n6 



BOTANICAL GAZETTE 



[august 



temperature record is somew 



6 



been under perfectly normal and proper conditions. 

 From this experiment it is evident that a small 



ilask of good quality may be 

 germinating power of a much 



At the 



same time 



used in this experiment, we cannot know that the seeds were equally 



mature 



can only believe that they were equally well kept. But be these 

 things as they may, this experiment demonstrates the feasibility of 

 ascertaining the germinating: nower of a given lot of seed, whatever 



$ 





* 



.4 



B 



Fig. 7. — Hemp: A, 1908; B, 1910 



its age and experience may have been, and, furthermore, shows that 

 this is possible in a short time and even with small quantities of 

 material. But on the assumption that these four lots of seed 

 differed from one another only in age, an assumption which we 

 cannot make with any assurance, it is noticeable that the germinat- 

 ing power falls abruptly; thus the germinating power of the 191 1 

 seed in 19 12 was approximately 100 per cent, of the 19 10 seed 

 approximately 95 per cent, of the 1908 seed 40 per cent, and of the 

 1907 seed less than 1 per cent. We did not have any 1909 seed, 

 and we regret this. The results recorded in table VII are graphi- 

 cally represented in tig. 8. 



