formations, often being the most brilliant points on their walls, when
they are found in this position. Very frequently too they are not only
very bright themselves, but stand on bright areas, whose borders are
generally concentric with them, which shine with a glistening lustre, and
form a kind of halo of light around them. Euclides and Bessarion A, and
the craters east of Landsberg, are especially interesting examples. It
seems not improbable that these areas may represent deposits formed by
some kind of matter ejected from the craters, but whether of ancient or
modern date, it is, of course, impossible to determine. Future observers
will perhaps be in a better position to decide the question without
cavil, if such eruptions should again take place. Like the larger
enclosures, these smaller objects frequently encroach upon each other--
crater-ring overlapping crater-ring, as in the case of Thebit, where a
large crater, which has interfered with the continuity of the east wall,
has, in its turn, been disturbed by a smaller crater on its own east
wall. The craters in many cases, possibly in the majority if we could
detect them, have central mountains, some of them being excellent tests
for telescopic definition--as, for example, the central peaks of
Hortensius, Bessarion, and that of the small crater just mentioned on the
east wall of Thebit A. A tendency to a linear arrangement is often
displayed, especially among the smaller class, as is also their
occurrence in pairs.

CRATER-CONES.--These objects, plentifully distributed on the lunar
surface, are especially interesting from their outward resemblance to the
parasitic cones found on the flanks of terrestrial volcanoes (Etna, for
instance). In the larger examples it is occasionally possible to see that
the interiors are either inverted cones without a floor, or cup-shaped
depressions on the summit of the object. Frequently, however, they are so
small that the orifice can only be detected under oblique illumination.