314 ARTHUR I. KENDALL 



general upon material likely to be exposed to its action), and the 

 possibility of leaving poisonous residues. Of these factors, the cost 

 and supply are factors quite without the pale of any experimental 

 data, aside from the question of cost in so far as it is affected by 

 questions of relative efficiency, and need not concern us here. 



The question of killing power, chemical changes, and action upon 

 substances exposed to the action of the fumigant are points of the 

 greatest importance. 



It should be stated that there are two factors involved in the 

 lethal action of fumigants upon mosquitoes: the first, a stupefying 

 effect, which is, or was, overlooked for a time, and the actual death 

 of the insect. Almost invariably stupefying precedes the death of 

 the mosquito, although the latter may follow the former so quickly 

 as to appear almost as a simultaneous phenomenon; hence it is 

 necessary so to place insects upon which one decides to try the action 

 of a fumigant that they may be freely exposed to the culicide, yet 

 be available for examination at any period of the experiment. 



The apparatus described below has been constucted with these 

 points in view; in addition, provision is made for the introduction 

 of samples of the various fabrics, paints, finishings, and, in general, 

 any sort of material likely to be exposed to fumigation. 



The latter is more important than would at first seem possible. 

 For example, it has been found that certain paints, containing lead 

 or similar metals, if poorly applied, or exposed to certain chemicals 

 in the presence of excessive moisture, turn yellowish, or even dark- 

 colored, due to the formation of sulphides. While this does not 

 necessarily spoil the protective action of the pigment, it detracts 

 greatly from the esthetic appearance and renders the fumigat- 

 ing squad liable for damages. This is a particularly important 

 point in tropical countries, where the humidity is always excessive, 

 and where much fumigation is necessary. 



The principle involved in this apparatus is simple. The essential 

 parts are a box having a content of 100 cubic feet, provided with 

 a series of holes through which may be introduced cages, made of 

 wire gauze (20 mesh) six inches long, one and one-half inches in 

 diameter, closed at one end with wire netting of the same mesh as 

 that forming the body of the cage, at the other end by a tapering 



