
a 
Parr I. Szcr. i. § 3.) FISSURE ERUPTIONS. O55 

Volcanic islands, unless continually augmented by renewed 
eruptions, are attacked by the waves and cut down. The examples 
of Graham’s Island and Sabrina above cited show how rapid this 

Fig. 62. View or Sr. Pau Isuanp, Inpian OCEAN, FROM THE Hast (Carr. 
BiLack woop 1N ADMIRALTY CHART). 
a, Nine-pin Rock, a stack of harder rock left by the sea; b, entrance to crater lagoon 
(see Fig. 60); ¢, d, e, cliffs composed of bedded volcanic materials dipping towards 
the south, and much eroded at the higher end (c) by waves and subaerial waste ; 
f, southern point of the island, likewise cut away into a cliff. 
disappearance may be. The Island of Volcano has the base of its 
slopes truncated by a line of cliff due to marine erosion. The 
island of Teneriffe shows in the same way that the sea is cutting 
- back the land towards the great cone (Fig. 61). The island of St. 
Paul (Figs. 60, 62) brings before us in a more impressive way the 
tendency of volcanic islands to be destroyed unless replenished by 
continual additions to their surface. At St. Helena lofty cliffs of 
voleanic rocks 1000 to 2000 feet high bear witness to the enormous 
denudation whereby masses of basalt two or three miles long, one or 
two miles broad, and 1000 to 2000 feet thick, have been entirely 
removed,’ — 
ii. Fissure (Massive) Eruptions. 
Under the head of massive or homogeneous volcanoes some 
geologists have included a great number of bosses or dome-like 
projections of once-melted rock which, in regions of extinct volcanoes, 
rise conspicuously above the surface without any visible trace of 
cones or craters of fragmentary material. They are usually re- 
garded as protrusions of lava, which, like the Puy de Dome in 
Auvergne, assumed a dome-form at the surface without spreading 
out in sheets over the surrounding country, and with no accompanying 
fragmentary discharges. But the mere absence of ashes and scorize 
is no proof that these did not once exist, or that the present knob 
or boss of lava may not originally have solidified within a cone 
of tuff which has been subsequently removed in denudation. 
The extent to which the surface of the ground has been changed 
by ordinary atmospheric waste, and the comparative ease with which 
loose volcanic dust and cinders might have been entirely removed 
require to be considered. Hence, though the ordinary explanation 
is no doubt in some cases correct, it may be doubted whether a large 
proportion of the examples cited from the Rhine, Bohemia, Hungary, 
1 Darwin, “ Volcanic Islands,” p. 104. 
