RECENT ADVANCES IN SCIENCE 



37i 



the time : the region near the sun was partially obscured by- 

 thin cloud during the whole of totality at Principe and during 

 a part at Sobral. Consequently, the plates obtained at Sobral 

 show more stars than those obtained at Principe and give a 

 better determination of the deflection at the limb. 



The astrographic object glasses of the Greenwich and 

 Oxford Observatories were used respectively at Sobral and 

 Principe in conjunction with 16-inch ccelostats : in addition, 

 a 4-inch lens of 19 feet focal length was used at Sobral with an 

 8 inch ccelostat. 



The results obtained are given in a paper by the Astronomer 

 Royal and the observers, read at a joint meeting of the Royal 

 and Royal Astronomical Societies on November 6, 1919. 

 They may be summarised as follows : 



(1) The best results were obtained from the long-focus 

 4-inch lens at Sobral, which gave images of good definition. 

 The photographs were measured against other photographs 

 of the same region taken at the same altitude from the same 

 station a few months subsequent to the eclipse, when the sun 

 had moved away. A reliable determination of scale was in 

 this way obtained, the scale-value naturally being of funda- 

 mental importance. The mean displacements of the seven 

 brightest stars actually obtained, and those to be expected on 

 Einstein's theory, are given in the following table : 



A very satisfactory accordance in both co-ordinates is thus 

 obtained. 



(2) The images on the astrographic plates taken at Sobral 

 were diffuse, this being attributed to a change in figure of the 

 ccelostat minor ; there was a rather large discordance between 

 the mean results from the individual plates, but from the whole 

 series an outward deflection at the limb of o"-93 was obtained. 



(3) The plates taken at Principe showed only a few stars 

 and the scale could not be directly determined. Plates of an- 

 other region of the sky taken at the same altitude were used, 

 and compared with plates of the same region and of the eclipse 

 field obtained previously at Oxford. The determination of 

 scale was therefore somewhat weak, though the uniformity of 



