PHYSICAL PROPERTIES OF WATER, ETC. 7 



The extension of my formulae to very low pressures, though it agrees in a remarkable 

 manner with some of the best of accepted results, such as those of Buchanan and of 

 Pagliani and Yincentini, is purely conjectural, and may therefore possibly involve error, 

 but not one of the least consequence to any inquiries connected with the problems 

 to which the Challenger work was directed. 



The piezometers, which had been for three years employed on water and on sea- 

 water, were, during the end of last summer, refilled with solutions of common salt of 

 very different strengths, prepared in the laboratory of Dr. Crum Brown. The 

 determinations of compressibility were made at three temperatures only, those which 

 could be steadily maintained, viz. 0° C, 10° C, and about 19° C, the two latter being 

 the temperature of the room, the former obtained by the use of an ice-bath. Here 

 great rapidity of adjustment of the scale to the mercury was requisite, even in the 

 experiments made near 0° C, for the salt solutions (especially the nearly saturated one) 

 show considerable expansibility at that temperature. In these salt solutions, however, 

 the hair indices behave very irregularly ; so that this part of my work is much inferior 

 in exactitude to the rest. 



Besides the determinations briefly described above, there will be found in this 

 Report a number of experimental results connected with the effect of pressure on the 

 temperature of water and on the temperature of the maximum density of water. 

 Though I afterwards found that the question was not a new one, I was completely 

 unaware of the fact when some experiments, which I made in 1881 on the heat 

 developed by compressing water, gave results which seemed to be inexplicable except 

 on the hypothesis that the maximum-density point is lowered by pressure. Hence 

 I have added a description of these experiments, since greatly extended by parties of 

 my students. 



And I have appended other and more direct determinations of the change of 

 the maximum-density point. I also give, after Canton, but with better data than his, 

 an estimate of the amount by which the depth of the sea is altered by compression. 

 Also some corresponding inquiries for the more complex conditions introduced by the 

 consideration of the maximum-density point, &c. 



An Appendix contains all the theoretical calculations, the results of which are 

 made use of in the text ; as well as some speculations, not devoid of interest, which 

 have arisen in the course of the inquiry. 



II. Some former Determinations. 



There seems now to be no doubt that Canton (in 1762) was the first to establish 

 the fact of the compressibility of water. But he did far more ; he measured its apparent 

 amount at each of three temperatures with remarkable accuracy, and thus discovered 



