364 The A»icrican Geologist. June. i898 
good evidence that the rate of denudation is exceedingly slow 
in a forest covered country, and that the land is and has been 
far from stable, when long periods of time have been involved. 
Alternate Hypotheses. — Two hypotheses have been sug- 
gested to explain the facts considered above: (i) that of 
marine denudation; (2) that of subaerial denudation. Al- 
though American geologists in general consider marine de- 
nudation possible only in rather restricted areas, this hypothe- 
sis is certainly not an improbability. (So, also, subaerial de- 
nudation, if continued long enough, with land level maintained 
somewhat uniformly throughout, would vmdoubtedly reduce 
any area to a level condition, though the places most likely to 
be so reduced are those near the sea, or those in which the 
rocks are soft, or the elevation slight. The possibility of these 
two causes for reduction is not questioned, although the prob- 
ability of general reduction by such causes is called in ques- 
tion. This doubt is still further strengthened by the belief that 
the evidence of ancient peneplains, upon which the entire ar- 
gument of former base-leveling is founded, is far from con- 
vincing. 
Summarizing this evidence it seems that the ancient plain 
is constructed upon the basis of the existence of moderately 
uniform hill crests, whose total area is not more than 10 to 25 
per cent of the entire area. liven among these, the hill tops 
reach to considerably dififerent elevations. Of the remair^^- 
ing 75 to 90 per cent the greater part is sunk below this upland 
level. The hill tops are mainly of hard rock, while the valleys 
are mainly located in the areas of less resistant beds, and the 
streams are, in general, in quite close accord with the rock 
structure. There are. moreover, numerous localized eleva- 
tions, called monadnocks, reaching well above the upland 
level; and, in the western and northern portions of New 
England, at no very great distance from the coast, the region 
is elevated and very irregular and mountainous, as for instance 
in Maine, the White mountains. Green mountains, Berkshire 
hills and Adirondacks. These greater elevations do not cor- 
respond with marked differences in rock structure. The ad- 
vocates of the peneplain admit past irregularity, in the form of 
monadnocks rising above the ancient peneplain, and they 
also admit present irregularity in a marked degree, but ex- 
