The results of experiments with two different specimens given above show that a current applied under the given conditions has practically no effect on the latent period, the slight variation being of the order of one-hundredth part of a second. This is quite negligible when the total period observed for transmission is, as in the following cases, equal to nearly 2 seconds. _Induced changes in the Velocity of Transmission._--Having found that the average value of the latent period in summer is 0.1 second, we next proceed to determine the influence of the direction of current on velocity. _Experiment 41._--As a rule, stimulus of induction shock was applied in this and in the following experiments on the petiole at a distance of 15 mm. from the responding pulvinus. The recording writer was tuned to 10 vibrations per second; the space between two succeeding dots, therefore, represents a time-interval of 0.1 second. The middle record, N in Fig. 46, is the normal. There are 17 spaces between the application of stimulus and the beginning of response. The total time is therefore 1.7 seconds, and by subtracting from it the latent period of 0.1 second we obtain the true time, 1.6 seconds. The normal velocity is found by dividing the distance 15 mm. by the true interval 1.6 seconds. Thus V = 15/1.6 = 9.4 mm. per second. We shall next consider the effect of current in modifying the normal velocity. The uppermost record (1) in Fig. 46 was taken under the action of an ‘up-hill,’ or ‘against’ current of the intensity of 1.4 microampères. It will be seen that the time interval is reduced from 1.7 seconds to 1.4 seconds; making allowance for the latent period, the velocity of transmission under ‘up-hill’ current V_{1} = 15/1.3 = 11.5 mm. per second. In the lowest record (3) we note the effect of ‘down-hill’ current, the time-interval between stimulus and response being prolonged to 1.95 seconds and the velocity reduced to 8.1 mm. per second. The conclusion arrived at from this mechanical mode of investigation is thus identical with that derived from the electric method of conductivity balance referred to previously. [Illustration: FIG. 46.--Record showing enhancement of velocity of transmission “up-hill” or against the current (uppermost curve) and retardation of velocity “down-hill” or with the current (lowest curve). N, normal record in the absence of current; ← indicates “up-hill” and → “down-hill” transmission.] That is to say, _the passage of a feeble current modifies conductivity for excitation in a selective manner. Conductivity is enhanced_ against_, and diminished_ with_, the direction of the current._ The minimum current which induces a perceptible change of conductivity varies somewhat in different specimens. The average value of this minimal current in autumn is 1.4 microampères. The effect of even a feebler current may be detected by employing a test stimulus which is barely effective.