THE MOLECULAR FORCES IN PLANTS. 195 



the commencement of the experiment, and subsequently at in- 

 tervals of five minutes by means of a millimetre scale. We ob- 

 serve that for some time the water rises higher and higher. This 

 goes on for three-quarters of an hour, or sometimes even for an 

 hour and a half. Then the water sinks for a short time, or 

 under some conditions even for a few hours, finally again 

 rising. In exact investigations the influence of temperature on 

 the position of the water must be taken into consideration. 



If seeds are placed in contact with water, and imbibition takes 

 place, the water entering the seeds must suffer condensation. 

 Such a process would, however, lead to a diminution of the 

 aggregate volume of seeds and water. Of this, however, nothing 

 is to be seen in our experiment. On the contrary, at the com- 

 mencement of the experiment, the water rapidly rises in the tube, 

 clearly indicating an increase in the aggregate volume of the 

 seeds and soaking fluid. The causes underlying this are therefore 

 at all events preponderating in their effect on the position of the 

 water, and they are to be sought in a folding of the testa, which 

 takes place in the first stage of the soaking. The testa raises 

 itself from the cotyledons of the seed, and between the cotyledons 

 and the seed-coat are formed cavities filled with rarefied air, so 

 that the aggregate volume of the seeds and the water must 

 necessarily be increased. If, in fact, we employ peas whose testa 

 we have injured, the ascent of the water in the tube, as I have 

 found, does not take place during the whole of the first period of 

 the experiment. 



During the second period of the experiment the water sinks in 

 the tube, indicating a decrease in the total volume of the seeds 

 and water, and this must be referred to the penetration of water 

 into cavities of the seeds. I do not here proceed to consider the 

 causes leading to the renewed rise of water in the tube during the 

 third period of the experiment. As to this, and also as to the 

 behaviour of other seeds in this respect, my work cited below is 

 to be consulted. 1 



1 See Detmer, Vergleichende Physiologic des Keimungsprocesses der Samen, 

 Jena, 1880, p. 71, and Mattirola's work cited below. See also Botan. Central- 

 Matt, 1892, Bd. 52, p. 155. 



