382 



This shear-like action of the chitinising perioptic mem- 

 brane results also in a curious change in the shape of the 

 ommatidia, which is very easy to recognize (in spite of their 

 very close clustering together in this region), on account of 

 the rows of red pigment granules which run along them 

 (pigment of the sheath cells). The lower ommatidia, which 

 were originally straight, now become bent, and, as the pass- 

 ing inwards and upwards of the perioptic membrane 

 increases, become bent more and more, and eventually come 

 to curve back upon themselves, in order to maintain con- 

 nection with the- optic nerve. This recurving is exceedingly 

 characteristic of the lower ommatidia of the eye ; from the 

 above description it necessarily follows that the higher the 

 ommatidia are in the eye, the less will they be bent; in the 

 upper ones, indeed, the bending has been only very slight 

 (fig. 80). 



The outstanding feature, then, of the development of the 

 eye during the last day and a half of pupal life is the bending 

 outwards and compression of the optic "nerve," and the con- 

 sequent curving of the ommatidia, movements which are pro- 

 bably to be explained as due to the compressing action of the 

 perioptic membrane, as it begins to cliitinise. 



By this complex process there is gradually produced the 

 eye as we see it in the adult wasp, with its corneal lenses, 

 pigment layers, and ommatidia resting upon the fenestrate 

 membrane, which admits the fibres from the optic "nerve," 

 and in intimate relation with which has been produced a disc 

 of chitin which protects the eye from within, and the whole 

 organ covered internally by a very feeble membrane — the 

 mesodermal somatopleure. 



It is necessary to refer now to the work of others on the 

 development of the eye. It is in Weismann's great memoir 

 (1864) that we find the first correct account of the develop- 

 ment in its main outline. He regarded the layer of lenses 

 and ommatidia as arising directly from the surface ectoderm, 

 while the optic ganglion ("bulbus") he regarded as being a 

 direct outgrowth from the brain. He apparently even saw 

 the perioptic membrane, of which he says: "Between the 

 bulbus and the disc there penetrates a thin layer of fat and 

 granule cells, from which the cells which unite the two surfaces 

 very probably develop." He summarises his description thus : 

 "The morphological value of the different parts of the eye is 

 as follows : the cornea is the chitinous skeleton ; the other 

 parts of the eye-chamber (the crystalline cones, nerve rods, 

 and their investments) are modified hypodermis; all the 

 central structures (the ganglion layers and bulbus) are formed 

 as outgrowths from the nervous system" (quote from Lowne). 



