On the Expansive Power of Liquids. 103 



whose expansion had just raised it. Contraction would im- 

 mediately take place ; and expansion beginning in the oppo- 

 site vessel, the piston would move to the other extremity. If 

 the expansion should become more rapid than the opposite 

 contraction, the apparatus would probably give way. In or- 

 der to prevent this, there must be attached to the vessels safe- 

 ty-valves. These may be of a very simple kind. Part of the 

 vessels may consist of cylinders, in which pistons are placed, 

 free to rise only when the great piston has reached the extre- 

 mity to which it has been sent by the expansion of that parti- 

 cular vessel. Too rapid expansion in one of them, will mere- 

 ly have the effect of raising this piston in the other, and the 

 hot water may not issue into the pipes until the contraction is 

 complete and the piston is again shut. This arrangement, 

 however, might have the effect of keeping the piston station- 

 ary at one end of the cylinder for a certain time. A mode of 

 obviating this difficulty will occur at once. There may be 

 two or more similar vessels at each end of the cylinder, one of 

 which will always be in the proper state of contraction, and 

 from it may the machinery cause the expanding alcohol to 

 flow. By a few other simple contrivances, a regular recipro- 

 cating motion may be ensured, and the machinist put in pos- 

 session of the greatest power which can be derived from the 

 agency of heat. The foregoing calculations were made on 

 rather approximating principles, but they are sufficient for the 

 general purpose I have in view. I have merely intended to 

 throw out some hints, which I conceived might lead to im- 

 portant practical results. My errors will therefore be par- 

 doned. 



I beg to subjoin the result of part of some investigations, in 

 which I am now engaged with respect to a kindred subject. 

 The operations of the high pressure engines in Cornwall have 

 now entirely set to rest the question regarding the eligibility 

 of such engines. The ratio of the increase of the density of 

 steam along with its elastic force, is likewise quite decisive of 

 this question. It is stated, if I recollect aright, by Mr South- 

 ern, and several other engineers of authority, that the density 

 increases in the exact ratio of the elastic force. I have been 



