William Thomson to Faraday   10 January 1850

2 College, Glasgow | Jan 10, 1850

My dear Sir

By the same post I send two papers, one by myself, and the other by a brother of mine¹, which were communicated to the Royal Society of Edinburgh last spring².

In the paper by my brother some very remarkable reasoning founded on the principles of “Carnot’s³ Theory of the Motive Power of Heat”, is adduced to show that the freezing point of water must be lower under a high pressure than it is when only subjected to the pressure of the atmosphere.

I have today succeeded in verifying by experiment this conclusion; and I think the fact that an increase of pressure lowers the freezing point of water may be considered as satisfactorily established.

My brother gives in his paper an estimate of the amount by which an additional atmosphere of pressure will lower the freezing point of water, wh[ich] he deduced, by Carnot’s method of reasoning, from Regnault’s experiments on the pressure and latent heat of vapour of water near the freezing point. The number which he gives is .0075 of a degree centigrade or .0135 of a degree Fahrenheit, of lowering due to 1 additional atmosphere of pressure.

After several unsuccessful attempts to test this result by experiment, I got a thermometer made with ether as the liquid, wh[ich] is so sensitive that the column of ether in the tube rises by more than two inches for 1˚ Fahr. of elevation of temperature. After graduating the tube very rudely, I had the whole thermometer sealed up in a glass tube to prevent the bulb from being exposed to the pressure to be applied to the water in the experiment. I then took an Oersted’s apparatus for compressing water and filled it with lumps of clean ice and water; and I put it in a guage [sic] for the pressure, and my ether thermometer in its glass envelope.

Although the thermometer is excessively sensitive, it remained absolutely steady in the mixture of ice & water, until a pressure of about 9 atmospheres was applied. The column of ether then ran rapidly down, & nearly settled about 7 1/2 divisions of my scale lower than its primitive position. I increased the pressure up to about 19 atmospheres above the original atmospheric pressure and the column of ether descended farther, very rapidly. After giving it ample time to settle, during which the pressure was kept constant, I found that it stood 17 1/2 divisions below its primitive position. When the pressure was suddenly removed, the column of ether instantly began to run up, and it ran above its primitive position on account of a good deal of ice having been melted, and a lead ring I employed to keep a space clear for reading the thermometer having descended below the top of the bulb.

I have found by a very rough comparison with a common mercurial thermometer, that a degree Fahr. corresponds to about 70 divisions of my scale. Hence 17 1/2 of my divisions are equivalent to .25 of a Fahrenheit degree or a quarter <_><_><_>. According to my brother’s prediction, the lowering of temperature would have been .256 of a degree, if the pressure added had been exactly 19 atmospheres. This is only six thousandths of a degree more than what I found by experiment.

When I found how very close the agreement of my results with Theory was, I was very much surprised; and, when I consider the roughness of my experiments to find the value of a division of my scale; and the uncertainty of my pressure guage, I cannot but attribute it partly to chance that I have fallen upon a result agreeing so very closely with the indication of Theory.

I hope you will excuse my having troubled you with all this. I have done so because I thought that, the lowering of the freezing point of water being a very remarkable fact apparently well established now, you might possibly feel some interest in having it brought under your attention[.]

I remain, My dear Sir, | Your’s most truly | William Thomson

Prof. Faraday