58 BULLETIN OF THE 



explanation of the way in which the distal and proximal retinulse have 

 arisen, but it also accords well with the fact that the proximal retinulse 

 end distally in fine fibres which stretch toward the cone, and are ap- 

 plied to the fibres of the distal retinulse. 



If what has been said of the growth of ommatidia be true, the course 

 which the development of these structures takes in the case of the 

 lobster must be somewhat different from that by which they arose 

 phylogenetically. In the lobster the division of the nuclei is entirely 

 completed before the ommatidia are differentiated. Consequently, after 

 differentiation has occurred, no further cell-division ensues among the 

 elements of an ommatidium. This fact, however, is by no means a 

 serious objection to the view which I have expressed, for of the two 

 processes, cell-division and the differentiation of ommatidia, it is only 

 necessary to imagine that in successive generations the differentiation 

 was retarded until a stage was reached in which all cell-division was 

 accomplished before the differentiation of ommatidia began. It is of 

 interest to observe, however, that in the lobster the planes of division 

 among the nuclei, which eventually enter into the formation of omma- 

 tidia, correspond in direction to the plane in which the nuclei of the 

 ommatidium in Gammarus would divide, should this ommatidium be 

 converted into that of the lobster. Thus the plane of nuclear division 

 which was so characteristic of the distal superficial part of each optic 

 disk in the lobster may have a phylogenetic significance. 



In how far the ommatidia of all the Crustacea can be brought into 

 relation by a process of cell-division such as I have outlined, and what 

 constitutes the simplest form of an ommatidium, are questions which 

 require careful and extensive comparative study, and which I am not 

 able to discuss here. Suffice it to say, that between the ommatidia of 

 the higher and lower Crustacea there is reason to conclude that such a 

 relation as I have pointed out probably exists. 



Cambridge, October 1, 1889. 



