226 
THOMPSON YATES LABORATORIES REPORT 
In the first place, experimental modifications designed only to affect the physiologically interesting 
units of structure of the nerve (the axis cylinders) are also capable of producing great modifications in 
recorded electrical changes by altering the conditions of the complex wrappings which surround the axis 
cylinders. There is often reason to suppose that modifications in recorded electrical changes obtained 
after immersion of nerves in solutions of electrolytes, and even of non-electrolytes, are merely variations 
in the relative value of the outwardly demonstrable change. The real phenomenon being, perhaps, unaltered 
in value or else altered in a manner quite different from that in which its 'externally visible ' moiety is 
affected. 
The need of such cricticism is obvious when variations are produced by such extreme means as the 
immersion of nerves in saturated salt solutions, but it is also applicable and almost invariably neglected in 
other less striking instances. 
The relative value of the outwardly demonstrable change to the real phenomenon in the axis cylinder 
depends upon at least three factors — the physical characters (electrical conductivity, etc.) of the lymph, 
the nerve fibre sheath, and the axis cyclinder. It is more than conceivable that a reagent which affects 
only the nerve sheath should, even if not in an}- way modifying the condition of the axis cyclinder, produce 
an important variation in the outwardly accessible phenomenon ; and there is no difficult}- in extending 
such a proposition to include also the effect of reagents which alter the surrounding 'lymph' without 
affecting any part of the nerve fibre itself. 
This proposition also holds good for experiments in which modifications are produced in the so- 
called 'excitability' by immersion of the nerve in solutions, etc., which may produce their apparent effects 
by altering the conditions under which the stimulating current arrives in or leaves the axis cylinder with- 
out affecting the axis cylinder itself in any manner. 
Immersions in solutions, exposure to gases, variations of temperature, all alike may produce the major 
effect observed by altering the quantity of change which can make its way to the surface, or in the case of 
stimulation, by varying the amount of stimulation which reaches the axis cylinder from the surface of the 
nerve. 
Nor is this the onlv consideration which places negative results under suspicions unless stringently 
examined, for some of the experiments which are presumed to present results of value in this connexion 
have involved other complications still more undesirable. 
The ordinary diphasic record obtained by the usual means does not, it is acknowledged, present in 
either of its phases a correct, or even approximately correct, idea of the magnitude and duration of the 
electrical change traversing successive sections of the nerve. The record is the algebraical sum of two 
such real phases. In the generally chosen conditions of experiment, when the interval of nerve between 
the electrodes is a short one, and traversed by the nervous impulse in a time which is a fraction of the 
whole time occupied by the complete passage of the changes accompanying it past either point ; under 
these circumstances large portions of the two opposite phases occur at the same time, and to a great extent 
are, therefore, eliminated from the record. 
When statements are made as to the non-occurrence of portions of the anticipated electrical change, 
it is as well to immediate!}' consider the possibilities of such elimination. Important as this is when only 
a single nervous impulse is presumed to have passed the two electrode points, it becomes of vital interest 
when a second has been despatched in rapid succession to the first, for then the record becomes the 
algebraical sum of four phases, and is unrecognizable for an}' useful purpose. It is an easy matter to 
demonstrate that such an algebraical sum may apparently present four phases, or three, or even two ; 
and in fact may so closely imitate a diphasic record, the record resulting from the passage of a single 
nervous impulse, as to be mistaken for this. Such a record may then be, and probably has actually been, 
held out as a proof that the second nervous impulse did not in its passage give rise to a second travelling 
electrical change. 
