ted in the return of the galvanometer needle to its original
position.

[Illustration: FIG. 4.--ELECTRIC RECORDER
(_a_) M muscle; A uninjured, B injured ends. E E' non-polarising
electrodes connecting A and B with galvanometer G. Stimulus produces
'negative variation' of current of rest. Index connected with
galvanometer needle records curve on travelling paper (in practice,
moving galvanometer spot of light traces curve on photographic
plate). Rising part of curve shows effect of stimulus; descending
part, recovery.
(_b_) O is the zero position of the galvanometer; injury produces a
deflection A B; stimulus diminishes this deflection to C; C D is the
recovery.]

#Two types of response--positive and negative.#--It may here be added that
though stimulus in general produces a diminution of current of rest, or
a negative variation (e.g. muscles and nerves), yet, in certain cases,
there is an increase, or positive variation. This is seen in the
response of the retina to light. Again, a tissue which normally gives a
negative variation may undergo molecular changes, after which it gives a
positive variation. Thus Dr. Waller finds that whereas fresh nerve
always gives negative variation, stale nerve sometimes gives positive;
and that retina, which when fresh gives positive, when stale, exhibits
negative variation.

The following is a tabular statement of the two types of response:

I. _Negative variation._--Action current from more excited to less
excited--cuproid change in the excited--e.g. fresh muscle and nerve,
stale retina.

II. _Positive variation._--Action current from less excited to more
excited--zincoid change in the excited--e.g. stale nerve, fresh
retina.[4]

From this it will be seen that it is the fact of the electrical response
of living substances to stimulus that is of essential importance, the
sign _plus_ or _minus_ being a minor consideration.

#Universal applicability of the electrical mode of response.#--This mode
of obtaining electrical response is applicable to all living tissues,
and in cases like that of muscle, where mechanical response is also
available, it is found that the electrical and mechanical records are
practically identical.

The two response-curves seen in the accompanying diagram (fig. 5), and
taken from the same muscle by the two methods simultaneously, clearly
exhibit this. Thus we see that electrical response can not only take the
pla