52 BULLETIN OF THE 



Several investigators have already described the development of the 

 ommatidia in the compound eyes of the higher Crustaceans. The differ- 

 ent accounts disagree especially in two particulars ; first, as to the source 

 of the different structures in the retina, and, secondly, as to the number 

 and arrangement of the cells in an ommatidium. In describing the 

 development of the retina, I have already discussed the first difference, 

 and need here only recall my conclusion ; namely, that the retina, includ- 

 ing the corneal hypodermis, crystalline cones, retinulse, and rhabdome, 

 originates as a simple hypodermal thickening, and that no part of it is 

 derived from the deeper ectoderm, which becomes the central nervous 

 system. As to the number of cells which constitute an ommatidium, it 

 will be recalled that in the lobster there are at least sixteen in each 

 ommatidium ; two in the corneal hypodermis, four cone-cells, two distal 

 retinuloe, eight proximal retinulse one of which was rudimentary, and a 

 small, but variable number of accessory pigment-cells. The last are 

 probably mesodermic in origin ; all of the others are derived from the 

 ectoderm. 



The several accounts of the corneal hypodermis given by various 

 authors differ principally in the number of cells which are said to be 

 found in each ommatidium. Reichenbach ('86, p. 91) and Nusbaum 

 ('87, p. 179) state respectively that in Astacus and Mysis there are 

 four hypodermal cells in each ommatidium. Nusbaum's statement 

 is further supported by Grenacher ('79, p. 118), who describes four cells 

 under each facet in Mysis. Herrick ('89, p. 167) has found two hypo- 

 dermal cells in the ommatidium of Alpheus. Patten states that the 

 number of corneal cells in the ommatidia of all Decapods which he has 

 examined is two. 



All the direct evidence that I have seen points to the conclusion that 

 the ommatidia of the Decapods possess two cells in the corneal hypo- 

 dermis. Reichenbach's observation directly opposes this view. I have 

 not had the opportunity of examining the same species as Reichenbach 

 did, but I have studied a representative of the fresh-water crayfishes, 

 Cambarus Bartoni, and there is no question that in the ommatidium 

 of this species only two corneal cells are present. The nuclei of these 

 two cells lie in the angles of the hypodermal squares, each one directly 

 above a nucleus of a cone-cell. When viewed from the surface, it is 

 difficult to say whether there are two or four hypodermal nuclei, be- 

 cause the four nuclei of the cone lie so near to the hypodermis and 

 resemble its nuclei so closely. It seems to me possible that in 

 his surface view Reichenbach ('86, Fig. 226) may have drawn with 



