#Apparent fatigue when stimulation frequency increased.#--If the rhythm of stimulation frequency be now changed, and made quicker, certain remarkable modifications will appear in the response-curves. In fig. 20, the first part shows the responses at one minute interval, by which time the individual recovery was complete. The rhythm was now changed to intervals of half a minute, instead of one, while the stimuli were maintained at the same intensity as before. It will be noticed (fig. 20, _b_) that these responses appear much feebler than the first set, in spite of the equality of stimulus. An inspection of the figure may perhaps throw some light on the subject. It will be seen that when greater frequency of stimulation was introduced, the tissue had not yet had time to effect complete recovery from previous strain. The molecular swing towards equilibrium had not yet abated, when the new stimulus, with its opposing impulse, was received. There is thus a diminution of height in the resultant response. The original rhythm of one minute was now restored, and the succeeding curves (fig. 20, _c_) at once show increased response. An analogous instance may be cited in the case of muscle response, where 'the height of twitch diminishes more rapidly in proportion as the excitation interval is shorter.'[11] [Illustration: FIG. 21.--FATIGUE IN CELERY Vibration of 30 deg. at intervals of half a minute.] From what has just been said it would appear that one of the causes of diminution of response, or fatigue, is the residual strain. This is clearly seen in fig. 21, in a record which I obtained with celery-stalk. It will be noticed there that, owing to the imperfect molecular recovery during the time allowed, the succeeding heights of the responses have undergone a continuous diminution. Fig. 22 gives a photographic record of fatigue in the leaf-stalk of cauliflower. [Illustration: FIG. 22.--FATIGUE IN LEAF-STALK OF CAULIFLOWER Stimulus: 30 deg. vibration at intervals of one minute.] It is evident that residual strain, other things being equal, will be greater if the stimuli have been excessive. This is well seen in fig. 23, where the set of first three curves A is for stimulus intensity of 45 deg. vibration, and the second set B, with an augmented response, for stimulus intensity of 90 deg. vibration. On reverting in C to stimulus intensity of 45 deg., the responses are seen to have undergone a great diminution as compared with the first set A. Here is seen marked fatigue, the result of overstrain from excessive stimulation.