of Determining Moisture and Carbonic Acid in Air. 315 



Although Regnault proved by experiment that the absorp- 

 tion of moisture was complete with the chemical method, yet 

 he was prevented from bringing it into general use in hygro- 

 metry by two difficulties. These were the inaccuracy of the 

 weighings, and the slow rate at which the air could be 

 aspirated through the absorption-tubes *. All other observers 

 who have used the chemical method have had to encounter 

 these same difficulties. Even the mercury-joints and glass- 

 stoppered absorption-tubes used by Shaw have not removed 

 the errors arising from these causes f. Each determination 

 made by these observers required about one hour, and generally 

 longer. By what simple means these serious errors can be 

 avoided has been shown above. 



I must here express my hearty thanks to the Radcliffe 

 Observer for giving me every facility to carry out these 

 experiments. To his assistants Messrs. Wickham, Robinson, 

 and Bellamy, my thanks are also due. 



Two determinations were made each time and as nearly 

 simultaneously as possible. Two aspirators similar to the one 

 described above were used. Four pairs of absorption- tubes 

 were employed each time : — pair 1 for determination A ; pair 

 2 for determination B ; pair 3 as a test pair to show if any 

 alteration in weight was caused by carrying the tubes from 

 the Physiological Laboratory (where the tubes were weighed) 

 to the Radcliffe Observatory and back ; whilst pair 4 was the 

 counterpoise. 



The tubes were carried in a small box with partitions of 

 copper wire to prevent them from knocking against each 

 other. The entrance tube, by which the air to be examined 

 passed into the pair 1 (or 2), was fixed through a small per- 

 foration in a rubber partition covering a hole in the box, so 

 that there was no possibility of air being taken from the inside 

 of the box. The box and its tubes were placed about 1 foot 

 below the wet and dry bulbs. That the air in this position 

 might be fairly compared with that in the shed containing 

 the psychrometer is shown by experiment 6, in which pair 2 

 was placed in the shed close to the bulbs, and pair 1 in the 

 usual place below. 



In making a determination the following was the order of 



* Ann. de Chim. et de Phys. xv. pp. 153 and 164. " La mtfthode 

 chimique esttrop embarassant, et elle exige une manipulation trop longue 

 pour qu'on puisse l'employer souvent dans les observations mlteorolo- 

 giques." 



T Shaw on Hygrometric Methods, Phil. Trans. 1888, A, pp. 83 and 

 84. The variations in weight range from — -0009 to +-0023 gram, or still 

 higher. 



