11 



than those placed in water twelve hours later. After the second 

 set was placed in water all were under identical conditions. 



In tabulating the results (tables 5^6) it was found that 

 the combined results gave a generally uniform increase in weigh 

 However, there was a fluctuation with the different sets; those 

 which were at a higher temperature during the first twelve 

 hours of the soaking had gained more than the others. This is 

 clearly seen in the curves immediately following the tables, 

 in which the broken line represents the space that would have 

 been occupied by the unbroken line of the other curve had the 

 two sets of seeds given one uniform curve. 



Three complete experiments with each kind of seed were 

 carried on simultaneously as will be seen from the tables. The 

 mean per cent gain is calculated from the average weights for 

 the three experiments, and the average dry weights of the same 

 seeds. 



In the curves ordinates represent mean per cent gain and 

 abscissas the time of soaking, in hours. 



In the tables the numbers in horizontal rows close together 

 are the dry weights; the upper row of those weighed after six- 

 teen, twenty and twenty four hours soaking; the lower row gives 

 those weighed after four, eight, and twelve hours soaking. 



