Modified Theory of Gravitation. 87 



the suspended system turns about a vertical axis. Now in 

 virtue of the earth's orbital motion, the highest value attained 

 by the velocity v must be at least as great as 32 kilometres 

 per second ( = 3*2 x 10 6 cm. per second), so that by (33) our 

 suspended system, when most favourably placed and oriented, 

 could experience a couple as great as 



1(3-2 xl0 6 ) 2 K = 1-28 xl0 12 K. . . . (34) 



Also the restoring moment when the system is displaced 

 through one second of arc about a vertical axis is 4*4 X 10^ 7 

 dyne-cm.; and for the experiment to give a null result, we 

 should accordingly have 



K > 3*4 x 10" 19 in c.G.s. measure. . . (35) 



Assuming, in advance, that a null result would be obtained 

 in this case, the value which, in the sequel, will be provisionally 

 assumed is 



K = 2xl0- 20 (36) 



25. This tentative value of K may be compared with one 

 suggested by another way of regarding the question. If the 

 negative electron could be regarded as having a vacuous 

 spherical nucleus of radius 10~ 13 cm., and if we assume the 

 atom of hydrogen to contain only one negative electron, 

 while^the number of molecules in 1 c.c. of gas at 0° C. and 

 1 atmosphere pressure is taken to be 3*9 x 10 19 ; then it 

 appears that, on the score of negative electrons alone, F, the 

 extrusion of aether per unit mass of matter (§ 7), would be 

 4 x 10~* 5 c.c. of aether per gram of matter ; equivalent to 

 •004 gram of aether per gram of matter, when an aetherial 

 density of 10 12 is assumed*. Though the existence of a 

 vacuous nucleus in an electron does not appear reconcilable 

 with some of the views adopted in this paper, it may be 

 worth while to remark that the value of F just mentioned 

 leads to K=4x 10 -18 , which is 200 times as great as the 

 estimate (36), and about 12 times as great as might just 

 suffice to give positive results in the hypothetical experiment 

 of § 24. 



Conjectural Estimates. 



26. The values of a further number of physical magnitudes 

 will have to be conjectured before a quantitative view of the 

 theory can be attempted. Some of these magnitudes are 



* Cf. § 35 below. 



