steps at the floor levels, partly by exterior batter, the interior wall surface being approximately vertical. Some writers, noting the inclination of the outer wall surface, and not seeing the interior, have inferred that the walls leaned considerably away from the perpendicular. This inference has been strengthened, in some cases, by an examination of the interior, for the inner wall surface, while finely finished, is not by any means a plane surface, being generally concave in each room; yet a line drawn from floor level to floor level would be very nearly vertical. The building was constructed by crude methods, thoroughly aboriginal in character, and there is no uniformity in its measurements. The walls, even in the same room, are not of even thickness, the floor joists were seldom on a straight line, and measurements made at similar places, e.g., the two ends of a room, seldom agree.

PLATE LV

WEST FRONT OF CASA GRANDE RUIN

A series of precise measurements gives the following results: Outside eastern wall, at level 3 feet above center of depressed area adjoining the ruin on the east, 59 feet; western wall at same level, 59 feet 1 inch; northern and southern walls, at same level, 42 and 43 feet respectively. These measurements are between points formed by the intersection of the wall lines; the northeastern and southeastern corners having fallen, the actual length of standing wall is less. At the level stated the northern wall measures but 34 feet 4 inches, and the southern wall 36 feet 10 inches. A similar irregularity is found in the interior measurements of rooms. The middle room is marked by an exceptional departure from regularity in shape and dimensions. Both the east and west walls are bowed eastward, making the western wall convex and the eastern wall concave in reference to the room.

Precise measurements of the middle room at the second floor level, 8 feet above the base previously stated, are as follows: Eastern side, 24 feet 8½ inches; western side, 24 feet 2 inches; northern side, 9 feet 3½ inches; southern side, 9 feet 1 inch. The eastern room is a little more regular, but there is a difference of 11 inches between the measurements of the northern and southern ends. A similar difference is found in the western room, amounting there to 6 inches. The northern and southern rooms do not afford as good bases for comparison, as a corner is missing in each; but measurements to a point where the interior wall surfaces would intersect if prolonged, show variations of from 6 inches to a foot. The statement that the ruin exhibits exceptional skill in construction on the part of the builders, is not, therefore, supported by facts.

DETAILED DESCRIPTION.

The Casa Grande ruin is often referred to as an adobe structure. Adobe construction, if we limit the word to its proper meaning, consists of the use of molded brick, dried in the sun but not baked. Adobe, as thus defined, is very largely used throughout the southwest, more than nine out of ten houses erected by the Mexican population and many of those erected by the Pueblo Indians being so constructed; but, in the experience of the writer, it is never found in the older ruins, although seen to a limited extent in ruins known to belong to a period subsequent to the Spanish conquest. Its discovery, therefore, in the Casa Grande would be important; but no trace of it can be found. The walls are composed of huge blocks of earth, 3 to 5 feet long, 2 feet high, and 3 to 4 feet thick. These blocks were not molded and placed in situ, but were manufactured in place. The method adopted was probably the erection of a framework of canes or light poles, woven with reeds or grass, forming two parallel surfaces or planes, some 3 or 4 feet apart and about 5 feet long. Into this open box or trough was rammed clayey earth obtained from the immediate vicinity and mixed with water to a heavy paste. When the mass was sufficiently dry, the framework was moved along the wall and the operation repeated. This is the typical pisé or rammed-earth construction, and in the hands of skilled workmen it suffices for the construction of quite elaborate buildings. As here used, however, the appliances were rude and the workmen unskilled. An inspection of the illustrations herewith, especially of plate LV, showing the western wall of the ruin, will indicate clearly how this work was done. The horizontal lines, marking what may be called courses, are very well defined, and, while the vertical joints are not apparent in the illustration, a close inspection of the wall itself shows them. It will be noticed that the builders were unable to keep straight courses, and that occasional thin courses were put in to bring the wall up to a general level. This is even more noticeable in other parts of the ruin. It is probable that as the walls rose the exterior surface was smoothed with the hand or with some suitable implement, but it was not carefully finished like the interior, nor was it treated like the latter with a specially prepared material. The material employed for the walls was admirably suited for the purpose, being when dry almost as hard as sandstone and practically indestructible. The manner in which such walls disintegrate under atmospheric influences has already been set forth in detail in this report. An inhabited structure with walls like these would last indefinitely, provided occupancy continued and a few slight repairs, which would accompany occupancy, were made at the conclusion of each rainy season. When abandoned, however, sapping at the ground level would commence, and would in time level all the walls; yet in the two centuries which have elapsed since Padre Kino's visit—and the Casa Grande was then a ruin—there has been but little destruction, the damage done by relic hunters in the last twenty years being in fact much greater than that wrought by the elements in the preceding two centuries. The relic hunters seem to have had a craze for wood, as the lintels of openings and even the stumps of floor joists have been torn out and carried away. The writer has been reliably informed that as late as twenty years ago a portion of the floor or roof in one of the rooms was still in place, but at the present day nothing is left of the floors except marks on the vertical walls, and a few stumps of floor joists, deeply imbedded in the walls, and so high that they can not be seen from the ground.

PLATE LVI

INTERIOR WALL OF CASA GRANDE RUIN

The floors of the rooms, which were also the roofs of the rooms below, were of the ordinary pueblo type, employed also today by the American and Mexican population of this region. In the Casa Grande ruin a series of light joists or heavy poles was laid across the shorter axis of the room at the time the walls were erected; these poles were 3 to 6 inches in diameter, not selected or laid with unusual care, as the holes in the side walls which mark the places they occupied are seldom in a straight line, and their shape often indicates that the poles were quite crooked. Better executed examples of the same construction are often found in northern ruins. Over the primary series of joists was placed a layer of light poles, 1½ to 2 inches in diameter, and over these reeds and coarse