THE MILKY WAY. 261 



a certain period the cluster would return to its original 

 configuration, and that configuration would be stable. 

 Unfortunately the clusters do not appear homogene- 

 ous. We observe a condensation at the centre, and 

 we should still observe it even though the sphere were 

 homogeneous, since it is thicker at the centre, but it 

 would not be so marked. A cluster may, therefore, 

 better be compared to a gas in adiabatic equilibrium 

 which assumes a spherical form, because that is the 

 figure of equilibrium of a gaseous mass. 



But, you will say, these clusters are much smaller 

 than the Milky Way, of which it is even probable that 

 they form a part, and although they are denser, they 

 give us rather something analogous to radiant matter. 

 Now, gases only arrive at their adiabatic equilibrium 

 in consequence of innumerable collisions of the mole- 

 cules. We might perhaps find a method of reconciling 

 these facts. Suppose the stars of the cluster have just 

 sufficient energy for their velocity to become nil when 

 they reach the surface. Then they may traverse the 

 cluster without a collision, but on reaching the surface 

 they turn back and traverse it again. After traversing 

 it a great number of times, they end by being deflected 

 by a collision. Under these conditions we should still 

 have a matter that might be regarded as gaseous. If 

 by chance there were stars in the cluster with greater 

 velocities, they have long since emerged from it, and 

 have left it never to return. For all these reasons it 

 would be interesting to examine the known clusters 

 and try to get an idea of the law of their densities and 

 see if it is the adiabatic law of gases. 



But to return to the Milky Way. It is not spherical, 

 and would be more properly represented as a flattened 



