APPENDIX II. 



While my paper was in press, Professor Biitschli [2] published a short 

 note upon the eye of salpa. I discussed briefly in the Zoologischer 

 Anzeiger [11] the theoretical conclusions of his paper. I wish here 

 also to review it. Lest my criticisms should seem severe, let it be 

 borne in mind that Professor Biitschli expressly says that when he 

 wrote his brief note he did not have access to the material upon which 

 he had worked, or to his preparations, notes, or figures, but that he wrote 

 entirely from memory. Moreover, his work upon this subject was con- 

 fessedly incomplete, so that his suggestions were not the expression of 

 his judgment carefully matured after thorough investigation, but were 

 rather the conclusions toward which the work seemed to be leading him. 

 The paper is written, however, with Professor Butschli's characteristic 

 clearness and suggestiveness, and his conclusions are so plausible that 

 they deserve careful examination. 



First, let me call attention to one anatomical point. Professor 

 Butschli says that the retina is composed of two sorts of cells, the visual 

 cells and supporting cells (" Stiitzzellen "), and that the latter by their 

 anastomosing processes form a network in whose interspaces the visual 

 cells are held. As I have shown, this network is not a network of cell 

 processes, but is really the darkly staining cell walls of the more deeply 

 lying portion of the visual cells. In some cases (see Fig. 14, Plate LII) 

 the visual cells wholly, or in part, lose their columnar form and regular 

 arrangement in a single layer. The arrangement may even become 

 wholly irregular (see Fig. 10, Plate LIV), while at the same time they 

 assume very irregular shapes. They still retain, however, the tendency 

 to thicken their cell walls. The appearance seen in stained sections, 

 which resembles an irregular network of cells, is not a network of 

 branched cells, but is formed by the thickened walls of the irregularly 

 shaped, probably degenerate cells, which once functioned and perhaps 

 still function as visual cells. Because of this tendency to thicken their 

 cell walls, I have throughout my paper called them rod cells. They are 

 undoubtedly closely related structurally and physiologically to the rod 



