4 
under definite conditions, and to determine exactly the amount of pro- 
tection required to prevent the penetration of the gas to the mosquitoes 
in sufficient strength to kill them. 
The pill boxes were small cylindrical cardboard boxes, one-half an 
inch high by one and one-half inches in diameter. The lid of each box 
was replaced with gauze mosquito netting. 
Mosquitoes thus confined often die within twenty-four hours under 
favorable conditions of temperature and moisture. Therefore, if one 
survives in a box containing several insects, the death of the others 
must not be attributed to the effects of the disinfecting gas. All the 
mosquitoes in the box must be dead to draw positive conclusions. 
Another advantage of these boxes was that we were enabled to com- 
pare the germicidal with the insecticidal power of the gases used. 
Cultures of bacteria were exposed upon slips of paper in similar pill 
boxes, which were sterilized for this purpose. These boxes containing 
the cultures were placed alongside the boxes containing the mosquitoes, 
so that the two were subjected to exactly similar conditions as far as 
that was possible. 
The cultures were exposed upon sterile strips of white filter paper. ; 
The spores of the hay bacillus and of anthrax were thoroughly dried in ; 
the incubator at 37° C. Eich garden earth was emulsified with water 
and the strips impregnated with the suspension. These were always dried 
before being exposed. Fresh young cultures of a staphylococcus, pyo- ; 
cyaneus, and bacillus icteroides were used to inoculate the slips and were i 
exposed moist and freshly prepared. These were selected as represent- ! 
ative of the non- spore-bearing group of microorganisms. 
In some of the experiments the bacteria and the mosquitoes were i 
not expsed under precisely similar conditions. In such cases, noted 
hereafter, the slips of paper impregnated with the cultures were 
placed in Petri dishes, in envelopes, and in small boxes. The Petri i 
dishes were opened so that the gas could gain free access to the test i; 
slips. The envelopes were not sealed. The boxes were small cardboard i 
cylinders about 2 inches long by 1 inch in diameter, with 5 small 
holes punched in both top and bottom to permit the entrance of the gas. 
FORMALDEHYD GAS. 
It is a well-known fact that formaldehyd gas readily enters into com- 
bination with the protoplasm of the lower forms of vegetable life, I 
which makes it a very efficient germicide. It is, however, not toxic i 
to the higher forms of animal life. It is very irritating in its effect , 
upon the mucous membranes of rats, mice, guinea pigs, rabbits, and i 
mammalian animals generally, but not necessarily fatal, even after pro- ! 
longed exposures. Many insects, such as roaches and the like, may be 
exposed to strong concentrations of the gas a long time without apparent - 
effect. - I 
