194 Journal of Agricultural Research voi. xxvi, no 4 
gave with tap water from 7.8 to 10 per cent, with an average of 8.5 per 
cent of arsenic oxid. Thus, samples of calcium arsenate which appeared 
practically identical, according to the two tests usually applied in the 
chemical control of calcium arsenate, actually differed markedly. 
The explanation of this difference must be sought in the relative 
quantity of some other constituent, or in the physical nature of the 
calcium arsenate, methods for investigating which are being developed. 
The most plausible chemical explanation is that insufficient free lime is 
present. Since free lime keeps the water-soluble arsenic content low, it 
is possible that a certain minimum requirement to react with the con- 
stitiltents of the dew is necessary. A method for determining free lime 
in calcium arsenate is being studied at the present time, with a view to 
testing this point. 
SUMMARY 
Dew on Upland cotton plants contains large quantities of salts which 
in laboratory tests so act upon calcium arsenate as to increase greatly 
the water-soluble arsenic content. That this is the cause of the erratic 
injury noted remains to be proved. What is true of the cotton plant 
may be true of other plants. Excretions from leaves, therefore, must be 
considered in explaining arsenical injury to plants. This factor may be 
of value in estimating the suitability of an arsenical for a particular 
purpose. 
ADDITIONAL COPIES 
OF THIS PUBLICATION MAY BE PROCURED PROM 
THE SUPERINTENDENT OF DOCUMENTS 
GOVERNMENT PRINTING OFFICE 
WASHINGTON, D. C. 
AT 
10 CENTS PER COPY 
Subscription Price, $4. 00 Per Year 
PURCHASER AGREES NOT TO RESELL OR DISTRIBUTE THIS 
COPY FOR PROFIT.—PUB. RES. 57, APPROVED MAY II, 192a 
V 
