each series. The dilutions shown in the tables given below are 

 based on freezing-point determinations made by Dr. Shive, espe- 

 cially for this work. Other plans of dilution (which cannot alter 

 the volume-molecular salt proportions in any given case) may of 

 course be employed. The osmotic values are to be considered as 

 much less precise than the salt proportions. 



In this general connection, it may be noted that actual ionic 

 partial concentrations of the various salts in any of these three- 

 salt solutions is at present impossible to determine (with the 

 single exception of the hydrogen-ion, for which a method is of 

 course available) , and any mental picture of assumed values 

 of the degrees of ionization of the various kinds of salt molecules 

 must depend upon more or less probable assumptions, for the 

 actual testing of which no methods have yet been devised. The 

 physical chemistry of such three-salt solutions as these is still 

 far beyond us, excepting in its most general aspects. But physi- 

 ology and agriculture need not wait for the advance of physical 

 chemistry in this connection ; we aim to determine the physio- 

 logical properties of our solutions (by plant tests) and merely to 

 define our solutions in such a way that they may be reproduced 

 at any time in the future, for advanced physical-chemical study, 

 etc. Any solution is sufficiently defined for exact reproduction 

 when the volume-molecular partial concentration is stated for 

 each of the salts used. 



For a beginning, we wish to compare the twenty-one different 

 sets of salt proportions for each of the six types of solution and 

 for the same total osmotic concentration throughout the entire 

 series. The uniform total concentration adopted should be one 

 that promises to be suitable for good growth of the plants, and, 

 at the same time, it must be such that all of the solutions may be 

 possible in this concentration without the formation of precipi- 

 tate. The osmotic value of 1.00 atmosphere at 25° C. is chosen 

 as the index of this uniform total concentration. There are thus 

 126 different solutions (really only 123, see below) to be com- 

 pared in the beginning, all having the same osmotic value (1.00 

 atm.), but all differing in other ways (salt proportions or kinds 

 of salts used). 



The formulas for the 126 solutions, and fpr the universal con- 

 trol solution (Shive's IR5C2, increments of 1/10, 1.75 atm.) are 

 given in the accompanying tables, which have been prepared by 

 Dr. Shive for this project. It will be noted that there is a sepa- 

 rate table for each of the six types of solution and that each table 

 presents the twenty-one different sets of volume-molecular salt 



28 



