1 82 Prof. Forbes's Researches on Heat, 



the smaller effect. It seemed reasonable to expect, that an 

 apparatus constructed on the principle of determining the 

 critical angle of total reflection of heat from different sources 

 within a prism, would afford much more definite information 

 as to the refrangibility of heat than any other method. After 

 much consideration, an apparatus of the following kind was 

 adopted. 



It is fundamentally composed of a jointed frame, resembling 

 a box exactly square, ten inches in the side, without top or 

 bottom, and having hinges at every angle, so that it may be 

 formed into a lozenge of any degree of obliquity. This is 

 seen in Plate IV. fig. 1, and marked A B. By an arrange- 

 ment presently to be described, the rays of heat are made to 

 pass parallel to the edge a c of one of the sides of the box, 

 and to fall upon the prism P, whence, after undergoing re- 

 flection (total or partial) at the posterior surface of the prism, 

 they proceed parallel to the line a d, and fall upon the sen- 

 tient extremity of the pile at p. Now, in order that this 

 course may be taken by the reflected rays, it is necessary that, 

 supposing the prism to be an isosceles one, the posterior re- 

 flecting surface a' Z/, fig. 2, should form equal angles with the 

 incident and reflected rays c e and yd. It was to effect this 

 that the arrangement of the jointed lozenge was adopted. 

 The prism P (fig. 1.) rests on a column O, moveable round 

 the line of junction of the sides C and D of the lozenge. The 

 column O has connected with it a tail-piece of brass a E 

 passing through the diagonal of the frame, and preserved con- 

 stantly in that position by a slit parallel to its length, through 

 which passes a clamping screw b, serving at once to maintain 

 this constancy of direction, to secure the form of the move- 

 able lozenge, and by means of an index pointing to a gradu- 

 ated scale of inches reckoned from a, along a E, to determine 

 the length of the diagonal a b at any moment, and consequently 

 the angles of the lozenge. 



A little consideration of this mechanical arrangement, will 

 show how it is adapted to the end in view. The rays from a 

 source of heat S, rendered parallel by the lens of rock-salt L, 

 fall upon the prism P, and after undergoing two refractions 

 and one reflection, they fall upon the sentient surface of the 

 pile p. This will always take place so long as the posterior 

 surface of the prism forms equal angles with the lines ac, ad^ 

 which will be secured by making it truly perpendicular to the 

 tail-piece a E, by which it is guided, and which of course al- 

 ways bisects the angle cad. Now, it is evident that, whilst 

 the angle cad remains small, the reflection will continue 

 partial^ but that as the diagonal a 6 is shortened, a point will 



