364 BOTANICAL GAZETTE [may 







plants by all their external characteristics were unmistakably 

 true X. pennsylvanicum. These were used in the later work to 

 give an idea of the variability to be encountered in the moisture 

 intake by a given kind of substance. 



The absorption took place in test-tubes of distilled water which 

 were kept at the desired temperature by standing them in a water 

 bath. Care was taken, particularly in the later work, to have the 

 seeds at the same temperature as the water when they were first 

 brought together. Three temperature curves are discussed in 

 the present paper, 5, 20, and 35 C. Tests were run at 5 intervals 

 from 5 to 50 C, but these three stand near to the temperatures 

 used by Brown and Worley, and afford a satisfactory basis for 

 comparison. The others have been omitted. In all cases the 

 fluctuation rarely exceeded 0.25° on either side of the chosen 

 temperature during the significant period of intake. 



At close intervals the seeds were removed from the water, 

 dried uniformly and quickly on filter paper, and weighed with 

 analytical accuracy. The time periods of immersion were made 

 as sharp and accurate as possible, and the time during which the 

 seeds were out of the water was reduced to the lowest possible 

 limit. The drying required 10-20 seconds usually, and the weighing 

 was done as rapidly as accuracy permitted. During this period 

 the seeds had some opportunity to change from the temperature 

 of absorption in the 5 and 35 tests, but histeresis of the seed 

 colloids would tend to prevent serious alterations in colloidal 

 aggregation during the brief interval involved. The errors due to 

 such changes would be slight. The intervals between weighings 

 were made short throughout the work. The first weighing was 

 always made at the end of 1 minute to catch the very rapid initial 

 intake. Succeeding intervals were usually 10 or 15 minutes, or 

 longer when continuous attention could not be given to the work. 

 The time intervals used will always be indicated in the tables with 

 the absorption data. In all cases the time needed for drying and 

 weighing was subtracted. This weighing at intervals was con- 

 tinuous in the case of Xanthium seeds until the intake was well 

 above 35 per cent out of a possible 50-55 per cent. By the time 

 40 per cent of water had been taken in, the velocity of intake always 



