Russell] CORRUGATIONS AND ARCHES. 95 
J. D. Dana as occurring in the lava flows of the Hawaiian Islands, 
and the explanation is suggested that they are due to the force of steam 
or gases generated beneath a still plastic crust. This explanation is 
not favored in the instances observed about the Cinder Buttes, although 
it might seemingly apply to the larger hollow ridges or elongated 
domes if the}^ did not form part of a series. The transition from 
simple corrugations to those having cavities in them, and from these 
to arches with coves beneath, is complete, and includes also the ridges 
with sloping roofs which have cracks along the top. The generation 
of steam or gases beneath a crust can not be accepted as explaining 
the origin of either the earlier or later members of the series and is 
unnecessary for explaining the origin of the intermediate forms. 
That steam may cause arches to rise in the manner referred to by 
Dana may be true, but it does not seem to have been the cause of the 
arching seen in the lava streams about the Cinder Buttes. 
The ridges with raves beneath, as stated, are frequently from 15 to 
20 feet across and rise from 5 to 10 feet above the bordering surfaces; 
their arched roofs arc between 1 and 2 feet thick. Still larger swells 
on the surface of the pahoehoe lava were observed, but as they were 
unbroken, it was not evident that openings occur beneath them, and 
it is possible they are hardened streams of lava instead of pressure 
ridges. 
It is but a step from the larger ridges on the fresh lava streams, 
just described, to similar but frequently much larger ridges, which 
form a characteristic feature of many portions of the older sheets of 
Snake River lava. The forms referred to are dome-like ridges, com- 
monly from 10 to 30 feet high, 20 to 70 feet wide, and from 50 to pos- 
sibly 500 feet long, which are usually cracked open at the top through- 
out their length, the cracks being open fissures, usually 3 to 4 feet wide, 
but sometimes as much as 10 feet (PI. XV). The fissures referred 
to narrow downward, and the basalt forming their walls is frequently 
columnar, the columns being at right angles to the outer surfaces of 
the ridges. Where these ''cracked ridges," as I have termed them in 
my notes, are in groups, as is frequently the case, their longer axes are 
nearly parallel, but this is not an invariable rule. In some instances 
the sides of the ridges, instead of forming an arch in cross section, 
as is commonly the case, present flat slopes like the roof of a house, 
with an open fissure along the top. The steeply inclined sides of these 
A-shaped ridges frequently carry well-defined corrugations, which were 
formed when the surface of the cooling lava was horizontal. 
The similarity between the large ridges on the surfaces of the older 
sheets of the Snake River lava and the hollow folds, corrugations, 
etc., observed at the Cinder Buttes, leaves no room for doubting that 
