324 ANNUAL KEPORT SMITHSONIAN INSTITUTION, 1908. 



Some preliminary work with the spectroheliograph attachment of the SO-foot 

 Littrow spectograph of the tower telescope, in which I had the assistance of 

 Mr. Adams, indicated that bright flocciili are more numerous and extensive when 

 photographed with Ha than when H5 is used. I then tried Ha with the 5-foot 

 spectroheliograph of the Snow telescope, and immediately obtained excellent 

 results. The images were stronger and of much better contrast than those given 

 by HS. Moreover, the curved and radial structure surrounding sun spots was 

 so striking as to lead to the hope that important advances might be expected to 

 follow from the systematic use of the Ha line. * * * 



This is so definite in form and so unmistakable in character as to satisfy 

 the hopes aroused by the earlier photographs. It seems evident, on mere 

 inspection of these photographs, that sun spots are centers of attraction, draw- 

 ing toward them the hydrogen of the solar atmosphere. Moreover, the clearly 

 defined whirls point to the existence of cyclonic storms or vortices. * * * 



In the present paper I wish to illustrate the phenomena photographed with 

 the aid of Ha in the neighborhood of a spot which reached the east limb of the 

 sun at 8" 16"' a. m. on May 26, 1908. A photograph of this spot, made by myself 

 with Ha on May 29, at 4'' 26'" p. m. Pacific standard time, is reproduced in figure 

 1, plate 1. The whirl structure, which is clearly shown by this photograph, is 

 also very distinct, though of somewhat dift'erent form, on the photograph of 

 May 28. It is interesting to inquire as to the probable level of the region in 

 which this whirl occurred, and the height of the long dark flocculus south of 

 the spot. For this purpose we may examine photographs of the chromosphere 

 and prominences at the limb, taken on May 25, 26, and 27. In the first of these, 

 made on May 25 at 9'' 18'" a. m. (No. 4142), a long narrow prominence, extending 

 toward the north, rises from the limb at position angle 92°, a point about 

 one degree north of the spot. It makes an angle of about 12° with the limb, 

 and fades out at the upper end, its length being approximately 90" (geocen- 

 tric). There are other small filamentary prominences in the region extending 

 about 7° north of the spot, and smaller elevations in the chromosphere to the 

 south. At position angle 98° a bright prominence rises to a height of about 

 20" and then slopes to the chromospheric level at position angle 107°. Near 

 its southern end is an independent filamentary prominence about 55" high. 

 On May 26, at 6" 38'" a. m. (No. 4144), the prominences were photographed at 

 the east limb. The lowest point in the chromosphere on this photograph corre- 

 sponds to the position (position angle 93°) where the spot crossed the limb about 

 two hours later. It will be seen that these prominences, which extend from posi- 

 tion angle 82° to 106°, cover much of the region in which the whirl structure 

 of plate 1 appears. The prominence south of the spot is very bright and its 

 highest point reaches an elevation of about 35". On May 27, at 5" 22'" p. m. 

 (No. 4152), a prominence about 25" high extends from position angle 105° to 

 109°. This is doubtless the eastern extremity of the strong flocculus in plate 1, 

 which may be there seen curving toward the spot. 



We may now pass in rapid survey the more important photographs of the 

 disk. On May 28, at &" 58"" a. m. (No. 4157), the spot is near the east limb and 

 the whirls are well shown. To the east of the spot is a long, narrow line of 

 bright hydrogen. On May 29, at 6" 24'" a. m. (No. 4171), the whirls are very dis- 

 tinct and differ in many respects from those shown on May 28. Eruptive 

 regions of bright hydrogen are seen southeast and west of the spot. The 

 eastern end of the long, dark flocculus is changing in form, and bridges are 

 appearing over the spot. Negative No. 4175, taken one hour and nineteen minutes 

 later, seems to show distinct changes in the whirls, though they are not measur- 

 able. On May 29, at 4'' 26'" p. m. (No. 4176), the whirls resemble those shown 

 in negative No. 4175, but exhibit some marked changes. An eruption which 



