5G Prof. Potter on the Fourth Law of the Relations of the 



had other experiments in view on the same subject, and soon 

 after I procured one of M. Breguet's exceedingly sensitive me- 

 tallic helix thermometers, by means of which I hoped to ascer- 

 tain directly the temperature of an expanding jet of air entering 

 the exhausted receiver of an air-pump, but was disappointed in 

 obtaining anything more than confirmatory results, and should 

 not by means of it have ascertained the law. This arose from 

 the unsteadiness and vibrations of the helix and index-needle 

 when a jet passed through them, even when an addition had 

 been made to the instrument to steady the index-needle at the 

 expense of a small loss of sensibility. This addition consisted of 

 a small brass vertical pillar screwed into the pedestal under the 

 centre of the helix, with a small cylindrical hole down a part of 

 its axis, which was directly over a small polished agate cup on a 

 part of the pillar which had been filed away so that it was directly 

 under the cylindrical hole. The vertical needle of the helix 

 which carries the horizontal index-needle being raised and passed 

 down the fine cylindrical hole, its sharp point rested on the agate 

 cup. A very small loss of sensitiveness was found, but great 

 additional steadiness, from this arrangement. The helix in the 

 experiments was surrounded with a cylinder of gilt paper rather 

 larger than itself, which was attached air-tight to the part of the 

 instrument where the upper end of the helix is screwed fast. 

 This part turns round in a step to adjust the index-needle to any 

 point in the horizontal circle of degrees, and has a hole down its 

 axis. This hole was used in the experiments to pass a jet of air 

 through. A long fine brass tube, small enough to go into the 

 hole just named, passed through a stuffing-box in a brass plate 

 covering the opening in the top of a large glass receiver of an 

 air-pump within which Breguet's thermometer was placed. The 

 fine brass tube had its lower end prepared with an aperture 

 w T hich could be opened and shut by a plug at the end of a brass 

 wire passing down it, which was somewhat longer than the tube. 

 The tube being smaller than the hole in the moveable part of the 

 thermometer, was packed tight in it at its lower end with lint. 

 A light cup of paper was, after some experiments, also placed to 

 rest on the index-needle, to turn with it and envelope the lower 

 end of the vertical cylinder of gilt paper without touching it. 

 This was intended to retain the expanded air of the jet around 

 the helix. When the receiver of the air-pump was partially ex- 

 hausted, and the plug raised by means of the long wire passing 

 down the brass tube, the external air rushed into the receiver 

 through the cylinder of gilt paper, and communicated its tempe- 

 rature to the helix ; the temperature was thus shown by the 

 degree on the scale to which the index-needle pointed. It will 

 be easily conceived that the helix, of which the thickness (con- 



