Melaphyy. 
The melaphyrs throughout the Nantasket area differ from 
the diabases in being of a finer grain, in that their feldspars 
are less decomposed, and in never, so far as observed, contain- 
ing a recognizable unaltered trace of either augite or olivine ; 
also in being, as a rule, more or less amygdaloidal. They 
also carry a considerable proportion of amorphous base. These 
characteristics are alone sufficient in every ease examined to 
distinguish the so-called melaphyrs from the diabases (or the 
surface flows from the later dike rocks). The melaphyrs also 
carry a larger proportion of secondary epidote, and less of the 
viridite, these two substances occupying the space of the de- 
composed feldspars and augites and filling the amygdaloidal 
cavities. The feldspars have the appearance described by 
Denton! ; and, although still showing twinning striw, are more 
or less kaolinized or show traces of the viriditie and epidotic 
alteration so common in the dike rocks. The black material 
of the ground-mass is in part magnetite, as shown by its bluish 
reflections and by its being strongly attracted by the magnet ; 
but it is often accompanied by blood red scales of hematite, 
The amygdules in these rocks are in part true steam-holes 
or vesicles, as shown by their regular oval form, sharply de- 
fined walls, and the arrangement of the enclosed and adjacent 
minerals. In other cases, however, they are merely cavities 
left by the decomposition of some constituent of the rock, and 
are irregular in outline, poorly defined and show frequently the 
remains of some less easily decomposed mineral, as a feldspar, 
projecting from the cavity wall inward. These pseudo-amyg- 
dules, like the true amygdules, ате filled with a variety of 
secondary minerals, as feldspar, quartz, epidote and chlorite, 
often without definite order, or again showing a quite regular 
concentric arrangement. I have not, however, been able to 
find the marked regularity in the arrangement of these minerals 
in the pseudo-amvgdules that I was led to expect from a peru- 
1 Proc, B. 8, N. H., Vol, xx., page 416. 
