373 



upon their distance from the central ommatidium. Under this 

 pressure, of course, not only the somatopleure and its adjacent 

 membranes, but also the cuticle, will bend, and this would 

 account for the increase of convexity of the eye as the cells 

 gradually grew. (The obvious question as to why, under this 

 pressure, the cells might not be expected to converge just as 

 readily towards the exterior as in the opposite direction finds 

 its reply in the fact that the ommatidia are much broader 

 at their distal than their proximal ends; indeed, they are 

 really cone-shaped structures, so that such an arrangement 

 would not be possible.) 



A short digression may be of interest here. If the above 

 suggestion is correct, it will follow that the convexity of the 

 eye of Nasonia depends upon the ratio between the tension 

 of the somatopleure and its adjacent structures and that of 

 the cuticle covering the eye. Insect eyes vary greatly in 

 convexity; one has but to compare the almost spherical eye 

 of a Cicada with the rather flat eye of many flies. If it 

 should be of advantage to a species to have an eye of greater 

 or lesser convexity, it follows that it would not be necessary 

 to postulate, in the germ cells, a factor for increased eye 

 convexity, but that the result could be obtained simply either 

 by a strengthening of the germinal representative of the 

 membranes underlying the optic discs, or by a weakening of 

 that of the optic cuticle. 



It is necessary now further to consider the histological 

 changes undergone by the developing eye. In transverse 

 sections of the ommatidia the rhabdome cell is seen to be 

 fairly thick, especially at its base, where its nucleus lies. 

 The seven sheath cells can generally be clearly made out, 

 surrounding it (fig. 58c). Tlie pigment cells are long, and 

 extend through the thickness of the disc; the nucleus is in the 

 middle of the cell, although the distal end is still generally 

 the widest part of it. The four vitreous cells have now 

 become fairly distinct; occasionally the lens cells appear to 

 be differentiating, presenting at times a rather vacuolated 

 appearance. They are also seen embracing the outer end of 

 the visual cells more closely. 



But it is not till after pupation that any really marked 

 changes appear. In the pupa of about four hours the cells 

 of the developing eye have already increased considerably in 

 length, the thickness of the disc being now SO/x, of which 

 about 25jui represent the length of the sheath cells. The 

 rhabdome cell has meantime narrowed considerably, the 

 proximal end, in fact, having become developed into a rather 

 long narrow filament. The nucleus, which is situated in its 

 lowest portion, is visible only with difficulty. Distally it 



