lines of
the chromosphere spectrum. In addition to it the corona shows a few
other bright lines, belonging, no doubt, to the same unknown element
("coronium"), and also a faint continuous spectrum, in which even a few
of the more prominent dark lines of the solar spectrum have been
sometimes detected. This shows that in addition to glowing gas
(represented by the bright lines) the corona also contains a great deal
of matter like dust, or fog, the minute particles of which are capable
of reflecting the sunlight and thereby producing a feeble continuous
spectrum. This matter seems to form the principal constituent of the
long coronal rays and streamers, as the latter are not visible in the
detached images of the corona which appear instead of the bright lines
when the corona is viewed, or photographed, during an eclipse, in a
spectroscope without a slit. If the long rays were composed of the gas
or gases which constitute the inner corona, it is evident that they
ought to appear in these detached images. As to the nature of the forces
which are continually engaged in shooting out these enormously long
streamers, we have at present but little information. It is, however,
certain that the extensive atmospheric envelope round the sun, which
shows itself as the inner corona, must be extremely attenuated. Comets
have on several occasions been known to rush through this coronal
atmosphere without evincing the slightest appreciable diminution in
their speed from the resistance to which they were exposed.

We have accumulated by observation a great number of facts concerning
the sun, but when we try to draw from these facts conclusions as to the
physical constitution of that great body, it cannot be denied that the
difficulties seem to be very great indeed. We find that the best
authorities differ considerably in the opinions they entertain as to its
nature. We shall here set forth the principal conclusions as to which
there is little or no controversy.

We shall see in a following chapter that astronomers have been able to
determine the relative densities of the bodies in the solar system; in
other words, they have found the relation between the quantities of
matter contained in an equally large volume of each. It has thus been
ascertained that the average density of the sun is about a quarter that
of the earth. If we compare the weight of the sun with that of an
equally great globe of water, we find that the luminary would be b