Mason —Growth and Abscission in Sea Island Cotton . 481 
The next phase of the inquiry dealt with the effect of certain 
environmental factors upon the daily growth-rate of the main axis and 
the rates of shedding. It was found that dark, humid days, during which 
the rate of evaporation remained low and on which a great deal of rain 
generally fell, resulted in marked retardation in the growth-rate of the 
main axis, and that in the later stages of the plants’ development they were 
the invariable precursors of augmented rates of boll-shedding. It was 
assumed that the occurrence of external conditions of this nature must 
operate to check the rate at which food was elaborated, and that this check 
was the cause both of the retardation in the growth-rate of the main axis 
and of the augmented rates of shedding. The absence of laboratory 
facilities rendered it impossible to place this hypothesis on an experimental 
basis. It was shown, however, that the removal of all the foliage leaves 
was followed within from three to eight days by the dropping of almost all 
the young fruit. It was also found that the removal of all except a few 
fruits considerably prolonged the interval elapsing between the application 
of the stimulus and the completion of abscission. Moreover, it was found 
that pruning off the fruit, and thereby diminishing the number of competing 
centres, resulted in augmented growth-rates of the main a:£is. 
Inasmuch as previous work has consistently pointed to a water-deficit 
in the body of the plant as being the most constant cause of shedding, an 
experiment was made to determine whether incipient drying was sufficiently 
pronounced, under the humid conditions of St. Vincent, as to markedly 
retard the growth of the plant. As the result of measurements which 
were made both in the morning and the evening, it was found that 
growth was only slightly inhibited while the sun was up. This was not 
unexpected, for the desiccating power of the aerial environment is normally 
very small during the period in which cotton is cultivated in St. Vincent. 
In view of this, and the considerations just advanced, the conclusion seems 
inevitable that the growth-inhibition which precedes shedding in St. Vincent 
is not the result of pronounced incipient drying, but is due rather to 
a retardation in the rate at which assimilates are produced by the plant. 
It should be emphasized, however, that a series of abnormal saturation deficits 
would also check the rate at which carbohydrates are elaborated. 
Finally, it was shown that the growth-rate of the boll fell normally 
into two distinct stages. During the first period growth continued for 
some four to five days after anthesis, but at a declining rate. Growth 
throughout this period was apparently in no way dependent on fertilization. 
During the second stage the rate increased until finally it declined. The 
initiation of this stage was found to be dependent on fertilization. It was 
concluded that the stimulus supplied by fertilization operated in some way 
to ensure a movement of assimilates into the boll. In the absence of 
fertilization the boll underwent a steady decrement in growth until it was 
