04 
western section becomes the base of the central section, the two 
together ineluding four beds of conglomerate and four beds of 
lava. 
Strictly speaking, the basal conglomerate is not wanting in 
the central area, since it is undoubtedly represented by the 
patches of clastic material on the granite south of Round Hill; 
and this hill thus admits of comparison with Great Hill. The 
relations to the granite and the boundary fault are the 
same ; but the Round Hill conglomerate is the third bed and 
the Great Hill conglomerate is the second bed. 
The satisfactory correlation of the central and coastal areas 
presents still greater difficulties, and appeared practically im- 
possible, until, fortunately, a partial clue was afforded by the 
ledges on Rocky Neck. The completion of the Nantasket 
column must, therefore, await the detailed description of the 
Rocky Neck section; and it will suffice here to state that all 
the melaphyr of the coastal area and the underlying conglomer- 
ate of Long Beach Rock belong above the highest bed, the 
third melaphyr, in the central area. 
Melaphyr Peninsula.—This most northerly and most nearly 
isolated part of the western area is the only part that does 
not admit of ready correlation without appealing to external 
evidence. For this reason, and because it is probably the new- 
est part of the area, its description has been reserved to the 
last. Melaphyr Peninsula is quite certainly made up of two 
prineipal flows of highly basie and amygdaloidal melaphyr. 
The northern and older flow shows, at the west end, three fairly 
distinct bands of coarse amygdules, averaging about. three 
feet thick. The lowest is near the water, and the 
second is separated from it by three feet and from the third 
by two feet of compact melaphyr. The exposed thickness 
of this flow is about 15 feet, and it is clear that we 
see only the superficial, vesicular part of it. Immediately 
ast of the dike (675, the second and third bands of amygdules 
