June, '21] headlee: bean weevil and moisture 265 



were raised. This line then left the experimental chamber and passed 

 over to the sink, where it was connected with a constantly functioning 

 glass suck piimp. The other lines, all of which contained air that 

 must receive further conditioning as to atmospheric moisture, passed 

 into different experimental chambers, all of which were held at a con- 

 stant temperature of 80 Fanr., where the air was led through con- 

 centrated aqueous salt solutions; each stream passing through the par- 

 ticular salt solution which gave to it the desired amount of atmospher- 

 ic moistiire. Each line was then led to its particular group of five 

 containers in which the beetles and their progeny were living. From 

 the containers each air stream was led by a separate pipe to the sink, 

 where it was joined to its own particular glass suck pump. The one 

 in which the air must be conditioned to approximately 100% atmos- 

 pheric moisttue was led through a flask of distilled water and thence 

 to its group of five containers. 



The \\Titer has tried various methods of conditioning the air, such 

 as: (1) constantly raising the atmospheric moisture of the experimen- 

 tal chamber by the introduction of sponges or wet cloths and holding 

 the percentiim of moisture desired by passing the air over calcium 

 chloride as often as it rose above the desired percentum; (2) passing 

 moisture laden air through low temperatures and then passing it into 

 a chamber, the temperature of which bore such relation to the prev- 

 ious low temperature that the percentage of atmospheric moisture was 

 such as was desired. The first method is unsatisfactory, because there 

 was no provision by means of which the air could be changed with suf - 

 fiicient rapidity to maintain the normal equilibrium of its component 

 gases, especially as regards the proportion of carbon dioxide. The 

 second method is ideal, but involves so large a quantity of expensive 

 apparatus as to render it impracticable. 



In view of the, to him, insurmountable objections to the two meth- 

 ods just described his attention was directed to the possibilities of util- 

 izing the different vapor tensions of various aqueous salt solutions. 

 He found no difficulty whatever in obtaining data on the vapor ten- 

 sions of different aqueous salt solutions, but, in view of the fact that 

 none of the vapor tension data were derived from saturated aqueous 

 solutions, and in view of the further fact that without saturation the 

 maintenance of a constant concentration was apparently out of the 

 question, it became necessary to determine the amount of moisture 

 which the different saturated aqueous solutions of various salts would 

 give off under a constant temperature. Inasmuch as the machines 

 with which he was working were well adapted to the maintenance of a 



