[STAREEY & BARNES] DEFICI KNT HUMIDITY OF THE ATMOSPHERE 207 



Iiumiditv indoors with the windows closed and open respectively. 



During the months of October and November the marked deviation 

 of the records is not so noticeable as in January, Februarj^ and Marcii. 

 Tliis is owing to the fact that during October and November the fur- 

 naces are not as yet in full swing, and consequently a great deal of 

 unaltered outside air gains admission to the house. 



In these tables the relative humidity has always been calculated by 

 means of the wet and dry bulb hygrometer, which we now proceed to 

 show yields results very much too high for low relative humidities. 



WET AND DRY BULB HYGROMETER. 



The instrument most generally in use for the determination of rela- 

 tive humidity is the wet and dry bulb hygrometer, or psychrometer. 

 This type has the advantage of giving continuous readings, and its use 

 seems to have been tirst proposed by Sir John Leslie. 



It is of great importance to have an instrument for meteorological 

 work which gives at any time, without manipulation a reading of the 

 humidity. There are two forms in which this instrument is met with; 

 the usual one consists of two stationary thermometers, one of which gives 

 the air temperature and the other the temperature of a piece of absor- 

 bent material from which water is continually evaporating. The second 

 form is the instrument provided with a handle so as to rotate the two 

 thermometers rapidly and cause the maximum evaporation from the 

 moistened wick over the wet bulb. 



So many complex circumstances have effect on the indications of 

 the instrument that it seems impossible to deduce any satisfactory the- 

 oretical formula for it. The well-kno^vm formulas of Apjohn and of 

 August are not reliable over an extended range. 



The elaborate tables compiled by Glaisher, which are universally 

 used, were constructed from a large number of simultaneous readings 

 with the wet and dry bulb and the Daniell's hygrometer. It is stated 

 that these were made at the Greenwich Observatory and in India and at 

 Toronto. 



During the progress of our work a grave doubt was raised in our 

 minds as to the reliability of the instrument for air of very low moisture 

 content. Thus the comparison which we give of the humidity in an average 

 house in winter with the direct determination by the absorption method, 

 was far from satisfactory. This was shown also on two or three occa- 

 sions in the Macdonald Physics Building by comparing the dew point 

 obtained on the Regnault hygrometer with the dew point deduced from 

 a wet and dry bulb hygrometer. 



