September 16, 1922] 



NA TURE 



589 



The Total Solar Eclipse of September 21. 

 By J»r. A. C. D. Crommelin. 



THERE are at present in the Saros Cycle two series 

 of eclipses which have unusual length of totality ; 

 one including those of 1865, 1883, 1901, 1919, the other 

 including the great Indian eclipse of 1868, in which the 

 spectroscope was first applied to the prominences ; 

 also those of 1886 (West Indies) and 1904 (Pacific). 

 The forthcoming eclipse, September 21, being three 

 Saroses after that of 1868. is in nearly the same longi- 

 tude, but has moved southward, the only land stations 

 available being the Maldives, Christmas Island, and 

 Australia. 



The Maldives have the disadvantages of a rather 

 low sun, some difficulty of access, owing to the reels 

 surrounding the islands, probability of high wind, 

 together with a poor health record for European 

 visitors : they are, however, being occupied by Mr. 

 Evershed. Christmas Island lies in the longitude of 

 maximum totality (6 minutes), but being near the 

 northern limit of totality it will enjoy only 3 J minutes. 

 This is, however, amply long enough for the programme 

 planned. The station is occupied by Messrs. Jones 

 and Melotte from Greenwich, their equipment con- 

 sisting of the 13-inch astrographic equatorial, on a 

 mounting specially constructed for the low latitude of 

 the station. When the same instrument was used in 

 Brazil in 1919 the star-images were diffused, owing 

 probably to slight warping of the ccelostat mirror by 

 the heat of the sun before totality. The unsuitability 

 of the ccelostat had been foreseen, but the short interval 

 between the armistice and the departure of the ex- 

 pedition made it impossible to provide an equatorial 

 mounting. 



On the present occasion it is desired to secure a 

 completely satisfactory check on the 1919 results ; 

 these tended to confirm the amount of shift of light 

 by the sun's gravitation predicted by Einstein; the 

 difference in the results given by the two instruments 

 in Brazil was, however, too large to permit the results 

 to be taken as absolutely final, and a further test is 

 desirable. The star-field at this totality is, unfortu- 

 nately, much less favourable than that in 1919, which 

 was probablv the field containing the largest number 

 of bright stars close to the ecliptic. There are, however, 

 a fair number of stars of the eighth magnitude or 

 brighter in the present field, and it is hoped that these 

 may be photographed with somewhat longer exposures 

 than those given before. The corona will probably 

 be of the " Minimum " type, with little extension near 

 the poles ; this should enable stars fairly near the sun, 



which will have a large factor of shift, to be 

 photographed. 



Christmas Island is occupied by a Phosphate Com- 

 pany, under Scottish management, which has given 

 great assistance to the expedition in transporting their 

 baggage, in erecting huts, providing workmen, etc. 

 Reports received in July stated that the adjustment 

 of the instruments was complete, but that the weather 

 during May had been very wet, and little observing 

 was possible ; check plates of the eclipse field had, 

 however, been secured. The rainy season was, how- 

 ever, nearly at an end, and it was hoped that more 

 work would shortly be possible ; in addition to the 

 eclipse programme it was planned to take a series of 

 photometric plates, to connect the magnitude scales 

 of the northern and southern hemispheres. Profs. 

 Freundlich and Einstein also arranged to observe from 

 Christmas Island, their programme being much the 

 same as that of the British observers. 



The station on the coast of West Australia has a 

 high sun, long totality, and excellent weather prospects ; 

 but it is difficult of access, it being necessary to anchor 

 some miles out, and land in small boats through surf. 

 Several parties are there ; that with the largest equip- 

 ment is from the Lick Observatory, under Prof. 

 Campbell. This party also makes the Einstein problem 

 the chief item of the programme. To avoid a long stay 

 at the eclipse camp the check plates were taken at Fiji 

 on the voyage out. Other parties at this station are ■ 

 from Canada and from Perth (Australia). 



The observatories of Adelaide, Melbourne, and 

 Sydney are sending expeditions to stations in Central 

 Australia and in Queensland. The weather prospects 

 are good at both, but the sun in Queensland is rather 

 low. They are understood to be attempting the 

 Einstein problem, in addition to the older eclipse work 

 of photography of the corona and its spectrum. 



There is every reason to hope for success at some 

 of the stations ; fine weather at all of them should lead 

 to results of a decisive character on the Einstein 

 problem. The results will not be available for some 

 time, as the plates will not be measured till the return 

 of the different parties. They will, however, be 

 developed, in situ, which will permit a good idea of 

 their character to be formed. In this connexion it 

 may be noted that there is no cable to Christmas 

 Island, but it is expected that a Dutch man-of-war 

 will be there, which might send a wireless message to 

 Java. 



The Deflection of Light in a Gravitational Field. 



Bv Herbert Dingle. 



FROM an experimental point of view, Einstein's 

 general theory of relativity is at present in an 

 ambiguous position. It is well known that there are 

 three conceivable tests between its conclusions and 

 those of the traditional ideas which it attempts to 

 displace. With regard to the first of these — the move- 

 ment of the perihelion position of Mercury — the success 



NO. 2759, VOL. I ioj 



of the theory is decidedly impressive ; all the more 

 so, perhaps, because the result was stumbled upon, 

 as it were, involuntarily. In seeking first the gravita- 

 tional field of the sun, Einstein found the true orbit 

 of Mercury added unto him. On the other hand, the 

 predicted displacement of the solar spectrum lines 

 certainly conjures up a serious obstacle. The evidence, 



