20 THE ROYAL SOCIETY OF CANADA 



service of water to the growing fruit. It seems entirely probable 

 that the consequent considerable deficit of water, resulting, as it does, 

 in a corresponding degree of wilting of the fruit, procures an internal 

 condition, probably overheating, which constitutes the stimulus to 

 abscission. 



The sudden change of behaviour on the two subsequent days, 

 during which a somewhat higher growth rate with normal amounts of 

 shrinkage obtained, is to be charged to an amelioration of temperatures 

 and humidity due to high fog, characteristic of the summer season on 

 that part of the coast of California where this study was made. We 

 may probably infer that the process of abscission had begun in the 

 parenchyma, but had up to this time scarcely affected the water- 

 carrying tissues. That this, however, occurred in the ensuing period 

 is evident, for during the last three days there was only one period of 

 slight enlargement, while the duration of the shrinkage periods 

 became prolonged (inset. Fig. 2). During the last two days there 

 was no enlargement at all. The process of abscission was found to 

 have been completed shortly before the record terminated.^ 



That the total amount of shrinkage recorded was not very great 

 was due simply to the circumstance that the trees, which are low and 

 shrubby in habit, were situated in a gulch where the humidity is 

 generally high, especially in the night, and the nut was exposed, during 

 the last part of the abscission period, to this high humidity. A further 

 record would have been of withering only. 



It may be concluded that among the various conditions which 

 lead directly or indirectly to abscission, water deficit beyond a certain 

 limit is one, and that this is the condition which led to the abscission 

 of the fruits under consideration. After abscission has proceeded 

 sufificiently to break down the vascular tissues, no growth at all is 

 possible, and, at this point, the process may be regarded as complete, 

 so far as the fate of the part involved is concerned. The usual loss of 

 all but one fruit in a raceme is to be charged to the inadequacy of the 

 stem in supplying the water lost by transpiration from the exposed 

 fruit surfaces. 



It may, therefore, be concluded that the active period of abscission 

 was included within the last three days of the record. This is sub- 

 stantiated by the second record (Fig. 3), replotted in the graph 



^ At one point the record is not quite readily to be interpreted, namely, for the 

 period 8-14 hr. beginning the 11th day (Fig. 1). It seems entirely probable that this 

 irregularity was due to some shifting of the fruit caused by the loosening. At all 

 events, the irregularity represents a total movement of 0-025 mm. of the taking arm 

 of the lever. 



