146 BOOKS AND CURRENT LITERATURE 



one (2.50 atm.) near the optimal concentration for plant growth, while 

 the lowest (0.05 atm.) was well below and the highest (8.15 atm.) was 

 well above the optimal concentration. For each of these three series 

 the total osmotic concentration was considered as divided into ten 

 equal parts, and these tenths were distributed among the four com- 

 ponent salts in all possible ways; each series thus contained 84 cultures, 

 all of which had the same total osmotic concentration. Thus in the 

 various solutions of a series, from 1 to 7 tenths of the total osmotic 

 concentration was due to each of the four constituent salts, but no two 

 solutions contained the same proportions of the four salts. To repre- 

 sent the different salt proportions in the various solutions of a series a 

 tetrahedral diagram, composed of seven plane triangular diagrams, was 

 used. These diagrams also served as charts upon which the plant 

 measurements, such as dry weight, etc., were plotted. 



A trace of iron, as ferric phosphate, was added to each culture, and 

 a control culture in distilled water was included in each series. The 

 evaporating power of the air was measured during each of these experi- 

 ments by means of a cyHndrical porous cup atmometer, and the rela- 

 tive water loss by transpiration was measured for selected cultures for 

 a period at the close of each experiment. 



In the series of optimal concentration (2.50 atm.) the solution pro- 

 ducing the greatest dry weight of tops was 11% superior, by this cri- 

 terion, to Knop's solution of the same osmotic concentration. The 

 volume-molecular partial concentrations of the four salts in this best 

 solution were: KNO3, 0.00488 m; KH2PO4, 0.01551 m; Ca (N03)2, 

 0.01440 m; and MgSO, 0.01660 m. The solution giving the greatest 

 dry weight of roots was 29% superior to Knop's solution, but this was 

 not the same solution as that giving the greatest dry weight of tops. 

 In the suboptimal series (0.05 atm.) a still more marked improvement 

 over Knop's salt -proportions was obtained. The best solution for the 

 growth of tops gave a top yield 39% greater than Knop's, while another 

 solution of this series gave 15% better root growth. 



Similarly, salt proportions superior to those used by Knop were 

 found in the series of supra-optimal concentration, which had a calcu- 

 lated osmotic concentration of 8.15 atmospheres. Shive's^ cryoscopic 

 determinations showed that Totcingham's calculation gives too high 

 concentrations for these solutions, the average value obtained by Shive 

 being 7.22 atmospheres. For the optimal and suboptimal series Shive's 



^Shive, J. W., The freezing points of Tottingham's nutrient solutions. Plant 

 World 17: 345-353. 1914. 



