Longevity of Seeds 123 



which it was grown; (3) the degree of ripeness; (4) 

 the manner in which it is stored and handled ; (5) 

 the condition as respects mechanical or insect injuries. 

 There is a limit to the life of every seed, yet when 

 this limit is reached there may be no evidence in 

 the seed itself, either external or internal, to indicate 

 the loss of vitality. The seed must be placed under 

 circumstances most favorable to its germination, and 

 its condition tested by its ability or inability to grow. 

 In practice, the seeds of a given plant are tested 

 in a greater or less quantity, the percentage of the 

 quantity which germinates being adopted as the meas- 

 ure of germinative vitality. This results in a test of 

 the species, variety, or the sample as a whole, rather 

 than in a test of individual seeds. Hence it follows 

 that the limit of viability in commercial tests does not 

 represent the extreme age at which it is safe to plant, 

 there being a constant failure of individual seeds from 

 the first instance of loss of life until no one seed in 

 the sample remains alive. It therefore becomes appar- 

 ent that, as a rule, the fresher the seeds the better the 

 results, independent of the figures representing extreme 

 duration of vitality. Haberlandt shows that there is an 

 increasing failure in seeds kept from year to year. In 

 the tests 100 seeds were used in each case: 



1 yr. old 2 yrs. 3 yrs. 4 yrs. 6 yrs. 



Wheat ... 96 germinated. 84 60 73 4 



Barley ... 89 92 33 48 , . 



Oats .... 96 80 32 72 48 



The germinative power of seeds is sometimes tested 

 by placing the seeds on live coals. Those which snap 



