1889.] On the Early Development of Lepidosteus osseus. 117 



cells persist for a long time, lying outside the cord, and on its dorsal 

 or neural surface, just over the posterior fissure. 



Gradually they undergo a series of degenerative changes. The 

 stumpy processes vanish ; the cells shrink, and so get smaller. 

 Finally, they become glassy, having lost all traces of nucleus and 

 nucleolus, and disappear. 



In fact, the series of changes undergone by these cells corresponds 

 exactly to that degeneration and death of nerve-cells, ivhich the pathologists 

 term simple atrophy (einfache Atrophic, see Ziegler's ' Pathologische 

 Anatomie,' 3te Auflage, 2ter Theil, pp. 603 — 606). Though the 

 presence of giant ganglion cells in the embryos of certain Scylliidee 

 was known,* till now we had no idea of their distribution over the 

 whole spinal cord as stated above. Like Dr. Mayer, I find them in 

 Scyllium and Pristiurus, in both of which forms they are exceedingly 

 large, numerous, and well developed. Along with Mayer I failed to 

 detect them in Torpedo, but met with them in Raja. I also miss them 

 in Acanthias, but in Mustelus, where as a normal thing they do not 

 develop, one may find about a dozen of them well developed in a 

 single embryo, but then in abnormal situations, lying free in the 

 formative tissue of the mesoblast, and outside the central nervous 

 system. They are very obvious in 10 millimetre embryos of Salmo, 

 and may easily be detected in Labrax, Esox, and Rhodeus embryos of 

 the proper age. They are very numerous in newly hatched Lepidosteus. 

 Only with difficulty can they be demonstrated in young Petromyzon 

 embryos, on account of the yolk filling the cells, but they are certainly 

 present in this form. At present I do not possess sufficient material 

 to follow their fate in Petromyzon, but doubtless it is the same as in 

 the three groups represented by Scyllium, Sulmo,f and Lepidosteus. 



In Raja, Labrax, Esox, and Rhodeus, I have not followed all stages 

 of their degeneration, but I have studied this sufficiently to be sure 

 that their fate is that of the ganglion cells in the groups just 

 mentioned. They may be found in larvae of Bana and Triton, and 

 in the latter form they have the usual fate — that I have determined. 

 It is very significant to notice that the forms in which they normally 

 occur are, ivithout exception, oviparous. 



The abnormal occurrence of a dozen or so giant ganglion cells in 

 Mustelus and their presence in Raja, coupled with their absence in 

 Torpedo, are interesting facts, which point to the conclusion that the 

 viviparous Elasmobranchii once possessed them as a normal develop- 

 ment. 



The giant ganglion cells which occur in adult Amphioxus and 



* Paul Mayer: "Die TJnpaaren Mossen der Selachier." ' Mittheilungen Zool. 

 Stat. Neapel,' vol. 6, pp. 228—229. 



f They probably occur in nearly all Teleostei. According to Eisig they have 

 been seen by P. Mayer in many marine forms belonging to this group. 



