On the Expansive Power of Liquids. 101 



Aet. XX On the Application of the Expansive Power of 



Liquids to produce a Reciprocating" Rectilinear Motion. In 

 a Letter to the Editor. 



Sir, 



It has sometimes occurred to me that the almost entire in- 

 compressibility of liquids, employed with so much effect in 

 Bramalfs invention, might likewise be used to produce an in- 

 controllable reciprocating motion, and be thus made to com- 

 municate uniform motion to machinery of any extent. I was 

 curious enough to make some calculations upon the subject in 

 the rough way, and if you think they will answer any of your 

 readers, they are at your service. The raising and lowering of 

 the temperature of a liquid produces an alternate expansion and 

 contraction which may move a piston. Let us take a particular 

 example in order to form definite ideas respecting the expence of 

 such a motion. Suppose that to a cylinder, A B, Plate I. Fig. 

 7, whose interior diameter is three inches, there is fitted a pis- 

 ton, the rod of which is in diameter two inches. Suppose also 

 that the cylinder is five feet long. Now, if the openings com- 

 municate with two vessels containing a liquid, whose alternate 

 and opposite expansions and contractions are to produce the 

 alternate rise and fall of the piston, it is plain that the liquid 

 communicating with the cylinder at a must expand a quanti- 

 ty equivalent to the content of the cylindric space betwixt the 

 cylinder and the rod ; and that the liquid communicating with 

 the cylinder at b must expand a quantity equivalent to the 

 content of the cylinder itself. According to our supposition, 

 this in the former case will be 235.6 cubic inches, and in the 

 latter 425.11. Fix for an instant the alternate increase and 

 diminution of temperature at 60°. Alcohol expands about 

 j 7 of its bulk for an increase of temperature to this amount. 

 Hence the requisite expansion will be provided for if the ves- 

 sel communicating with»i contains about 6.6 cubic feet of alcohol 

 and the vessel communicating with a about 3.6 feet. A stroke' 

 of the piston then will be effected by a waste of heat capable 

 of raising about ten cubic feet of alcohol 60°. And this stroke 

 will (neglecting the small allowance for the compressibility of 



