RECENT ADVANCES IN SCIENCE 273 



rolling friction of one link upon another. The attracting 

 masses were large leaden weights, which contained a heating 

 coil of nichrome for raising their temperatures. 



Reference should be made to the original paper for an 

 account of all the details of the apparatus finally adopted as 

 the result of experiments extending over several years. The 

 final series of experiments undoubtedly point to an increase in 

 the Newtonian constant with temperature, for no point seems 

 to have been left unexamined which might impair the validity 

 of this conclusion. 



The results obtained are so unexpected and the magnitude 

 of the effect found is so large, that one naturally turns to con- 

 sider to what extent fundamental conceptions are concerned, 

 and whether any other evidence can be adduced to support or 

 disprove this result. It may first be mentioned that Prof. 

 J. H. Poynting and Mr. P. Phillips had previously made careful 

 investigations into this question (Proc. Roy. Soc. ser. A, vol. 

 Ixxvi. p. 445, 1905) with negative results. The method adopted 

 was to counterpoise a weight upon a balance and to vary its 

 temperature over a very wide range, the attracting mass being 

 the earth. The difference between these experiments and 

 those of Dr. Shaw is that in the one case the small (attracted) 

 mass, in the other the large (attracting) mass was heated. If 

 then the attraction between two masses M , m of respective 

 temperatures /, t' and distance apart d is of the form 



, n T , Mt + mt'~\ Mm 



the two results are not in opposition. Other forms for the 

 attraction can of course be found which will reconcile the 

 two series of experiments. Another mode of expressing the 

 results is to say that the attraction of one body by a second 

 depends upon the temperature of the attracting body only, 

 but not upon that of the attracted body. 



The astronomical aspect of the question has been discussed 

 in The Observatory, vol. xxxix. p. 318, 1916. It is known that 

 the planets have widely differing temperatures, and the accuracy 

 with which Kepler's laws for the motion of the planets around 

 the Sun are obeyed proves that the attraction is independent 

 of the temperature of the attracted (small) body, in agreement 

 with the result found by Poynting and Phillips. But it seems 



