216 Transactions of (he Royal Microscopical Society. 



307° C. would be exact if the law of solubility of salt in water 

 already referred to, was well established for high temperatures. 

 Unfortunately experiments are wanted here. The law of solubihty 

 of sodic chloride remains constant to 120^ C. ; above this we 

 are ignorant of its behaviour. Considering that superheated water 

 becomes a powerful dissolvent of artificial glass in the experiments 

 of MM. Daubree and Sorby, we are led to believe that its action 

 upon sodic chloride is greatly augmented at 200° or 300°. The 

 doubt upon this fundamental point permits us to assign to our 

 number 307° only an approximative value. Accepting this as 

 such, we will continue our examination of the physical conditions 

 under which the crystallization of this rock has taken place. 

 This is an example of calculation which will hereafter doubtless 

 give results on which we can confidently rely. Knowing the 

 temperature at which the cavity was formed, we can determine 

 the pressure necessary to prevent, at this temperature, the com- 

 plete evaporation of the water. It suffices to app)ly the formula of 

 M. Koche.* 



The equation [a] gives p + — - =0,000200644 



- 35,5 



„ [6] „ w = 0,3033 i) 



p = 0,000178 

 o) = 0,000064 

 q = 0,000022 



'^' = 0,483. 

 a 



Hence t = 24 (48,3 - 35,5) = 24 x 12,8 



t = 307°. 



* The theoretical formula of M. Koche is the same as that found by Clapeyron, 

 August, De Vrede, Holtzmann. " This formula," says M. Eegnault, " represents 

 the elastic force of aqueous vapour for a great extent of temperature with remark- 

 able accuracy ; it, indeed, between 100° and 220°, gives a result for the elastic 

 force too great, but the greatest error only amounts to 35 millimetres. It is 

 applicable perfectly to the vapour of water, and also to the vapours of alcohol 

 and ether." M. Roche's formula is as follows : 



F = aa' + "'-' 



In this formula x represents t + 20°, t being the Centigrade temperature 

 counted from the melting point of ice as zero, and according to Regnault's 

 calculations : 



m = 0,004884085 



log. a = 0,0386182275 



log. a = 1,9590414 



