48 EXPERIMENTS WITH PLANTS 



whole seeds do not. The rotting is perhaps due to 

 lack of air, and the seed-cover may prevent this by 

 keeping a certain amount of air around the seed. The 

 3over likewise protects the germ from mould. But, 

 under favorable conditions, and probably in the great 

 majority of instances, the seed-cover is a hindrance 

 to germination. 



Inasmuch as our experiments indicate that in Very 

 many instances the seed-cover is a hindrance to germi- 

 nation, it would seem to be important for the plant to 

 get rid of it as soon as possible. This is the next 

 problem to consider. 



What covers seem hardest to burst open ? How much 

 force does the seed exert in bursting the covers? We 

 may get an approximate idea of this by placing the 

 seed in a pair of pliers, as shown in Fig. 38; one of 

 the two handles of the pliers is firmly wedged in a 

 hole bored in a block of wood, while the other is 

 attached by means of a wire nail, which is bent as 

 shown in the figure and passed through the hole at 

 the end of the rod attached to the spring in an ordi- 

 nary balance. The balance is to be wired at the other 

 end to a bolt which passes through a hole in a small 

 vertical piece of wood which is firmly screwed to the 

 block. On the end of the bolt is a nut, by turning 

 which the position of the balance may be adjusted. 

 If a seed is placed between the shorter arms of the 

 pliers and water is poured into the tumbler until the 



