NO. 1865 



OUR KNOWI^EDGE; of the sun HALU 



353 



The measurement of the hydrogen flocculi is complicated by their 

 changes in form, which are much more rapid than in the case of 

 calcium. It is not surprising that this should be true, if the hypoth- 

 esis provisionally adopted to account for the nature of the flocculi 

 is correct. According to this hypothesis, the calcium flocculi shown 

 by the spectroheliograph correspond to three different levels, defined 

 in any case by the position of the second slit with reference to the 

 H or K line. These lines are of complex structure, as Plate xxxiv 

 illustrates. H consists of a broad hazy band, designated as H^ ; 

 superposed on this is a narrow bright line, called Hg ; and near the 

 center of this bright line is a very narrow dark line, called H3. K is 

 similar to H (though somewhat stronger) and contains the con- 

 stituents Ki, K2, and K3. If the second slit of the spectroheliograph 

 is set at some point on the broad H^ or K^ band, only the low-lying 

 calcium vapor which is dense enough to produce a band of this width 

 is capable of showing its presence on the photograph (Fig. i, Plate 

 xxxv). When the second slit is set so as to include H, or Ko, the less 

 dense vapor, lying at a higher level (a few thousands of miles above 

 the photosphere), produces the calcium flocculi measured in the 

 above mentioned determination of the solar rotation (Fig. 2, Plate 

 xxxv). The Ho photographs frequently show evidences of the ab- 

 sorbing efifect of vapors lying at the H3 level, which give rise to dark 

 calcium flocculi. When the spectra of these flocculi are photo- 

 graphed, the H3 and K3 lines are found to be greatly widened and 

 strengthened in them. There can therefore be but little doubt that 

 they correspond to absorption effects produced at a comparatively 

 high level. Independent evidence in favor of this view is afforded 

 by the fact that spectroheliograph pictures of the Sun's limb fre- 

 quently show prominences, many thousands of miles in height, to be 

 present at points where dark flocculi extend on to the disk. This 

 question has been specially investigated by Michie Smith and Ever- 

 shed, at Kodaikanal, India, and their conclusion that these dark 

 flocculi are prominences, absorbing the light of the disk, is in perfect 

 harmony with the Mount Wilson results. In some cases, however, 



