28 liXPElilMENTS WITH PLANTS 



air-tight joint with the glass cover. The apparatus 

 shown in Fig. 27 is very convenient for this purpose: 

 the inverted bottle has a cork in the sides of which 

 two grooves are cut, so that as soon as the water sinks 

 below the mouth of the bottle air enters through the 

 grooves and water runs out until the level rises and 

 closes the grooves: the water-level thus remains 

 almost constant. 



As the experiment is to last for some time, it is 

 advisable to take precautions against the growth of 

 mould. Place the seeds in a tumbler and set them in 

 a jar or pail in the bottom of which is a little forma- 

 lin: cover the pail so as to confine the formalin vapor, 

 and let it stand five or six hours. 



Control seeds (treated in the same way) should be 

 planted in moist earth or sawdust. 



The experiment should be continued two or three 

 months if necessary. The percentage of moisture in 

 the seeds may be determined at the beginning and 

 again at the end of the experiment, to see how much 

 moisture they have absorbed and also the minimum 

 amount which will enable the seed to germinate. 



The power to absorb water from moist air may be 

 of great practical importance in storing seeds, etc., 

 since if kept in a damp place they may easily germi- 

 nate or decay. 



Of especial interest, in considering the absorption 

 of water, are the mucilaginous covers of Flax, Quince, 



