APPENDIX III. 



Since my paper was in press, Dr. E. Goppert has published an excel- 

 lent paper [3] upon the eye of salpa. The author had evidently not seen 

 my preliminary abstract, so it is the more gratifying to see how com- 

 pletely, so far as he goes, he confirms my results. He gives a good 

 account of the anatomy of the eye of both the chain and solitary form 

 of Salpa Africana-maxima, Salpa runcinata-fusiformis, Salpa (Cyclo- 

 salpa) pinnata, Salpa democratica-mucronata and Salpa scutigera-con- 

 federata. His description of the minute internal anatomy of the rod 

 cells, especially as regards the so-called phaeospheres, is more elaborate 

 and therefore better than my own. 



There are but few points upon which I must take exception to his 

 statements, and few others upon which I need comment. There is, how- 

 ever, one fundamental disagreement between our observations, namely, 

 in regard to 'the innervation- of the rod cells. I may state at once that 

 if I had not had opportunity to study the development of the eye I 

 would have been unable to speak with any certainty of the innerva- 

 tion of one portion of it, that portion upon which our observations 

 disagree. In his description of the larger dorsal eye of the chain 

 individuals of Salpa Africana-maxima, Salpa runcinata-fusiformis and 

 Salpa (Cyclosalpa) pinnata, Goppert says that in the posterior por- 

 tion of the eye the rod cells receive the innervating fibers into their 

 thin-walled ends, while in the anterior portion they receive the inner- 

 vating fibers into their thick-walled ends. Reference to my Figs. 7, 8 

 and 9, Plate XLVIII, will show that this is a mistaken statement. 

 Fig. 7, on, shows the large optic nerve above one of the posterior 

 limbs of the eye (e'), in which region it directly innervates the thin- 

 walled ends of the rod cells which lie just beneath it. Fig. 8 shows 

 the innervation of the third region of the eye (e'"). That the rod cells 

 of this region receive their innervating fibers into their thin- walled ends 

 is more clearly shown in Fig. 7, Plate XLIX. Fig. 9, Plate XLVIII, 

 shows another bundle of fibers from the optic nerve, on", passing down 



