xi ELECTRICAL FISHES 367 



According to Fritsch (who is again contradicted by Iwanzoff), 

 " it is possible, in specially favourable parts of the cross-sections 

 of a plate, to discover fine threads of nerve at right angles to the 

 direction of the plates, entering at the granular layer and dis- 

 appearing between the granules." Beyond these again, in the 

 palisade border, they are plainly visible, and form the direct 

 boundary of the palisade (Fig. 236). Their proper ending seems 

 to be at the dorsal border of this layer, in " soft protoplasmic 

 bodies, which in the preparation cohere in spherical (berry -like) 

 masses." Beyond the border of the palisade comes the dorsal 

 (" muscular," Fritsch ; " metasarcoblastic," Babuchin) part of the 

 plate (Ranvier's couclie intermddiaire). This layer, which is 



FIG. 236. Torpedo ocellata. T.S. of electrical plate with dependent nerve. 1 = dorsal border, or 

 layer, of connective tissue; m stratum moleculare ; p= palisade-border with nerve-endings; 

 gl = stratum granulosum ; n nerve. (G. Fritsch.) 



developed by the transformation of embryonic muscle-substance, 

 exhibits nothing further of the characteristic structure of striated 

 fibres. Krause, indeed, describes " striated, bowed fibres " as a 

 residue of muscle-fibrils, but other observers have not detected 

 them. According to Fritsch, this layer, like the stratum 

 granulosum, " is composed of minute particles, set in rows parallel 

 with the axis of the prisms, and little more refractile than the 

 intermediate substance" (Fig. 236, m). 



Fritsch inclines to regard the regular structure (as observed 

 by him) as a confirmation of du Bois-Reymond's molecular 

 theory, though he does not go so far as to affirm that the rows 

 of particles are actually " the required electromotive molecules." 

 Iwanzoff, on the contrary, views them merely as a kind of honey- 

 comb or foam-like structure of the plasma of which the "inter- 

 mediate, layer" consists. 



