50 Exri:inMtjyTs with plants 



that while the iorce with which the seed swells may 

 be great, the bulk of swelling may be small; hence, if 

 set at zero it may raise the pressure only to 4, w^iile 

 if it had been set at 6 it would have raised it to 10. 



If it w^ere convenient to watch the apparatus con- 

 stantly, we might get still better results by varying 

 the pressure so as to just prevent the seed from swell- 

 ing. For this purpose a short section of ruler should 

 be attached to the block near the movable arm and 

 any motion of the arm at once counterbalanced by 

 turning the nut until the arm returns to the place 

 at which it stood at the beginning. 



Owing to the fact that the pliers act as a double 

 lever, the pressure must be calculated as follows: 

 Divide the distance a b, from the wire to the rivet, 

 by the distance c h, from the (center of) the seed to 

 the rivet, and multiply by the registered pressure. In 

 order to get the pressure per square inch, we must 

 divide the calculated pressure by the area of the seed 

 which is in contact with one arm (if the contact areas 

 of the two arms differ, an average must be taken). 



We shall probably find that individual seeds of 

 the same kind var}^ somewhat in the amount of 

 pressure they exert; for this reason it is better to test 

 a considerable number at once and get an average 

 result. We may do this by means of the apparatus 

 shown in Fig. 39. It consists of a half- pint agate- 

 ware cup fitting into another slightly larger one in 



