598 
MR. J. NORMAN" LOCKYER ON THE 
No hydrogen rings were seen during the eclipse of 1878, and as the slitless spectro¬ 
scopes employed by Draper, Danyard, and myself were of considerable dispersive 
power, no appearances of rings due to continuous spectrum could have been produced. 
The presence of hydrogen rings about the time of sun-spot maxima in the eclipses of 
1871 and 1882, and their absence near the time of minimum in 1878, suggests a 
relation to the sun-spot period, similar to that which has been established in the case 
of 1474 K, but as the rings are not shown in the photographs taken at the maximum 
in 1893, a final verdict cannot yet be given as to whether hydrogen varies with the 
sun-spot period. 
Reference to Helium. 
As shown in the table, the D 3 ring was stated to have been photographed in Egypt 
in 1882, but not on any other occasion.* From the description of the ring given in 
the report of this eclipse,! it seems very probable that it is similar to the broad yellow 
ring photographed on the isochromatic plates in 1893, which, as already shown, is due 
to the action of continuous spectrum acting on plates with several maxima of 
intensity in different parts of the spectrum. It is described as being more uniform 
all round the sun than the 1474 ring, and this is the case also in the 1893 photo¬ 
graphs, on which the ring in question is a little more refrangible than D 3 ; but in the 
smaller scale photograph of 1882 this difference of position may not have been 
evident, or the 1882 plate may have had one of its maxima of photographic action 
still nearer to D 3 . In the absence of a copy of the photograph it is difficult to deter¬ 
mine how far this explanation holds good, but, if it should be established, this, in 
combination with all the other eclipse records, would prove that D 3 does not form 
part of the coronal spectrum. 
Reference to Unknown Radiations. 
As to the variation in intensities of other radiations of the corona, no evidence is 
furnished by the prismatic camera results alone. The table of coronal rings already 
given (p. 593) suggests a similarity of the 1886 spectrum with that of 1893, but the 
conditions under which the two series of observations were made w T ere so different, 
that it would be unwise to draw any conclusions as to the variation of the feebler 
radiations. 
At present, then, there is no continuous and distinct instrumental evidence of 
a periodic change of specific parts of the spectrum, other than the varying brightness 
of 1474 K. 
* As stated in the preliminary report on the eclipse of 1893 (‘ Phil. Trans.,’ vol. 185, 1894, A, p. 716), 
a yellow ring was seen by Mr. Shackl ETON in Brazil; the subsequent discussion has shown that this 
must be ascribed without hesitation to chromosphere and prominences, and not to the corona. 
t ‘Phil. Trans.,’ vol. 175. 1884, Part I., p. 264 
