 

 between the wing-membranes, and those of the blind glands 

 rest on a substructure which presents a horizontally 



