In this peaceful and industrious life they were exposed, however, to raids from the nomadic and more savage tribes of the moun tains and plains north and east of them, and hence were compelled to concentrate their homes into defensive villages, to build fort-like walls and erect watch-towers along their fron tiers and on surrounding heights. These struc tures, as indicated by their ruins, appear essen tially lilce those now occupied in New Mexico and Arizona. The habits and general culture of these prehistoric people, or peoples, were apparently much the same as those of their descendants when first visited by Spanish ex plorers in the 16th century.
The walls of the canyon-valleys were com posed of sandstones in strata of varying hard ness, so that in many places the wearing away of a softer layer left a long horizontal recess, sometimes several yards in depth, overhung by harder rock. These natural shelters were some times low down and easily accessible, and were naturally taken advantage of as good situa tions for storehouses for grain and for resi dence. Often there was room for only one or two houses, which could be built economically, as little more was required than a front and one or more side walls, enclosing a space of the ledge-floor and roofed by the overhanging rock. Such single houses were often found occupying niches hundred.s of feet above the valley-floor. In several places, however, the recesses, or shallow caves, were of sufficient length and depth to accommodate a large num ber of houses, forming a real village, with granaries, lcivas and protective front walls. The most extensive of these towns yet known is that in Walnut Canyon, a tributary of Mon tezuma Creek, Colorado, called Cliff-Palace. This contained about 100 rooms. The terrace had been leveled and extended by building sup porting walls along its irregular front, and long and prosperous occupancy is evident. This and another remarkable ruin nearby called Spruce Tree House, lately cleared of rubbish, are now national reservations and in the custody of the Smithsonian Institution. Such reservations also ensure the continued safety of similar remains in various places elsewhere. See CASA GRANDE, etc.
the construction of these houses is described in the Handbook of American as follows: "In the large shelters the buildings are much diversified in plan and elevation owing to the irregularities in the conformation of the floor and walls. The first floor was the rock surface, or, if that was uneven, of clay and flagstones, and upper floors were constructed of poles set in the masonry, often projecting through the walls and overlaid with smaller poles and willows, finished above with adobe cement. The masonry is excellent, the rather small stones, gathered in many cases from dis tant sites, being laid in mortar. The stones were rarely dressed, but were carefully selected, so that the wall-surface was even, and in some cases a decorative effect was given by alternat ing layers of larger and smaller pieces and by chinking the crevices with spalls. The walls
were sometimes plastered inside and out and finished with clay paint. The doorways were small and squarish, and often did not extend to the floor, except an opening or square notch in the centre for the passage of the feet. The lintels were stone slabs, or consisted of sticks or small timbers. Windows or outlook-aper tures, were numerous and generally small. . . . Where the way is very steep niche-stairways were cut in the rock-face, making approach possible. Ladders of notched logs were also used.* In some parts of this district, as, especially on the eastern side of the Jemez Plateau, in New Mexico, great numbers of dwelling-places have been dug out of the rock, or crevices have been enlarged and the front walled up. Ari zona and Mexico show these also. They are associated with the cliff-houses, but are more properly classified and described elsewhere.
The age of these ruins, by whom built, and when and why they were abandoned, are matters of conjecture, with few substan tial bases for hypothesis. They certainly antedate the coming of white men, for no objects of metal nor any evidences of domestic animals have been found. Esti mates as to antiquity vary from 500 to sev eral thousand years— the latter probably very excessive. Many places have evidently been occupied within comparatively recent times, and 1,000 years seems enough to allow to the oldest. The best opinion is that pressure by enemies led to the making of these inconvenient habitations, and that this with the increasing dryness of the climate (of which other evi dence exists) finally compelled retreat. Speak ing of the Mesa Verde district, Dr. J. Walter Fewkes, of the bureau of ethnology, who is perhaps the best informed man in the country on this subject, expresses the subjoined opinion: "The inhabitants of these buildings strug gled to gain a livelihood against their unfavor able environment until a too exacting nature finally overcame them. . . . One of the primary reasons was change of climate, which caused the water-supply to diminish and the crops to fail; but long before its final deser tion many clans abandoned the place and drifting from point to point sought home-sites where water was more abundant. . . . Where the descendants of Cliff Palace now dwell, or whether they are now extinct, can be determined only by additional research.* Bibliography.— Annual Reports United States Geological Survey (Hayden's, 1874, 1876); Reports Bureau of Ethnology, and Bul letins, especially Nos. 30, 41, 50, 51; Hodge, 'Handbook of American Indians' (Washing ton 1907); Nordenskiold, 'Cliff-dwellers of the Mesa (Stockholm 1893).