DEVELOPMENTS IN PHILOSOPHY OF BIOLOGY 



83 



thermodynamics are no longer applicable 

 to them. Such laws must be replaced by 

 the statistical theory of molecular fluctua- 

 tion and in the last resort by the theory of 

 individual action. 



It must be noted that to admit this is not 

 to go so far as Johnstone and Lewis, who 

 assert the existence of disentropic phases 

 within the living cell. The question may, 

 moreover, be approached in another way. 

 If a machine, living or otherwise, is per- 

 fect it should yield up the same amount 

 of work as is put into it, in other words 

 it should waste nothing of the driving 

 force supplied to it. The ratio between 

 the energy put in and the energy got out 

 may be called the ' 'free energy efficiency' ' 

 of the process, and obviously for a perfect 

 machine this should be unity or 100 per 

 cent. For a system containing an imagi- 

 nary being such as Maxwell's demon, 

 capable of defeating the second law of 

 thermodynamics by causing a decrease in 

 entropy the efficiency would be greater 

 than 100 per cent. For all known 

 machines it is less than 100 per cent. The 

 parallel between the numerical figure for 

 the efficiency of living and non-living 

 machines is often very close. To compare 

 the animal to a heat-engine, is, as Guye 

 points out, misleading, for the energy of 

 the steam engine is changed in form as it 

 passes through u, but the comparison 

 between the growing animal and a boiler 

 or an electric battery is much closer. The 

 "apparent energetic efficiency" of the 

 developing embryos of the chick, the 

 silkworm, the minnow, and the frog has 

 recently been calculated (Needham (4Z)) 

 and works out on the average to 77 per 

 cent. This means that out of 100 gram 

 calories presented to the organism in the 

 form of yolk and white 1^ have to be 

 wasted and 77 can be stored. Similarly, 

 the business of the electric accumulator is 

 to store energy and this cannot be done 



with an efficiency greater than 75 per 

 cent (Davidge and Hutchinson (10)). The 

 boiler, again, is designed to transfer as 

 much heat as possible from coal to steam, 

 but this can never be done with an effi- 

 ciency greater than 77 per cent (Brownlie 

 (4)). It is doubtful, however, how much 

 stress can be laid on any numerical corre- 

 spondence between the efficiency of the 

 growing embryo and such non-living 

 machines. 



What of course would show up the 

 presence of the Maxwell demon in living 

 matter would be an efficiency greater than 

 100 per cent. Such an efficiency has never 

 been reported for any of the higher ani- 

 mals under any conditions, but it might 

 be thought that the place to look for it 

 would be the iron and sulphur bacteria, 

 maintaining as they do a precarious 

 existence on the most unpromising food- 

 stuffs. This problem has recently been 

 considered by Baas-Becking and Parks 

 (1). They have calculated the free energy 

 efficiencies of many forms of autotrophic 

 bacteria, and though here, if anywhere, 

 living matter might be expected to need 

 demonic assistance, the efficiencies were 

 always very low. The hydrogen bacteria 

 operated at an efficiency of x6 per cent, the 

 methane bacteria at 30 per cent, the nitrite- 

 forming bacteria at 6 per cent. As for the 

 sulphur bacteria the figure obtained was 

 8 per cent, and for the iron bacteria (sup- 

 posing that such organisms really exist) 

 it would also be very low. These ob- 

 servations certainly seem to be in favor of 

 the views of Hill and against the possi- 

 bility indicated by Donnan and assumed 

 to hold in practice by Johnstone, Lewis, 

 and Driesch. But it must be remembered 

 that an appeal to colonies of bacteria 

 gives no less a statistical result than the 

 appeal to Atwaters' calorimeter. Refer- 

 ence must here be made to a very important 

 recent paper of R. S. Lillie's (34). Lillie 



