90 THOMPSON YATES LABORATORIES REPORT 
anterior in an overlapping series, so that each member contributes to a sloping edge, and the 
middle membei juts the farthest. It results, therefore, that the posterior as well as the anterior 
side of the limb is as regards skin sloping, and that neither side is covered by a single segment only. 
When previously pointing out this discrepancy between the arrangement of the fields of 
distribution of the motor and sensory roots, I added:* 'it must be remembered that the fields of 
the sensory roots as delimited in the present research are cutaneous and literally skin-deep only. 
In such glimpses as we obtain of the distribution of sensory nerve-fibres to muscles, they seem to 
correspond segmentally with the motor supply.' ' The diflference between the arrangement found 
for the motor-roots and that for the sensory may really be due less to the comparison being of 
efferent with afferent distribution, than to the comparison being of muscle with skin.' I have now 
observations which, I think, prove the above suggestion to be correct. The observations are as 
follows : — 
I. The phrenic, which may be taken as a muscular nerve, contains fibres from the IVth, 
Vth and Vlth cervical spinal ganglia, i.e.^ from the sensory roots of exactly the same spinal 
segments as those whence motor fibres are furnished to it. 
II. Afferent fibres in the nerve to vastus medialis and femoralis muscles are traceable to 
the spinal ganglia of the Vth and IVth lumbar nerves of Macacus, that is, to exactly the nerves 
which furnish the motor innervation of those muscles. I showed previously that the afferent 
nerve-fibres of these muscles pass through the Vth lumbar root, but I failed to demonstrate them 
with certainty in the root of the IVth lumbar, although in this latter root the corresponding 
motor-fibres are easily detected. Returning to the experiments again, I have since found that 
by employing strychnia it is possible to distinctly detect the existence of afferent fibres from 
vastus medialis and femoralis in the root of the IVth lumbar, in addition to those in the Vth. 
The proof is furnished by the persistence of the knee-jerk under strychnia poisoning, after 
severance of the obturator and sciatic trunks and of all branches of the femoral nerve, except that 
to vastus medialis and femoralis, together with section of the sensory root of the Vth lumbar nerve. 
The jerk persists, although in a crippled manner ; it is completely (with the reservation explained 
in next paragraph) abolished by subsequent section of the sensory root of the IVth lumbar nerve. 
The same appears true for the Cat, if for Vth and IVth lumbar nerves Vlth and Vth be 
substituted. But in the Cat I have twice observed a phenomenon, recorded by Westphal, which might 
appear to invalidate the above evidence. Westphal states, on the strength of two observations on the 
Dog, that after exhibition of strychnia the knee-jerk is obtainable after section of all the afferent roots of 
the plexus ; these observations form a part of his evidence that the jerk is not a reflex. In spite of their 
importance I cannot find the observations to have been ever repeated by Westphal or by others. 
Certainly, in the Cat, after section of all the afferent roots of the plexus, if strychnia be exhibited, a tap 
upon the patellar tendon occasionally evokes an extension of the knee, indistinguishable, as far as inspection 
can judge, from a true knee-jerk. But the phenomenon is exceptional, not the rule ; thus it occurred in 
two out of six experiments. The exceptional production of this condition by strychnia need not really 
confuse its regular efJect of rendering sufficient for production of the jerk the comparatively slender 
afferent path in the IVth lumbar {Macncus) root, after section of which the jerk, even under strychnia, is 
as a rule absolutely lost. 
* ' Phil, Trans.,' B, vol. 184, !oc, clt, I have since examined the sensory nerve-fibres in muscles, 'Journ. of Physiol.,' 
vol. 17, 1894. 
