134 Mr. Dana on Areas of Subsidence in the Pacific, &c. 
Washington Island, (coral,) in lat. 5° N., is the last spot of land 
as we recede from our boundary line to the north-northeast. 
Beyond isa bare sea, to the Sandwich Islands. Is not this an 
area where the subsidence was too rapid for the corals to keep 
the islands at the surface ? 
It appears then that during this era, the Pacific from 30° N. to 
30° S.—and perhaps beyond—was one vast region of subsidence: 
that subsidence took place most rapidly over the bare area be- 
tween the Sandwich Islands and the equator, and less and less so 
as we go from this, to the south-southwest. At the boundary 
line pointed out, it was not sufficient to submerge many of the 
mountain summits, and south of this, the effect was still less. 
This area covers at least five thousand miles in longitude and 
three thousand in latitude. ‘The seas about the northwest coast 
of New Holland, show by their reefs, a contemporaneous subsi- 
dence, and they should probably be included, as well as some 
parts of the Hast Indies. Fifteen millions of square miles is not 
then an overestimate of the extent of the region that participated 
in this subsidence. 
The region of greatest subsidence lies nearly in a west-north- 
west line, for we may trace it along by Washington Island far 
towards the arctic coast. 'The whole broad area of subsidence 
has nearly the same direction ; for this is the course of the boun- 
dary line we have laid down as separating the high basaltic and 
the low coral islands. It is highly interesting to observe that the 
trend of the principal groups of islands in the Pacific, corresponds 
nearly with this course. ‘The Low or coral Archipelago, the Soci- 
ety Islands, the Navigators, and the Sandwich Islands, lie in the 
same general direction, nearly west-northwest and east-southeast. 
It should be remarked that the Sandwich group, does not contain 
merely the seven or eight islands usually so called; eight or ten 
others stretch off the line to the north; some, small rocky islets, 
and others, coral, and the whole belong evidently to one series. 
I will not say that there isa connection between the trend of 
these groups and the area of subsidence; yet it looks much like it. 
A further point may be worthy of consideration. 'The Sand- 
wich group consists of basaltic islands of various ages. ‘The isl- 
and at the northwest extremity, Tauai, is evidently more ancient 
than the others, as its rocks, its gorges and broken mountains, 
indicate. By the same kind of evidence it is placed beyond 
