132 Mr. S. U. Pickering on the Densities 



(d) In the last column the volumes of water, taking that 

 at 4° as unity, are given. These volumes, like all the numhers 

 deduced from formula No. 1, are referred to a pressure of one 

 atmosphere. In order to obtain the volumes at a pressure of 

 p atm. ; we must divide the numbers in the table by 



1+^(^-1), 



just as was done previously when examining Hirn's figures. 



In conclusion, I think it necessary to repeat that, whenever 

 I am able I shall endeavour to make a series of fresh deter- 

 minations, taking into consideration all the necessary conditions 

 of the variation of the density of water with a change of 

 temperature, because the sum of modern information on this 

 subject has been already amassed, but suppositions have been 

 admitted (for example, the constancy of the coefficient of ex- 

 pansion of glass and mercury irrespective of a change of 

 temperature, the absence. of the influence of pressure, &c), 

 which cannot be held to agree with our existing knowledge. 

 And should fresh determinations, made with all possible 

 accuracy, confirm the aspect of the formula 



S-1- (*- 4 ) 2 

 D *— (A + t){B-t)C 



or lead to a more correct formula, then we may hope by its 

 means to arrive at a better understanding of the true law of 

 the expansion of all liquids, and consequently of gases also. 

 The correct idea of the influence of heat on densities and 

 volumes began with the study of water, and, in my opinion, we 

 may expect, by means of investigations upon water, to make 

 further progress in the study of matter under the influence 

 of a rise of temperature. 

 St. Petersburg, April 1891. 



XII. The Densities of Sulphuric-Acid Solutions. 

 By Spencer Umfreville Pickering, F.R.S.* 



A SHORT time ago Mr. Lupton (Phil. Mag. xxxi. p. 424) 

 attempted to disprove one of the changes of curvature 

 in the figure representing my " first differential " of the den- 

 sities of sulphuric-acid solutions by bridging it over by a 

 straight line. As, however, this figure is evidently curvi- 

 linear, it was not surprising that he failed, even though he 

 selected for the attempt that particular change which, as I had 

 pointed out, was more doubtful than any other (that at 58 per 

 cent.), and for the same reason it is evident that he would 

 * Coniniunicated by the Author. 



