M. H. Sain te- Claire Deville on Affinity and Heat. 371 



ing the coefficient of expansion k of the liquid from 0° to a tem- 

 perature a little higher than 6, the specific heat c of the liquid 

 being supposed constant between the same temperatures, and m 

 its weight, we shall have for the heat of contraction Q the value 



e-Of 



Q, 



the temperature 6, at which the mixture resumes its original 

 volume, being 



(HH 



I have determined, by methods which I cannot describe here, 

 the heats of contraction of a great number of liquids resulting 

 from the combination or the solution of two elements also liquid, 

 such as sulphuric acid and water in very variable atomic propor- 

 tions, sulphuric acid and soda of various degrees of dilution, 

 water and alcohol, water and acetic acid, water and formic acid, 

 varying the proportions, which are always atomic, in a great num- 

 ber of modes*. 



The following is the result of my experiments : — 



(1) When two liquids combine or dissolve and give a product 

 which is also liquid, the highest temperature resulting from the 

 mixture is generally smaller than the temperature 6 which con- 

 traction could give if the liquid disengaged all the heat corre- 

 sponding to this contraction. 



(2) Hence the quantity of heat disengaged in these kinds of 

 combinations or solutions is always less than the heat of con- 

 traction. 



It follows that, in all the cases which I adduce, the mere phe- 

 nomenon of contraction is sufficient, and more than sufficient, to 

 explain the development of heat in chemical combinations. 

 Hence part of the heat which contraction disengages becomes 

 latent in the new compound, and there plays an important part 

 which I shall afterwards point outf. 



This quantity of heat, which is latent or lost to the thermo- 



* Compare Comptes Rendus, vol. 1. pp. 354 & 584. 



t The heat absorbed has served to reduce the liquid from the original 

 volume corresponding to the level A and the temperature 6, to the volume 

 corresponding to the level B and the temperature t ; it is indicated by the 

 temperature 6—t. The original liquid has therefore been forced upon itself, 

 owing to combination. Knowing its coefficient of compressibility between 

 6 and t, and its specific heat C between these temperatures, its mass being m, 

 we might calculate the weight P with which the liquid at A must be charged 

 to make it sink, on being compressed, to the level B. The number of 

 kilogrammetres obtained on multiplying this weight by the distance AB 

 would correspond to the work of a quantity of heat equal to {6— t)mC, and 

 would enable us to obtain the mechanical equivalent by a chemical method 



2 B 2 



