460 On the Adiabatics and Isothermals of Water. [June 18, 

 whence for all points on the curve =0, we have 



Cvv 



fd?v\ /^ 

 W/Q W 

 and therefore the contact is of the second order. 



P.S. Since the above was written, a paper has been published in the 

 'Annales de Chimie et de Physique 7 for March 1874, in which the 

 author, M. J. Moutier, is led, from thermodynamical considerations, to 

 the conclusion that it is impossible for aqueous vapour in contact with 

 ice to have the same tension as when it is in contact with water at the 

 same temperature ; and as some conclusions have been pointed out in the 

 preceding pages which follow on the assumption that at the triple point 

 the tension of the vapour is the same in each case, it may be well to show 

 that his arguments do not really touch the question as to which of the 

 two hypotheses is the true one. 



M. Moutier discusses the case of a body which can exist in two dif- 

 ferent states, M and M', such as the solid and liquid ; and supposing that 

 the tension of the vapour is different according as it is in contact with 

 the first or second, he obtains a general formula for the heat of trans- 

 formation from M to M', from a consideration of the quantities of heat 

 gained or lost if the body is compelled to undergo a definite series of 

 changes constituting a closed cycle (p. 348). 



The second operation in this cycle is that the body M' passes from the 

 pressure r to the pressure p' ; and in the application of the general 

 formula to the case of water, M is taken to represent ice at 0., M' 

 liquid water at the same temperature, r the atmospheric pressure, and p' 

 the tension of aqueous vapour over liquid water at 0. (p. 362). 



If, however, the symbols have these meanings, the prescribed operation 

 is, in the case of water, impossible ; for as water cannot exist at C. 

 in the liquid state at less than the atmospheric pressure, the body M' 

 would be converted into M as soon as the pressure <ar was diminished, 

 and no conclusions can be drawn from the cycle in question in the case 

 of water. 



M. Moutier employs a second argument which can be shown to have 

 no greater weight than that already discussed, and which may be stated 

 as follows : 



If Q is the latent heat of conversion of ice into water, and L and L' 

 the latent heats of conversion of ice and water respectively into steam, 

 then at the triple point we must have 



Q=L-L'. 

 L and L' are given by the well-known formulae 



L=AT(t l -)|, 



