24 BULLETIN OF THE 



together. On comparing opposite faces of the rhabdome, it will be seen 

 that the crests of one side are in the same horizontal plane as those of 

 the other ; but on comparing adjoining faces, it will be observed that the 

 crests of one correspond to the furrows of the other. 



In the rhabdome of the lobster I have not found a complicated system 

 of plates, such as Patten describes in Penaeus. The substance of the 

 rhabdome in the fresh condition is apparently homogeneous, but in har- 

 dened preparations it is finely granular and stratified. The strata are 

 at right angles to the long axis of the rhabdome, and the rhabdomes 

 often break transversely. The stratified condition of the rhabdome, and 

 the close relation which the proximal retinulse bear to it, support the 

 conclusion that the rhabdome is the product of the proximal retinulse. 



In the structure of the rhabdome there is one peculiarity which, 

 although I cannot explain it, requires some comment. In transverse sec- 

 tions the square area of the rhabdome is often divided into four smaller 

 squares by two intersecting lines (Figs. 13 and 43). I made this obser- 

 vation before I had studied the relation of the distal tip of the rhab- 

 dome to the cone-cells, and I concluded then, that, if Patten was correct 

 in believing that the cone-cells and rhabdome were continuous, these 

 four divisions of the rhabdome must correspond to the four cone-cells. 

 I recognized the fact, however, that, if such was the case, the group of 

 cone-cells in the region of the rhabdome was turned through an angle of 

 45° as compared with its position at the surface of the retina (contrast 

 Figs. 4 and 13). After having satisfied myself that the cone-cells and 

 rhabdome were separate structures, I was forced to the conclusion, that 

 the four segments of the rhabdome were independent of the four cone- 

 cells. That there can be no question of the independence of these two 

 structures is shown by the condition of the cone-cells and rhabdome 

 in Mysis. In all Decapods, so far as I am aware, the cone is formed of 

 four cone-cells, and the rhabdome has four segments; in Mysis, how- 

 ever, the cone is formed of only two cells, although the rhabdome has 

 four segments. 1 



I can offer no explanation of the cross lines which occur in the rhab- 

 dome. As Grenacher ('79, p. 124) has observed, one might at first take 

 them for the outlines of such parts of the rhabdome as were produced by 

 individual retinulae. There are, however, seven retinulse, and only four 

 segments. Not only do the numbers disagree, but the position of the 

 lines in the rhabdome is difficult to explain. If the lines are related 

 to the retinula?, it would be natural to expect that they would coincide 



1 My attention was called to this fact by my friend, Mr. H. H. Field. 



