240 Mr A. Macpherson on a Self- Acting .Apparatus 



the atmospherical pressure on the surface of a very small piston ; 

 and, lastly, it would, according to the determination of Dulong and 

 Petit, require about 400 cubic inches of mercury to produce one 

 single cubic inch of vacuum in its depression between those two de- 

 grees of temperature. In short, it would not answer the purpose. 



The next idea was kindly suggested to me by Sir David Brewster. 

 It was to employ the expansion of metallic rods, on the principle of 

 the pyrometer, to effect the same purpose ; and something analogous 

 to the original pyrometer of Muschenbrock would probably be the best 

 adaptation. A bar of zinc, or any other expansive metal, of convenient 

 length, may have its expansion or contraction increased to any extent, 

 by a series of compound levers ; and this might be applied as the mo- 

 tive power in a manner precisely similar to that described. But the 

 mechanical difficulty of regulating the action of the pyrometer be- 

 tween two specific degrees of temperature again interposes ; other- 

 wise, I. have no doubt but that a properly-constructed apparatus of 

 this kind might be the most unfailing means of action I could adopt. 

 It has these advantages over the mercury, that its power is irresisti- 

 bly great, its cost would be trifling, and it would be of much easier 

 application. 



Finding these instruments to be practically inapplicable, I was 

 driven to think of something more resembling the sudden expansion 

 of water when freezing. It is obvious that this remarkable property 

 of water is unlike the regular expansion of metals. What was ne- 

 cessary, I then thought, must be some liquid that will solidify a 

 degree or two higher than water, and of its expansion being applied 

 to shut off the water. Glacial acetic acid assumes the state of a 

 solid at a temperature so high as 45°, and sulphuric acid of the 

 density of 1*78 does the same at 46° Eahr. I was proceeding to 

 experiment upon these, or similar bodies, when it occurred to me 

 that of all the applications of this kind the best might be found to 

 exist in the expansive force of water itself when freezing. 



Water, according to the experiments of Boyle and Dalton, in- 

 creases in bulk about one-ninth in its transition from the fluid to the 

 solid state. This expansion occurs with an irresistible degree of 

 force. The Florentine academicians burst a hollow brass vessel with 

 a cavity of only 1 inch in diameter, by freezing the water it con- 

 tained. The force produced was equal to 27,000 lb. Major Wil- 

 liams, in his experiments at Quebec, in 1784-5, besides bursting 

 bombshells, projected an iron plug, 2| lb. in weight, from one of the 

 shells, to a distance of 415 feet by the expansive force of the water 

 it contained when freezinor. These facts afford sufficient grounds 



o o 



for establishing the most important points of the desired motive 

 power ; those, namely, of the amount of expansion the water sus- 

 tains, and the prodigious force with which that is accompanied. I 

 calculated, that I would require in a vessel, such as I have described, 

 27 cubic inches of water, which would produce 3 inches of expan- 

 sion, to act effectively on a stop-cock. 



