84 SCIENCE PROGRESS 



individual chemical and physical properties — to depend largely 

 — if not entirely — upon the number and distribution of the 

 electrons making them up. The increased intellectual re- 

 spectability of the structural formulae of chemistry has reacted 

 upon physiology, and most physiologists to-day would be pre- 

 pared to believe in a spatial distribution of the parts or processes 

 of the living cell ; the days of a homogeneous, undifferentiated 

 protoplasm with certain mysterious labels called " properties " 

 are gone. Clearly, therefore, the spatial characteristics of the 

 processes of a living cell are related to the spatial characteristics 

 of the atom, molecule, and particles concerned, and will be 

 understood only when the latter are understood, and when the 

 methods by which the latter were discovered are applied to the 

 living cell. The living cell has still many secrets to yield to the 

 older mass-methods of physics and chemistry, but the most 

 fundamental advance waits for means which can take account 

 of the distribution in time and space of the parts and events 

 concerned in vital processes. At present one is forced to infer 

 the mechanism by analogy from mechanical systems, showing 

 the same general physical features. 



The laws of physics come under two main categories, those 

 which are true — within limits — of all systems, large or small, 

 and those which are statistical in nature and true only of 

 systems large enough to allow the laws of aggregates to apply. 

 Many of the most fertile generalisations of physics, such as the 

 Second Law of Thermodynamics, the Gas Laws, the Laws of 

 Osmotic Pressure, and the Laws of Mass Action, are of the 

 latter restricted kind, applicable only to systems where the 

 number of parts is so large that the laws of averages work out 

 correctly. A particle showing Brownian movement is already 

 beginning to be too small to obey these statistical laws, and 

 the whole system of co-ordinated mechanisms of many living 

 cells is confined within a space no larger than that of a particle 

 showing vigorous Brownian movements ; we may well doubt 

 whether the laws of averages will always work out correctly 

 in such a system. It is possible therefore, indeed probable, 

 that, in the organised system of minute mechanisms making 

 up the living cell, such statistical laws will not, in the last 

 resort, be found to hold. The laws of mass action apply cer- 

 tainly to the case of a ferment action occurring in a large 

 homogeneous system : the number of working parts is so large 

 that the statistical rule works out correctly. But in the living 

 cell, where a single ferment molecule or particle may be an 

 essential link in themechanism devoted to some specificfunction, 

 there is no a priori reason why such a statistical rule should 

 hold at all. Indeed, it would seem likely that an explanation 

 of the behaviour of the living cell will finally have to rest on 



