NATURE 



[Dec. 9, ic 



The article opens with a discussion of the sense of heat, 

 and of the distinction between heat and temperature. We 

 are thus introduced to the conception of latent heat, 

 which is explained at some length. The two leading 

 methods of calorimetry, viz. calorimetry by latent heat, 

 and therniometric caloiimetry, are then discussed in 

 general terms ; and the results of the comparison of 

 the different calorimetric units by Regnault and others 

 are given. Then follows a full account of the origin of 

 the modern theory of heat, which regards it as energy, 

 and measures it by the equivalent amount of work. \Ve 

 thus have a third method of calorimetry, which is 

 called dynamical calorimetry. Of the thirty-five pages 

 occupied by the whole article eighteen are devoted 

 to thermometr)'. This is the most important, and 

 certainly the most interesting part of the article. After 

 discussing a theoretical (and to some extent prac- 

 tical) system of thermometry by mixtures of hot and 

 cold water, the thermoscope being the sense of heat 

 in the hand, the author gives an elaborate classifica- 

 tion of the different possible kinds of thermoscopes. 

 Then comes an extremely interesting discussion of the 

 merits of the different kinds of thermometers with 

 arbitrary scales. The defects of the mercury-in-glass 

 thermometer, and the advantages which led Regnault to 

 prefer the (constant volume) air-thermometer are fully 

 explained. We do not remember to have anywhere seen 

 so full, and, it is needless to say, so philosophical an 

 account of Regnault" s results of the comparison of the 

 different therniometric scales. The rest of the part 

 devoted to thermometry is more or less speculative. The 

 absolute thermodynamic scale of temperature, invented 

 by the author himself, is defined ; and its great advantage 

 pointed out, viz., that it gives us a definition of tempera- 

 ture " such that, if a thermometer were graduated accord- 

 ing to it from observation of one class of thermal effects in 

 any one particular substance, it would agree with a thermo- 

 meter graduated according to the same thermodynamic 

 law from the same class of effects in any other substance." 

 Thermodynamic formula; are investigated in a variety 

 of cases for graduating thermometers, according to the 

 absolute scale, from experimental data concerning the 

 thermometric substance. A number of instruments are 

 described in detail which are intended to realise these 

 cases in practice. We are thus introduced to the water 

 steam, mercury steam, and sulphurous acid steam thermo- 

 meters, and the constant pressure hydrogen thermometer. 

 These instruments are mostly new as to their details, 

 and all of them are new in the sense that they have not 

 been practically used hitherto. Nevertheless a great 

 future is predicted for them. It would appear that Sir 

 William Thomson has himself constructed models of 

 them all ; but whether he has used any of them in 

 practical work he does not say. It has doubtless occurred 

 to many of our readers, as it has to us, to have doubts 

 and difficulties about thermometric measurements. No- 

 where could we find better reasons for our scepticism 

 than in the earlier part of Sir William Thomson's discus- 

 sion of the systems of thermometry at present in use ; 

 we shall look, therefore, with all the greater interest for 

 some farther account of the practical working of these 

 new instruments. Their success, were it even but 

 partial, would be an immense gain to thermal science. 



Thermal capacity and specific heat are next defined; 

 and a brief account of the leading features of the results 

 of different experimenters is given, without detail as 

 to the methods employed in obtaining them. For 

 further information we are referred to the articles on 

 •'Theimodynamics," "Matter," "Liquid," "Steam." 

 The remaining five pages of the article deal with the 

 transference of heat. Radiation is explained and dis- 

 tinguished from other modes of transference ; but to our 

 great regret is dismissed very briefly. A criticism of the 

 work of the various experimenters in this department 

 from an authority like Sir William Thomson would have 

 been most interesting. There is still much doubt and 

 difficulty hanging over the subject of the diathermancy of 

 gases, for instance ; we need scarcely mention as an 

 illustration the famous controversy which has raged over 

 water vapour. It may be however that these and 

 kindred matters are to be treated under " Radiation " or 

 " Light" ; although we are not referred to these articles. 

 The general principles of the theory of the conduction 

 or diffusion of heat, as laid down by Fourier, are ex- 

 plained ; and a most interesting critical account is given 

 of the earlier attempts to measure conductivity. The 

 explanation of the causes of the failure of Cldment and 

 Feclet in measuring high conductivities, such as that 

 of copper, is very instructive, and should be closely 

 studied by those engaged in like researches. Of the 

 methods in use for measuring the thermal conduc- 

 tivity (or dift'usivity as the case may be) of solids, Sir 

 William Thomson prefers that of Angstrom, and recom- 

 mends along with it the use of thermoelectric methods 

 for determining the temperatures along the experimental 

 bar. The mathematical theory of this method is given, 

 and its connection with the researches of Forbes and 

 Thomson on underground temperatures pointed out. 

 The method of Forbes for measuring the conductivities 

 of metals in absolute measure is described in general 

 terms ; and the results obtained with it by Tait are 

 given, and compared with those of Angstrom and Thalen. 

 We regret that no mention is made of the recent 

 attempts to measure the conductivities of liquids and 

 gases. The only result given is that of Bottomley for 

 water, and no description of the method accompanies 

 it. It is quite true that the success of many of these 

 attempts has been somewhat doubtful ; but, for that 

 very reason, a criticism of the methods by a com- 

 petent and impartial authority would have been most 

 opportune, and useful as a guide to future experimenters. 

 Appended to the article are a series of ten tables of 

 thermal constants, and a reasoned synopsis of the prin- 

 cipal mathematical formulae that occur in the theory of 

 diffusion. This last is a most valuable part of the 

 article ; for it could be given only by a master of the 

 subject, and it is likely to be extremely useful to many 

 physicists, who have sufficient knowledge to enable them 

 to use such formula;, but not sufficient mathematical 

 power to find them for themselves, or sufficient time to 

 hunt them up from ponderous treatises and half-forgotten 

 memoirs when they want them. 



We sincerely congratulate the editor and publishers of 

 the Einyc/opcrdia on the high degree of success which 

 continues to attend their great undertaking. 



