November 4, 1909] 



NATURE 



observed at the sun's limb, the bright calcium flocculi, as 

 these luminous clouds are called, are so low, in comparison 

 with the prominences, that they can hardly be detected as 

 elevations. Thus our knowledge of the calcium flocculi 

 must be derived mainly from the study of spectroheliograph 

 plates, which show them in projection on the disc. 1 must 

 not omit to mention, however, that the calcium vapour 

 rises to the highest parts of the prominences, and that 

 this higher and cooler vapour frequently indicates its 

 presence on spectroheliograph plates in the phenomena of j 

 dark flocculi. These are relatively inconspicuous, however, 

 and need not be discussed here.' 



It soon appeared that the average photograph of bright 

 calcium flocculi could not be counted upon to indicate the j 

 e.\istence of definite streams or currents in the solar atmo- 

 sphere. In 1903 the hydrogen flocculi were photographed 

 for the first time. By comparing these flocculi with the 

 corresponding calcium flocculi we see that, in general, ' 

 dark regions on the hydrogen image agree approximately [ 

 in form with bright regions on the calcium image. This 

 might appear to indicate that hydrogen is absent in the 

 regions where cJUcium is most abundant. \v\ investigation 

 of the question, however, does not lead to this conclusion. 

 Dark hydrogen flocculi seem to mark those regions on the 

 ^.un's disc where hydrogen is present as an absorbing 

 medium, which reduces the intensity of the light coming 

 through it from below. In certain areas, where the 

 temperature is higher or the condition of radiation other- 

 wise different, the hydrogen flocculi are bright. In manv 

 cases eruptions are in progress at these points, but in 

 others the difference in brightness is apparently not iIt 

 direct result of eruptive action. 



The hydrogen flocculi, thus photographed with the lin- - 

 H/3, H7, or H5, differ in many respects from the calcium 

 flocculi. Not only do they usually appear dark, where tli 

 calcium floccuU are bright : their forms exhibit strikin- 

 peculiarities, which are absent or much less conspicuous 

 in the case of calcium. The appearance of the calcium 

 flocculi resembles that of floating cumulus clouds in our 

 own atmosphere: their capricious changes in form revf.il 

 the operation of no simple law. But the hydroi;iii 

 flocculi, on the contrary, exhibit a definiteness of structui- 

 in striking contrast to this appearance. Some of ih- 

 photogra[ihs strongly remind us of the distribution of imi: 

 filings in a magnetic field, and suggest that some unknow i 

 force is in operation. 



Such was the condition of the subject when the red lU. 

 line of hydrogen was first applied to the photography .t 

 the flocculi, on Mount Wilson, in March, iqoS. Ili 

 calcium and hydrogen flocculi had been studied for sev.r.i 

 years, and much had been learned as to their nature an^l 

 their motions. It had been found, for example, that tli- 

 calcium flocculi observe the same law of rotation th.ii 

 governs the motions of sun-spots, while the hydrogen 

 flocculi apparently follow a different law, in which the | 

 decrease in the angular rotational velocity from the equator ] 

 toward the poles is much less marked. The latter result j 

 is in harmony with the investigations of .Adams, whose 1 

 accurate measures of the approach and recession of the | 

 hydrogen at the eastern and western limbs of the sun i 

 offer but little evidence of equatorial acceleration on the 

 part of this gas. For this and other reasons it had been 

 concluded that the hydrogen shown in such photographs 

 reaches a higher level than the vapours of the bright (H,) 

 calcium flocculi. The region of the atmosphere previously 

 explored with the spectroheliograph was nevertheless con- 

 fined (except in the case of eruptions and dark calcium 

 flocculi) to a comoaratively low level, lying within a few 

 thousand miles of the photosphere. What might be ex- 

 pected if a still higher region could be satisfactorily photo- 

 granhed in projection on the disc? 



The red line of hydrosjen offered the means, of disclosing 

 the phenomena of this higher atmosphere. As it may not 

 immediately appear why different lines, caused by the 

 r.-wiiation of the same gas, should not give precisely similar 

 photographs, a brief reference to the aspect of a promin- 

 ence in the red and blue hydrogen lines may be 

 advantageous. Here are two photographs of the same 

 prominence, seen in elevation at the sun's limb, one made 



1 Eruptive prominences .ire also recorded on the disc as bright fioccul'. 

 NO. 2088, VOL. 82] 



with Ha, the other with H5. As the red line is very 

 bright, even in the highest regions, the photograph taken 

 with its aid shows the entire prominence. riS, on the 

 other hand, is relatively weak at the higher levels, and 

 consequently only the lower and brighter parts of the 

 prominence are well recorded when this line is used. If,, 

 now, we suppose ourselves immediately above such a 

 prominence, at a point where we observe it in projection 

 against the disc, it is evident that the character of the 

 hydrogen lines must depend upon their brightness at 

 different levels. As we know that, speaking generally, 

 absorption is proportional to radiation, the amount of light 

 absorbed in the upper part of the prominence will be 

 much greater for Ha than for HS. Hence the average 

 level represented by the absorption of Ha will be higher 

 than the average level represented by H6, since the higher 

 gases play a more important part in the production of 

 the former line. We may therefore expect that photo- 

 graphs of the sun's disc, taken with the Hght of Ha, will 

 show the dark areas corresponding to absorption in the 

 prominences much more clearly than photographs taken 

 with H5. Moreover, since Ha is stronger than H5 in the 

 upper chromosphere, in regions where no prominences are 

 present, the average level represented by this line will, in 

 general, be higher than that representee! by H5. k com- 

 parison of two photographs of the sun's disc, made with 

 the lines in question, 'will <ul'li. u u. maki:- this clear. This 



er.oimous group of prominences, stretching for several 

 hundred thousand miles across the sun, is much more 

 clearlv indicated by Ha than by H5. In general, the 

 hvdrogen flocculi are stronger and more distinct when 

 pi-.otographed with Ha, and there are some regions which 

 appear bright with Ha and dark with HS. This latter 

 peculiarity probably has an important bearing upon the 

 similar behaviour of hydrogen in certain stars and nebulze, 

 but a discussion of this question cannot be undertaken 

 here. 



The first of the Ha photographs gave strong hopes of a 

 substantial advance in our knowledge of the solar atmo- 

 sphere. The sharpness and comparatively strong contrast 

 of these flocculi, and the evidences of definite structure anJ 

 clearlv defined stream lines which they revealed, were 

 highlv encouraging. The work w^as begun during the dis- 

 turbed weather of the rainy season, when the definition- 

 of the solar image is never of the best. On April 30, iqoS, 

 the first photographs were secured under the fine atmo- 

 spheric conditions which prevail in the dry season. !\ 

 direct photograph (Fig. i) shows a small and. insignificant 

 gioup of sun-spots, whicii would not seem, without other 

 indications, to merit special attention. The next phulo- 

 graph (F'ig. 2) shows that an enormous calcium floc-'ruhis. 



