WITHOUT pretending to furnish an exhaustive treatment of the subject proposed for this chapter, we desire to make observations on some phases of the development of American civilization in the Pre-Historic period. One of the most natural fruits of the study of the arts and customs of any people, is a disposition on the part of the investigator to institute a comparison with corresponding features of civilization in all parts of the world. Unfortunately this disposition has led many writers on America into wild and fanciful speculations, which tend only to deceive the reader and add nothing to true investigation. In a few instances pronounced old world analogies have been proven to exist in ancient American institutions and arts, but their number bears a small ratio to the multitude of fancied analogies which never existed, except in the imaginations of their discoverers. To discuss the subject in hand without transcending the limits of the period which is treated in previous chapters, namely, the Primitive period—that which antedates the era of the annals of those ancient peoples, is a somewhat difficult task, since the question of dates is a very uncertain one in the absence of any sufficient key to the hieroglyphic and picture records. The customs and political organization, together with the Aztec civilization, have been often treated, and by none better than our own Prescott and Bancroft. The repetition of their labors here would be highly superfluous. We shall, however, ask the attention of the reader to some considerations upon the following divisions of the subject:

1. Architecture. 2. Sculpture and Hieroglyphics. 3. Chronological and Astronomical Knowledge. 4. Religious Analogies.

Architecture.—The works of the Mound-builders and Pueblos have already been described and their transitional forms or stages noted. To seek for parallelisms or analogies between the Mound-builders and the people of Asia because mounds are common to both continents, or to seek to identify them with the people of Northern Europe because the shell-heaps of our sea-board resemble those of Denmark, would certainly be an unjustifiable use of the imagination, in anything like a serious discussion of the question. We have no disposition to speculate on this subject, since such speculation cannot furnish any satisfactory results. Certain resemblances between American and Hindoo-mounds have been supposed to exist, but the resemblance, if any, proves nothing.[491] That more fruitful and wonderful field of ancient architecture in Central America, Yucatan and Mexico, furnishes abundant opportunity for the discussion of our subject. Detailed descriptions of the remains found in different localities have been given by travelers, artists and authors, the latter availing themselves of several accounts and instituting comparisons between the statements of different explorers. Such works, savoring somewhat of the critical, cannot be underrated, since their development of the true facts has contributed largely to our knowledge of the subject. It has been generally the rule for writers to undertake the description of remains in a particular locality and treat them in detail, thus presenting to the mind a pleasant picture of the whole, together with the relation of parts. This is certainly a satisfactory plan to many readers, but it seems to us that such a course is unnecessary, after it has been once pursued by the explorer. By repetitions nothing is gained, unless the work of classification (by which certain architectural forms and methods are woven into a style and their variations noted) receives attention. In preceding chapters we have treated of the Maya, the Quiché, and the Nahua peoples, and in this, it is our purpose to briefly note the main features of their styles of architecture, sculpture, etc., as indicated in the divisions above laid down.

Maya Architecture furnishes evidence of growth, and may be classified into the Chiapan or ancient and the Yucatanic or modified styles. The Chiapan or ancient style is exhibited in the imposing remains of Palenque, with which the reader is supposed to be already familiar, from the descriptions of several explorers.[492] Palenque is situated in the Usumacinta River region in Chiapas, on a small stream sometimes called the Otolum, a tributary of the Tulija, which is itself a branch of the Usumacinta. The ruins are situated in a small valley of the foothills, from which rise the high table-lands of the interior. They are known as the Palace, with a pyramidal base measuring two hundred and sixty by three hundred and ten feet and forty feet high; Temple of the three Tablets; Temple of the Beau Relief; Temple of the Cross, and Temple of the Sun. The most conspicuous feature of the architecture employed, and seen in most of the Central American structures, is the massive pyramidal foundation. The sides of the pyramid of the Palenque palace are faced with regular blocks of hewn stone, with extensive flights of stairs, upon the east and north leading to its summit.[493] Mr. Bancroft has analyzed the structure of the American pyramid in a philosophical way, and no doubt has in part explained its object. “I think,” he remarks, “that perhaps with a view to raise this place or temple above the waters of the stream, four thick walls, possibly more, were built up perpendicularly from the ground to the desired height; then, after the completion of the walls, to strengthen them, or during the progress of the work to facilitate the raising of the stones, the interior was filled with earth, and the exterior graded with the same material, the whole being subsequently faced with hewn stone.”[494]

