377 



consist of the central rhabdome cell, surrounded by six slieath 

 cells all embedded, it appears, in a gelatinous (? ) matrix. 

 In almost mature pupae the rhabdome cell appears as a 

 brownish rod; workers with larger insects seem to agree that 

 this consists of chitin, and the appearance of the rod in 

 Nasonia, with its sharply defined outline, certainly lends sup- 

 port to this view. The rod has evidently been secreted 

 directly from the rhabdome cell. (This view has, of course, 

 been assumed throughout in applying the name to that 

 cell('^).) It can be seen entering the distal cell complex, but, 

 as far as I could observe, does not reach to the exterior. 



Tlie development of the lens cells has already been 

 described. The development of the vitreous cells which they 

 enclose is completed at about this time. This consists in a 

 curious movement of their nuclei upwards to lie very close 

 below the chitinous lens, and the four cells arrange them- 

 selves in such a way that an opposite pair is in contact for 

 a considerable distance, so that the remaining two do not 

 meet each other (fig. 64c), but so far as I could observe, the 

 rhabdome no longer extends right through this distal cell 

 group. 



Meanwhile the cells have not lost their property of 

 secreting cuticle. During larval life the outer layer of cells 

 of the optic imaginal disc, from which the pigment cells, 

 vitreous cells, and lens cells later develop, secrete the larval 

 cuticle (fig. 55), while towards the end of larval life they 

 secrete the cuticle of the pupa. But when once this cuticle 

 has been secreted, the cells commence to differentiate into 

 pigment cells, vitreous cells, and lens cells, and it is in the 

 last alone that this property of cuticle secretion is retained. 

 Already in the pupa of twenty-one houi^, at a time, namely, 

 when the lens cells have scarcely surrounded the vitreous cells, 

 an outwardly convex cuticle is being secreted by each 

 ommatidium. Since the rhabdome cell at this stage forms 

 part of the external boundary of the ectoderm, it seems diffi- 

 cult to deny that it plays a part in this process. Indeed, 

 since the rhabdome cell secretes a chitinous rhabdome over 

 the greater part of its length there is no apparent reason 

 why the distal part, included among the vitreous cells, should . 

 lose this property. If we grant that this portion of the cell 

 assists the lens cells in the early stages of lens formation, we 

 might have a suitable explanation for the otherwise unex- 

 plained disappearance of the rhabdome cell from the end of 



(4) The sheath cells are usually regarded as aiding in the 

 secretion of the rhabdomes. I could, hoAAever, find no evidence 

 for this in Nasonia. 

 L 



