kussell.] RECENT VOLCANOES. 51 
across. Some of the lava in this pool, however, came from the crack 
mentioned above, along which there is situated a row of driblet cones. 
After the opening of a breach in the west side of the crater the open- 
ing was partially closed by a discharge of lapilli, but not rebuilt to 
its former height. On the southeast side of the crater the portion of 
its wall which has disappeared includes about one-third of its original 
extent, and through the break there was outpoured a flood of lava 
which inundated an area of about 50 square miles. The lava cooled 
in the breach it had formed in thin sheets, so as to produce a steep- 
sided ridge, which completely occupies the break and unites with the 
remaining fragments of the lapilli and cinder cone at each end. This 
ridge of lava is of the same nature as the walls of typical lava cones, 
being formed of thin sheets of highly vesicular rock, each one seem- 
ingly formed by the congealing of a thin overflow from the crater and 
succeeded by another similar overflow, until a steep ridge composed 
of overlapping lava sheets was produced. The main discharge of lava 
took place beneath the crust thus formed, after the manner so common 
in highly liquid lava streams, and escaped through tunnels. 
The final drawing off of the molten material, which once filled 
Crater No. 4 up to the level of the top of the rim of congealed lava on 
its eastern side, allowed the crust of scoriaceous and rop}^ pahoehoe 
lava formed within it to subside, and although in part broken it still 
rests at the bottom of the irregular pit it left about 100 feet lower 
than its former position. Since the subsidence of the floor of the 
crater there has been some tumbling in of its unsupported walls, but 
the cakes of corrugated lava on its floor are still to a large extent 
unconcealed. 
The rise of liquid lava within a cinder and lapilli cone, as in the 
example just described, the building of a raised rim in the breach in 
the crater walls composed of fragmental material, and the subsequent 
lowering of the lava crust formed on the pool within the crater are all 
features which are found in numerous volcanoes in Idaho and Oregon 
which built lava cones. The greater number of the lava cones that have 
been examined, however, are situated on generally plain surfaces, 
and the floods of outwelling highly liquid rocks, which were extruded 
from them, spread in all directions, and the cinder and lapilli cones, 
which probably in all the instances referred to were formed at an 
early stage in the eruptions, were completely destroyed or buried. In 
the instance here considered, however, owing to the inclination of the 
surface on which the volcano originated, the lava extruded flowed 
almost entirely in one direction, leaving a large portion of the pre- 
ceding cinder and lapilli cone intact. No. 4 of the Jordan Craters is 
thus in part a cinder and lapilli cone and in part a lava cone. Had 
the slope down which the lava flowed been steeper it is probable that 
the liquid rock would have flowed away too quickly for a lava ridge 
to have been formed about the pool in the summit of the conduit from 
