10 JSXPERIMENTS WITH PLANTS 



opening is suspected to exist may be sealed to the 

 end of a tube, which should then be placed under 

 water and blown into forcibly (see page 31). We 

 may also place the seeds in an air-pump and exhaust 

 (see page 187). 



Does most of the water enter through these open- 

 ings ? One way to find out is to stop up the opening, 

 then submerge the seed and note how much the ab- 

 sorption of water is thereby hindered (as compared 

 with untreated seeds which are submerged at the same 

 time). A good substance for closing the opening is 

 rubber cement (the kinds used for repairing bicycle 

 tires are good and easily obtainable) , or melted rubber 

 applied hot (prepared by melting good black rubber 

 in a spoon) ; if the latter is used, place whiting or 

 flour on the surface of the rubber (after it is applied 

 to the seed) to prevent it from sticking to other ob- 

 jects. Try to cover the opening only: it is difficult, 

 however, to avoid covering at the same time the adja- 

 cent portions of the seed. Thus, in the case of the 

 Horse-bean, the scar is apt to get covered. Since 

 this is apparently more porous than the rest of the 

 cover, it may be that it admits water readily; hence 

 in covering it we are perhaps introducing an error. 

 This may be offset if we cover an equal area (includ- 

 ing the scar) in the control seeds (i. e., those which 

 do not have the openings covered) without interfer- 

 ing with the opening. 



