136 Mr. G. P. Scrope on the Formation of Craters, 
and unyielding*. So also during the Vesuvian eruption of 
1753, persons who ventured to the summit of the cone observed ~ 
jets of liquid lava thrown up from the surface of a mass which 
occupied the bottom of the crater, and conducted itself exactly 
in the manner of a liquid in ebullition. Spallanzani remarked 
a similar appearance within the great crater of Etna in 1788. 
In the volcano of the Isle of Bourbon, Bory de St. Vincent de- 
scribes a source of very liquid and glassy lava ceaselessly and 
somewhat tranquilly boiling over in concentric waves from the 
summit of a dome-shaped hillock composed of its overflowings. 
Circular form of Craters.—A consideration which has not, 
perhaps, been sufficiently adverted to by geologists speculating 
on the origin of volcanic craters, is the cause of their invariably 
circular or nearly circular figure. If I am right in attributmg 
their formation exclusively to aériform explosions, it follows that 
each is, in fact, simply the external orifice of a more or less cy- 
lindrical bore drilled through the pre-existent rocks by repeated 
discharges of highly expansive aériform fluids (probably for the 
most part steam) forcing their way upwards at some weak point ; 
and that it is to the equal pressure in all directions of the ex- 
panding fluid that the circular form of the section of this orifice 
is due,—the same cause, in fact, which gives a spherical form 
to a bubble of air or gas rising through water. Indeed the 
eruptive explosions must be considered as occasioned by the rise 
of a succession of enormous bubbles from a great depth in the 
fluid lava below. ach single explosion attests the bursting of 
* Dana, American Journal, 1850, vol. ix. p. 383. 
[Nore sy J. D. Dana.—I do not regard the origin of the crater of 
Kilauea essentially different from that of other craters. But there is this 
peculiarity, that the lavas have not in modern times, at least, overflowed 
the pit; and moreover the country around, neither in its height or slopes 
or scoria bears evidence of long-continued overflows. There is no cone 
about the crater. It is a pit, which certainly overflowed at first, but for a 
long period has discharged itself by lateral fissures. There are several 
other large pit-craters in the vicinity of Kilauea which are without any 
cones or slopes around them, being literally pits; they once contained 
boiling lavas to their top like the small active pools in the bottom of Kilauea, 
but a subterranean opening discharged them, and left a deep pit with ver- 
tical walls like Kilauea. ‘The sides of the walls in such a case are as clear 
from scoria as a cliff of stratified limestone, because the undermining, owing 
to the drawing-off of the lavas, caused the sides to a certain distance around 
to fall from want of support, and so leave fresh fractures. I have attributed 
the origin of the Val di Bué (Bové) of Etna to the same cause that has pro- 
duced Kilauea, and I still believe the view right. There is, in a certain 
-sense, an “ engulfment;” and so there is in the eruptions of Vesuvius. 
Mr. Scrope writes as if I had described from the observations of others 
alone, and does not appear to have seen my Report on the Volcanoes of 
pe are in my Geological volume connected with the Exploring Expe- 
ition. ‘ 
