and its relation to Latent Heat. 17 



Betv^'een 28° and 95° C. I observed the capillary column in flat 



tubes and plates todescend r—, tlie height at 28° oi-yrj-p, the height 



at middle temperature 61 ^° [§ 10] ; hence -j- = -087. The value 



SI 



of -Y- at this temperature is from Regnault's formula •0785, and 



— from Despretz is 0117. These give -087 --0785 = -0083 



instead of •0117. If we assume the decrement of capillarity -j-,-, 



we should have -0909 — ^0785 = •012-4, which is very close to the 



observed value of — . The value of this -r-, computed from M. 



WolPs empirical formula, is -. This formula is derived from 



observations made in a tube of round bore yijydth of an inch 

 in diameter from 0° to 25°. 



Dr. Young, in his essay on the cohesion of liquids, casually 

 mentions that he observed the capillary column descend at the 

 rate of jo'ojyth for each degree of Fahrenheit's scale above 50'. 



This would arive -r= ~„. In a tube of xVth of an inch round bore 

 * h 17 ^-^ 



2 . . 2 



I found it to be— [^ 10]. The equation requires it to be .yy. 



§ 19. In the cases of alcohol, tether and turpentine, we have 



ST 

 no direct information as to the value of -j- ; but we may com- 

 pute it for each of these liquids by means of the equation, from 

 their differentials of volume and capillarity. Future observation 

 must decide as to their accuracy. Enough for the present if 

 they are such as may be deemed probable. 



the data for alcohol (85 per cent, pure spirit) are, from 30° 



to 79° C. column descends from 113 to 103, or ~r=^TFo^ 



^ = J[^ -.0189. Hence ^=0737, and L = 253° (allowing 

 m 971 o 1j 



for thelo per cent. water); henco8L= (253 + ^oL) ^0737 = 19^3, 



19-3 

 and the specific heat of alcohol bcing'6 J, we have j^r — ^rrr^ =0'61 



as the proportion of each degree that goes to diminish the latent 

 heat, while alcohol is heated from 30° to 79° C. The corre- 

 spondnig proportion in water is, according to Rcgnault, ()-0!)5. 

 This proportion may be called 8\, it being understood that the 

 latent heat is cxpn^ssed in degrees, which have a mtchaiiical 

 equivalent corresponding with the specific heat of the litpiiil. 

 J'/ut. May. S. 4. Vol. 15. No. 97. Jan. I S5S. C 



