EXPERIMENTS WITH ALCOHOL 135 



large tanks and, in the absence of any reagent, will show no 

 signs of discomfort from lack of air for a period of at least three 

 hours. 



Regarding the mode of administration of the poisons used it 

 was found early in the work to be undesirable to depend entirely 

 upon the evaporation of the reagent from cotton in the chamber 

 at the bottom of the tank. This process took altogether too 

 long a time to saturate the air of the tank with the vapor. Prac- 

 tically from the beginning we have used a combination of this 

 method plus a preliminary saturating of the air with the vapor 

 of the substance used by means of an atomizer. The routine 

 procedure is this: there is placed in the reagent chamber at the 

 bottom of the tank a piece of absorbent cotton soaked with 

 the reagent to be used, ethyl or methyl alcohol or ether, as the 

 case may be. Then the operator quickly but thoroughly fills the 

 whole of the tank proper by means of an atomizer with a satur- 

 ated \'apor of the same substance. The birds to be treated are 

 then introduced quickly, allowing as little as possible of the 

 vapor to escape in the process. When the birds have been in- 

 troduced the cover of the tank is tightly closed and left in that 

 condition for one hour. It is to be understood throughout this 

 paper that every bird designated as a 'treated bird' has spent 

 one hour every day in one of these tanks subjected to the fumes 

 of the reagent specified in the particular case. At the be- 

 ginning of the experiments it was thought desirable to accustom 

 the birds gradually to the vapor treatment, and consequently in 

 some cases the treatments for the first week were only one-half 

 hour in duration. It was soon found, however, that a sound 

 healthy bird could stand the treatment for an hour, even from 

 the first. Consequently in all later work it has been the rule 

 to make the treatment extend over one hour each day from the 

 very beginning. 



It is impossible, as Stockard has pointed out, to make any 

 very precise statements regarding dosage when a reagent is 

 used by the inhalation method. The time of subjection to the 

 fumes is perhaps the most significant measure that one can get. 

 It may not be without interest, however, to note the amounts 



