650 
Journal of Agricultural Research 
Vol. XXVIII, No. 7 
early October, 1918, the owners discovered that while the bulk of the consign¬ 
ments was uninfested, many carload lots of the seeds showed heavy infestation 
and were heating badly as a result of bruchid attack. The centers of infestation 
were, unfortunately, so distributed throughout the warehouses that rapid spread 
in infestation was taking place, and the heating caused by the infestations was 
offsetting the normal checks of colder winter weather upon weevil activity and 
spread. Numerous sacks of seeds had already been rendered unfit for food both 
by the bruchids and the Rhizopertha. Since the seeds had a retail value of 
about $5,000,000, action was imperative. As the warehouses, with one excep¬ 
tion, were within the congested business districts of the city only a noninflam¬ 
mable gas could be used. Hydrocyanic-acid gas was chosen by the owners, 
who engaged a professional fumigator. This operator fumigated the ware¬ 
houses, using 23^ pounds of sodium cyanid 98 to 99 per cent pure for each 1,000 
cubic feet of space. The senior writer was present at the time of the fumiga¬ 
tions and took numerous samples, the results of the examinations of which 
are only in part reported upon in this paper. The fumigations were allowed to 
extend over a period of 48 hours, although the warehouses were not as a rule 
tight enough to confine the gas that long. The writers were not permitted to 
experiment with the dosage chosen by the owners. While the cost of using so 
large a dosage greatly increased the cost of fumigation above what was necessary 
in the opinion of the writers, the cost was nothing in the eyes of the owners when 
compared to the value of the commodity. The fumigations were conducted 
when the average mean temperatures ranged from 58° to 75° F. 
In taking samples both before and after fumigations, the writers used the 
usual grain car probe, about 5 feet long, as illustrated (PI. 1, C). This probe 
was thrust from end to end of the sack through the central axis and the samples 
(usually six) placed separately in sacks and numbered according to their relative 
positions in the bulk of the sack sampled. In making the intensive examination 
later only 50 of the seeds of a sample were used, the remainder being held for 
future observation. 
Diagrams are presented hereafter showing the location of the sacks from 
which samples were taken after fumigation, together with data on the degree 
of infestation and penetration of the gas and the effect of the fumigant upon 
the insects. It happens that the data on infestations serve to indicate also 
both the penetration and the effectiveness of the fumigant, for the reason that 
in no sample recorded in the following pages were any insects found alive, and 
no development occurred subsequently from eggs on the seeds or from larvae 
or pupae within the seeds at the time of fumigation, as shown by examination 
made at intervals during two years following the fumigations. No extended 
examinations were made until about one month after the fumigations, when the 
dead larvae and pupae had had time to discolor if killed. All seeds examined for 
larvae and pupae were first softened and then sliced with scalpels to reveal infes¬ 
tation. The references to the degree of larval development, such as one-eighth, 
one-fourth, one-half, or full-grown, are of course approximate statements based 
upon the general appearance of the individual. 
WAREHOUSE NO. 1 
The 30,000 sacks of chick-peas stored in this warehouse were in a room on the 
third (top) floor, measuring 150 by 150 by 20 feet. The warehouse was of brick 
construction, with a tar and pebble roof. Leakage of gas about the roof during 
fumigation was rather great, little or no discomfort being experienced when the 
room was opened for ventilation 48 hours after fumigation was started. The 
fumigator used 80 fifty-gallon oil barrels, each generating a charge containing 
15 pounds of sodium cyanid. The capacity of this warehouse was greatly over¬ 
taxed (PI. 1, A), the sacks being piled from 6 to 17 deep, whereas the usual 
