RECENT ADVANCES IN SCIENCE 461 



constitute fewer than one per cent, of the whole cluster, and 

 that these stars are also the reddest stars and belong largely 

 to one spectral class. Their large luminosities may be associ- 

 ated with other peculiarities, such as peculiarities of mass or 

 of motion, which may affect the observed densities and give 

 fictitious results. 



He points out also the danger of systematic errors in 

 counting dependent upon the distance from the crowded centre 

 and the possibility also of errors due to the photographic 

 interaction of contiguous images, and error varying with the 

 distance from the centre and causing the limiting magnitude 

 correspondingly to vary. 



Shapley also remarks in reference to Jeans's theory, referred 

 to above, that observations have not as yet either proved or 

 disproved the existence of galactic planes in globular clusters. 

 The appearance of the few brightest stars is not sufficient to 

 decide, because these may be uniformly distributed, but the 

 fainter stars may be arranged in a belt. He finds some ten- 

 dency in 00 Centauri towards a cluster galaxy, but it is not 

 very pronounced. We await with interest further observa- 

 tions in this direction to see how they agree with theory, which, 

 for once, has outrun observation. 



PHYSICS. By James Rice, M.A., University, Liverpool. 



Relativity and Gravitation. — The first of the two organised 

 discussions arranged for the Mathematics and Physics Section 

 of the British Association was devoted to the subject of 

 Gravitation. 



A considerable part of the discussion was occupied with the 

 recent work of Einstein and Grossmann which attempts to 

 bring gravitation within the scope of the principle of relativity. 

 This latter principle, now widely accepted, denies the possi- 

 bility of determining absolute motion in space by means of any 

 physical phenomenon. The various attempts to determine 

 the velocity of the earth relative to an assumed immovable 

 ether and the consistent failures of these experiments to detect 

 such relative motion either by optical or electrical methods, have 

 accumulated a considerable body of evidence in favour of the 

 principle. 



In Einstein's earliest formulation of the principle of rela- 

 tivity (Ann. der Physik, xvii. 913 1905) he stated it in these 



