383 



If the perioptic membrane in Nasonia and the blow-fly are of 

 similar origin, then Weismann's view as to its origin is in- 

 correct; his very recognition of the membrane, however, in 

 hand dissections of fly pupae is itself a remarkable instance 

 of his power of observation. His descriptions are supported 

 by the work of G. H. Parker (1890) and of Carriere (1884). 

 Later workers, using much more accurate methods, have con- 

 tradicted Weismann's statements ; their descriptions, unless 

 the process is different in the material which they used from 

 what we see in Nasonia^ are, however, quite erroneous. 



A number of writers, e.g.^ Reichenbach (1886) or Patten 

 (1886), regard the Arthropod eye as having arisen as an 

 invagination of the ectoderm, with subsequent fusion of the 

 rim of the depression. Tlie upper and lower layers of the 

 invagination then meet and produce, between them, the 

 vitreous and lens cells and ommatidia. According to Eeichen- 

 bach (working with the crayfish), two other layers are formed 

 between these two. The superficial and outer of these two 

 layers then fuse and produce the layer of vitreous and lens 

 cells; the third layer forms the rhabdomes, and the inner 

 layer is actually regarded as forming the ganglion. 



Patten (1886), on the other hand, regards the superficial 

 layfer as forming the cornea; the outer layer of the flattened 

 vesicle disappears, and the rest of the invagination forms the 

 ommatidia. The more recent work of Giinther (1912) on 

 Dytiscus marginalis supports Weismann's original account. 



Lowne (1893-1895) partly accepts Weismann's views, but 

 disagrees with him in certain important points, in which, 

 however, he is undoubtedly incorrect. In support of his 

 Dioptron Theory of Insect Vision he wholly denies the pene- 

 tration of the fenestrate membrane by nerves ; but there can 

 be no doubt as to its occurrence in Nasonia. His view of the 

 origin of the rhabdomes is very remarkable; he regards these 

 structures as arising from the mesoderm and developing in a 

 manner analogous to that of the tracheae of the eye (thes6 

 are highly developed in the blow-fly) ; the perioptic membrane 

 he regards as growing out from the brain, although the occur- 

 rence of so much neuroglia tissue in that organ has not been 

 demonstrated. An examination of fig. 71 (p. 546) of his work 

 shows the perioptic membrane communicating with the ecto- 

 derm on either side of the optic disc ; the cells stand off from 

 the fenestrate membrane, and nerve cells are seen m^igrating 

 into the fibres of the optic stalk, which may possibly have 

 been formed from the cells of the perioptic membrane. In 

 fig. 6, pi. xxxviii. (p. 548), he actually shows a branching 

 cell of the perioptic membrane attaching itself to the fene- 

 strate membrane at the base of the ommatidia, exactly as I 

 have described it above in Nasonia. There seems, then, to 



