I20 



BOTANICAL GAZETTE 



[FEBRUARY 



It will be noticed even when magnesium greatly preponderates in 

 the mixed solutions the plants live twenty-five times as long as in pure 

 MgClj, and over eight times as long as in pure KCl. Increasing the 

 proportion of potassium increases the length of life. 



The same relation is seen more completely in the next table (Table 

 V). Decreasing the amount of Mg causes increased growth up to a 

 certain point (loo K + io Mg). Still further decrease of the relative 

 amount of Mg beyond this point is unfavorable. The optimum 

 relation is therefore not far from lo Mg + loo K. 



TABLE V 



Liverworts 



All quantities given axe cubic centimeters of .037SW solutions 



It will be noticed also that the gemmae made no growth whatever 

 in pure MgClj or pure KCl, while in mixtures of the two a good growth 

 occurred. 



For the study of flowering plants wheat was chosen. The seeds 

 were supported in the solutions on strips of filter paper as described 

 in a previous paper.'* The results agree with those already given. 

 Table VI shows that certain mixtures of potassium chlorid and mag- 

 nesium chlorid are much more favorable than either of the pure salts 

 (see also figs. 2, j). 



4 0sTERH0trr, BoT. Gazette 44:259-272. 1907. 