In the above cut Mr. Bancroft illustrates his opinion. Stephens and Waldeck, who excavated from the summit downwards, imply that the interior D is of earth. Twenty years later Charnay found a perpendicular wall on the eastern side, quite contrary to the observations of all previous travelers. Mr. Bancroft accounts for this on the supposition that the stone facing, loosened by the growth of trees which covered it, had fallen from B to F, and that the earth which filled the sides at E E had been washed away by the rain and left the perpendicular wall exposed at B. Such a supposition we consider to be perfectly probable in view of the rapid dilapidation of the ruins since Dupaix’s visit in 1806. The ancient model thus established in the construction of this, perhaps oldest of existing American cities, may have determined the style of many similar edifices. A plan of the palace has been furnished by several authors.[495] The accompanying restoration from Armin’s Das Heutige Mexiko, employed by Mr. Bancroft, may serve to give an idea of the proportions of the structure. The edifice occupies the entire summit platform of the pyramid except a narrow passage-way around the edge, and measures 228 feet by 182, and about 30 feet in height. The doorways, of which there are forty in the outer wall, are wider than the piers intervening between them, and were constructed originally with flat wooden lintels, all of which have disappeared. The main architectural features will be observed in the accompanying plate from Waldeck. The lower right-hand figure shows the angle of the foundations of one of the interior buildings and the manner in which the stones were laid. The left-hand figure affords a sectional view of the eastern stairway descending from the principal corridor into the grand court. It will be observed that the height of the steps considerably exceeds their width. Waldeck illustrates this singular disproportion by a diagram in which a native is represented as sitting upon the stairway. The perpendicular face of a step is shown to be considerably higher than the Indian’s knee, and must have measured two feet. The upper left-hand figures represent the forms of niches, which are of frequent occurrence. The T shaped niche is the representative of a numerous class so resembling the Egyptian tau or cross as to excite no little interest in its origin. M. Waldeck found the marks of lamp-black upon the tops of some of them, and supposes them to have held torches which illuminated the corridors; others, which extend through the walls, may have served for the purposes of ventilation; while others perhaps contained idols.[496] The right-hand upper figures represent the highly artistic double cornices employed. Nothing of a definite nature is known of the style of roof with which the palace was covered, since every vestige of it has disappeared. Castañeda represents it as sloping and plastered, while Dupaix refers to it as consisting of large stone flags, carefully joined together.[497]

The neighboring buildings, such as the Temple of the Three Tablets, the Temple of the Cross, and the Temple of the Sun, each have well-preserved roofs of masonry, which are quite remarkable. The first of these stands upon its lofty pyramidal base, measuring one hundred and ten feet on the slope, with continuous steps on all sides. The temple, which is thirty-five feet high, is crowned with a sloping ornamental roof of great beauty. Stephens illustrated the temple in several views, subsequently copied by Bancroft.[498] The roof is divided into three parts; the lower section recedes from the cornice with a gentle slope, and resembles the corresponding section of a French or Mansard roof. The stucco decorations of this lower section, which is also painted, add considerably to the general effect. Five solid square projections with perpendicular faces suggestive of the attic windows of a modern French roof are found on this section, corresponding to the several doors of the temple immediately below. The second section, which slopes back at a more acute angle, is of solid masonry. The crowning section seems to have been purely ornamental, consisting of a line of pillars of stone and mortar, eighteen inches high and twelve inches apart, surmounted by a layer of flat stones with projecting sides. The Temple of the Cross and Temple of the Sun both have roof-structures which may be described as resembling a lattice-work of stone.

The most interesting feature of Palenque architecture is the arch, of which there are two styles, if one of them may be classed as an arch at all; of this we have doubts. The style to which we allude is that which has been designated as the Yucatan arch. A section of the double corridor of the palace furnishes an example as shown in the cut from Mr. Bancroft’s work.

This so-called arch is nothing more than the approach of two walls toward each other in straight lines, nearly forming an acute angle at the top. These inclining walls are constructed of overlapping stones, with a small surface of exposed ceiling, produced by a lintel-like covering. The principal doorway, which is eighteen feet high, is constructed in the form of a trefoil arch, while niches or depressions of the same trefoil form are ranged along the inclined face of the gallery on each side of the entrance. This arch is suggestive of the Moorish pattern, though the latter probably is the more modern. The accompanying cut—a photographic reduction from Waldeck—will convey a clear idea of its form.

The tower situated in the southern court is considered by Waldeck as the crowning work of all. The frontispiece is a photographic reduction from Waldeck’s drawing, and no doubt indicates the true number of its stories, as well as the remarkable growth of vegetation upon its roof. The descent of the little roots and tendrils of the trees above in quest of nourishment, furnish a striking illustration of the luxuriant vegetable growth which pervades the region. The very air is laden with life, though the remains of man’s handicraft and power are but the lifeless monuments of his vanished glory. The gentle evening breeze which plays upon the tendrils stretching themselves down the tower’s wall, produces a soft melodious sound, resembling that of the Æolian harp, and gives rise to the apprehension in the minds of the natives that the place is enchanted.[499]

The second division of Maya architecture, namely, the Yucatan or modified style, presents some variations from the ancient or Chiapan. Probably the most remarkable group of ruins in that richest of American architectural fields—Yucatan—is situated at Uxmal, in Lat. 20° 27′ 30″, thirty-five miles south of Merida. The reader is of course acquainted with the detail of the survey of this remarkable city of antiquity through the work of Stephens and Catherwood.[500] These indefatigable explorers examined about forty ruined cities, nearly all of which were previously unknown to others than the natives, and many of them were unknown at Merida, the capital of the country. While these travelers are pre-eminently the explorers of Yucatan, there are others whose services have been of great value in the same field.[501]

