104 J. D. Dana— Volcanic Action. 
have happened since illustrate the same fact. Similar testi- 
mony comes from other volcanoes. If the ascent in the upper 
part of the volcanic conduit is aided by some additional up- 
thrust force, then the facts just stated make the deep-seated 
action by so much the slower, since in that case it is accounta- 
ble only for a part of the whole rise required for an eruption. 
4. Superficial ascensive action.—The deep-seated ascensive 
force in a lava-conduit should raise the contained liquid rock 
to a common level for all its places of discharge in acrater. It 
was to mea strange fact that the lava-pools over the bottom of 
Kilauea underwent independent changes of level—a sinking of 
a hundred feet or more in one, unaccompanied by a like change 
orany changein another. But this admits of explanation on the 
ground of occasional discharge by local fissures opened beneath, 
and on the ground of much friction along the local conduits, 
and a depth to the common junction as great perhaps as the 
space between the pools (half a mile or so in the great pit crater). 
It was more of a marvel that while the pools were in their 
usual state of ebullition, lavas should be lifted at times toa 
level 500 feet or more above them in a fissure intersecting the 
walls of the crater. Here was an effect of ascensive force, but 
of a force of relatively superficial origin. A similar fact is 
mentioned by Prof. Alexander in his notes on the summit 
erater Mokuaweoweo, cited on page 235 of volume xxxti (1886) 
of this Journal, where the lift was 800 feet above the bottom of 
the crater, and the eruption a large one. Some superficial 
agency must be concerned in such eruptions; and the same may 
occasion local oscillations in associated lava-vents. 
5. Surface action in the lava; Projectile effects—In an active 
volcanic vent the feeblest action possible is the quiet escape of 
vapors. ‘I'he second stage is the aerial projection of lava in 
portions, masses, or fragments; which fragments may fall back 
as cooled cinders or as liquid lava, according to the height of 
projection. The latter stage may.be the final one before the 
return to a feeble or an inactive state in the crater. But the 
third may follow, namely: the outflow. As long as the liquid 
lavas are exposed in the bottom of the crater, aerial ejections 
are possible and probable; and if sunk below the bottom, they 
may continue provided there are cavities large enough to admit 
of the action. This projectile force is evidently distinct in ori- 
gin from the supplying force. 
2. Kilauea. 
6. Character of the lavas of projectile ejections and of those flow- 
ing from the open vents.—The lavas of projectile ejections when 
cooled are usually scoriaceous and often semi-glassy, and the 
