ON THE TOTAL ECLIPSE OF THE SUN, MAY 28, 1900. 
409 
In connection with the intensities it is important to note that the relative intensi¬ 
ties are not the same at different stages of the eclipse. Thus in photographs Nos. 1 
to 4, the helium line 4026'34 is very much less intense than the adjacent strontium line, 
4077’89, while in photograph No. 5, which shows the spectrum of the chromosphere 
only, the two are of practically equal intensity. As both arcs are of the same length, 
this change indicates that while the strontium vapour extends down to the sun’s limb, 
the helium exists only in an elevated shell concentric with the photosphere. A 
similar behaviour is noted in the lines of asterium; in numbers 1 to 4, tor instance, 
the asterium line 4922T0 is only very slightly stronger than the adjacent barium 
line 4934‘24, whereas in No. 5 the latter is scarcely visible, while 4922T0 is a well- 
marked arc. 
The same feature is observed in other arcs due to asterium and helium, and also in 
the unknown line at wave-length 4687. Arcs due to other substances, however, 
gradually become shorter and less bright as the moon eclipses more and more of the 
chromosphere. 
The Spectra of the Prominences. 
The prominences photographed during the eclipse are few in number, and with two 
exceptions were of no considerable magnitude. For convenience of reference the 
accompanying diagram (fig. 7) has been prepared from photographs 5 and 7 of the 
20-foot series, by painting out all parts except the images in K light, and the 
various prominences have been numbered as shown. 
Fig. 7. The chromosphere and prominences in K light. 
The prominences numbered 1, 2, 4, 6, 7, and 9 have spectra in which H and K 
are very bright, the chief lines of hydrogen less bright, and those of helium com- 
VOL. cxcviii.—A. 3 G 
