446 



NATURE 



[March n, 1880 



bodies the molecules are swinging to and fro about 

 positions of equilibrium ; that " heat " is the energy of 

 these molecular vibrations ; and that the " temperature " 

 of the body is the mean amplitude of the vibrations. If 

 more energy is imparted to a solid, the more energetically 

 will its particles oscillate, the longer will be the mean 

 amplitude of their oscillations, and the higher the tem- 

 perature. If we allow that the gravitation law of attrac- 

 tion, namely that the attraction between two masses varies 

 inversely as the square of the distance between them, 

 holds good not only on the grandest scale but also on the 

 most minute, we must admit that the force acting on a 

 vibrating particle at the furthest limits of its swing, and 

 tending to attract it back, will be relatively weak as the 

 amplitude of the swing is great. Hence too long a 

 vibration may carry the particle right beyond the field of 

 molecular attraction ; and the particle will not return but 

 will carry off with it in the form of potential energy part 

 of the heat furnished to the body. The sum of these 

 small quantities of potential energy which must necessarily 

 disappear from the body during its change of state from 

 the solid to the liquid condition constitute that which we 

 usually term " latent heat." 



Now consider a solid body at the absolute zero of 

 temperature to which new quantities of heat are con- 

 tinuously imparted. What will be the successive changes 

 to be observed ? At first the temperature of the body will 

 rise proportionately to the quantity of heat imparted to 

 it When the vibrations of the particles have attained 

 a certain amplitude, fusion will take place, not all at once 

 but gradually, each molecule passing away from the 

 attraction of its neighbours, as soon as its vibration is 

 sufficiently energetic. Each solid particle will thus be 

 split up into two or more liquid molecules exactly resem- 

 bling each other. Every one of these molecules will 

 require potential energy, hence during the entire process 

 of liquefaction, the whole of the heat imparted will be 

 employed in producing the change of state ; so that the 

 temperature will be stationary in spite of the continual 

 addition of heat. But when the whole substance has 

 melted, the temperature will again rise up to a certain 

 point determined by the commencement of ebullition, a 

 point which will vary with the conditions of external 

 pressure. This second change of state arises from a 

 further splitting up of the molecules into two or more 

 portions each, every separated portion again carrying off 

 with it a further quantity of potential energy, the "latent 

 heat" of vaporisation. If the gaseous molecules thus 

 produced receive still further quantities of heat, the 

 temperature will go on rising until another point is 

 reached, corresponding to a first chemical dissociation, 

 when, as the lengths of oscillations become excessive, the 

 separate atoms are successively thrown apart. This 

 process, like those of liquefaction and vaporisation, will 

 be accompanied by the absorption of heat. The extent 

 to which energy must be furnished in order thus to 

 produce chemical separation, will be proportional to the 

 chemical affinity of the separated atoms ; and if the body 

 consists of several chemical constituents it is probable 

 that some of these will be dissociated at lower tempera- 

 tures and some at higher. The limits of dissociation will 

 have been reached when the body has been separated 

 into its ultimate particles or true elements. 



The striking feature of this series of changes is that 

 while the addition of quantities of heat goes on continu- 

 ously, the rise of temperature is discontinuous, having 

 several stationary points in the range between the absolute 

 zero and the highest possible temperature ; each fresh 

 stationary point corresponding to a change of state, or 

 a decomposition of the particles into simpler forms. 



Suppose next that we could reverse the order of 

 operations, and could abstract the heat continuously from 

 the dissociated bodies, we might expect to find the same 

 series of changes occurring in the inverse order. But 



this expectation would not be realized, for reasons which 

 are not difficult to find. In the two changes of state 

 which are of a nature usually termed physical changes^ 

 namely liquefaction and vaporisation, the result of the 

 splitting up is to produce particles all of the same kind. 

 In a liquid — water, for example — all the liquid particles 

 are water. In a vapour — steam, for example — the particles 

 are all particles of steam. But in the case of dissociation T 

 which is a chemical change of state, the result of the 

 splitting up is to produce particles not all of the same kind. 

 Thus, if steam is passed through a white hot platinum 

 tube, the dissociated matters are of two kinds, oxygen 

 particles and hydrogen particles. In the changes denom- 

 inated " physical " which produce homogeneous particles, 

 the recombination does not depend on the relative positions 

 of the constituents but only on pressure and temperature. 

 In the changes denominated "chemical" which, as we 

 have seen, produce heterogeneous particles, the recombi- 

 nation of the constituents depends on their relative 

 positions and on the way in which they have to be grouped 

 in the compound, as well as on pressure and temperature. 

 This most important distinction must not be overlooked. 



Again, the dissociated chemical atoms carry away with 

 them in a potential form the heat which has disappeared 

 during the process of dissociation, exactly as a liquid 

 carries in a potential form the "latent heat" which 

 disappeared during the process of liquefaction. If we 

 collect the separated chemical constituents — the oxygen 

 and hydrogen for example — and make them recombine, 

 they will evolve this potential energy and the heat will 

 reappear. The limit of temperature, therefore, which can 

 possibly be reached by the combustion or chemical 

 combination of any bodies is precisely the temperature of 

 the dissociation point of the substances formed. Hence 

 there is obviously, as we remarked at the outset, a limit 

 to the power of the chemist to dissociate bodies ; a limit 

 determined simply by the temperatures he can artificially 

 produce. 



It will be remarked, however, that we have in the 

 electric current a means of obtaining many decompositions 

 which without its aid would have been unknown to us. 

 We may even assert upon the certain evidence of the 

 spectroscope that the temperatures attained by the electric 

 spark are far higher than those of any known combustion. 

 Nevertheless there are here also limits which cannot be 

 passed. If in the circuit of the most powerful battery 

 we interpose a conductor of considerable resistance its 

 temperature will rise ; and, if the conductor be reduced 

 in thickness to augment its resistance, will continue to rise 

 until the conductor itself is either liquefied, volatilised or 

 dissociated, when of necessity a practical limit is reached 

 in the entire stoppage of the current. Again, with the 

 discharges from induction coils and Leyden jars, which 

 take place even across gases, there must be a limit, 

 determined by the absorption of energy by the very 

 molecules which are concerned in the discharge, and 

 whose resistance to the electrical action will increase with 

 their temperature. It is a point which may admit of some 

 further discussion. But, on the whole, one is led to the 

 conclusion that the dissociations we have shown to be 

 theoretically possible are in a very large number of cases 

 absolutely beyond the practical limits of experimental 

 achievement. 



One course yet remains open. We have not hitherto 

 considered the connection between temperature and 

 radiation in its bearings upon this question. It appears 

 that every temperature, as defined above, corresponds to a 

 definite kind of radiation. Every calorific oscillation of a 

 particular rate is then associated with the propagation of a 

 wave of disturbance in the surrounding ether ; this wave 

 having a particular frequency, or, what is the same thing, 

 a particular wave-length. When these calorific waves in 

 passing through space meet a body they tend to set its 

 particles vibrating ; and, what is more important, tend to> 



