NO. 17.] HYGROMETERS. 19 



ing table shows the error 4r in the relative humidity, corresponding to an 

 error de in the vapour-tension at different temperatures, t, below zero. 



The vapour-tension may accordingly be assumed to be determined with 

 an accuracy of one tenth of a millimetre. 



The Relative Humidity has been computed 



(1) from the psychrometrical observations in conjunction with the vapour- 

 tension. 



(2) from the readings of the hair-hygrometer. From October, 1893, to 

 March 17, 1894, only Hygrometer I. was observed. From the latter date to 

 the 20th August, 1896, both hygrometers were as a rule observed, and the 

 mean of their corrected readings provided their corrections were considered 

 trustworthy adopted for the relative humidity. The values of the relative 

 humidity computed from the psychrometrical observations have been checked 

 throughout by the simultaneous observations of one or two hygrometers. 



The mean error of a single determination of the relative humidity entered 

 in the Tables, resulting from the different comparisons of simultaneous ob- 

 servations of two instruments, may be taken as hardly greater than 5 /o. 



