036 
DE. TEMPEST AXDEESOX OX THE 
[XoY. 1910 . 
Then, again, there is the high temperature of the lavas and their 
power of remelting, dissolving, and carrying away previously solidi¬ 
fied lava, and even surrounding rocks. At Kilauea this has been 
described by various observers, and I myself saw the incandescent 
molten lava in the ‘ working pit ’ eating into the walls and forming- 
caves. Two of these, during my stay, undermined considerable 
portions of the so-called ‘ black ledge,’ that is the most recentlv 
solidified portions of the lava-lake. The roofs of these fell in and 
formed recesses scores of yards across, which really deserved the 
name of bays. Prof. Paly, of the proposed observatonq told me 
that there was a cave in the walls of the working crater above the 
black ledge and pit evidently formed in the same way, and I saw 
landslides which by all analogy had been due to the undermining 
of the foot of the walls and their consequent subsidence. The 
vertical or overhanging walls of this crater had apparently been 
left standing in this manner. The exposed -ends of the beds of 
lava were quite sharp except in places low down, where they 
showed signs of remelting, and there was no trace of ejected 
material round the lip of the crater. The same reasoning would 
apply to the landslides on the walls of the large encircling crater 
near the Yolcano House, and even, I think, to the curious pit- 
craters (so called) in the neighbourhood, such as Kilauea Iki and 
several others. 
Matavanu supplies abundant evidence of the same action. It is 
true that there is no old encircling crater with its landslides and sub¬ 
sidences ; but the working crater has Avails in many parts vertical 
or overhanging, and cracks about the lip shoAved that portions 
Avere ready to fall in, and rendered caution necessary in approaching 
the edge. The strata in the walls in the upper parts, Avhere not 
obscured by recent lava-splashes, were broken off sharply like those 
iu Ivilauea. Caves and tunnels were visible in the walls, and 
especially the great tunnel at the western end appeared to be strictly 
comparable to the caves in the black ledge of Kilauea, and not a 
tunnel of outflow like that at the other end of the crater. Frozen 
ledges on the Avail indicated former ledges of lava like the black 
ledge in Kilauea, and showed that the molten river was now 
flowing out at a loAver level than during an earlier phase of the 
eruption. The mouth of the tunnel into which the lava flows on 
leaving the crater appears to have undergone a corresponding lower¬ 
ing. The crater-wall above it is much less regularly stratified than 
the adjacent parts, and appears to consist of agglomerate, or perhaps 
partly of material frozen on from below. I saw this process 
going on in one of the caves of Kilauea, where stalactites 
were forming from the roof. This may be the place from which 
the great outburst of lava took place in February 1906 (see 
map, p. 622). 
This brings us to one of the most curious apparent differences, 
but perhaps real resemblances, between the two volcanoes. The 
lava in Kilauea, as is well known, is in constant motion; and, at first 
sight, it looks as if it were flowing, out at one end of the crater. 
This appearance is certainly deceptive : the direction of the flow and 
