310 SIR F. W. DYSON, PROF. A. S. EDDINGTON AND MR. C. DAVIDSON ON A 



the stars shown were measured. They were reduced by the same method as that 

 employed for the " 4-inch " photographs. With the exception of plates Nos. 15 and 

 16, taken through clouds, the stars numbered 3, 4, 5, 6, 10, 11 and 12 are shown on all 

 the plates ; the fainter stars 2, 7, 8 and 9 are sometimes shown, but No. 1, which is 

 very near the sun, is always drowned in the corona. These plates were only measured 

 in declination, as the right ascensions were of little weight. 



24. In the following table is given the value of a, the constant of the gravitational 

 displacement, as calculated from the measures ; the apparent difference of scale e 

 between the eclipse and comparison plates ; d the difference of orientation of the plates 

 given by the measures of y, and depending on the adjustment of the plates in the 

 measuring machine. 



TABLE IX. 

 (l r 12"- 3). 



Thus the mean value of a obtained from all the astrographic plates is 0"-86, a figure 

 considerably less than that obtained from the 4-inch photographs. 



25. Reference to the diagram shows that the measurement of displacement depends 

 essentially on the position of the stars Nos. 3 and 4 relative to 5 on one side and 6 

 and 10 on the other. These are all bright stars, and in this respect their images are 



