the first time without uttering a shout of wonder." It requires a very
powerful telescope completely to resolve this fine nebula, but the
outlying streamers may be resolved with a good 3-inch telescope. Sir W.
Herschel considered that the number of the stars composing this
wonderful object was at least 14,000. The accepted views respecting
nebulæ would place this and other clusters far beyond the limits of our
sidereal system, and would make the component stars not very unequal (on
the average) to our own sun. It seems to me far more probable, on the
contrary, that the cluster belongs to our own system, and that its
components are very much smaller than the average of separate stars.
Perhaps the whole mass of the cluster does not exceed that of an average
first-magnitude star.

The nebulæ 92 M and 50 H may be found, after a little searching, from
the positions indicated in the map. They are both well worthy of study,
the former being a very bright globular cluster, the latter a bright and
large round nebula. The spectra of these, as of the great cluster,
resemble the solar spectrum, being continuous, though, of course, very
much fainter.

The star [delta] Herculis (seen at the bottom of the map) is a wide and
easy double--a beautiful object. The components, situated as shown in
Plate 3, are of the fourth and eighth magnitude, and coloured
respectively greenish-white and grape-red.

The star [kappa] Herculis is not shown in the map, but may be very
readily found, lying between the two gammas, [gamma] Herculis and
[gamma] Serpentis (_see_ Frontispiece, Map 2), rather nearer the latter.
It is a wide double, the components of fifth and seventh magnitude, the
larger yellowish-white, the smaller ruddy yellow.[5]