178 ANNUAL REPORT SMITHSONIAN INSTITUTION, 1920. 



his return from Germany through England that Russell showed us 

 his stepladder diagram at the Royal Astronomical Society, and 

 expounded his views on the evolution of a star, which were that 

 its life began at the foot of the upright leg, the ascent of which 

 signified that the star was growing continually hotter and changing 

 its spectral type meantime from M upward toward B, that at B 

 the increase of temperature was arrested and after a time cooling 

 began, carrying the star down the inclined leg of the ladder through 

 changes in the reverse order. The only weak spot in the evidence 

 arose from the small number of observations. To determine the 

 actual or intrinsic brightness of a star we must know, its distance, 

 and there are not many stars of which the distance can be easily 

 measured, and though Russell had himself increased the number", the 

 total was still not large. To get further evidence he had recourse to 

 indirect estimates of distance, especially those of clusters of stars. 

 We have lately become more and more aware of the association of 

 stars in clusters represented by their common movement, somewhat 

 in the way that the movements of a flock of birds migrating from 

 one place to another are associated. If we may accept this evidence, 

 and if we can determine the distance of any one star in the cluster, 

 the distances of the others can be inferred. In Russell's skilful hands 

 this evidence was collated and found to strengthen his conclusions. 



Let us pause here for a moment to reflect on the inherent proba- 

 bility of the suggestion. Is it not after all much more likely that a 

 star first rises in temperature and then falls rather than that it 

 should be permanently either rising or falling? Now that the idea 

 has been put forward, and that there seems to be not only good 

 evidence of this change in the sky, but, as we shall presently see, also 

 good theoretical reason for it, we wonder why the idea was not the 

 most natural one to adopt from the first. But curiously enough it 

 was not the one adopted by astronomers, with the notable excep- 

 tion of Sir Norman Lockyer, who made the same suggestion as 

 Russell's (though on different grounds) many years before. May I 

 give a crude illustration from our ordinary life of the mistake that 

 was made by many of us ? It is as though we had taken the amount 

 of hair .as an indication of the age of a man. In very early life 

 the amount of hair is small, it increases with age up to a certain 

 point, but then it begins to decrease until a very old man often has 

 as little hair as a new-born baby. We could give Shakespeare's 

 seven ages of man according to the amount of hair in the same 

 diagrammatic form as Russell's stepladder, beginning with the baby 

 at the foot of the upright leg, ascending to the man in the middle life 

 with maximum hair (corresponding to the maximum temperature), 

 and placing the greater ages down the inclined leg till we arrive 



