446 H. A. Hazen — Condensing Hygrometer 



the evaporation from ice is different from that from water, and 

 hence the value of A must change suddenly at 32°. If there 

 be such a change allowed for, the distinction should be between 

 water and ice and not above or below 32°, for frequently the 

 wet bulb may be lowered to below 26° without the formation 

 of ice. This is an important problem, and has been practically 

 solved in various ways. In March, 1884, the writer, with the 

 assistance of Mr. Fassig, performed some experiments in a room 

 with the temperature just below freezing. It was found possi- 

 ble to have two wet bulbs in operation at the same time as 

 sling thermometers, the one being covered with water and the 

 other with ice. The difference between the two was practically 

 inappreciable The results in detail may be found in the Amer- 

 ican Meteorological Journal for June, 1884. Later comparisons 

 between a condensing hygrometer, water- and ice-covered 

 bulbs confirmed the above conclusion. The results of these 

 may be found in the journal just quoted, for February, 1885. 

 Even before that time somewhat the same conclusion had been 

 arrived at in Europe, though the experiments here were entirely 

 independent of those. 



Value of A. 



Having given the dew-point, t, l! and h, we can obtain the 

 constant of the formula. As i' enters the formula twice, it has 

 by far the greatest influence and it is partly because it has 

 been found impossible to obtain an invariable reading of t' that 

 so many differing values of A, have been obtained; a few 

 examples will suffice. 



In still air Eegnault found A— '00128; in the open air 

 •00074: Doyere with the sling psychrometer in 1855 from a 

 verv few observations in hot weather found '000687 ; Swory- 

 kin^ with fair ventilation '000725; Blanford '000827; Angot 

 from over 3,000 observations '000851 ; Chistoni '000851. The 

 latter three results are probably without any- uniform ventilation. 

 The first comparisons by the writer were made in Feb. 1883, 

 with a bellows for a ventilator of the dry and wet bulbs ; these 

 results were published in Science, June 8, 1883. A longer series 

 of experiments were instituted in October, 1884. From 700 

 sets of observations, both indoors and out, with high and low 

 temperatures, and with large and small differences between t 

 and t', a value of A was obtained at '00068 and upon this value 

 was computed the table for relative humidity to be found at 

 the end of this paper. The value of A that has been adopted 

 may seem rather small, nevertheless, it agrees exactly with that 

 adopted by Eegnault for temperatures below freezing, so that 

 no change is made in the table below that point. This value 

 depends upon the separate determination of three quantities, 



