M. M. METCALF ON THE EYES AND SUBNEURAL GLAND OF SALPA. 369 



rod cells of the anterior horns also are innervated at their thin-walled 

 ends. 1 



Goppert's description of the histology of the eye of the solitary Salpa 

 Africana-maxima, which I was unable to study in section, agrees very 

 closely with my description of the histological condition of the eye of 

 the solitary Salpa runcinata-fusiformis ; thus confirming my view that 

 these two species are very closely related. 



Of the connection between the rod cells and intermediate cells in 

 the larger eye of the chain Salpa Africana-maxima, Goppert says: "Ich 

 glaube mit aller sicherheit behaupten zu konnen, dass jede der grossen 

 Zellen [rod cells] durch ihren, die mittlere Retinaschicht durchziehenden 

 Fortsatz mit einer dieser kleinen Zellen [intermediate cells] verbunden 

 ist"; thus agreeing with the opinion I have stated in my account of the 

 eye of Salpa runcinata-fusiformis. His figure illustrating the point is, 

 however, more diagrammatic than any of my sections. 



That Goppert fails to see the homology between the regions of the 

 eye of the chain Salpa scutigera-confederata with those of the eye of any 

 other species is only natural, since he did not clearly grasp the idea of 

 the reversal of the eye in most species and the consequent change in the 

 course of the optic nerve. It is interesting to note that he does suggest, 

 though not with much confidence, that there has perhaps been a for- 

 ward bending of the eye in the chain individuals, a thing which I had 

 already shown to be a fact [12]. 



Leuckart's statement, which Goppert quotes apparently with ap- 

 proval, that the pigment layer is always superficial, immediately under 

 the epidermis, is clearly shown to be inaccurate by my Figs. 7 and 8, 

 Plate LI ; 2, Plate LII ; 10, px, Plate LIV. 



Dr. Goppert emphasizes the statement that at least in the dorsal eye 

 of salpa the rod cells of the different regions receive light from but one 

 direction, thus enabling the animal to perceive direction. Whether the 

 facts of the anatomy lend support to this view I am unable to decide. 

 It is not improbable. I am much less able to accept as true the 

 author's assumption that the masses of rod-like cells which occur in 

 the ganglion and are not associated with pigment, are functional optic 



1 Goppert describes in Cyclosalpa pinnata, solitary form, a connection of the nerve 

 fibers with the ends of the rod cells near the pigment layer, while in the anterior part 

 of the retina of the chain Salpa Africana-maxima he says the nerve fibers connect 

 with the cells of the intermediate layer. His whole account of the innervation is 

 anomalous. 



