102 



C. V. L. Charlier 



The agreement between the values of a and 1 : ]/^ is very good with excep- 

 tion for the stars of type M. These stars seem to have a larger mass~than that 

 deduced from formula (101), if the deviation is not to be explained in other way. 



The variation in the velocities of the stars with the spectral type is usually 

 formulated in such a way that it is said that »the velocity of the stars is increasing 

 with the age». If the explanation given above holds good '), the velocity is not 

 directly correlated with the age but with the mass of the stars. The correlation 

 found between age — it would be more logical to say temperature — • and velocity 

 had to be considered as a secondary phsenomenon, being only an expression for 

 the fact that the younger (hotter) stars have, upon an average, a larger mass than 

 the elder (cooler) ones. Let us suppose, for a moment, that all stars had been 

 » created* at the same time out of a nebulous chaos, so that all stars would have 

 the same age, then it might happen that, after the lapse of some time, the largest 

 masses still retain their high temperature and appear as stars of type B, whereas 

 the smallest masses have already passed through all stages of their evolution from 

 B to M. 



It must be remarked that the relation of Maxwell between mean velocities 

 and masses, obviously, does not hold good for the very smallest masses such as 

 those of the comets or the meteorites^). Whether this remarkable fact shall be ex- 

 plained in such a way that the Milk}'- Way has not still reached a statistical equi- 

 librium, or whether the statistical mechanical laws tend to associate the very smallest 

 masses to the larger members of the stellar system, are questions that do not belong 

 to the line of research of this memoir. 



^ When these lines were in print, I have found that this explanation of the small velocity 

 of the Orion-stars has already been proposed by Halm (M. N. 1911) and that Eddington in his 

 interesting exposé on >the stellar distribution and movements* at the Meeeting of the British 

 Association in Portsmouth 1911 advances certain objections against this hypothesis. 



^) This has been remarked by Poincaeé in his » Leçons sur les hypothèses cosmogoniques» 

 P. 264. 



