Structure and Polarity of Electric Motor Nerve-Cell in Torpedoes. 239 
25 and 30, we find fish that were killed without electrical activity and 
show 99 and 95 per cent of ventral orientation respectively. Other 
experiments not recorded here have also shown that the fish which die 
quickly do not always show a general or central distribution of the 
plasmosomes. 
The third condition can now be discussed, namely, that this orienta- 
tion of the plasmosome is a permanent condition. To arrive at a con- 
clusion it must be shown that there are no temporary movements of 
the plasmosomes during the electrical activity of the fish. 
Fresh slices of the electric lobe were cut with a Valentine’s knife in a 
vertical direction and after being covered and put under an immersion 
lens they were subjected to electrical and other stimuli that would 
have been sufficient to cause their discharge in a living fish. While 
thus being stimulated they were closely watched and it was noted 
that no movements of the plasmosomes occurred. 
The fact that the plasmosomes exhibit a widely differing percentage 
of orientation in different individuals and that this percentage appears 
to be the same in all parts of the electric lobes of each individual might 
also seem to indicate that there were no temporary movements of the 
plasmosome. This has been further controlled by observing both fresh 
material as mentioned above and different fixations of the same speci- 
men, after further stimulation followed by fixation and sectioning. Thus 
No. 10 was killed with the knife and fresh material was at once put 
under the microscope, where it was determined that the percentage 
of ventral orientation was 5 per cent. This was a very weak ventral 
orientation, so the remaining tissue, still capable of physiological action, 
was further stimulated both mechanically and electrically and then 
bits were fixed and sections cut. These preparations showed no 
futher orientation, so it was concluded that the excess stimulation had 
not moved them, and since this stimulation was applied to cells that 
were not ventrally oriented, and in which there was still room for 
orientation, this was considered to be strong proof that there is no 
ventral movement during the nerve activity which results in the 
electric discharge. 
Another strong point must be considered at this time. Magini 
erroneously assumed that the functional axis of these cells lay in all 
cases in a dorso-ventral direction, with the implantation cone and neu- 
raxis on the ventral side of the cell (text-fig. 4); and he therefore states 
in some places that the plasmosome (nucleolo) moves ventrally, and 
in other places he states that it moves toward the origin of the neuraxis. 
The writer, as stated in the description of the cells above (page 216), has 
found that a certain number of cells in each electric lobe, about 20 per 
cent of them, are so situated that the neuraxis leaves in a lateral or 
even in a dorsal direction (text-fig. 5), and that in these cases the plas- 
mosome acts in accord with the majority of usual cases, 7. e., those with 
