abandoned; a fourth, 43 ft. from Building No. 1 and 18 ft. from pillar of subterranean apartment, went down 5 ft. in sand.. I then set all my men to work cleaning out the passage, and finally succeeded in opening it completely. It is only a room, and a very small one at that. Its rough dimensions (for accurate figures vide ante measures at Mitla) are: Length, 9 ft. 6 in.; height, 4 ft.; width, 3 ft. 7 in. Roof of very large stones. South end once closed by a large stone still lying in the large patio (vide plan). Floor hard polished cement. Walls ornamented in the usual mosaic pattern in a band 1 ft. 10 in. wide, 1 ft. and 1 in. from the floor. Stones forming pattern much decomposed by dampness. I broke through the floor to find solid rock underneath. In fact a great hole was dug in the rock and this chamber constructed therein. The approach is by a flight of three very rough steps hewn out of the living rock, each 1 ft. high. The passageway is a chamber which from its size and position seems to have been simply a sepulchral chamber. If I may be permitted an hypothesis, I should say this chamber served as a grave or vault for the remains of great dignitaries, either chiefs in war, rulers or priests, and that when a ruler died, his bones were collected and deposited in the rooms of the "cruzero." I am now confident that the "30 leagues" of Burgoa and the "88 feet by 26" of Mr. Mill are both wrong.

Leaving the South Group we go to the northeast and reach the finest building left in Mitla. It occupies the north side of a "patio" or courtyard rather larger than that of the South Group. On the east side all that remains of the building once there is a portion of the centre doorway, still holding up two of the huge monolithic lintels; the third lies prone, but entire, at their foot, and within lie two large columns 54 ft. apart. The west and south sides are mere shapeless mounds. The North

Building is a very large construction containing five rooms, a central court, and a passageway. The first room one enters is that marked M on the plan. It is a noble apartment 125 ft. long and 23 ft. wide, and its most striking feature is the row of six large columns it contains. This room has as usual three doors with the usual three monolithic lintels

and two jambs. It is perfectly plain, save the niche. The floor is well covered with cement. The columns are arranged from east to west, they do not run in a line but are eccentrically placed as regards the room; they also vary in their distances from each other and from the end wall. They are six in number (one of them does not appear in the picture owing to the location of the camera), monoliths, and are not of porphyry. The eccentricity of these columns is notable as a type of the ruling style of construction of Mitla. The great court-yards are not perfectly square; lines drawn from the centres of doorways do not intersect in the centre, but at the side; the ornament on one side of a door is never duplicated on the other side, no two door lintels are of the same size. In a word, careful attention has been paid to make the whole asymmetrical. The effect in the matter of ornament is bizarre and striking, in the architectural position of the buildings, rooms, and parts of rooms; it is only revealed on measurement. In Chichen-Itza, Uxmal and Kabah, on the contrary, we find the most perfect symmetry. This asymmetry of Mitla is not accidental I am certain, but made designedly; what that purpose was I do not know. M. Désiré Charnay tells me that he has observed the same thing at Palenque. To return to our columns. They are 11 ft. 1 in. high, nearly perfectly cylindrical, slightly tapering toward the top which is flat. They did not probably support a roof at any time, for the

The

position of the two pillars 54 feet apart, with a smooth cement floor stretching between them in the ruins of the East Building, shows conclusively that they had a special significance. Further, there are two in front of another ruined building in Mitla, and a third in the village, standing in the open air. Authors, including Burgoa, say that they are of porphyry, and served to support the roof. Both of these statements are erroneous. A new feature in this room is a veritable window (vide plan). A curious point I noticed is a notable diminution in the width of the doorways of Rooms M, P and R in the later constructions, and also the window just mentioned. These later additions are sometimes, as in Room R, of adobe, in the other cases of brick and cement. At first I thought that it was done as supports, as many, in fact all but two of the huge lintels are broken, apparently by their own weight, but close examination showed this hypothesis to be untenable.

