104: REPORT—1845. LNT KO 
or cooled (in comparison with the surrounding space), the influence on de- 
viation was just as the above-mentioned effects of friction might lead us to 
expect. Iwas confirmed in this position by operating on many groups or 
eombinations of the substances which form the thermo-electric series. Thus, 
for instance, the sulphuret of molybdenum, which when joined to bismuth 
gives no deviation by difference of temperature, appeared likewise with- 
out any influence when rubbed on the same metal. The sulphuret of lead 
(galena), which alone in the whole series makes bismuth negative by heat, 
renders it also negative by friction. Omitting for the present some very 
interesting details, which I reserve for a monograph of tribothermie elec- 
tricity, it seems evident, therefore, that in these experiments the metallic 
conductors of electricity are thoroughly devoid of such specific or direct 
faculty of producing positive or negative electrodynamic actions, as the iso- 
lating substances possess for producing electrostatic effects ; if you should not 
incline, with some of our philosophers, to regard even the electricity pro- 
duced by friction of isolators as but a modification of heat. But postponing 
this question, let us see in what manner the theory, and perhaps even the 
practical application of electricity, may be promoted by the researches on 
tribothermic electrization. For this purpose we must enter into some further 
details :—1. The tribothermical effect is an instantaneous one. Indeed, at 
the very beginning of friction of any intensity, the needle moves. There 
is no trace whatever of the retardation undergone by heat when spreading 
through the mass of any substance. 2. The tribothermic effect is likewise in- 
dependent of the masses putin action. The point ofa needle rubbed against 
a considerable heterogeneous mass, gives immediately the deviation ; and an 
increase of extent of the surfaces in friction does not appear even to add 
materially to the intensity of electrization. 3. The deviation vanishes quite as 
instantaneously as it commenced, and the immediate return of the needle to 
its primitive station is even one of the most striking features of the phzno- 
menon. These three facts are very instructive, and seem by far more likely 
to be effected by a vibratory motion of molecules, than by the continuous 
efflux of a calorific fluid. Indeed, if we suppose any mass imbued with a 
given quantity of heat, and producing, when brought into contact with the 
other elements of a couple, a certain deviation proportionable to this quan- 
tity ; the slightest increase of deviation would then require a considerable 
addition of heat, and, such addition taking place, the deviation could but 
augment very slowly, while, on the contrary, we find by experiment that the 
slightest friction produces a strong deviation. Moreoyer, supposing once 
more that the very quantity of heat, represented by the temperature and by 
the mass of the whole body, were the efficient cause of the deviation, the in- 
crease of deviation produced should be durable, while by experiment we 
always see it instantaneously vanish when friction ceases, just as should be the 
case were it produced solely by a molecular action of the rubbing-points. In 
the event of the refrigerated metal giving a western deviation, which a mo- 
mentaneous friction inverts into an eastern one, but only as long as the fric- 
tion lasts, the result is still more paradoxical, and we have probably no other 
explanation of it, but by admitting a specific difference between the mode 
of production of heat in this-case on the one hand, and in that of heat per- 
manently residing in the body on the other. The type of molecular vibra- 
tions will once more, and very naturally, be recalled by this remarkable fact. 
4. The tribothermic deviations attain in every case amaximum, which under 
similar circumstances is different for different couples of metals. Indeed the 
friction produces, while it exists, new increments of heat which must give rise 
to increments of deviation. These latter however become more and more 
ae 
