Mr. M. M. P. Muir on Chemical Notation. 15 



must be smaller than the average velocity of the molecules in 

 a gas ; nevertheless this does not prevent us from believing 

 that the velocity of individual liquid molecules may be greater 

 than that of individual gaseous molecules. Supposing that 

 liquid molecules of such great velocities are moving at the 

 surface of the liquid from the liquid, they will tend to escape 

 from the liquid into the surrounding vapour. Supposing that 

 a gaseous molecule of great velocity strikes the surface of the 

 liquid, it will tend to become entangled in the liquid. Hence 

 we shall have evaporation of the liquid and condensation of 

 the vapour going on continuously. 



30. So also I might go on to show that many of the facts of 

 spectroscopy are in keeping with the molecular theory — that 

 the vibration of each molecule may well be supposed to com- 

 municate itself to the ether in a regular manner, that an in- 

 crease in density of a gas will cause the introduction of irre- 

 gular motion among the molecules, and hence the appearance 

 of a more or less continuous spectrum ; but, for a considera- 

 tion of the bearings of the molecular theory of matter upon 

 the explanation of spectroscopic phenomena, I must refer to a 

 lecture by Mr. J. Norman Lockyer, reported in ' Nature,' vol. 

 x. p. 69.' 



31. In those points which I have briefly touched on, the 

 molecular theory is, it appears to me, not only in accord with 

 known facts, but it also enables us to group together and so to 

 explain what had appeared to be isolated phenomena. But 

 this theory does more than this: we can deduce from it certain 

 conclusions which may then be proved true by experiments. 

 Thus the law of Boyle, the law of Charles, and the law of Gay- 

 Lussac may be deduced from a consideration of the dynamical 

 conditions which must prevail among a group of particles sup- 

 posed to constitute the mass of a gas, the pressure and tempe- 

 rature of which vary from time to time. 



32. It has been possible to predict certain results from a 

 consideration of phenomena viewed in the light of the mole- 

 cular theory, which results have afterwards been verified. 

 Thus the rate at which diffusion of energy must take place 

 through air when one portion is heated was predicted from 

 data derived from experiments on viscosity. The actual mea- 

 surement of the diffusion-rate, made by Prof. Stefan, of 

 Vienna, has proved to be in close agreement with the pre- 

 dicted rate. 



33. If we interpret chemical notation in the light of the 

 molecular theory, we assume that the symbol of each chemical 

 substance represents a molecule of that substance, and that 

 chemical equations represent the results of the mutual actions 