In Room R a very curious feature is to be noticed. As the whole of the inner walls was decorated, in this building, with the usual mosaic pattern, whenever the solid stone rendered it necessary the pattern was sculptured across the stone; so we find it on the south end of the lintels. A glance at the plan will show that the north end is in a dark corner. Now the sculpture should have run clear across the stone; it does not, but for a great space the stone is left blank. Whether the sculptor became tired or forgot, or whether the stone was too hard, whatever the reason, there we have an evidently unfinished piece of work. That they were not afraid of work is shown by the care used and great labor expended on the east lintel of Building No. 2, South Group (vide Fig. 3). There are examples of apparent great care and as great carelessness. In the plan the winThis is wrong; it should have

dow is shown of

this form:

6' 4"

of large stones covered with hard, bright, dark red cement; it opens on a central court surrounded by four rooms. The court-yard and rooms are all floored with hard cement. The patio is very well preserved. Room N is poorest of the four, and Room O is in a very perfect state. Room P comes next, and Room R. is in good condition. The patio and rooms, and indeed the whole building, except the Hall of Pillars (Room M), is covered with ornaments. These ornaments are variations from a few simple types. The walls are faced with large stones and the ornaments are arranged in panels. The photographs sent with this paper show the ornaments well. Still further to the north are the remains of a very large edifice, but it is now part and parcel of a church and of the buildings thereto annexed, and it is hard to separate the old from the new. Still here I found on a cornice remains of mural paintings. The subject seems (from memory) to be identical with that sculptured on the so-called "Sacrificial Stone" in the patio of the Museo Nacional at Mexico, and on the walls and roof of the chamber of the

Gymnasium at Chichen-Itza. You see the same file of chiefs, all bent forward and bearing lances and other insignia. Directly west of this building is a high mound, with a very much ruined flight of steps leading to the summit. This mound is surmounted by a modern building. Further south and in and near the village, often occupied as houses, are other ruins, in one of which I saw a long strip of painted cornice. The centre evidently represented a calendario," but unfortunately it was broken in parts. The two groups I measured were accurately oriented to the magnetic points of the compass at the date of my visit. In speaking of the lintels, I find that I have omitted to state that they repose on very large blocks, which have in every case but two, circular holes made in them. (Vide Fig. 1.) They also project considerably thus, in cross-section. See also photograph of the Hall of Pillars.

Fig 7.

Lintel

Support.

This concludes my notes on such build

ings as I could accu- Inner rately measure and Face. study. The natives

(Zapotecos) tell me that everywhere, but chiefly to the north and east, are large mounds, and during my stay I was continually pestered by women and children bringing me

idols, etc., to buy. One word as to the probable age. The buildings are carefully looked after by the Government, and have an intelligent guardian in the person of Don Felix Quero, to whose hospitable courtesy and beautiful house I can heartily recommend all archæologists who may visit Mitla. As he truly says, the big stones are too big, the small stones too small, to move with profit; hence they have suffered little ravage from men. An inhabitant of the village since 1849 assured me that the ruins are in exactly the condition they then were. But Burgoa in 1674 says: "This work being most ancient, beyond the memory of any one living, has yet lasted to our times. I saw them very leisurely over thirty years ago," and then goes on to describe them as they stand to-day. Therefore, in 1644, two hundred and thirty-eight years ago, they were practically as they are to-day. I can see no reason why they should not last for centuries still.

I beg that too severe criticism of my notes may not be exercised, as they were hastily thrown into shape, but every measurement may be relied on as correct. I have a large plan of the buildings measured by me, and tracings of some of Mühlenpfordt's plans. They are useful, but not absolutely correct. They are four in number. No. 1. Plan of Buildings remeasured by me. No. 2. Section through cruzero and subterranean passage. No. 3. Ground plan of cruzero and subterranean passage. No. 4. General plan of Mitla.

I found Mühlenpfordt's original plans in the library of the Institute at Oaxaca, but my short stay was shortened by violent attacks of fever, and I could only copy these. Mühlenpfordt seems to have understood the construction and size of the supposed passage. I also send three

Wor

photographs. No. 1. Hall of Pillars. No. 2. North Group, north building, from northwest. No. 3. North Group, north building, from northwest.

Hoping this humble contribution may be of some use to all American archæologists, I lay down my pen with a feeling of regret akin to that I felt when I cast my last glance at the beautiful ruins I have so feebly described.