1 
62 THOMPSON YATES AND JOHNSTON LABORATORIES REPORT 
Aside from minor series of experiments not here reported, one hundred 
and thirty-five sciatic and fifty-five ulnar nerves were immersed and measured. 
Each of the eleven strengths of solution was tried on at least ten sciatic nerves 
and four ulnar nerves. The number of cases and the uniformity of results are 
sufficient to guarantee the average determinations from large chance errors. 
The lengthwise resistance of the sciatic nerve of the cat, when examined 
a few minutes after death, at i8°C, was found to average 200*7 ohms per 
cubic centimeter. The specific resistance of the ulnar nerve was somewhat 
smaller, 186*0 ohms on the average. Otherwise expressed, the specific conductivitv 
of the sciatic nerve is 0*0050 mhos, and that of the ulnar nerve 0*0054 mhos. 
The substance of the ulnar nerve is thus a better conductor by about eight per 
cent, than the substance of the sciatic. This is not a chance result, but is almost 
invariable for every cat used, though a few exceptions occur in which the sciatic 
has as high conductivity as the ulnar. 
The specific conductivity (lengthwise) of the posterior spinal roots is con- 
siderably less than that of the sciatic nerve, being about 0*003 m hos, and that 
of the cord is somewhat less still. Gothlin 1 has found the specific conductivity 
of the corpus callosum to be much less than that of a peripheral nerve, and 
supposes 2 the difference to be due to the presence, in the sheath of the peripheral 
nerve, of well-conducting lymph spaces. If this were so, then freeing the peri- 
pheral nerve fibres of part of their envelopes and lymph spaces would increase the 
longitudinal specific resistance of the nerve. It is easy, in case of the sciatic, to 
pull out without violence one of the large strands of which the trunk is composed. 
I have done this with eleven nerves, and find the specific conductivity to be 
increased, not diminished. The specific conductivity of the strand is about one- 
eighth greater than that of the whole nerve. Now since the strand contains a 
larger proportion of nerve-substance proper than the whole trunk, the higher 
conductivity of the strand proves that nerve-substance proper is more conductive 
than the connective tissue envelopes with their lymph spaces. Moreover, in 
' nerve-substance proper ' is contained the non-conductive medullary sheath, as 
well as the axis cylinder, each occupying about half of the cross-section. The 
specific conductivity of the axis cylinder must, therefore, be more than twice as 
high as that of the outer envelopes with their lymph spaces. This result goes 
to substantiate Macdonald's view that the interior of the nerve-fibre contains a 
more conductive, and therefore more concentrated, solution of electrolytes than 
the lymph bathing the outside of the fibres. 
The high resistance of the posterior roots, and of the cord, is probably 
to be explained by the lack of parallelism of the fibres. It is known 5 that the 
1. G. F. Gothlin. Upsala Lcikareforenings Forhandlmgar, 1902, VIII, 156, 163. 
2. Gothlin, op. at., p. 165. 
3. Tereg. Archif.fur {Anat. u.) Physiol. 1899, p. 319 ; Gothlin, op. cit., p. 163. 
