the west twenty-four feet six inches, and the north side, which is irregular and oblique, eighteen feet. The orientation is N. 40° W.* The floor averages five feet below the surface of the ground.

In these excavations charred wood, ashes, and other evidences of fire were found. The most important discoveries made in this room were a human skull and other bones, which were found four feet six inches below the surface in the middle of the chamber, directly under the place where the old sky-hole formerly opened, through which the relentless Hopi may have thrown down the burning fagots and chile upon their helpless victims. After the Indians had found these bones, which en passant it must be said they would not touch with their hands, one of their number, whose father-in-law is In'-ti-wa, the ka-tci'-na moñ'-wi, returned to Walpi that night. The next day a curious custom was carried out, by the advice of In'-ti-wa. Several na-kwa-ko-ci, strings with feathers attached, were deposited in the trenches as propitiatory offerings to Ma'-sau-wuh, the Death God of whom I have elsewhere spoken. Observing the anxiety of the Hopi workmen, I abandoned excavations in the po-wa'-ko kib-va from policy, for I did not wish the report to be circulated among their people that I desired to find the skeletons of the wizards, as it might prejudice them against me. The number of bones of human beings thus far taken from this excavation is too small to answer the requirements of the legend, but the mass of earth still filling this chamber is great enough to cover many more. New excavations alone can shed light on this question.

I believe that I have excavated the po-wa'-ko or sorcerer's kib-va, and tradition supports this identification. Still I have not proven that to be the case, although the discovery of human bones supports the legend. It seems pretty true that a tragedy took place in this kib-va, for there is no evidence that the bodies were buried there, but whether this chamber is the original po-wa'-ko kib-va or not will always be in doubt. †

The Mission. The walls standing above ground belong to the Mission and mark the most recent portion of the pueblo. In addition to the standing walls, there are others which undoubtedly belong to the period of the Spanish occupation, and still others of adobe.

*For the orientation of the Walpi estufas see my article on Tusayan ceremonials, Jour. of Amer. Eth, and Arch., vol. ii, No. 1. It is interesting to find that the orientation of this chamber corresponds so closely with that of the kib-va of Walpi.

The Indian workmen called this chamber a sorcerer's kib-va, and its relative position adds weight to the acceptance of the above identification, but I am not sure that the Indians regarded it as an estufa before my excavations.

which may have been added after the destruction of A-wa'-to-bi. The accompanying chart of the ground plan (Pl. II) gives all the data necessary to know the size and altitude of the standing walls. No attempt was made to excavate within the inclosure of the Mission. Mindeleff, in his description of A-wa'-to-bi, gives a fair account of the Mission, and my map of the ground plan supplements his measurements. The differences in our plans may be due to changes since he made his survey-falling of walls, drifting of sand, and other causes.

Old shrines near A-wa-to-bi.—Of the several shrines in the vicinity of A-wa'-to-bi there is one near the outline ruin to the west of the main pueblo, not far from the sand-hill burial place. This shrine faces to the east and is a rude pile of stones, in which is a piece of petrified wood and two water-worn bowlders, called the "Old Chiefs." It is not peculiar in any respect and no offerings were found near it. A figure of this shrine is given in plate III, figure 1. At the extreme west end of the A-wa'-to-bi mesa is a simple shrine composed of eight bowlders, with a reddish-colored stone, "Chief," within it. This shrine is figured in plate III, figure 2. A shrine which has evidences of recent as well as ancient use lies under a huge bowlder at the base of the extreme west end of the mesa, among the foot-hills. This shrine is more complicated than those mentioned above, and the stones which compose it are nicely joined together. The crescent of this shrine opens toward the south. It has been used in recent times by the Mi-coñ'-iñ-o-vi people, but many of the bung-shaped, squash and melon, ba'-ho(s) which cover the ground bear evidence of considerable age (Pl. III, figure 3).

The entrance to the shrine of the A-lo'-sa-ka faces south, and is situated about fifty feet below the old ruin at the most precipitous point of the mesa. It is a shallow chamber eroded in the side of a large bowlder, in front of the entrance to which stands a roughly laid stone wall only a few feet high (Pl. III, figure 4). A flat stone formerly closed it, and sticks of wood, now irregularly placed, show that it was once more of a structure. The floor was covered with soil and there were offerings of ba′-ho(s) in the shrine when I visited it. This is the shrine out of which the A-lo'-sa-ka were taken, but they were afterward returned to the Mi-coñ'-iñ-o-vi people, as I have elsewhere mentioned. * Rough sketches of the A-lo'-sa-ka were

*Jour. of Amer. Eth. and Arch., vol. ii, No. 1.

made when they were taken from the shrine, but the images themselves have not been seen since they were restored to the Indians.

Cemetery of A-wa'-to-bi.-The sand-dunes somewhat back from the mesa and to the west of the ruin of A-wa'-to-bi served as an aboriginal burial place in former times, and from it has been taken some of the best pottery found in Tusayan. These sand-hills are sparsely covered with stunted bushes, and their contour is continually changing on account of the drifting sand. Every sand-storm alters the surface, bringing into view the skeletons of the dead and fragments of food-vessels, some of which are covered with most interesting symbolic decorations. It was not always the habit of the A-wa'-to-bi people to "kill" these food-vessels on the interment of the dead, as many unbroken bowls have been found there. The dead were buried in a sitting posture, the knees drawn up to the breast.*

EXPLANATION OF PLATES.

PLATE I.-Map of ruin of A-wa'-to-bi. This map was plotted by Mr. A. M. Stephen during our visit to the ruin. The heavy black line indicates the edge of the mesa; the lighter lines the contours.

a. Highest point of the western portion of the ruin.

b. s. Site of the room in which the bear's skeleton was found.
c. Rooms with stacks of burnt corn.

e. Northeastern corner of ridge of rooms of the eastern division.
h. Site of northwestern corner of main ruin.

n. s. Narrow room.

p. Po-wa'-ko kib-va.

p. h. Room with fragments of pigment and niche with hair strings. w. s. Room with southern window.

The figures indicate the size of rooms excavated. Ground plans of buried chambers are to be found in all the higher mounds, but the more important are indicated by special lettering.

PLATE II.-Ground plan of the standing walls of the Mission, with measurements and altitudes.

PLATE III.-Old shrines near A-wa'-to-bi.

I. Shrine near the burial sand-hills.

2. Shrine at the extreme west end of the mesa.

3. Shrine with offerings, situated among the foot-hills at

the extreme west end of mesa.

4. Shrine of the A-lo-sa-ka.

*The Walpi people now bury their dead in the foot-hills to the south of the mesa. Food-vessels, with food, a planting-stick, and a twig, with feathers, are placed above the grave.