342 JOHNS HOPKINS UNIVERSITY MOEPHOLOGICAL MONOGRAPHS. 



cells, as I have called them. These cells are not constant, being present 

 in some species and absent in others, and in certain species being 

 present in the solitary form and absent in the chain form. I do not 

 think they can be regarded, in every case where present, as simply 

 certain of the cells of the pigment layer which have not become pig- 

 mented. In some cases this may be the explanation, but in Salpa run- 

 cinata-fusiformis they are separated from the pigment layer by the 

 definite membrane which encloses the rod cells, and also they send up 

 processes toward the rod cells which appear to connect with the proto- 

 plasmic core of the thickened ends of the rod cells (Plate LIU, Fig. 8). 

 What may be the meaning or function of these cells I am unable to say. 



The variation found in the eyes of salpidae is itself an interesting 

 feature. The fact that this variation is slight in the solitary forms while 

 it is very great in the chain forms is an important fact. So far as we 

 now know, there is nothing in the conditions of life of the chain forms 

 which would create a greater need for eyes of diverse structure than is 

 found in the solitary form. It may be that there is less rigorous selection 

 of optic organs in the chain individuals, owing to the interdependence 

 and mutual assistance of the members of the community. 



We have noticed the great variety of shape, position and histological 

 structure shown by the larger eye of the chain individuals of different 

 species. An even more remarkable series of variations is shown in the 

 smaller eyes. In some species, e. g., Salpa cylindrica, Cyclosalpa pinnata, 

 these show the normal optic structure. In other species, e. g., Salpa hexa- 

 gona, Salpa scutigera-confederata, they bear no resemblance to eyes, yet 

 the gradations between the two conditions shown in other species, e. g., 

 Salpa runcinata-fusiformis and Salpa costata-Tillesii, seem to prove 

 beyond question that all these structures are homologous. Because of this 

 homology I have called them all eyes, not meaning that they all function 

 as visual organs. What is the meaning of this great diversity of struc- 

 ture I am unable to say. If it was not for the embryological evidence to 

 the contrary we would seem to be dealing with a series of degeneration. 

 If this be so I am unable to conceive what change in the life conditions 

 of the chain individuals has caused them to have less need of numerous, 

 well developed optic organs than formerly. It is useless to speculate 

 upon the matter. 



We have seen that in the most primitive form of salpa eye (that 

 found in the solitary individuals) the thick- walled ends of the rod cells 

 and the pigment layer are normally toward the brain (or, what is essen- 



