142 RUMFOED SPECTROHELIOGRAPH. 



ered Avitli columns of bright calcium vapor, A^arying in diameter 

 from less than a second to several seconds of arc, separated by darker 

 spaces, which correspond in appearance to the darker spaces that 

 separate the photospheric '' grains." The summits of these columns 

 seem to lie in the second stratum, corresponding to the bright lines 

 H, and K,. 



In the larger flocculi the surmise of a structure composed of ex- 

 panding columns of calcium vapor seems to be borne out by the pho- 

 tographs. Compare, for example, figures 1 and 2 in plate vii. At 

 the lower level (fig. 1) the flocculus is resolved into a series of Avell- 

 defined elements, of comparatively small area. At the higher level 

 (fig. 2) the area of the entire flocculus is greatly increased, and there 

 seems to be evidence (hardly visible in the cut) that the columns 

 composing it have arched over, so that they are no longer seen end 

 on. Few photographs are sufficiently well defined to bring out such 

 details, and it can not be said with certainty that the effects seen at 

 the higher level are always due to separation and bending of the col- 

 umns, as well as to expansion of each of the individual columns. In 

 any event, the increase in area at this level is sometimes very great, 

 in many cases sufficient to cover not only the penumbra, but also 

 entire spots. Another illustration of the expansion at increasing 

 altitudes may be seen in the four photographs, coi-responding to dif- 

 ferent levels, which are reproduced in plates vii and \iii. A much 

 finer illustration is afforded by the photographs in plates xi to xviii 

 of the recent great sun spot. 



HYDROGEN FLOCCULI. 



The method of i^hotographing the sun with the aid of the dark 

 Fraunhofer lines has already been explained. The spectroheliograph 

 is employed exactly as in the case of the bright calcium lines, but the 

 dispersion is increased sufficiently to insure that the width of the 

 dark lines shall be greater than that of the second slit. Under 

 such circumstances photographs corresponding to the hydrogen lines, 

 or to any other dark lines of sufficient width, may be obtained. 



The first photograph made with a dark hydrogen line {H ft) was 

 taken with the Rumford spectroheliograph on ]\Iay 16, 1903. On 

 developing the plate we were surprised to find a structure differing 

 materially from that obtained with H^ and Ko. Closer examination 

 and a comparison of the i)hot()graph with a K. photograph made on 

 the same day showed that the bright calcium flocculi were replaced 

 on the up photogra,ph by dark structures of similar, though by no 

 means identical, form. There could be no doubt about the adjust- 

 ment of the FIft line on the second slit, since a prominence was shown 

 on the photograph extending above the sun's limb. At a point near 

 a sun spot a brilliant object appeared. The same bright object was 



