116 Professor Forbes' s Researches on Heat. 



suggested its use), and yet it does not appear that the general 

 conclusion which I intended is entirely warranted. 



37. It is well known that Sir Isaac Newton overlooked the 

 variable dispersive power of bodies for light, in consequence 

 of having compared two, in which the dispersion happened to 

 be proportional to the mean infraction. A similar haste to 

 generalize would have led to error on the present occasion, 

 had not a simultaneous investigation led me to re-consider 

 the subject of powders. Whilst waiting for the arrival of 

 fine wire-gauze from Paris, it occurred to me to try the effect 

 of metals in a state of extreme division. It seemed, however, 

 first desirable to ascertain whether the metals are as incapable 

 of transmitting heat as is commonly supposed. 



38. For this purpose, I stretched a piece of the thinnest 

 gold-leaf across a wide diaphragm of pasteboard, and suffered 

 an intense parallel beam of heat from Locatelli's lamp to fall 

 directly upon the pile. A screen of glass was interposed, 

 which, by experiment, was known to stop 43 per cent, of this 

 sort of heat. The needle of the galvanometer deviated 31°*2, 

 the glass being interposed ; the equivalent direct effect would 

 have been 31-2 x -j^ — 72°. When the glass was removed 



and the gold-leaf put in its place, on the brass screen being 

 alternately introduced and removed, not the faintest motion 

 was perceptible in the needle ; had it amounted to ^th of a 

 degree, that is, had y^^dth of the incident heat been trans- 

 mitted by the gold-leaf^ I considered that the effect would 

 have been perceptible. Yet this gold-leaf was so thin that 

 the features of a landscape could be distinctly seen through 

 it, of the usual bluish-green tint. No more convincing proof 

 certainly can be desired, that conduction plays no sensible 

 part in these experiments, since it did not sensibly act on a 

 film of one of the best-known conductors of heat, and per- 

 haps not more than ^nooo o c ^ tn °f an mcn tWdk I thought it 

 worth while to repeat the experiment with dark heat, and with 

 the same results. The analogy of the action of split mica on 

 light to metallic reflexion led me to suspect, that if any kind 

 of heat were transmitted by metallic leaves, it would be that 

 of low temperature. 



39. The imperviousness to heat of gold-leaf, the thinnest 

 continuous film of metal which we can obtain, satisfied me 

 of the importance of obtaining the metals in a condition to 

 verify my experiments with the powder of other substances. 

 When the hope diminished of obtaining wire-gauze of a de- 

 gree of fineness (I mean fineness in the wire, not closeness of 

 texture, for that was comparatively immaterial) which might 



