16,5 Espino: Salt Requirements of Young Rice Plants 481 
series as far as these were obtained. This information will aid 
in making comparisons with similar series that have been carried 
out by other students or that may be carried out in the future, 
and will also furnish some fragmentary evidence regarding the 
sort of aerial conditions that prevail in a greenhouse at Balti- 
more for the seasons considered. They are set forth in Table 4. 
PLANT RECORDS 
The numerical plant data derived from these experiments and 
to be considered below include the following values, though these 
values were not all obtained for all cultures or for all series : 
(1) Dry yield of tops (six plants). 
(2) Dry yield of roots (six plants). 
(3) Green weight of tops (six plants) and percentage of 
water content. 
(4) Total absorption (six plants). 
(5) Water requirement per unit of dry weight. 
(6) Approximate length of tops (mean for six plants) . 
(7) Approximate length of roots (mean for six plants). 
The results will be presented below, the three types of solution 
being considered in order. 
THREB-SALT TYPE I (MONOPOTASSIUM PHOSPHATE, CALCIUM NITRATE, MAG- 
NESIUM SULPHATE) 
Numerical data were not generally obtained for the rice plants 
grown in the 3-salt solutions of type I, since it was evident 
throughout these series that none of the cultures were producing 
what might be called good growth; all exhibited much poorer 
growth than was obtained from the better solutions of the two 
types with ammonium. As to the comparative physiological 
values of the fifteen different sets of salt proportions tested, 
solutions RlSl, R2S1, R3S1, R4S1, and R5S1 always appeared 
to be nearly alike and much better than any of the others, while 
solutions R1S3, R1S4, R1S5, and R2S4 were nearly alike and 
always belonged to the poor group. These statements refer to 
all the different total concentrations tested. The comparative 
values of the fifteen different sets of salt proportions may be 
represented by the length values for tops, which were obtained 
for series 4. The relative quantities standing for these top 
values are shown in the diagram of fig. 1, for which the largest 
value (actually 20 centimeters) is considered as unity. It is 
seen at once that the region nearest the left margin of this 
diagram represents the best-balanced sets of salt proportions, 
while the region of the right apex represents the poorest sets. 
