1891.] Heat produced by compressing Liquids and Solids. 133 



Description of Apparatus and Experimental Methods. 



The compression apparatus used in all these various series of 

 experiments was one originally constructed for testing the effect of 

 pressure on thermometers. Its construction is shown in fig. 1. The 

 cylinder, called " the gun," is of wrought iron, about 16 inches 

 long and 4 inches diameter, with a bore of 1 inch. It is fixed 

 vertically, and the lower end is flanged and closed by a very thick 

 flange plate, bolted upon soft leather washers, and drilled to admit 

 water from the pump. The pump-barrel consists of a steel ingot, 

 If" X 6", drilled with a hole j inch diameter ; in this works a plunger, 

 with a steel cup at the lower end, through a stuffing-box and cup 

 leather. The stroke of the pump is about 2|- inches, and it is moved 

 by a handle nearly 3 feet long, with a leverage of about 10 to 1. 

 The upper end of the gun is closed by a solid plunger, turned to fit 

 very accurately, and rendered water-tight by means of a steel cup, 

 turned to a very thin, knife-like edge, slightly belled out, so as to 

 press against the sides of the gun. When pressure comes upon the 

 cup, the sides expand sufficiently to form an almost perfectly tight 

 joint. The plunger is held in position by a key fitting into a hole cut 

 through the sides of the gun and through the plunger or ram. The 

 cup is filled with tallow or lard to avoid leaving an air space inside 

 the gun. The connexion to the pump is formed by a solid-drawn 

 copper tube with a steel connecting piece, which is screwed upon a 

 washer of soft copper. 



For measuring the pressure, a compression gauge, designed by Pro- 

 fessor Tait, and used by him in his experiments on the " Challenger " 

 thermometers, was employed. It is represented in fig. 2. Essen- 

 tially it consists of a steel tube, about 5 inches long, \ inch in 

 diameter, and -^ inch thick, filled with mercury, and bearing a glass 

 capillary, A. This tube is enclosed in a steel vessel communicating 

 with the pump by a solid-drawn copper tube. The pressure is 

 measured by the rise of the mercury in the tube A. The gauge in 

 the first instance was graduated by means of air manometers, and was 

 afterwards compared with an Amagat mercury gauge. Its indications 

 have been found extremely constant. The long bulb contains an 

 inner bulb, which nearly fills up the whole space, leaving only a thin 

 shell of mercury, which is very little affected by temperature. This 

 gauge marked 22'35 mm. per ton pressure. The amount is small, and 

 the reading cannot be said to be very accurate, but no other gauge 

 could be obtained equally trustworthy and which required so very 

 small a bulk of water a matter of great importance, as will be per- 

 ceived on referring to the description of the experiments given 

 below. 



The wires of the thermo-electric junction are introduced by placing 



