

HYGROMETRIC RESEARCHES. 545 



In most experiments the aspirators were iron cylinders with conical tops, through 

 the bung of which a glass connection-tube and a thermometer were inserted. The 

 volume of water run out was measured at first by noting the amount in the aspirator, 

 which had been determined previously when filled to a mark on the glass tube, and 

 either completely emptying the aspirator or measuring the water left ; or, in later 

 experiments, by running the water from the aspirator into another similarly-shaped 

 vessel until it reached the level of a mark made on a glass tube projecting from it. In 

 later experiments, the iron cylinder was wrapped in flannel to keep the temperature 

 more equable. 



The volume of water run from the aspirator does not precisely represent the volume 



of air drawn through the tubes. The air entering the tubes contains some water vapour, 



which is arrested in the tubes, and that entering the aspirator is dry air. The atmosphere 



inside the aspirator, however, is saturated with vapour from the water which was in the 



aspirator. The ordinary atmospheric pressure, p, is ultimately that inside the aspirator. 



In the free atmosphere the pressure of the dry air is p — e, where e is the vapour pressure 



calculated from the tables, from the dry and wet bulb temperatures t, t' ; whereas that 



of the dry air inside the aspirator is p — E, where E is the pressure of saturated vapour 



at the aspirator temperature T. Hence, to obtain the volume v of dry air drawn 



through the tube in the current atmospheric conditions, the volume of water V run from 



aspirator must be corrected as follows : — 



, r w-E 273 + * 



v=\. x . — . 



p-e 273 + T 



The gain in mass of the drying tubes is divided by this corrected volume, and the 

 mass of water vapour is expressed in grammes per cubic metre, or milligrammes per 

 litre of space dried. 



At the Ben Nevis Observatory the weight of the tubes used varied from about 3 to 

 25 grammes. The increase in the weight of the first tube during an experiment varied 

 according to the humidity of the air and the volume of air dried from # 2691 grammes 

 to 0*0064 grammes. Except in the earlier experiments, when 23 6 litres of water were 

 run from the aspirator, the volume of air dried varied from 26 '3 to 29 "5 litres. The 

 water was run from the aspirator at the rate of from 1*7 to 4*2 litres per minute ; the 

 average rate being slightly over 3 litres per minute. 



In the case of Fort- William the weight of the tubes used was from 9 to 25 grammes. 

 The first tube gain varied from 0"2715 to 0*1080 gramme during an experiment, and, 

 in the case of the experiments in January 1894, as low as 0*0597 gramme. The 

 number of litres of air dried in an experiment varied from 21*9 to 24*5 ; in experi- 

 ments 8 to 97 and after experiment 117, from 26'6 to 28*5. The rate at which the 

 water ran from the aspirator was from 2 '2 to 6*0 litres per minute; the average 

 being about 4 litres per minute. 



At Montpellier the apparatus was not quite so suitable for the work as that used at 

 the Ben Nevis Observatories, and the aspirators were smaller. The tubes varied from 



TRANS. ROY. SOC. EDIN. — VOL. XLIII. 3 Z 



