slide in the White Mountains in 1826, and the still greater one which descended into Biraki Ganga River, near Gohna, India, in 1893, and gave origin to Gohna Lake, are instances of this nature. A large number of landslides, however, are due to geological rather than climatic causes, and it is to this class that I wish to invite attention.
It frequently happens that a layer of hard rock rests on softer or more easily soluble beds. When steep escarpments are formed of two strata having this arrangement, conditions are produced which favor the breaking off of masses of the hard upper layer and their descent to the foot of the escarpment in landslides. There are other conditions, such as the thickness of the hard layers and the manner in which they are jointed, and the consistence of the soft beds—whether slippery clays, loose volcanic lapilli, etc.—which modify the process. The main or controlling conditions referred to are furnished at hundreds of localities in the region occupied by what is known as the Columbia lava, and drained by Columbia River. In that region sheets of basalt, ranging from two to five hundred feet and more in thickness, alternate with or overlie sheets of clay, shale, volcanic tuff, etc., which in many instances are hundreds of feet thick. The basalt was poured out in a molten condition and spread over the land in horizontal sheets. On cooling, these sheets acquired a columnar structure due to joints, usually at right angles to the top and bottom surfaces. This columnar structure facilitated the breaking away of great masses of basalt, when, for various reasons, portions of the sheets form the summits of bold escarpment.
In part, the Columbia lava, and the softer beds interleaved with it, have remained undisturbed, and are now practically horizontal over thousands of square miles. In adjacent areas of a great extent the beds have been broken by extensive fractures and the blocks thus formed variously tilted. Over still other extensive regions, particularly in the Cascade Mountains, the originally horizontal sheets, aggregating several thousand feet in thickness, have been raised into dome-shaped uplifts, at least one of which is nearly circular, while others are more nearly elliptical, the major being several times as long as the minor axis. Some of these domes, if unaffected by erosion, would have an altitude above their immediate bases of from five to eight thousand feet.
Where the Columbia lava is still essentially horizontal, as in southeastern Washington, it has been dissected by streams which now flow in magnificent cañons with clifflike walls from a thousand to four thousand feet high. In the region referred to as having been extensively fractured, cliffs have been produced by the tilting of the blocks thus produced. The dome-shaped uplifts have been broken by weathering and by the work of streams in such a manner as to
