» 
Oct., 1922] TURNER EFFECTS OF CERTAIN MINERAL SALTS 423 
The solutions were changed once a week, but distilled water was added 
every other day to make up for water which may have passed off through 
transpiration or evaporation. At the beginning of the experiments, it was 
thought advisable to change the solutions more frequently than once a 
week because of a possible exhaustion of nitrates, particularly in the low- 
nitrate solution; however, qualitative tests for nitrates by the diphenylamine 
method made at intervals of two days showed that this precaution was not 
necessary. All experiments were carried on in the greenhouse of the de- 
partment of plant physiology of Cornell University. 
After a series of preliminary experiments with different plants (including 
peas, corn, barley, buckwheat, and flax), barley was selected for the first 
work as it showed in comparison with the others the slightest apparent 
variation under greenhouse conditions; both the cultures as a whole and 
the individuals in a culture made a fairly uniform development. 
The variety of barley used was known as Silver King; it was secured 
from Professor H. H. Love of the department of plant breeding and was a 
pure line which had been selfed for several generations. 
It was assumed that any marked variations in growth would be due to 
the varying amounts of nitrate at the disposal of the plant, since the other 
components were approximately equal in the three solutions and since the 
plants grew in the greenhouse under the same external conditions of light, 
temperature, and relative humidity. 
The general method employed for obtaining quantitative data was to 
harvest a series from each solution at the same stated time, to wash off any 
salt crystals that may have adhered, then to separate carefully tops from 
roots and desiccate them at a temperature of 102° C. to constant weight. 
Other features such as length and branching were sometimes taken into 
consideration. In later experiments green weights were also considered. 
Experimental Data 
Experiment I 
The duration of this experiment was 49 days. It was begun August 23 > 
1920. The first series was harvested at the expiration of 14 days, and 
another was taken down every seven days thereafter until all were removed. 
Table 4 gives the comparative results of the six series conducted with 
barley. Since growth has been quite uniform, the culture has been taken 
as the unit for calculation. The ratios of tops to roots per culture have 
been computed, as well as the probable errors of the means of these ratios 
for each series. In some cases the difference between the means of the 
ratios has been given together with the probable error of these differences. 
This will serve to show the significance of results obtained by adding 
nitrates. As an aid in comparing the absolute efificiencies of the three 
solutions, total dry weights of the four cultures are given also. 
