n6 Presidential Address 



be expected that it would be split up into so many equal 

 fractions that the translatory energy of the molecules 

 i.e. the motion upon which "temperature" depends 

 would be increased to only an insignificant extent. But it 

 is not so. Experiment on the velocity of sound through 

 various gases shows that the energy imparted by acoustic 

 compression raises the temperature quite decidedly, and is 

 therefore only divided into quite a moderate number of 

 parcels. The number is found to be 3 for monatomic 

 molecules, which therefore to that extent behave like 

 smooth spheres; 5 for diatomic molecules, which therefore 

 behave like smooth dumb-bells; and 6 for molecules con- 

 sisting of 3 or more atoms, which therefore behave like the 

 general rigid body. 



How it can come about that vibratory degrees of free- 

 dom are excluded from the sharing, even after the lapse of 

 some time, has long been a subject of controversy; and the 

 fact that it is so has thrown doubt upon the equi-partition 

 theorem itself. The answer suggested by Professor Planck, 

 the eminent physicist of Berlin, is that fractions of a unit 

 of energy cannot exist, so that the attempt to subdivide a 

 given portion of energy into an immense number of frac- 

 tions must fail; only those degrees of freedom that can re- 

 ceive a whole unit can be effective: it is a case of all or none, 

 and those subsidiary modes of motion that can only receive 

 a fraction will not get any. 



This is an exceedingly crude and partial account of the 

 initial stages of a large and complex subject, which has 

 been highly developed and is arousing much interest. The 

 unit of energy called "Planck's constant "enters into very 

 many parts of Physics, and appears to be a reality, how- 

 ever it be accounted for; but I consider that the explana- 

 tion is to be found in specific properties of matter, and not 

 in the really atomic character of energy itself. 



The fact that vibratory molecular energies are not ef- 

 fective in an ordinary gas, are not evoked by ordinary 



