120 
Lord Kelvin on 
[r.s.e., sess. 
280° C. {Trans. Roy. Soc. Edin ., May 1854, “Dynamical Theory 
of Heat,” part vi. §§ 105-106; Froc. Roy. Soc. Lond ., May 1854, 
“ Experimental Researches in Thermo-electricity,” § II. Both 
republished in Arts. XLYIII. and LI., vol. i., Mathematical and 
Physical Papers. See pp. 239-241 and 464-465). 
But we must now consider, also, a quasi-Peltier effect produced 
by electricity crossing the border between air or ether at the sur- 
face of either metal, and the homogeneous metal inside. We have 
absolutely no thermodynamic or molecular-hypothetic guide to 
even guess the sign or magnitude of this effect at the surface of 
either metal. It is conceivable that it may have opposite signs for 
different metals, or that it is essentially of the same sign in all ; but 
it seems to me exceedingly improbable that it is non-existent, when 
I consider Pellat’s and Murray’s discoveries of change of Volta- 
surface-potential produced by scratching and by burnishing, without 
any change of chemical constitution of the surface layer of a metal. 
§ 5. Let Q denote the total quantity of heat produced by the 
Peltier effect at J and the quasi-Peltier effect at the surfaces II 
and CC, per unit quantity of electricity flowing from II through J 
to CC, in virtue of motion of CC outwards. The part of Q which 
is produced at J is, as we have seen, positive when X and Y are 
in the thermo-electric order stated in § 3 ; but the total amount of 
Q may be either positive or negative. 
§ 6. Our Carnot cycle will consist of the following four opera- 
tions : — 
I. (“Adiabatic” — according to Rankine’s nomenclature.) The 
whole apparatus being ideally coated with varnish im- 
permeable to heat, draw out CC so slowly that the 
temperature of the whole apparatus remains uniform 
throughout, while rising from t to t', on absolute thermo- 
dynamic scale. 
II. (Isothermal.) The whole apparatus being kept by proper 
surface appliances at constant temperature t\ let CC move 
inwards very slowly, until a certain quantity of heat, H', 
has been taken in to the apparatus from without. 
firm it in respect to the magnitude, of the Peltier effect. Le Roux (1867), 
Ann. de Chim. et de Phys. (4), vol. x. p. 201 ; Jahn(1888), Wied. Ann., vol. 
xxxiv. p. 767. 
