airly constant value after the third or fourth response.
After taking these records of normal response, I placed the specimens in
an ice-chamber, temperature 0 deg. C., for twenty-four hours, and
afterwards took their records once more at the ordinary temperature of
the room. From these it will be seen that while the responsiveness of
Eucharis lily, known to be susceptible to the effect of cold, had
entirely disappeared, that of the hardier plants, Holly and Ivy, showed
very little change (fig. 36).

Another very curious effect that I have noticed is that when a plant
approaches its death-point by reason of excessively high or low
temperature, not only is its general responsiveness diminished almost to
zero, but even the slight response occasionally becomes reversed.

[Illustration: FIG. 36.--AFTER-EFFECT OF COLD ON IVY, HOLLY, AND
EUCHARIS LILY
_a._ The normal response; _b._ Response after subjection to freezing
temperature for twenty-four hours.]

#Influence of high temperature, and determination of death-point.#--I next
tried to find out whether a rise of temperature produced a depression of
response, and whether the response disappeared at a maximum
temperature--the temperature of death-point. For this purpose I took a
batch of six radishes and obtained from them responses at gradually
increasing temperatures. These specimens were obtained late in the
season, and their electric responsiveness was much lower than those
obtained earlier. The plant, previously kept for five minutes in water
at a definite temperature (say 17 deg. C.), was mounted in the vibration
apparatus and responses observed. The plant was then dismounted, and
replaced in the water-bath at a higher temperature (say 30 deg. C.) again,
for five minutes. A second set of responses was now taken. In this way
observations were made with each specimen till the temperature at which
response almost or altogether ceased was reached. I give below a table
of results obtained with six specimens of radish, from which it would
appear that response begins to be abolished in these cases at
temperatures varying from 53 deg. to 55 deg. C.

TABLE SHOWING EFFECT OF HIGH TEMPERATURE IN
ABOLISHING RESPONSE

Temperature Galvanometric response
(100 dns. = .07 V.)

(1) {17 deg. C 70 dns.
{53 deg. " 4 "

(2) {17 deg. " 160 "
{53 deg. " 1 "

(3) {17 deg. " 100 "