202 Smith and Butler. — Relation of 
slight changes in elemental composition are less important than changes 
in concentration. 
Experiment 3. In this experiment Early pedigree Dent field Corn was 
grown in sand culture. The corn was carefully selected for size and only 
seeds weighing between 250 mg. and 350 mg. were used. The mean 
potassium content of a seed was determined by the analysis of three 
samples, each composed of several hundred seeds, and found to be 0-85 mg. 
The corn seed was therefore much richer in potassium than the wheat. 
The sand employed for the cultures was a very fine and pure acid- 
washed silicon dioxide with a mean water saturation-point of 22-5 percent. 
The pots used were made of glazed earthenware, and contained 1,400 grm. 
of the sand. Before planting the seed, water was added to the sand in 
sufficient quantity to give 60 per cent, of saturation. The water was added 
in the form of the appropriate nutritive solution and afterwards maintained 
by additions of distilled water as required. The full nutritive solution 
employed was solution F, the composition of which is given in Table XII. 
For the plants growing in the absence of potassium the same solution was 
employed, but without the addition of potassium. The full nutritive solu- 
tion F contained 3-6635 grm. salts per litre, the less potassium solution F 
only 2-4815 grm. ; but since wheat (see Table XX) yields identical results 
when grown in solutions free from potassium containing 3*6635 grm. salts 
and 2-4815 grm. salts per litre respectively, we assumed that the corn would 
behave similarly. 
Eight pots were used for the full nutritive solution and eight for the 
nutritive solution less potassium, one plant being grown in each. 
The experiment was begun on October 19, the seed, previously soaked 
sixteen hours in distilled water, being planted 1 cm. deep. Four days 
after planting the seedlings came through the sand, and after eight days 
measurements of rate of growth were taken periodically until the close of 
the experiment, the data obtained being given in Table XIII. A study of 
the table shows that potassium starvation gradually set in after the twelfth 
day from planting and that growth ceased after the thirty-fourth day in 
the absence of this element, while in the case of the plants growing in the 
full nutritive solution growth continued vigorously and uninterruptedly. 
The behaviour of the corn, therefore, whether growing in the presence or 
absence of potassium, resembles very closely that of wheat, the only 
difference worthy of note being that the symptoms of potassium starvation, 
which are in all points otherwise similar, appear more slowly, and extinc- 
tion of growth is longer delayed. 
After fifty-nine days the experiment was closed, and the plants from 
each pot weighed separately. The data obtained is given in Table XIV. 
It should be noted that the weights of the roots are approximate only, since 
it was neither possible to recover with certainty the entire root systems 
