EXPERIMENT STATION BULLETINS. 447 



and 403.7 

 401.1 



2.6 c.c. CO2 given off during test in the CS2— treated gas-container. 



Right-hand System (check). 

 420.0 c.c. =Vol. of right-hand system, exclusive of gas-burette (at start.) 

 20.5 =Vol. of right-hand system gas burette. 



440.5 



440.5(740-16.655)273 



760(273-1-19.3) 



420.0 c.c. 

 23.6 



=394.7 c.c. Vol. at start (reduced). 



443.5 c.c. =Total vol. at end of test (before CO2 had been absorbed.) 

 443.5(734-17.826)273 



=389.3 c.c. above vol. (reduced). 



760(273 4-20.4) 



420.0 c.c. 

 20.5 



440.5 C.C. =Total vol. at end after CO2 had been absorbed. 

 440.5(734-17.826)273 



760(273 -1-20.4) 



Then 394.7 

 386.6- 



= 386.6 C.c. above volume (reduced). 



8 . 1 c.c. O2 absorbed during test by the check. 



and 389 . 3 

 386.6 



2.7 c.c. CO2 given off during test by the check. 

 8.1 —5.5 =2.6 c.c. more O2 abosrbed In case of the check (i. e. not treated with CS2 -vapor). 



Note was kept of twenty-one experiments carried out in adjusting and 

 in learning to manipulate and use tlie apparatus to obtain trustworthy 

 results. It was found that the greatest care must be taken to be sure 

 that the two systems of the apparatus were at the same temperature. A 

 door, opened for a short time to admit air of a different temperature 

 from one side of the room, could cause a variation in the volume of the 

 gas in the two systems which would entirely vitiate results if readings 

 were made at that time. It was discovered also that a variation in the 

 influence of the insecticide-vapor occurred, depending on whether the 

 extract were used immediately after it had been prepared, or not until 

 a few hours later — this seemed to be especially true when gasoline was 

 used. This fact served to emphasize the necessity of taking, from the 

 prepared extract, the sample to be treated with the insecticide and the 

 check sample at the same time, and of carrying out the tests with the 

 two during the same period — although the reason for the variation men- 

 tioned above was not appreciated until later in the investigation. If the 

 insecticide vapor in the air used in an experiment was near saturation, 

 then it was necessary not to allow the room temperature to fall below 

 that at which the experiment was started. Otherwise some of the in- 

 secticide-vapor would be condensed and the vapor tension of the gas- 

 raixture changed. Correction for change in vapor-tension of water vapor 

 due to a change in temperature could be readily made, but no method 

 was at hand for making a similar correction in case of gasoline or carbon- 

 disulpbide vapor when the apparatus illustrated by Fig. I was being used. 

 Furthermore, it was found that high concentrations of the insecticide 



