THE MILKY WAY. 267 



be no reason for their emerging from the planes nor 

 for the symmetry to be altered. This configuration 

 would accordingly give equilibrium, but it would be an 

 unstable equilibrium. 



If there is rotation on the contrary, we shall get 

 an analogous configuration of equilibrium with four 

 curved radii, equal to one another, and intersecting at 

 an angle of 90°, and if the rotation is sufficiently 

 rapid, this equilibrium may be stable. 



I am not in a position to speak more precisely. It 

 is enough for me to foreshadow the possibility that 

 these spiral forms may, perhaps, some day be ex- 

 plained by the help only of the law of gravitation and 

 statistical considerations, recalling those of the theory 

 of gases. 



What I have just said about internal currents shows 

 that there might be some interest in a systematic 

 study of the aggregate of the individual motions. 

 This might be undertaken a hundred years hence, 

 when the second edition of the astrographic chart of 

 the heavens is brought out and compared with the 

 first, the one that is being prepared at present. 



But I should wish, in conclusion, to call your 

 attention to the question of the age of the Milky Way 

 and the nebulae. We might form an idea of this age 

 if we obtained confirmation of what we have imagined 

 to be the case. This kind of statistical equilibrium of 

 which gases supply the model, cannot be established 

 except as a consequence of a great number of col- 

 lisions. If these collisions are rare, it can only be 

 produced after a very long time. If actually the 

 Milky Way (or at least the clusters that form par^" 

 of it), and if the nebula- have obtained this C(}uilil)riuin, 



