178 GEOLOGY.
clay. They may be traced running in broken lines nearly east and west, or a little on
the south side of east and north of west.
The coral reefs along the coast are of two sorts ; one above low water mark, and in
which the animals are dead, is dark, cellular, and rough, similar to what generally forms
the rocky inland eminences ; the other is always covered by the sea, and generally presents
an arborescent surface of brown or white colour, and is a t this time occupied by the
living animals. B u t even these have a darker appearance than is usual in growing corals
among the islands nearer the equator, apparently from a quantity of mud and clay deposited
among the crevices, and which seems to be continually supplied from the soil and
the marly detritus of the island, washed down by rains and rivulets ; for I saw nothing
that was entitled to the name of a river, unless we may dignify with that appellation a
moderately-sized stream that comes down from the principal lime-stone ridge, a little to
the west of the village of Ishoomee, and of the building called Eepang-kwang.
The soil of the island in the vicinity of Nawha is in general light, arenaceous, and
marly. In a few places only does it approach to clay, and rarely is it of any considerable
depth. The rivulets are few in number ; and I observed no moist plains or marshes, excepting
a very few fields in which the water is retained by artificial means, chiefly among
the blue marly eminences to the south-east of Abbey Point, already mentioned.
Th e plants which grow wild are made up of the floras of the torrid and temperate
regions ; we have ferns and palms, compound and umbelliferous plants ; and the boer-
haavia, scoevola, toumefortia, &c., growing in similar situations as in tbe low islands of
Polynesia. Several rosaceæ, onagreæ, prunulaceæ, &c. associate the vegetation of
Loo Choo with that of the temperate continents of Asia, and even of Europe. The re markable
genus clerodendron is perhaps peculiarly abuudant here.
IS L A N D S OF T H E A RZOBISPO G R O U P ,
O F W H IC H T H E LA R G E S T IS P E E L IS L A N D ,
From lat. 26® 30' to 27® 45' N, ; long. 217® 48' W.
Th e Arzobispo Group, or Bonin Islands, are considerable, if we regard their number
alone ; we saw five groups, each composed of several islets. I f we take into account,
however, the superficial extent of land, and still more of productive soil, they
will be lessened in our estimation. They seem to be universally volcanic, if I may judge
of the similarity of those I saw at a distance to those which I had an opportunity of examining,
excepting some bare and bordering rocks of coral, raised most probably above
the level of the sea by subterranean power. The hills are peaked, the shores precipitous,
with deep water close to them ; but pointed pinnacles rise up from the bottom
nearly to the surface for some distance, and endanger the navigation.
The mountain ridges, as well as the groups of islands, lie in a north and south direction,
and appear to be only the more elevated tops of a subaqueous chain extending
in the same direction ; for in sounding we always found it shallower on a line connecting
the groups from north to south, than at the same distance from the shore on either
side of this line.
Coral reefs border some places of the shore, and have not yet reached the water’s
level; and corals, with sand and shells, from tbe bottom at a little greater distance.
The rocks o fth e more extensive volcanic formation are generally a grey tufaceous
basalt, often coloured with a greenish tin t; it contains numerous nodules of chalcedony
and of carnelian; zeolites often occur abundantly, particularly stilb ite ; olivine and
hornblende are also not uncommon. The geodes were often found containing w a te r;
and although frequently covered with the sea, the liquid, when obtained by breaking the
hollow stone, is said not to taste salt. I was not fortunate enough to see any o fit.
Angular basaltic columns were not unfrequent, and in one place they were divided
horizontally into joiuts at short distances, as at the Giant’s Causeway. In the bed of
the river at the bottom of tbe harbour in which we lay, there is a sort of tessellated
pavement, composed of upright angular columns placed side by side, each about one
inch diameter, and separated by horizontal fissures ; it resembled the lower part ofthe
Giant’s Causeway in miniature.
The two men who had voluntarily remained here after the wreck of the William
whaler in the month of September preceding our visit, informed us that they had felt
the shocks of earthquakes several times, but never so strong as to throw them off their
legs ; and that one night in the month of January, the tide suddenly rose about twenty
feet above its wonted level, overflowed their house, which was near the beach, and
made them fly to the mountains for safety. They said they had observed some of
the peaks, more especially after rainy weather, to vomit forth volumes of smoke like
that which arises from a foundry, but never any light, nor have they seen any ashes
falliug. Scorife are, however, pretty common near the surface. I t is possible that
these men mistook the light cloud formed near this elevated peak, by its cooling and
condensing the moisture of the passing air, for the denser smoke of a volcano. We saw
clouds forming in this way.
A luxuriant vegetation covers the larger of these islands, from the sea to their
highest peaks. Trees constitute nearly the whole of this vegetation, and the tall slender
trunks, crowned with the spreading bunches of leaves of the cabbage and fau-palms
(areca olearica and coryptia rotundifolia), communicate a characteristic feature to the
scenery. The pandanus odoratissimus, the tamanu of Tahiti, some species of laurel,
the terminalia, shrubby species of urtica, the dodonea viscosa, the el®ocarpus serratus,
and several others new to us, are common. A considerable variety of herbaceous plants,
and of ferns, also grow partly under the shade of the lofty trees, or freely exposed on the
jutting rocks.
PO R T C LA R EN C E , AND COAST B E TW E EN IT AND K O T Z EBU E ’s SO U N D .
P o rt Clarence is situated a t the end of an irregular and deep incurvation of the
coast, between Point Rodney and Cape Prince of Wales, aud in lat. 65° 17' N ., long.
166® 48' W. I t is formed of two wide basins; the outer one separated from the sea,
except at tbe entrance, by a low and long spit of alluvial formation, called Point Spencer
; and the inner basin, called Grantley Harbour, being divided from the outer by a