and Terrestrial Elements in the Sun. 407 
ground for believing that the sun contains any thing which is 
not to be found on earth. 
After all, there are very few of the lines developed in spots 
which do not appear at least as faint Fraunhofer lines in the 
ordinary solar spectrum. The maps of Fievez and Vogel 
give many such lines which are not in Angstrom's map. 
Whenever lines of two metals are coincident, or nearly so, 
there is of course a double chance of increased width; and this 
corresponds with the observations. Further, it is noticed that 
the expansion of some lines is unsymmetrical — they expand 
more on one side than on the other. This may be an effect 
of the upward or downward motion of that portion of the 
vapour which is under observation; but it may also be due to 
the development of a second line close to that ordinarily seen. 
Such an effect we have observed in the case of the blue line 
of lithium, which appears to expand with increased tension of 
its vapour more on one side than on the other; but the 
effect is really due to the development of a second line close 
to the first and the expansion of both lines. This second line 
is seen as a narrow line as the metal is dissipated, and then 
disappears entirely long before the usual blue line shows any 
sign of failing. This will, we believe, satisfactorily account 
for some of the apparent motions of lines. Where a displace- 
ment, by reason of rapid motion of the sun's atmosphere, of 
some of the lines occurs without any displacement of others, 
we must suppose that the displaced lines are produced by one 
layer of the sun's atmosphere, while the undisplaced are due to 
another. This will account for the long lines, or lines of vapour 
of low tension, being displaced, while those due to vapour of 
high tension are unmoved, and vice versa. There is, however, 
one line which appears to be very anomalous in its behaviour, 
and requires further notice, because Mr. Lockyer has more 
than once called special attention to it. This is the line with 
wave-length 4923. There is an iron line with this wave- 
length; but in iron it is certainly not a long line: it is not 
seen in the arc when there is only a very little iron present; 
and it comes out with increasing density of vapour. Corre- 
sponding with this character, it is frequently seen strengthened 
in spots like the other iron lines. On the other hand, a line 
of this wave-length is one of those most frequently (40 per 
cent.) observed by Young as a bright line high up in the 
solar atmosphere; and it has as frequently been observed by 
Tacchini in the spectrum of prominences. It should there- 
fore be a long line, or one of vapour of low tension, if 
judged by this character; and this is confirmed by the fact 
recorded by Mr. Lockyer that it is frequently absent from 
