338 
B. E. LIVINGSTON AND W. E. TOTTINGHAM 
of any kind of plant may be possible, all six of these salt combinations 
must of course be thoroughly tested with reference to the plant in 
question. Hitherto, only the first of these has received attention in 
the literature, the salts of group I being those employed in Shive's 
elaborate study. The present paper shows the results of a preliminary 
test of a series of solutions using group III — potassium nitrate (KNO3), 
magnesium sulphate (MgSOJ, and mono-calcium phosphate (Ca- 
(H2P04)2). 
In order to investigate in an adequate way the combination of 
these three salts in plant cultures, it is of course necessary to include a 
number of different sets of salt proportions and a number of total con- 
centrations or osmotic values, as both Tottingham^ and Shive have 
pointed out. Shive tested thirty-six different sets of salt proportions 
and three different total osmotic values, so that he dealt with 108 dif- 
ferent solutions, all made with the same three salts. The preliminary 
study here reported involved, however, only one total concentration, 
about the same as the one termed optimal by Shive (osmotic value 
about 1.75 atmospheres), and only twelve sets of salt proportions were 
here employed. These were selected to correspond to certain ones of 
Shive's thirty-six, rather evenly distributed over his triangular dia- 
gram. 
In preparing the solutions, it was assumed as approximately 
correct that the osmotic effect of dissociation occurring in these solu- 
tions is the same for Ca(H2P04)2 as it was taken to be for Ca(N03)2 in 
Shive's work. Likewise, it was assumed that the osmotic effect of the 
dissociation of KNO3 in these solutions is approximately like that cal- 
culated for KH2PO4 by Shive. Finally, it was assumed that the 
osmotic effect of the dissociation of MgS04 is the same in our solutions 
as it was taken to be in Shive's calculations for the same salt. Fol- 
lowing these assumptions, the partial volume-molecular concentra- 
tions of our salts are the same as those for the corresponding sets of 
salt proportions shown in Shive's table for the total concentration 
value of 1.75 atmospheres, but of course we employ Ca(H2P04)2 
instead of Shive's Ca(N03)2, and KNO3 instead of his KH2PO4. Using 
Shive's designations for the different solutions (referring to his trian- 
gular diagram), the partial volume-molecular concentrations of each 
of our three salts, in each of our 12 solutions, are given in table i. 
2 Tottingham, W. E. A quantitative chemical and physiological study of 
nutrient solutions for plant cultures. Physiol Res. i : 133-245. 1914. 
