602 



Scientific Proceedings, Royal Dublin Society. 



thesis of uniformity were approximately true, we should arrive at 

 the same result by both modes of computation ; but if light is ab- 

 sorbed by the medium, the value derived from the total light should 

 always be less than that derived from the total number of stars. I 

 accordingly tried this method on the stars from the first to the 

 ninth magnitude inclusive, taking the numbers of the stars from 

 Dr. Ball's Table, and computing the total light on the assump- 

 tion that the average light was represented by the multiplier f at 

 each stage of the descent. The first pair of figures shows a wide 

 divergence in the opposite direction from that which would be pro- 

 duced by the absorption of light, which I attribute to the great 

 richness of the region of the second magnitude stars, our field not 

 being yet wide enough to render the uniformity-hypothesis approxi- 

 mately true. At the next step the divergence is in the same direc- 

 tion, but much reduced ; at the third stage the figures are practically 

 equal ; and in the remaining five the figure derived from the num- 

 ber of stars is considerably in excess of that derived from the total 

 quantity of light, both figures displaying a good deal of steadiness, 

 as appears by the following Table : — 



The divergence in one direction, at the earlier stages, does not 

 compensate for the opposite divergence at the later ones. The 

 multiplier which represents the total change from the first to the 



