1888.] S. A. Hill — Psyclirometer and Condensing Hygrometer. 375 



altered from 0*429 to O'^S, though, for lower degrees of humidity, this 

 alteration made the air appear too dry. The same was found to be the 

 case with a series of observations made in the Pyrenees by M. Izaru at 

 a mean atmospheric pressure of 700 millimetres, the standard observa- 

 tions for comparison being those of a condensing hygrometer. 



In Regnault's class-room, with the doors and windows closed and 

 no sensible circulation of air, it was found that, by using the same for- 

 mula, the humidity deduced was much too high, or that the factor A 

 must for still air be considerably increased. 



Since 1845, many observers have attempted to verify or improve 

 upon Regnault's results — amongst them Mr. H. F. Blanford, F. R. S., 

 who, in 1876, published, in the Journal of this Society (Part II, VII), 

 an account of observations made in various parts of India with a view 

 to determine which formula of reduction was best suited to the condi- 

 tions obtaining in this country. Mr. Blanford's general conclusion was 

 that the dew point computed by August's formula, with Regnault's 

 constants, from observations of the psychrometer made under an open 

 shed, comes very near to that observed with a condensing hygrometer, 

 even when the dew point is more than 40° below the temperature of the 

 air. Both AjDJohn's formula and Glaisher's factors give too high a 

 result. This conclusion is verified by the observations now published, 

 of which the first series has already appeared in the Indian Meteorolo- 

 gical Memoirs, Vol. I. In Table II. the absolute and relative humidities 

 as given by the dew point instrument and deduced from the psychrome- 

 tric observations by August's formula are compared, and it will be seen 

 that, whereas in still air, whether in the interior of a room or in a veran- 

 dah without a thorough draught, the deduced humidities are invariably 

 too high, the formula gives results agreeing closely with the truth when- 

 ever there is fair ventilation. 



In very strong winds the formula gives results slightly too low, 

 for in such conditions the assumptions underlying the theory are more 



or less completely verified and the theoretical value of - should be used 



d 



instead of the modified value adopted by Regnault. 



