504 Scientific Proceedings, Royal Dublin Society. 



given a reason for thinking this figure too high, but the two re- 

 sults could evidently be reconciled by a large absorption. On 

 that view the loss of light would be so considerable that a star 

 should be placed at twenty-six times the distance of a first magni- 

 tude star, in order that the light which reached us should be that 

 due to thirty-nine times the distance, none being supposed to be 

 lost. The absorption would thus amount to f of the total light. 

 Probably the absorption is not so large as this, but if we halved 

 the difference between the two results the loss would still amount 

 to at least 36 per cent. If there is an absorption therefore a faint 

 star will always be nearer to us than we would infer from the 

 intensity of its light, on the hypothesis of uniformity. How much 

 nearer will of course depend on the amount of the absorption. 



