EXPERIMENTS WITH PLANTS 



to No. 2 twice as much as to No. 1, to No. 3 twice as 



much as to No. 2, and so on. 



Another method is to use fruit-jars (one-quart or 



two-quart), placing the seeds on the bottom of the 



jar and covering them with half an inch of sand. Add 



different quantities of water to 

 the different jars, and screw on 

 the tops tightly. 



Whichever method be used, 

 it is interesting to determine 

 the amount of water in the 

 seed at the stage when the cau- 

 licle begins to protrude. This 

 may be ascertained by weigh- 

 ing- the seed and then drying it 

 (on a water-bath; see Fig. 54) 

 until it ceases to lose weight; 

 dividing the loss in weight by 

 the weight of the undried seed 

 will give the percentage. The 

 determination is especially in- 



25. Seeds placed in a saturated , . . .> 



atmosphere. terestmg in the case of the 



seeds which germinate with the minimum amount of 

 moisture. 



It often happens that a seed has little or no water 

 at its disposal except the moisture of the air. Unless 

 it can absorb this in sufficient quantities, it cannot 

 germinate. 



