554 Royal Society : — 



view to determine the melting-points of a series of these alloys. 

 He adopted Deville's determination of the boiling-point of zinc 

 (1040° C.) as the basis of the inquiry, and ascertained, by the method 

 of mixtures, the mean specific heat of a mass of wrought iron 

 between 0° C. and the melting-point of silver, which, as Becquerel 

 showed, is the same as the boiling-point of zinc. 



The mean of three experiments, which were closely in accor- 

 dance, gave 0*15693 as the specific heat ; and it should be pointed 

 out that this number includes and neutralizes several errors which 

 would affect the accuracy of the subsequent determinations. 



The melting-points of several alloys were then determined by 

 plunging an iron cylinder into them and transferring the iron to a 

 calorimeter. These melting-points varied from 840° C. to 1330° C, 

 or through a range of 490° C. The alloys which occupy the lowest 

 portion of the curve contain from 60 to 70 per cent, of silver. 

 The results are interesting, as they show that the curves of fusi- 

 bility and electric conductivity are very similar. 



The author states that, in studying the phenomena of liquation, 

 the alloys were cast in red-hot moulds of firebrick in which the 

 metal (about 50 oz.) could be slowly and uniformly cooled. The 

 results showed that the homogeneity of Levol's alloy is slightly dis- 

 turbed by this method of casting • and, on the other band, that 

 alloys which contain more than 71 '89 per cent, of silver hardly 

 show signs of rearrangement when the solidification is gradually 

 effected. Two alloys were examined, which contained 63 and 

 33*3 per cent, of silver respectively. Both were found to be 

 far from homogeneous. In the case of the former the arrange- 

 ment was influenced by gravity, the base of the casting being 

 rich in silver. 



The density of pure silver and of Levol's homogeneous alloy 

 while in the fluid state were then determined bj the method de- 

 scribed by Mr. Robert Mallet'*, the metals being cast in conical 

 vessels of wrought iron. The results obtained were as follows : — 



Density fluid. Density solid. 



Pure silver 9-4612 10'57 



Levol's alloy 9-0554 9-9045 



In the case of silver, the mean linear expansion deduced from 

 this change of density is -00003721 per 1° C, which is nearly 

 double the coefficient at temperatures below 100° C. 



April 15. — Joseph Dalton Hooker, G.B., President, in the Chair. 



The following communication was read :— 

 " Researches upon the Specific Volumes of Liquids." By T. E. 

 Thorpe. 



I. On the Atomic Value of Phosphorus. 



Hermann Ivopp has shown that, as a rule, the specific volume of 



* Proc. Eoy. Soe. vol. xxiii. p. 209. 



