LIFE AND WORK OF LORD KELVIN — THOMPSON. 751 



reckoning." It was in this spirit that Thomson approached the sub- 

 ject of the transformation of heat. Joule had laid down on certain 

 lines the equivalence of heat and work, and had even measured the 

 numerical value of the equivalent. But before him, in 1824, Carnot, 

 though he proceeded on the fallacious assumption of the material 

 nature of caloric, had, in his remarkable book, Reflexions sur la puis- 

 sance Motrice du Feu, discussed the proportion in which heat is 

 convertible into work, and had introduced the very valuable notion 

 of submitting a body to a reversible cycle of operations such that, 

 after having experienced a certain number of transformations it is 

 brought back identically to its primitive physical state as to den- 

 sity, temperature, and molecular constitution. He argued, correctly, 

 that on the conclusion of the cycle it must contain the same quan- 

 tity of heat as that which it initially possessed. But he argued, 

 quite incorrectly, that the total quantity of heat lost by the body 

 during one set of operations must be precisely compensated by its 

 receiving back an equal quantity of heat in the other set of opera- 

 tions. We can see now that this is false; for if it were true, none 

 of the heat concerned in the cycle would be transformed into 

 work. Those who were investigating the subject at this time, among 

 them Clausius and Rankine, perceived this, and noted that since 

 the steam received into the cylinder must be hotter than that ex- 

 pelled from it, the degree to which the transformation is success- 

 ful must depend on the respective temperatures; a fact, moreover, 

 recognized by all engineers since the date when Watt discovered the 

 advantage of cooling the exhaust steam by a condenser. Carnot, 

 indeed, proved that the ratio of the work done by a perfect — that is, 

 a reversible — engine to the heat received from the source depends 

 on the temperatures of source and condenser only; and when these 

 temperatures are nearly equal the efficiency is expressible by the 

 product of their difference into a certain function of either of them, 

 called " Carnot's function." Eankine went further in pointing out 

 that this function was gTeater as the temperature in question was 

 lower. But here Thomson's exact mind seized upon the missing 

 essential. Temperatures had hitherto been measured by arbitrary 

 scales based on the expansion of quicksilver, or of air or other gas; 

 and the quicksilver thermometer scale did not agree precisely with 

 that of the air thermometer. He was not satisfied with arbitrary 

 scales. He had this in hand even before his first meeting with Joule, 

 and in June, 1848, communicated to the Cambridge Philosophical 

 Society a paper " On an absolute thermometric scale founded on 

 Carnot's theory of the motive power of heat, and calculated from 

 Regnault's observations." In this paper he set himself to answer 

 the question. Is there any principle on which an absolute thermo- 



