Manganese Sulphate upon the Growth of Barley . 575 
Since it was anticipated that the manufacture of sufficient glass distilled 
water would prove a difficulty when the investigations were extended, tests 
were made to find a satisfactory substitute. Pure glass distilled water was 
shaken up with pieces of metallic silver and aluminium respectively, and 
allowed to stand on the metals for at least an hour. Food solutions were 
then made up with the water so prepared, and test plants were grown 
in them, with parallel controls in pure-water culture solutions. It was 
found that there was little or no difference in the growth of the two sets of 
plants, the toxic effect of the metals, if any, being completely masked 
by the nutrient salts (PI. XLVII, Fig. 3). Acting on this discovery a Brown’s 
‘ chemist ’ still with a condensing arrangement of pure silver was obtained, in 
which the condensed water came into contact with nothing but the pure silver 
until it passed into a glass delivery tube. The water was finally filtered 
through charcoal, which has been found to remove the toxic effects of 
metals. All the later water-culture work was carried out with water thus 
prepared. At the close of the experimental work in 1907, two points stood 
out very clearly :— 
(1) Owing to the individuality of the plants, investigations dealing only 
with small numbers cannot be depended upon to give really reliable informa¬ 
tion as to the effect of various strengths of a toxic salt upon plant growth. 
To this end repeated experiments dealing with groups of 5, 10, or 20 plants 
for each concentration are necessary to give accurate information. In these 
experiments the question of individuality in barley was eliminated to the 
greatest practicable degree, as the seed used was a pure strain (obtained 
from Mr. E. S. Beaven, Warminster) which had been raised from a single 
seed. Further, the seeds were graded before being sown, so that they only 
varied within 0-02 gram from one another. 
(2) The presence of nutrient salts in a culture solution exercises 
a definite masking effect upon the action of the toxic salt, enabling a plant 
to make satisfactory growth in contact with a much more concentrated 
solution of the poison than it can otherwise endure, i. e. whereas a concentra¬ 
tion of 1:1,000,000 (and possibly less) copper sulphate solution alone 
entirely inhibits growth in barley, yet in the presence of nutrient salts 
a concentration of 1 : 250,000, at least four times as great, does not prevent 
growth, though it decidedly checks it. 
The preliminary year’s work had shown that the best results were 
obtained with plants grown between February and July, as during the early 
winter months practically no growth is made, and seedlings raised after the 
early part of June are very apt to fail. 
Early in 1908 extended experiments were started with twenty parallel 
series, each consisting of eight concentrations of C.S. ranging from 1 : 5o ; ooo 
to 1 : 10,000,006, with a control in distilled water, nutrient salts being added 
in each case. The seeds were graded between 0-04 and 0-06 grams and the 
