THE RECORD OF GROWTH-RESPONSE 41 I 



is characterised by the same cyclic variation which we have 

 observed in the case of multiple response ; (4) that just as the 

 application of appropriate stimulus renews pulsation in a 

 Desinodinni at standstill, so, in a plant with growth at stand- 

 still, appropriate stimulation renews pulsatory growth ; and, 

 lastly (5), that the modifying influence of external agents is 

 similar in both cases. 



From a series of observations, taken at intervals of several 

 minutes, on Spyrogym princeps, Hofmeistcr found that growth 

 undergoes fluctuation, the first and second maximal points in 

 his series of observations being separated by an interval of 

 forty-four minutes, and the second and third by an interval 

 of ninety-five minutes. Such experiments, however, have 

 laboured under the great disadvantage of discontinuity, and 

 in order to overcome this I undertook to devise some appa- 

 ratus by whose means growth-pulsations might be recorded 

 continuously, in such a way as to give not only the period, but 

 also the individual peculiarities of each pulsation. And I may 

 here forestall matters to say that by such means I have been 

 able to detect longitudinal pulsations two hundred times as 

 quick as those observed by Hofmeister. 



Conditions to be kept in view^ — Before attempting to 

 demonstrate the pulsatory character of growth-movements, 

 however, I shall point out certain facts which it is essential 

 to remember. We have seen that under rapidly succeeding 

 excitations, the separate responsive effects become merged, 

 and a response is produced, which is apparently continuous, 

 although the stimuli themselves were discontinuous. This 

 is seen, for instance, in the first part of the tetanic curve 

 (figs. 49, 50). But when once the maximum responsive 

 effect is produced, as there can be no further additive 

 effect, the subsequent responses show themselves in a series 

 of fluctuations of the top of the tetanic curve. In that 

 form of response, which we are now considering, as there is 

 no maximum limit, the additive effect of growth continues 

 indefinitely. It is thus clear that when rhythmic excitation 

 is very rapid, it may produce a growth-movement which 



