s keeping the rest of the circuit
untouched, merely moving the clamp from the left, past the balancing
point to the right, we get either a positive, or zero, or negative,
resultant effect.

In tin the current of response is from the less to the more excited
point. In the retina also, we found the current of action flowing from
the less stimulated to the more stimulated, and as that is known as a
positive response, we shall consider the normal response of tin to be in
like manner positive.

Just as the response of retina or nerve, under certain molecular
conditions, undergoes reversal, the positive being then converted into
negative, and negative into positive, so it will be shown that the
response in metallic wires under certain conditions is found to undergo
reversal.

#Anomalies of present terminology.#--When there is no current of
injury, a particular current of response can hardly be called a
negative, or positive, _variation_. Such nomenclature is purely
arbitrary, and leads, as will be shown, to much confusion. A more
definite terminology, free from misunderstanding, would be, as already
said, to regard the current towards the more stimulated as positive,
and that towards the less stimulated, in tissue or wire, as negative.

The stimulated end of tin, say the end A, thus becomes zincoid, i.e.
the current through the electrolyte (non-polarisable electrodes with
interposed galvanometer) is from A to B, and _through the wire_, from
the less stimulated B to the more stimulated A. Conversely, when B is
stimulated, the action current flows round the circuit in an opposite
direction. This positive is the most usual form of response, but there
are cases where the response is negative.

In order to show that normally speaking a stimulated wire becomes
zincoid, and also to show once more the anomalies into which we may
fall by adopting no more definite terminology than that of negative
variation, I have devised the following experiment (fig. 51). Let us
take a bar, one half of which is zinc and the other half copper,
clamped in the middle, so that a disturbance produced at one end may
not reach the other; the two ends are connected to a galvanometer
through non-polarisable electrodes. The current through the
electrolyte (non-polarisable electrodes and interposed galvanometer)
will then flow from left to right. We must remember that metals under
stimulation generally be