June 17, 1898.] 



SCIENCE. 



817 



It is, in any case, of the highest impor- 

 tance to determine the role played by the 

 nerves in this connection. In sections 

 stained with htematoxylin and picrocarmine 

 there is no difficulty in tracing the nerve 

 fibres through the corium, for these fibres 

 are meduUated and stain conspicuously, but 

 in the layer of chromatophores below the 

 epithelium the fibres lose their sheaths and 

 seem to blend with the bases of the proto- 

 blasts, giving rise to the meshwork just de- 

 scribed. The fibres can often be traced to 

 the immediate vicinity of the nuclei of these 

 cells, but because of the lack of contrast be- 

 tween the protoplasm of the cells and the 

 nerve fibres it is difficult to determine their 

 respective limits. Methylen blue prepara- 

 tion stained intra vitam, in which only the 

 nerve fibres are selectively impregnated, en- 

 able one to trace the medulated nerve fibres 

 through the corium and into a subepithelial 

 network, or in some cases into cells resem- 

 bling small non-pigmented chromatophores, 

 but which may be ganglion cells of the 

 plexus. From the plexus, non-medullated 

 fibres rise into the vicinity of the nuclei of 

 the protoblasts and appear to end in knobs 

 as described by Bethe. The results of the 

 methylen blue method are to be interpreted 

 with caution, but there can be no doubt 

 that the opportunity for nervous control 

 •over the reticulum or pericellular network 

 is most complete and extensive. The nerve 

 plexus entad of the corium is most distinctly 

 stained in methylen blue preparations and 

 is of exquisite delicacy and supplies the 

 fibres to the blood vessels and chromato- 

 phores. 



An entirely different type of nerve ending 

 is found in the skin of the head. The sense 

 buds give with methylen blue the usual ap- 



like those figured by Paulicki, but we also find in- 

 stances where the meshes of the network are broad 

 bands rather than rib-like bodies and their continuity 

 with the epithelium cell-protoplasm is perfectly ob- 

 vious. 



pearance of a perigemmular set of fibres, 

 but it seems to us that there is too great 

 haste apparent on the part of those who 

 have abandoned the classical results of 

 strictly histological methods which have 

 demonstrated an entirely different type of 

 intragemmal endings. It is customary at 

 present to deny the ending of peripheral 

 nerves in cells on principle, except in the 

 case of the olfactory nerves. To be con- 

 sistent, the same objections would prevent 

 us from recognizing the olfactory cells as 

 true cellular endings for the theoretical 

 considerations which have led authors to 

 set apart the olfactory nerve as distinct and 

 different from other sensory nerves are, it 

 is believed, based on false premises. In 

 fact, the olfactory nerve is simply a per- 

 sistent embryonic nerve, and its fibres are, 

 like the early stages of all nerves, monili- 

 form series of cells which have proliferated 

 from a common source. Each segment of 

 an ordinary nerve fibre is shown by embry- 

 onic and pathological evidence to be derived 

 from a neuroblast united to its neighbors at 

 either extremity. The nuclei of the sheath 

 of Schwann are morphologically the nuclei 

 of the neuroblasts which are represented in 

 the segments to which they belong. In the 

 second type of nerve endings it is possible 

 to take either one or two alternatives. 

 Either the terminal cell is an independent 

 neurocyte developed in situ, or it is a some- 

 what differentiated segment of the nerve 

 itself. There are some reasons for accept- 

 ing the latter alternative in the present 

 case. Such endings are best seen in the 

 skin of the head of the tree-frog. Here it 

 is easy to trace the nerves through the 

 corium in bundles of three or more, and the 

 fibres pass without interruption through 

 the chromatophore-layer and lie in a special 

 cavity of the skin in such a way that their 

 tips end free in a pore connecting with the 

 exterior. 



The terminal portion is continuous with. 



