PRESIDENTIAL ADDRESS. 811 



0)>ei'ative cycle. By Lis method a limit could be set, fixing the maximum of 

 mechanical energy to be obtained from a given heat quantity and a given 

 temperature range. To reduce this to numerical values it was necessary, how- 

 ever, to experiment on any one working fluid within the desired temperature 

 range in order to determine the work area in its relation to heat quantity and 

 temperature fall. Garnet's writings show that he intended to make such observa- 

 tions ; and, had he succeeded, thermodynamics would have become a science at 

 an early date. Garnot's death, however, in 1832, at the sadly early age of thirty-six 

 years, prevented this development. 



The name of Sadi Garnot will always be remembered by mankind as the 

 founder of one branch of the thermodynamics of the heat engine. 



His work remained practically without notice for thirteen years after his death, 

 when, fortunately, it attracted the attention of William Thomson during his 

 attendance at the Laboratory of Regnault in the year 1845. Thomson was then 

 twenty-one years of age, and had already attained a considerable scientific reputa- 

 tion. He took up the study of Garnot's work with enthusiasm. He became 

 Professor of Natural Philosophy in the University of Glasgow in 1846, and in 

 1848 he read a paper before the Gambridge Philosophical Society ' On an 

 Absolute Thermometric Scale founded on Garnot's Theory of the Motive Power 

 of Heat and calculated from Regnault's Observations.' Like Garnot, Thomson 

 accepted the ' material ' or ' caloric ' theory of the nature of heat, although, like 

 Garnot also, he had doubts as to its truth. Assuming its truth, however, he 

 carried Garnot's reasoning much further, and deduced from the Garnot cycle a 

 thermometric scale which was absolute in the sense that it defined the idea of 

 temperature independently of the properties of any particular body. 



It is very difficult to carry one's mind back to the material theory of heat, but 

 it is necessary to do so in order to appreciate the rigid accuracy of the reasoning 

 of both Garnot and Thomson ; and it is especially desirable to do so in order to 

 understand the great step made in this paper. A ccording to the ' caloric ' 

 theory, heat was supposed to be a subtle elastic fluid which permeated the pores 

 of bodies and filled the interstices between the molecules of matter. The funda- 

 mental quality imagined of this caloric or heat fluid was that of indestructibility 

 and uncreatability by any humanly controlled process. Bodies became warmer 

 when caloric was added to them, and grew colder as it left them. Galoric, how- 

 ever, might be added to a body without heating it. In this case the heat was 

 called ' latent,' and the state of the body changed from solid to liquid or from 

 liquid to vapour or gas. 



Galoric, too, was required in greater quantities for some substances than others 

 in order to warm the body equally. The capacity for caloric was thus greater 

 in some bodies than in others. 



If any particular body were heated without change of state it was hotter ; 

 that is, its temperature rose when the quantity of caloric present was increased. 

 It was not difficult to define equality of temperature. This was defined by a 

 constant condition when brought into contact. But it was very difficult indeed to 

 define temperature on any rational scale. 



To the acute and brilliant intellect of William Thomson it became apparent 

 that he had in the Garnot cycle a powerful instrument capable of widely general 

 use, apart altogether from the theory of heat engines ; and he here uses it in a 

 most skilful way to give definiteness and universal application to the idea of 

 temperature, as Professor Larmor states, ' elevating the idea of temperature from 

 a mere featureless record or comparison of thermometers into a general principle 

 of physical nature.' 



Thomson accordingly defines equal differences of temperature, in terms of the 

 reversible or Garnot engine. 



Equal temperature differences are to be differences between the temperatures 

 of the source of heat and the refrigerator, when the proportion of work produced 

 from a given quantity of heat is the same. Thermometers graduated in degrees 

 calculated in this way could naturally be treated as instruments based on definite 

 principles, independently of anj^ properties of any particular material. The idea of 



