128 
of rather coarse stars, and resembling Messier 13.” Any in- 
crease of brightness towards the centre seemed to proceed 
from the greater depth of stars there rather than from any 
notable difference of their magnitude. But the second class 
presents much more interesting phenomena: the appearances 
which previous observers had described as sudden condensa- 
tion, nuclei, or even single or multiple central stars, proving 
to be clusters of comparatively bright stars, surrounded by 
much larger collections of minute ones. A very beautiful 
example of this is 1456, fig. 41, M. 94, described in the cata- 
logue as ‘‘ very suddenly much brighter, almost up toa 
> 
nipple-shaped nucleus:” it proved, however, to be ‘‘a vast 
circular cluster of stars, with ragged filaments, in which, and 
apparently central, is a globular group of much larger stars, 
power 400.” The same system of arrangement (which seems 
very common) occurs also in 706, 748, 805, and many others: 
it is also found in the magnificent clusters 1663, M. 3; 1558, 
M. 53; and 1916, M.5. In these, the splendour of which 
is not to be described, besides the stars visible in other instru- 
ments (which here seem of the first or second magnitude), the 
whole field is crowded with others much smaller, to such a de- 
gree that, had the first been absent, these would still have been 
noted as remarkable objects. The interior group is not, how- 
ever, always central or symmetrical, but has knots of greater 
condensation, which sometimes (as in 1385) are“alone visible 
in smaller telescopes, and then look like ‘twin nebule ;” 
at others (as in 739), like stars. In 1622, fig. 25, M.51, 
which is so well known from a sort of resemblance to Saturn, 
and from the more exact analogy which, as Sir John Herschel 
has well remarked, it bears to the Milky Way, we have ano- 
ther different development of this arrangement. Here also 
the central nebula isa globe of large stars; as indeed had 
been previously discovered with the three feet telescope: but 
it is also seen with 560 that the exterior stars, instead of being 
uniformly distributed as in the preceding instances, are con- 
