P Y R M E T E R. 



Pyrometer. Let us now suppose that the temperature of the 

 two cast iron prismatic bars is kept unaltered, which 

 can be easily ascertained by keeping a thermome- 

 ter in the water of the troughs in which each is 

 placed. Let the first microscope with a point in its 

 field be set above one end of the metallic bar to be ex- 

 amined, so that this point or mark is directly coinci- 

 dent with the extremity of the metallic bar, or with a 

 point or mark near its extremity. In like manner, let 

 the second microscope be fixed above the other extre- 

 mity of the metallic bar, so that its moveable wire is 

 exactly coincident with the other extremity of the 

 bar, or with a point or mark near it. When these 

 adjustments are made, let us suppose that the tempera- 

 ture of the water in the copper trough, and conse- 

 quently the temperature of the metallic bar itself is ex- 

 actly 50. Place the twelve spirit lamps beneath the 

 copper trough, so as to raise the temperature of the wa- 

 ter in it, and consequently that of the bar gradually to 

 150 or any other temperature ; and look through the se- 

 cond micrometer microscope. From the instant that the 

 bar begins to experience the expansive action of the 

 heat, the mark at its extremity will be seen to lose its 

 coincidence with the wire of the micrometer, and ex- 

 tend beyond it. At any temperature, therefore, namely, 

 60, 70, 80, we can measure the expansion that has 

 taken place by observing how many turns and parts 

 of a turn of the micrometer screw are necessary to 

 bring the micrometer wire into coincidence with the 

 slowly moving mark on 'the expanding bar ; and as 

 the value in parts of an inch of each turn of the mi- 

 crometer screw is accurately known, we obtain a direct 

 measure of the elongation of the bar, free from all the 

 errors of wheels, levers, and pinions. By this instru- 

 ment, General Roy obtained the expansions of the 

 bodies to which his name is affixed under EXPANSION, 

 Vol. IX. p. 254, 255. See Phil. Trans. 17S5, vol. Ixxv. 

 p. 462. 



8. On Tiovgklon's Pi/romeler. 



Trough- In our article EXPANSION, Vol. IX. p. 254, we have 

 ton's pyro- mentioned the micrometer of that able artist, Mr. 

 meter. Troughton, and promised a description of it under the 

 present article. Mr. Troughton constructed this in- 

 strument in 1794, on a small scale, for trying pendu- 

 lums, and we believe that he does not wish any drawing 

 or minute description of it published till he has com- 

 pleted the large instrument which he has long planned, 

 for measuring the expansion of bars -ten feet in length. 

 We may mention however, that this pyrometer meas- 

 ures expansion by the indications of a fine level, which 

 is made to deviate from a horizontal position by the 

 direct influence of the elongation of the bar. 



9- Description of Dr. Brenster's Chromatic Pyrometer. 



Dr. Brews- In the different instruments that have been describ- 

 ter's chro- ed, the actual elongation of the metallic bar is either 

 matio py- ma gnified by levers, or by wheels and pinions, or it is 

 ^smeter. measured directly by a micrometer screw attached to 

 CCCCLXXI. a compound microscope. In the present instrument, 

 No. II. however, the elongation of the bar is measured by the 

 Fig 1 . 8. number or the intensity of the polarised tints, which 

 it produces by the inflexion of a plate of glass. If 

 AB, Plate CCCCLXXI. Fig. 8. for example, is the me- 

 tallic bar at a given temperature, whose end A touches 

 the surface of the glass-plate P Q, then, when it ex- 

 pands by heat, the end A will bend the glass plate in- 

 to a curve, the sagitta of which is equal to the elonga- 

 tion ; and by the principles already fully detailed in 



our article OPTICS, Vol. XV. p. 606, the polarised Pyrometer, 

 tints corresponding to the degree of inflexion of O P, V^py***/ 



and consequently to the length of the sairitta of curva- 

 ture in P Q, or the elongat.on of the bar, will be seen by 

 looking through the erlge P Q with a polarising and 

 analysing apparatus. The tints thus produced have an 

 accurate numerical value, and the scale by which they 

 measure the elongation, may be varied by akering the 

 thickness and length of the plate of glass P Q. 



There is one form of this instrument which seems 

 to have a peculiar application to the use generally made 

 of pyrometers for horological purposes. Let us sup- 

 pose that the clock-maker wishes te have the exact 

 length of a bar of zinc, which has the same expansion 

 as a bar of brass of a given length, for the purpose of 

 destroying the effects of the expansion by their opposite 

 action. The ends of the two bars cannot be made to 

 press on opposite sides of the plate of glass without 

 breaking it, but by using these plates, as shown in 

 Fig. 8, where XZ is a plate of glass placed between 

 two plates MN, PQ, which may either be of glass or 

 metal of equal thickness. If A'B'is the bar of zinc, 

 and CD the bar of brass, whose expansions are to be 

 equalised, each of them will produce the same inflexion 

 in the plates MN, PQ, which, if they are made -of. 

 glass, will show either equal or unequal tints ; or the 

 middle plate XZ will show no tints at all, if the expan- 

 sions of the metallic bars are equal. If the curvature of 

 one of these plates predominates, the plate XZ will be 

 concave towards that plate, and this concavity, though 

 too minute to be ascertained by the eye, will be ren- 

 dered obvious by the positive or negative character of 

 the polarised tints on each side of the neutral line. 



A pyrometer of another kind has been suggested 

 by Dr. Brewster. The metallic bar AB presses against 

 a piece of metal AC, which has its face AC ground 

 with emery. A small metallic, or glass cylinder D, 

 truly ground into a socket, in which it freely moves, is 

 turned round by. the action of the face AC, which is 

 kept against it by a spring. Instead of putting an in- 

 dex and graduated plate upon D, it is connected with 

 the axis of a goniometer, and its angle of rotation mea- 

 sured with great accuracy. The angle thus found will 

 be an exact measure of the arch of rotation of the sur- 

 face of the cylinder, and consequently of the elongation 

 of the bar. 



CHAP. II. ON PYROMETERS FOB MEASURING HIGH 

 DEGREES OF HEAT. 



ALTHOUGH the instruments which we propose to Pyrometers 

 describe under this chapter, are only thermometers f r measur- 

 adapted for measuring degrees of heat far above the In " high 

 indications of the common thermometer, yet as they e 

 are in reality different instruments, founded on differ- 

 ent principles, and distinguished by a different name, 

 we shall not scruple to introduce them here. 



1. Description of Wedgervood'j Pyrometer. 



In our article CHEMISTRY, (Vol. VI. p. 52.) we have Wedge- 

 already given a general description of this instrument, wood's py- 

 and of the principles on which it is founded. We meter ' 

 have, therefore, only to give a drawing of the gage, r ' AT] 

 and an account of the relation of its scale to the com- ^ 0> jj XX 

 mon thermometrical scale of Fahrenheit. pjj 9 ' 



In Plate CCCCLXXI. Fig. 9, which represents the 

 pyrometrical gage, A BCD is a smooth flat plate, and 

 EF and GH two rulers, or flat pieces, a quarter of an 

 inch thick, fixed upon the plate, with the sides that are 

 towards each other made perfectly true; and a little far- 



7 



