OF THE SUH OBSERVED AT CAROLINE ISLAND, 1883. 
123 
coronal, is important. It will be better in future eclipse expeditions to place the slit 
of the spectroscope tangential to the moon’s limb in preference to normally. This 
has been done in the recent eclipse observed in the West Indies (August, 1886), with 
most satisfactory results. The chief point to attain is to separate all prominence 
light from the coronal light, as it tends to mask the true spectrum of the latter. 
From the photographs I have examined I have come to the conclusion that not much 
more is to be learnt at present from them. It may be that as more eclipses come to 
be observed with the same instruments, or at all events on the same lines, the photo¬ 
graphs of the Caroline Island station will prove to be of greater value than they seem 
to be now. 
If we compare the corona of this eclipse wdth that of the eclipse in Egypt, perhaps 
the most striking feature is the absence of the hydrogen lines. In Egypt the photo¬ 
graph shows, besides the lines which may be presumed to be hydrogen at H, at least 
two other lines of hydrogen, \ 4340 and A. 4101. In the Caroline Island photographs 
these lines are entmely absent. It may be well to draw attention to the fact that in 
the former eclipse the prominences were very marked, and in the prismatic (slitless) 
spectrum the hydrogen rings were very powerfully shown. In the eclipse now under 
consideration the prominences were very small, and the prismatic (slitless) spectrum 
gave no result other than rings at H and K. It would seem, then, that the corona 
at the time of the Egyptian eclipse was illuminated more or less by the prominence 
light. If this be admitted, we ought to find that the corona during the Caroline 
Island eclijDse was illuminated by the light which emanated from the matter which 
gave H and K so strongly in the ring spectrum. Looking at the list of lines, we find 
that such is the case. Calcium was evidently present in the light, more especially 
near the limb of the moon. We find that three calcium lines are shown reversed 
across the dark moon, and two iron lines. It is somewhat hard to see how these 
reversed lines made their appearance in such a locality. It is quite evident that they 
must be due to reflected light. I can find no trace of Eraunhofee. lines about G 
outside the corona, such as Dr. Schuster and myself found in the Egyptian eclipse 
photograph, and which would be the first to appear in the photographic plate 
were any reflected sunlight as it reaches us present in those regions. It should be 
remarked that the reversed lines across the moon are extremely faint, but perfectly 
distinguishable and measurable. Most of the lines in the spectrum of the corona 
lie near the moon’s limb, and have quite a different aspect to those delineated in the 
Egyptian eclipse negative, and some of them are probably prominence lines, and 
I think it wmuld be dangerous to found any theory on the discovery of new lines in 
the coronal spectrum from the list of lines here recorded. 
In conclusion, I think I may say that the two English observers, Mr. H. A. 
Lawrance and Mr. C. E. Woods, deserve every credit for the amount of wmrk they 
did. The large number of instruments they were called upon to utilise during the 
eclipse, and wdiich they evidently most skilfully manipulated, could only have been 
R 2 
