hinges, of which the socket-holes of the hinges and holes for the admission of the arm behind the door to draw back the bolt are the only traces now to be seen. This structure had an upper story, but much less extensive than the lower story. Many other caves, similarly constructed but containing fewer rooms, are found all along the valley. At one of them the cliff is so well preserved as to show how access was gained. It was by means of holes cut in the rock for steps, which could be easily climbed by the aid of a rope hanging down from above.
Vestiges of cave dwellings are very abundant in America, but they have not been made the subject of special study to so great an extent as those of Europe. They are prehistoric, ancient, or relatively modern, and represent various stages of civilization in those who inhabited them. Some are found as far north as Alaska, where, according to Dr. Peet, who has published in the American Antiquarian excellent illustrated summaries of the results of the explorations of the cliff and cave dwellings, "they are associated with shell-heaps; others in the Mississippi Valley, where they are closely connected with the mounds; others in the midst of the cañons of Colorado and Arizona, where they are associated with structures like the Pueblos; others in the central regions on the coasts of Lake Managua, in Nicaragua; and still others in the valley of the Amazon in South America." According to Mr. William H. Dall, the cave-dwellers of Alaska were neolithic. The caves in Tennessee are described by Prof. F. W. Putnam as containing tokens of a neolithic character; but it is uncertain whether they preceded the mounds or were contemporaneous with them. Dr. Earl Flint has described caves in Nicaragua which strike him as being very ancient; and certain caves in Brazil are supposed to be palæolithic.
The most interesting of the American cave dwellings, and those which have received the most attention, are those which are associated and almost confounded with the cliff dwellings of the cañons of Arizona and Colorado. So nearly related are the cliff and cave dwellings of this region, in fact, that it seems to have been to a considerable extent a question of the shaping of the rock whether the habitation should be one or the other. Regarding the two as a whole, they were very numerous, and indicate the former existence of a large population. Major Powell is quoted as having expressed surprise at seeing in the region nothing for whole days but cliffs everywhere riddled with human habitations, which resembled the cells of a honeycomb more than anything else. Yet it is probable that only a small fraction of these singular dwellings have been seen, while the number of those that have been even only superficially explored is much less.
An excellent, finely illustrated description of some of these
