ON THE TOTAL SOLAR ECLIPSE OF MAY 17, 1882. 
269 
that there are indeed a number of lines in this region. Over the fainter portions of 
this continuous region lines can be distinctly traced. A few of them are sufficiently 
strong to be seen with the naked eye, but most of them require the use of a magnifying 
glass. The lines are no doubt very faint, but we have succeeded in showing them even 
to some whose eyes were not specially trained to examine faint photographic impressions, 
and there cannot be the slightest doubt as to the real existence of these lines. It is 
more difficult to say to what extent our measurements can be considered as accurate. 
They have been taken with a measuring microscope, but it was only when the light was 
good, and when the eyes were in their best condition, that satisfactory results were 
obtained. We give the wave-lengths as being as exact as we can hope for with the 
small dispersion employed. Four or five of the lines can be easily seen under ordinary 
circumstances. These are one line less refrangible than G, of very peculiar appearance. 
At its base it is a broad band, then contracting quickly it fades away into a sharp 
point; close to it is a sharper and longer line. A little less refrangible than H is an 
easily recognised line reaching to a good height away from the sun (more than a 
radius). Two or three lines in the ultra-violet are also easily recognised. 
The following list of lines does not pretend to be complete. There are some lines, 
especially in the ultra-violet, which have not been included, but they are so faint that 
their measurement would have been very difficult. As it is, even some of the lines 
given in the table can only be seen under specially favourable circumstances. On the 
northern side only a few of the principal lines can be traced. We have sometimes 
thought that a different set of lines appeared there, but the measurement and inter¬ 
pretation of the northern line present some difficulties. In the first place the spectrum 
of the corona is much fainter, and secondly the spectrum of the prominences is much 
stronger. Most of the prominence lines, like H and K, stretch across the moon’s 
disc, and therefore also across the corona. Amongst the number of lines we have 
measured it would not be difficult to point to some coincidences with the lines 
of spectra of known bodies, but other equally strong lines of the same bodies seem 
to be wanting, and we do not feel ourselves justified in suggesting any more than 
accidental coincidence. 
Another feature of the photograph is the fact that the Fraunhofer (dai'k) lines 
about G are clearly distinguishable in the coronal spectrum. About this region of 
the spectrum the photographic action attains its maximum, and it is in this locality 
that we should naturally first look for evidence of reflected solar light. That a small 
fraction of the coronal light is reflected light of this nature is now without doubt 
established, the photographic evidences being complete. We may remark that the 
lines are of greatest intensity a little distance from the moon’s track, though at the 
best they are extremely faint. 
