LATENT HEAT OF EVAPORATION 59 



does not hold for the normal molecular weight (C 2 H 4 O 2 = 60), 

 but for a higher value (about 97 according to Part II, p. 48), 

 so that q must be for that quantity also. If that is not 



the case, the quotient |, becomes abnormally small, thus : 



W = 97 y-273 = Il8 '5 fr = 14-87- 



If, as for ethyl alcohol, the vapour is normal, but the 

 liquid partly associated, then q is too large : 



W= 216.5 T-273 = 78.2 = 28, 



which might be expected for several reasons. On the one 

 hand, in addition to the latent heat, there is an absorption 

 of heat in breaking up the double molecules, by which q 

 becomes abnormally large. On the other hand, according 

 to p. 47, formation of double molecules causes an abnormally 

 rapid fall in the saturation pressure (dP) which according 

 to the thermodynamic relation of p. 57, necessarily implies 

 a high value of q. 



Since the molecular elevation of the boiling point t 

 stands in relation with the latent heat W (Part II, p. 56), 

 for: 



o-oi98T 2 0-OI98TW o-oi 9 8TM 

 t = - ^ -- = - - = - - = 0-00096 TM, 



q 20-63 



the molecular elevation of the boiling point also throws 

 light on the molecular magnitude M of the solvent. 



Normal solvents give actually values for ^^ close to 



0-00096 ; carbon disulphide 0-00096, benzene 0-00095. For 

 acetic acid, if M be taken as 60, we get the abnormally 

 high value 0-00108: for water and ethyl alcohol 0-00076 

 and 0-00072 respectively, abnormally small, as is to be 

 expected from the foregoing argument l . 



It may be added that deviation of the latent heat of 

 evaporation from Trouton's rule is one of the most com- 



1 Herz, Molekulargrosse der Korper (Ahrens, Sammlung), 1899. 



