294 Obituary Notices of Fellows deceased. 



hypotheses concerning the constitution of material bodies, we pursue 

 statistical enquiries as a branch of rational mechanics. In the present 

 state of science it seems hardly possible to frame a dynamic theory of 

 molecular action which shall embrace the phenomena of thermo- 

 dynamics, of radiation, and of the electrical manifestations which 

 accompany the union of atoms. Yet any theory is obviously inadequate 



which does not take account of all these phenomena Certainly 



one is building on an insecure foundation who rests his work on 

 hypotheses concerning the constitution of matter." These remarks > 

 and others of the same tenour, in the preface to the treatise above 

 mentioned, coincide with a general tendency in scientific exposition 

 which is now prominent. But it may be doubted whether there are 

 many who have pondered much on special physical theories, who have 

 not perforce realized the vastness of Nature, so as not to require any 

 reminder that they are merely following out analogies in one aspect of 

 its immense but rational scheme, or improving in one direction our 

 mental outlook on its operations. If we are dissuaded from framing 

 any dynamic theory of molecular action, we shall certainly not progress 

 in welding together the regions of phenomena which Prof. Gibbs 

 enumerates ; while to establish, or at any rate trace, their inter- 

 connexions, the unattainable ideal of complete knowledge is not 

 required, provided our hypotheses are held in a state of suspense so 

 as not to force us permanently in a wrong direction. And, moreover, 

 recent indications hardly bear out the hopelessness of direct 

 physical speculation in this very subject. The course of progress 

 has rather been usually an evolution from the special simplified 

 theory to the general formal scheme of relationship ; an analogy with 

 simple phenomena already understood suggests a special type of 

 hypothesis so to speak, a calculus adopting that theory as its 

 notation or mode of expression ; this is worked out and tested on the 

 facts; it is thus improved and generalised by dropping unessential 

 restrictions it may ultimately become strong and vivid enough to 

 drop all analogies, and emerge as a purified scheme of relations 

 embracing the sensible phenomena without requiring any forms of 

 expression that are not directly inherent in themselves. Such a 

 scheme of general relations is an improvement on the special analogical 

 theory, provided it does not become too abstract and intangible ; but 

 experience may be held to suggest that essential relations restricting 

 the excessive number of the variables in purely descriptive mathematical 

 formulations of experience are, in the first instance, suggested through 

 the analogies of simpler systems. This other point of view may in 

 fact also be put in Prof. Gibbs' own words relating to molecular 

 theory, again quoting from his notice (1889) of the work of Clausius. 

 " The origin of the kinetic theory of gases is lost in remote 



