August, '14] HYSLOP: SOIL FUMIGATION 307 



Mr. L. D. Larsen of the Hawaiian Sugar Planters' Experiment 

 Station has found that protozoa are very abundant in Hawaiian soils 

 and that soil fumigation with carbon bisulphid acted as a decided 

 stimulus to crop production. 



In the experiments so far carried on, the treatment has been so costly 

 as to be prohibitive in actual farm practice. However, this is not to 

 be regarded as a serious factor, for cyanide can undoubtedlj' be pro- 

 duced in a much cheaper form than that used in the trials, and smaller 

 dosage with cheap filler will very probably make the treatment econom- 

 ically available. 



The laboratory work was all carried on at the United States Field 

 Laboratory located at Hagerstown, Md., and in this work Mr. C. M. 

 Packard assisted materially. The field work was carried on at Wolf- 

 ville, Md., near Hagerstown, and at Bridgeport, N. Y., near Syracuse. 



The laboratory work was started late in May by a series of pot 

 experiments to determine the effect of the fumigant upon the crop 

 to be treated. Corn was selected as the crop because it is probably 

 more generally and seriously damaged by subterranean insects than 

 any of the cereal and forage crops. The cyanide used in these experi- 

 ments was not the c. p. sodium cyanide but a commercial mixture 

 containing from 39 per cent to 40 per cent of cyanogen, the equiva- 

 lent of 98 per cent to 99 per cent potassium cyanide. The mixture 

 was as follows: 



Sodium cyanide 74 per cent to 76 per cent 



Alkaline chlorides 16 per cent to 24 per cent 



Inert substances 2 per cent to 8 per cent 



The pots used in all the laboratory work were unglazed earthenware 

 flower pots six inches in diameter and six inches in depth. This size 

 pot was selected because the general depth of ploughing is about six 

 inches and by simply determining what fractional part of an acre was 

 contained in the exposed surface of earth in a pot, the dosage could be 

 roughly determined in terms of pounds per acre. The earth used in 

 these experiments was ordinary corn field earth from a neighboring 

 farm and the entire batch was thoroughly mixed to eliminate as far as 

 possible all variables but those intentionally introduced. Three 

 series were used, each series containing twelve pots divided into four 

 groups of three pots each. In the first series (Division A) all the pots 

 were seeded and treated on the same day, May 29. In the first group 

 of three pots in this series no cyanide was used and the group served 

 as a check. The second group contained one scruple of cyanide to the 

 pot, roughly 300 pounds to the acre; the third, one dram, 900 pounds 

 to the acre; and the fourth, two drams, 1700 pounds to the acre. In 

 the second series of twelve pots (Division B) the cyanide dosage and 



