its final abandonment as a place of residence. Analogous remains occur in conjunction with some large ruins in the north, and there they represent single rooms, parts of the original structure kept in a fair state of preservation by occasional repairs while the remainder of the village was going to ruin, and used as farming outlooks long after the site was abandoned as a place of residence. As these farming outlooks have been discussed at some length in another paper10 it is not necessary here to enlarge upon their function and the important part they play in Pueblo architecture. If the high mounds in question mark, as supposed, the sites of farming outlooks such as those which are found in the north, they indicate that the occupancy of the region in which they occur was continued after the abandonment of the Casa Grande structure by the people who built it or by people of similar habits and customs.

Fig. 329.—Map of hollow mound.

An inspection of the map will show a number of depressions, some of quite large area, indicated by dotted contour lines. The principal one occurs a little west of the center of the area, and is worth more than a passing notice since similar structures are widely distributed throughout this region. It may be roughly characterized as a mound with excavated center. The ground for some distance about the structure (except for two depressions discussed later) is quite flat. From this flat surface as a base the structure rises to a height of 5 feet. From the exterior it has the appearance of an ordinary mound, but on reaching the top the interior is found to be hollowed out to a depth which even at the present day is below the surrounding surface, although not below the depressions adjoining. The main structure or mound is shown in figure 329 (an enlargement from the map). It measures on top of the crest 150 feet from north to south and about 80 feet from east to west, but covers a ground area of 200 feet by 120 feet or over half an acre. The crest is of the same height throughout, except for slight elevations on the eastern and western sides and a little knoll or swell in the southwestern corner. There is no indication of any break in the continuity of the crest such as would be found were there openings or gateways to the interior. The bottom of the depression in the main structure is at present about a foot below the surrounding ground surface, but it must have been originally considerably more than this, as the profile indicates long exposure to atmospheric erosion and consequent filling of the interior. No excavation was made and the character of the construction can not be determined, but the mound is apparently a simple earth structure—not laid up in blocks, like the Casa Grande ruin.

PLATE LIII

GENERAL VIEW OF CASA GRANDE.

To the east and to the west are two large depressions, each about 5 feet below the surrounding ground surface, evidently the places whence the material for the construction of the mound was obtained. Yet the amount of material removed from these excavations must have been considerably in excess of that used in the construction of the mound, and this excess was doubtless utilized in neighboring constructions, since it is hardly to be supposed that it was carried away to any considerable distance.

The purpose of this hollow mound, which is a fair type of many similar structures found in this region, is not clear. Mr. Frank Hamilton Cushing, while director of the Hemenway southwestern archeological expedition, found a number of these structures and excavated some of them. From remains thus found he concluded that they were sun-temples, as he termed them, and that they were covered with a roof made of coiled strands of grass, after a manner analogous to that in which pueblo baskets are made. A somewhat similar class of structures was found by the writer on the upper Rio Verde, but these were probably thrashing floors. Possibly the structure under discussion was for a similar purpose, yet its depth in proportion to its size was almost too great for such use. The question must be left for determination if possible by excavation.

In the southern central part of the map is shown another excavation, covering a larger area than any of the others, of very irregular outline and from 3 to 4 feet deep. It is apparently older than the others and probably furnished the material for the house structures northeast and southwest of it. Bordering the depression on the south there are some low mounds, almost obliterated, which probably were the sites of other house structures.

Scattered about the area shown on the map there are several small depressions, usually more regular in outline than those described. The best example is situated near the northeastern corner of the area. It is situated in the point of a low promontory, is about 3 feet deep, almost regularly oval in outline, and measures about 50 by 100 feet. A similar depression less than 2 feet deep occurs near the northwest corner of the area, and immediately south of the last there is another, more irregular in outline, and nearly 3 feet deep. There are also some small depressions in the immediate vicinity of the Casa Grande ruin and of the mounds north of it.

With a single exception none of these depressions are so situated that they could be used as reservoirs for the storage of water collected from the surface, and the catchment area of the depressions is so small and the rate of evaporation in this area so great that their use as reservoirs is out of the question. It is probable that all of the smaller depressions represent simply sites where building material was obtained. Possibly the ground at these points furnished more suitable material than elsewhere, and, if so, the builders may have taken the trouble to transport it several hundred yards rather than follow the usual practice of using material within a few feet of the site. This hypothesis would explain the large size of the depressions, otherwise an anomalous feature.

CASA GRANDE RUIN.

STATE OF PRESERVATION.

The area occupied by the Casa Grande ruin is insignificant as compared with that of the entire group, yet it has attracted the greater attention because it comprises practically all the walls still standing. There is only one small fragment of wall east of the main structure and another south of it.

The ruin is especially interesting because it is the best preserved example now remaining of a type of structure which, there is reason to believe, was widely distributed throughout the Gila valley, and which, so far as now known, is not found elsewhere. The conditions under which pueblo architecture developed in the north were peculiar, and stamped themselves indelibly on the house structures there found. Here in the south there is a radical change in physical environment: even the available building material was different, and while it is probable that a systematic investigation of this field will show essentially the same ideas that in the north are worked out in stone, here embodied in a different material and doubtless somewhat modified