October 22, 19 14] 



NATURE 



203 



a 6-in. lens of 98 in. focal length, and a 6|-in. 

 flat mirror. A Gnib coelostat of 8-in. diameter 

 supplied a beam of light to a 5-in. Alvan Clarke 

 lens of 7 ft. focal length, which formed the sun's 

 image on the slit. 



With the assistance of the town electrician, 

 ^Herr Helenius, leads were run from the electrical 

 jlant in the technical school to operate an arc- 

 lamp with iron poles, the pressure being no 

 volts and the current 12 amperes. An auxilian.- 

 ;lens and a diagonal prism, which could be pushed 

 backwards and forwards as required, formed the 

 image of the arc on the slit. The slit could be 

 covered with a plate of zinc having a horizontal 

 aperture i,'io-in. in width, so that the comparison 

 spectrum could be placed over the position occu- 

 pied by the image of the dark moon. Owing to 

 the skill and energy of Mr. Gibbs, the engineer 

 to the party, the instruments were all erected, 

 the cases being utilised as bases, and adjusted, 



N 



.^^ w 



pictures to show the extension. With the 

 spectrograph the coronal spectrum was to be 

 photographed, principally in the red and yellow 

 regions, and if possible a comparison spectrum 

 of the iron-arc was to be placed on the plate. 



About half-past eleven on the morning of the 

 eclipse a goodly number of spectators, dressed 

 in their best clothes, assembled in the field near 

 the site of the instruments, and watched the 

 partial phases through dark glasses. In contrast 

 to the demeanour of the spectators in the eclipse 

 expedition to \'inaroz in Spain in August, 1905, 

 they were quite undemonstrative, even when the 

 magnificent sight of the corona in a perfectly dark 

 sky was presented to their gaze. For certainly it 

 was a magnificent spectacle, enhanced by the ap- 

 pearance of the planet Mercury brilliantly shining 

 to the north-west limb of the sun, and Venus near 

 the south-east horizon. To the naked eye the 

 most striking features of the corona were a long' 

 fish-tail streamer on the west limb of 

 the sun, and a spreading mass of 

 streamers on the east limb, the salient 

 phenomena beingf a streamer on the 

 north-east, another shorter streamer 

 south-east, and a long horizontal 

 streamer, less luminous than the other 

 two, placed equatorially. The promin- 

 ences were not visible to the naked 

 eye ; they were seen projected on the 

 slit, and a very fine arched prominence 

 is shown in the north-east quadrant 

 in the photographs. Five large scale 

 photographs were secured with the 

 20-ft. coronagraph, with exposures of 



Fig. 2.— Main features of the Solar Corona of 1914, August 21, roughly oatlined from photo- 

 graphs takeo with the 20-ft. coronagraph at Hernosanii. 



by Monday, August 17, on which day drills were 

 commenced. W^e were assisted by students of 

 the technical school under the direction of Mr. 

 Askling, the master of mechanical science. An 

 eclipse clock, designed by Mr. Gibbs, was also 

 erected, a large hand circulating round the dial 

 once in 129 seconds, the computed duration of 

 totality. Its face was lit up by an electric lamp. 

 The purity of the atmosphere at Hernosand 

 was remarkable, but there was a very consider- 

 able amount of cloud on every day, except one, 

 during the days preceding the date of the eclipse. 

 The barometer was high with a persistent north- 

 west wind, accompanied by a variety of types 

 of weather. On the evening, however, of August 

 20, the wind changed to the south-east, and a 

 beautiful night heralded the perfect weather con- 

 ditions for the eclipse. The programme settled 

 for me bv the Joint Permanent Eclipse Com- 

 mittee was to obtain large scale photographs of 

 the solar corona to show detail, and smaller scale 



NO. 2347, VOL. 94] 



4, 10, 25, 7, 3 seconds respectively. 

 These show a great amount of detail, 

 including beautiful sheafs of polar 

 rays, prominences, and several cross- 

 ing and interlacing streamers. The 

 main features are shown in the ac- 

 companying outline drawing (Fig. 2), 

 made from some of the negatives by 

 Mr. W. McKeon. With the 30-in. 

 Abney coronagraph, used in so many former 

 eclipses, four photographs were secured, with ex- 

 posures of 10, 50, 15, and 5 seconds, on plates 

 bathed by Mr. Crowther, of Carlisle, so as to be 

 not solarisable by over-exposure. These show the 

 extension of the corona for fully two solar 

 diameters. With the 12-in. coronagraph, work- 

 ing at //3*5, a single exposure was made for 

 95 seconds, on a Paget " Hydro " non-solar i sable 

 plate. This is a beautiful small scale picture, 

 showing, probably, the full extension of the 

 streamers. 



With the spectrograph, while I made the ex- 

 posures, Mr. Gibbs placed the sun's image 

 correctly on the slit, and also took charge of the 

 coelostat. Several exposures were made at inter- 

 vals of a few seconds on the tip of the cusps, 

 seven minutes before and five minutes after 

 totality. These mere lines of spectral light have 

 so far not received adequate study. The com- 

 parison iron spectrum is on each plate, which 



