I ir. Mini I y k-i.coHiu.Ks 



251 



This nn\cl iiii|)n>\fiiu-iil (Ii|hii(U. lor its tipi'iMlioii, <>ii tlic .ipproxi- 

 niiitc liiu'iirily and coiuiium iiiuiMc tion dl tin- (ndiiMiv Imriiidity 

 curves as shown in biij- '•'•' 



1 1 is appari-nt that each of tlu- linniidity curves is in elTecl a straiKlit 

 line and that, with an accuracy sutVuient for practical purposes, these 

 curves, representing; huniidilies of from IW.J to liW/'i, converge at a 

 point (a) whose coordinates are (h, c). Assuming that the humidity 



r BtlLB rCMPCIUTuU ■ 



Fig. Q 



curves are straight lines passing through |)()int (a), it is apparent 

 that the value of humidity is completely determined if the slope of 

 the particular curve is known, since each curve represents only one 

 value of humidity. It also is apparent that the slope is given by the 

 ratio of dry bulb temperature minus the ordinate of point (a), to 

 wet bulb temperatures minus the abscissa of point (a) ; or in other 

 woriis, the relative humidity is completely determined, if the dry bulb 

 and wet bulb temperatures are each known, above the datum coordi- 

 nates (b. c) of point (a). 



If then, a resistance is set olT, proportirjnal to the difference iK'tween 

 the temperature of the dry bulb and temperature (b), and another 

 resistance is set olT proportional to the difference between the tem- 

 perature of the wet bulb and temperature (c), the ratio between 



' Bur. Stands. Cir. No. 55, p. 116. 



