Davis.] 172 [t)ec. 6, 
loosened by decomposition. Its surface was firm and clean. 
Similar irregularity of surface was noted in some nondescript 
crystalline stones at South Yarmouth, in which the knots of 
harder minerals maintained a somewhat higher surface than their 
matrix. Granitic and gneissoid stones are unevenly carved ; a 
gieater impression being made on the weaker minerals, thus pro- 
ducing a very characteristic appearance of the surface. A coarse 
gneiss from Succonnessett has a fluted surface on its facets, with 
rather sharply-marked edges between the flutings. All kinds of 
stones, however complex in structure, exhibit a facetting more or 
less clearly ; each facet being minutely carved in a way character- 
istic of the kind of stone that bears it. 
The edges between two facets vary with the perfection of 
facetting. On the quartzites, the edges of well-cut specimens are 
straight and sharp. Some specimens have two curved facets, 
iueeting in a convex ridge line ; but this is exceptional. The 
gneisses and granites have irregular edge lines. The repetition 
of these characteristic features in stones of the same kind is very 
striking. 
The greatest depth now reported at which facetted pebbles 
have been found in our glacial gravels is twenty-five feet, in the 
cliff at Sankaty Head at Nantucket, as described by Professor 
Shaler. This would lend mucli support to the theory that the 
sand and gravel plains of the Cape, of Long Island, and of the 
outlying islands, are to be i-egarded as confluent fan-deltas built 
up by streams issuing from the ice sheet, at times when its edge 
lay along the northern margin of the plains. They are therefore 
homologous with the ''sand?-" of C4reenland, described by Hoist, 
and the gravel-fans of Alaska, described by Russell and others. 
They were built up in front of the ice margin, assuming such a 
slope as was determined by tlie size of the streams and by the 
quantity and quality of their load. It is quite possible that they 
were built outward into the sea ; but the originally submarine 
portion of these delta-plains may not now be above sea level. It 
has not been certainly identified. 
The gravel ])lains of the Cape and of the outlying islands are 
somewhat dissected by shallow valleys. It has commonly been 
assume<l that the land must have risen from the position in which 
tlie plains were formed, in order that the streams could cut these 
valleys. Tfiis, however, does not seem absolutely necessar3^ It 
