Apr., 1890 . 
OCEANIC ISLANDS. 
89 
our eyes. Tlie island of Hawaii is nearly one hundred miles 
in diameter, and contains the bulkiest, if not the loftiest, 
volcano in the world. The two masses of Manna Loa and 
Mauna Kea, with the subsidiary height of Hualalai and the 
crater of Kilauea, form a mass of black lava rock, rising by 
an extremely gradual slope to a height of over 18,000 feet 
above the sea; while the foundations below the sea level are 
correspondingly great, as that portion of the Pacific is one of 
the deepest areas. A chain of islands stretches away to the 
north-west, all mountainous, but exhibiting more traces of 
denudation the greater their distance from the present seat 
of activity in Hawaii. The action of the Hawaiian volcanoes, 
though not absolutely regular, is constant and invariable on 
the whole. At intervals an eruption occurs, never of stones 
or ash, but always of very hot fluid lava, which rushes down 
the side of the mountain to the sea, and leaves a permanent 
addition to the island. After a comparatively brief period 
this lava has sufficiently decomposed on the surface to allow 
of the spreading of the forest, and thus the lava streams are 
successively obliterated, and the general aspect of the 
island remains unchanged. When the period of activity has 
ceased, as it has in the other islands of the group, the 
scenery is broken by picturesque precipices and jagged hills, 
on which the forest flourishes undisturbed. It thus appears 
that the Hawaiian Archipelago is the result of an indefinitely 
vast succession of ejections of lava from a vent in the floor 
of the Pacific, and there is no indication of any junction 
with either Asia or California. Nor, from the appearance of 
the plants, would any such connection appear to Lave been 
possible ; and this brings me to the second portion of the 
subject. 
Up to this point I have indicated some of the phenomena 
which have led up to the induction that the origin of these 
volcanic islands may be assumed to be entirely sub-oceanic, 
and that no other hypothesis is either needed or admissible. 
I now pass on to show that this hypothesis receives a very 
remarkable confirmation from the plants and animals found 
on such islands, and also that it helps us very considerably in 
solving the problems of geographical distribution and its 
history in the geological part. Leaving Hawaii for the 
moment I must give you a brief account of some circum¬ 
stances connected with another island. Just off Cape San 
Roque at the extreme eastern point of South America, about 
120 miles from the coast and a day’s voyage from Pernam¬ 
buco, there lies the rocky islet of Fernando de Noronha. 
This little spot—it is not more than seven miles long by two 
