442 H. A. Hazcn — Condensing Hygrometer 



nault and Crova ; 2d, the mean of each set of three, and the 

 mean of all three sets. The value of A, in the psychrometric 

 formula (soon to be described) x=f—A(t-l')%h, has been as- 

 sumed as "00080 for the observations with the Eegnault appa- 

 ratus and the corresponding value with the Crova instrument 

 has been computed. 



Comparison of Regnault and Crova, 



~&o. of Dry Wet Dew-point observed. Value of A. 



Obs. observed computed Regn. Crova. Eegn. Crova. 



1 72°'5 58°0 44-1 46 ! 9 '00080 "00067 



1 67*5 56-8 46-9 49'6 '000S0 "00067 



1 77'9 65-4 56-7 59*2 '00080 '00059 



6 70'7 56'7 42'7 44'2 '00080 . '00073 



4 67-0 57-0 48-0 49'2 -00(180 '00073 



6 78'2 65'5 57'0 59M '00080 '00063 



16 . 72'7 60'0 49'4 51'2 '00080 '00069 



These results are extraordinary, and are probably due in large 

 measure to an improper condition of instruments rather than to 

 an actual difference in their normal indications. Certainly differ- 

 ences greater than '5° F. ought not to occur in a properly observed 

 Regnault apparatus, while these comparisons show an extreme 

 range in*the differences between the two instruments of 2° -9. 

 I have dwelt at length on these results as they show clearly 

 some of the difficulties to be met with in comparisons of this 

 kind. I add a remarkable comment by a prominent writer, on 

 the original observations. "In the observations given by 

 Crova we find, as was to be expected, that Eegnault's apparatus 

 gives dew-points too low by as much as l 0- 6 C." 



The Psychrometer. 



As is well known, this consists of two thermometers ; one of 

 them, the wet, hns a muslin over the bulb, which is kept con- 

 tinually moistened by a wick attached above the bulb and 

 dipping in a cup of water, and measures the temperature of 

 evaporation ; the other is dry and indicates the air temperature. 

 The amount of moisture makes but little difference ; Eegnault 

 found that if water dripped slowly from the bulb, he still 

 obtained accordant results ; this, however, was in still air. 

 Since the wet bulb may at times stand 20° below the dry, it 

 might be supposed that the stem of the thermometer immedi- 

 ately above the bulb would tend to raise the temperature. 

 This effect is very- slight, amounting in the case of a spherical 

 bulb to -3° when the air is perfectly still. The muslin should 

 be very thin and may be made to fit a cylindrical bulb very 

 tightly, if a piece the right size is first immersed in water and 

 then put on the bulb, a string being tied upon it just above 



