Solution of Atmospheric Nitrogen and Oxygen by Water. 323 



to obtain a complete set of observations under constant 

 conditions. The length o£ exposure varied from one hour 

 to 144 hours, and even aft^r six days' exposure, complete 

 saturation was not reached. In addition, it was found that 

 the variations of atmospheric conditions during these long- 

 periods of exposure exercised a very considerable influence 

 on the rate of solution, causing irregularities in the air- 

 content which could not be corrected for. 



In order to overcome this difficulty, it was thought ad- 

 visable to provide each tube with an enclosed atmosphere 

 that could be kept at least at a uniform temperature and 

 uniform state of humidity during the course of the expe- 

 riments. For this purpose a modification of the constant- 

 pressure apparatus, that was used by one of the authors to 

 measure the absorption of oxygen by polluted waters *, was 

 employed. 



The rate of absorption in these tubes was much the same as 

 before, but the conditions affecting the atmosphere above the 

 water-surfaces were under control. Thus some of the tubes 

 were wholly immersed in the thermostat to the level of the 

 corks; while others were only immersed to the same level as 

 the water- surface inside them. In the one case, therefore, 

 the air in the tube was kept at the same temperature as the 

 water, and evaporation from the surface was reduced to a 

 minimum; while, in the other case, the air was exposed to 

 all the fluctuations of temperature of the laboratory, and 

 considerable evaporation and condensation took place. There 

 was a marked difference in the rate of absorption under these 

 different conditions, it being much greater in the tubes liable 

 to fluctuation of temperature. 



In order to shorten the time during which it was necessary 

 to keep the tubes under observation, they were only half 

 filled and laid on their sides, so as to expose a greater area 

 of water-surface — equal to the longitudinal section of the 

 tube — and a considerably reduced thickness of water to be 

 saturated. 



The area of surface was in this way increased from about 

 12 square centimetres to about 120 square centimetres, and 

 the thickness decreased to a maximum of about 2 centimetres, 

 and, further, the actual volume of water used was only half 

 the previous amount. Under these conditions the aeration 

 took place in a much shorter time, and it was found possible 

 to saturate the water in about twenty-four hours. 



* Supplementary Volume VI. Fifth Report, Royal Commission on 

 Sewage Disposal, pp. 438-441. 



