386 BOTANICAL GAZETTE [may 



If the velocity of absorption were an exponential function of 

 the temperature, the relation between temperature and the rate 

 of entry of water into the seeds might be expressed by an equation 

 of the form v = ae ke in which 6 is the temperature. As I have 

 obtained evidence somewhat adverse to the assumption that 

 velocity of absorption is an exponential function of the temperature, 

 this equation does not hold. Wherever the logarithmic formula 

 , y = a log IO (fo:+i) +c holds for the curves of absorption, the velocity 

 of intake may be represented by the formula v = ae~ ktt> in which <f> 

 is the percentage of water already absorbed. In other words, the 

 velocity of intake is approximately an inverse exponential function 

 of the total preceding absorption. It is not claimed that this is true 

 for all cases of absorption, but that it is just as true as the logarith- 

 mic equation used. Wherever that equation holds, the velocity 

 formula holds. 



The chief interest centers in the temperature coefficient of 

 absorption. I have obtained coefficients ranging from 1.55 to 

 1.83 in Xanthium seeds, and 1.6 in split peas. These are all 

 above the temperature coefficient of physical changes, and below 

 that for chemical change. Brown and Worley obtained a 

 value above 2, and adopted the idea that absorption was 

 conditioned as to rate, in the case they studied, by some 

 chemical change. In seeking a chemical change to account 

 for their observations, they suggested that the semipermeable 

 seed coat of barley was involved in a special way, in its relation 

 to complex or simplified water molecules. They suggested the 

 possibility that the differential septum (semipermeable coat) 

 permits only hydrone to penetrate it, and that the temperature 

 rise increases the proportion of hydrone in solution. One of the 

 main difficulties in the way of accepting such a hypothesis as to 

 the relation of hydrone to semipermeable membranes, is its impli- 

 cation that all semipermeable membranes should behave alike. 

 Xanthium and Hordeum both have semipermeable membranes, 



the 



same 



ipermeable memb 



always individualistic. Each kind has its own behavior, no two 



