[ 348 J 



LVI. On Liquid Expansion. 



By John James Waterston, Esq.* 



[With a Plate.] 



IN the Philosophical Magazine for June 1861 there is an 

 account of a presumed law of liquid expansion which repre- 

 sents the proportionate differential of volume -7— at a given 



temperature t, as being simply proportional to the inverse of the 

 distance of that temperature from a fixed upper limiting tem- 

 perature 7. Expressed as an equation, this presumed law is 



-7— <X<y—t; and the integral equation corresponding to this dif- 

 ferential showed a seeming relation to the law of vapour-density 

 by an apparent constancy in the value of F (see above-mentioned 

 paper, § 3), the product of the coefficient of saturated vapour 

 density of a body by the exponent of the liquid density of the 

 same body. I have for some time past been employed in testing 

 this by means of M. Pierre's extensive series of experiments on 

 liquid dilatation, taken in conjunction with those on vapour- 

 tension by M. Regnault, lately published in his second volume 

 of Memoirs. A large proportion of the liquids were the same in 

 both, and chemically pure, so that the data are valuable for 

 testing the law supposed to connect the volume of a liquid with 

 the density of its saturated vapour at different temperatures, 

 although not sufficient otherwise to test the law of liquid expan- 

 sion by itself. But if F maintained its value at the lower range 

 (that is, under the boiling temperature of the liquids), to which 

 M. Pierre's researches were confined, as well as at the upper 

 range, to which my observations in sealed tubes were chiefly 

 directed, the integrity of the presumed law at all temperatures 

 might with some certainty be inferred. Now, on projecting the 

 coordinates of M. Regnault's vapour-tensions below the boiling- 

 points of the respective liquids, according to the mode previously 



described (viz. f ^ J laid off as ordinate to t), I find that 



the points for the most part range admirably in a line, and that 

 a large proportion of those lines (including those of mercury and 

 sulphur), when produced downwards, meet (below the axis of 

 temperature) at the zero of gaseous tension —274° (the apparent 

 parallelism previously noted being actually a convergence). 



This so far confirms the law of vapour-density, and gives a 

 very accurate value for the coefficient h of each vapour respect- 

 ively. The exponent p of the liquid, that ought to have a con- 



* Communicated by the Author. 



