66 KIMBERLITE FROM THE UNITED STATES
generally
is more rounded than in the last case, and a considerable amount of
the original mineral still remains, especially in the larger grains;
that is, they are about half olivine, half serpentine, exhibiting the
usual structure. There is one grain which appears to be a monoclinic
pyroxene, containing two or three flakelets of a micaceous mineral;
also, a small grain of the usual brownish mica, a garnet with the outer
zone of rich brown kelyphite, already described, with two others—which
possibly may be the same mineral more completely altered—and two or
three smaller deep-brown grains resembling chromite. One or two
irregular patches, consisting of a carbonate and some small minerals
(zeolites) are very like the irregular cavities in some scoriaceous
rocks, when these become converted into amygdules. I suspect the
presence of a small rock fragment consisting of little roundish grains
of olivine (or possibly malacolite) in a minutely granular matrix, of
which I can hardly venture to say more than ' probably basic igneous.'
(b) [435,
2.] The label with this specimen (a printed one on thick paper) is
partly effaced by rubbing; it has the words ' Kentucky Geological
Survey . . . and Bureau of (? emigration). Frankfort, Kentucky, U.S.A.'
(the effaced part being apparently a date, perhaps 1886). It is not
distinctly stated to be from the above locality, but I think there can
be no doubt on that point. It is practically identical with the
last-named, except that the yellowish-green spots run a shade larger,
occasionally almost one-quarter of an inch. Under the microscope the
rock appears very similar to the last described, but the olivine is
rather more serpentinised ; garnet as before ; one or two patches of
calcite with a tuft or two of an acicular zeolite, suggestive of
cavities of some kind; part of the matrix is tinged with green, this
apparently being due to the presence of minute scales of a chloritic
mineral.
(c) [437.]
This specimen bears the label, 'Elliott Co., Ky. (Diller).' The matrix
is slightly more compact than in the case of the last two the yellowish
spots are not quite