M. M. METCALF ON THE EYES AND SUBNEUEAL GLAND OF SALPA. 327 



eye of the chain Salpa runcinata-fusiformis, and yet more closely that of 

 the rod cells of the eye of the solitary form of that species. The rod cells 

 are roughly cylindrical, with irregularly thickened walls. They are not 

 arranged in a single definite layer, but are somewhat irregularly dis- 

 posed. In the cross-section (Fig. 14, Plate LII) we see the nerve fibers, 

 on, on the dorsal surface ; then an irregular layer of rod cells, s, r ; in 

 some places this layer is several cells deep. The dorsal ends of these 

 cells are thin-walled ; the ventral ends are thick-walled. Ventral to the 

 thick-walled ends of the rod cells is a considerable number of nuclei, , 

 the nuclei of the intermediate cells. The rod cells and intermediate cells 

 are enclosed in a very delicate membrane. Outside this membrane, ven- 

 tral to the intermediate cells, between them and the ectodermal optic 

 sheath, are the pigment cells, resembling the pigment cells of Salpa 

 runcinata-fusiformis, chain form. They are irregular in shape; their 

 protoplasm is so full of pigment granules that their nuclei can be seen 

 only after dissolving the pigment in acid. Fig. 14 represents a section of 

 the basal portion of the eye. A section of the apical portion would show 

 the same appearance, with the exception that the dorsal and ventral 

 sides of the section would be reversed. 



In the chain form of Salpa hexagona there is a pair of dorso-lateral 

 outgrowths from the ganglion, one on the right side, the other on the left 

 side of its dorsal face (Fig. V, Plate LVII, and Fig. 15, Plate LII). 

 These are spherical and composed of polyhedral cells with thick walls 

 and large nuclei, of the size of the large nuclei found in the periphery of 

 the ganglion or in the rod cells of the dorsal eye. The character and 

 homology of these outgrowths can best be discussed after describing a 

 simpler form of what I regard as the same organs in Salpa scutigera- 

 confederata. I will now call attention only to the fact that their position 

 on the dorsal surface behind the optic nerve corresponds to the position 

 of the smaller eye of the chain Salpa cylindrica or Salpa runcinata- 

 fusiformis, and the character of the cells of the outgrowths resembles 

 that of the imperfect rod cells of the smaller eye of the latter species. 



Professor Brooks has called my attention to a peculiar structure in 

 the young embryo of Salpa hexagona. It is seen in the embryo at a time 

 when the ganglion is well formed but has not yet attained the compact 

 structure seen in the latest stages ; the central cells of the ganglion have 

 not yet commenced to degenerate nor has the eye appeared. The 

 ganglion lies close to the dorsal ectoderm. Near the posterior end of 

 the ganglion there is a wide and deep invagination of the ectoderm, 