Mr. Bancroft has divided the architectural remains in Yucatan into four groups, classifying them geographically. We do not consider it necessary to follow such a course, nor enter into the detailed description of any group, but will content ourselves by simply noting any variations from the Palenque models. At Uxmal our attention is at once arrested by the irregular pyramidal base of the building known as the Casa del Gobernador. The base of the pyramid is a figure of an irregular rectangular form. The northern and eastern sides of the base are equal, and measure about six hundred feet each; the southern and western are, however, irregular. As all the angles are right angles, and two contiguous sides are equal, it will be understood that the figure of the base would have been a square, but for the irregularity of the remaining two sides. These irregularities fall within the figure of the square. The pyramid is terraced, the first promenade when observed being but three feet from the ground. The second terrace rises from this to a height of twenty feet, and supports a platform with sides 545 feet in length. A trifle west of the centre of this platform rises the third terrace, nineteen feet high, and supporting the summit platform, measuring about 100 by 360 feet, with an elevation above the ground of upwards of forty feet.[502] The pyramid is composed of fragments of limestone thrown together, but with the terraces substantially faced with walls of regular and smoothly-hewn limestone-blocks, laid in mortar which has become intensely hard. The corners of the pyramid differ from those usually met with in that they are rounded. The terrace walls incline slightly toward the centre of the pyramid. The second platform was reached by a long inclined plain on the south side one hundred feet wide. A regular stairway with thirty-five steps, and one hundred and thirty feet wide, furnished the means of ascent from the second platform to the summit. The crowning feature of the structure is the Casa del Gobernador, a characteristic Yucatan building, measuring three hundred and twenty-two feet long but only thirty-nine feet wide. The Casa is surrounded by a promenade thirty feet wide, and in its interior contains two parallel rows of apartments (a plan of which is given by Mr. Stephens).[503] A sectional view of the Casa resembles the sectional view of the palace corridors at Palenque, except that in the arches conspicuous in the latter, the irregularities produced by the square overlapping stones (which are filled up to an even surface by mortar and plastering), are avoided in Yucatan, by the overlapping stones of the arch being dressed carefully to the angle of inclination of the wall or ceiling, thus presenting a smooth surface. The roof is formed by filling in the space between the tops of the arches and between the arches and the outer walls with stone, up to the desired level; after which a perfectly flat covering of well-cut stones is laid over the whole, having a neat though small projecting cornice, as will be observed in the accompanying cut from Bancroft’s work. The rear wall is about nine feet thick and perfectly solid. The comparative modernness of the building may be realized when we state that Mr. Stephens found the top of each doorway supported by a heavy beam of zapote-wood. One of these, which was elaborately and beautifully carved, and measuring ten feet long and ten by twenty inches wide, he brought to New York, where, unfortunately, it was destroyed by fire with the remainder of his collection. It is presumed that the zapote-wood was prized for its rarity, as it is not found at present near Uxmal. Inside of and above the doors of the Casa were stone rings, which occur frequently in Yucatec structures, and are supposed to have supported curtains for closing the doorways. Stephens presents in a cut (page 346) a view of the imposing and elegant front looking toward the south.[504]

Of the several Uxmal edifices, one especially demands attention as representing the highest state of ancient architecture and sculpture in America. This is known as the Casa de Monjas, or Nunnery, and is situated nearly three hundred yards north of the Casa del Gobernador, on a pyramid with three terraces, and measuring three hundred and fifty feet square at its base. On the summit platform, only nineteen feet above the level of the ground, stand four of the characteristic Yucatan buildings upon four sides of a nearly square court. The northern building does not stand quite parallel to the building on the opposite side of the court. The plan from Stephens will present clearly the arrangement of the apartments, in which it will be observed that of the eighty-eight rooms contained in the Casa de Monjas, not more than two apartments open into each other, except in one instance, which occurs in the eastern front.[505] The court formed by these long narrow edifices measures 258 by 214 feet, and according to M. Waldeck was paved with 43,660 blocks of stone six inches square. In the centre stood the fragments of a rude column similar to others observed in the Casa del Gobernador.[506]

A cut of one of the beautifully sculptured façades of the Casa de Monjas will be found on a future page. Near the Casa de Monjas stands the pyramid and edifice generally known as the Casa del Adivino or Prophet’s house, and named by M. Waldeck the Pyramid de Kingsborough. The pyramid rises to a height of 80 feet from a base of 155 by 235 feet. The corners are rounded, and the sides, which are carefully faced with cubical blocks of stone, rise so steep that the ascent and descent by the grand stairway on the eastern face is giddy and dangerous. The stairway measuring one hundred and two feet on the slope is inclined at an angle of eighty degrees.[507]

About a dozen miles south-eastward from Uxmal are the remains of the ancient city known as Kabah, where ruins quite similar and nearly as extensive as those already described are found. However, new architectural features here meet the observer. In one instance the structure which surmounts a terraced pyramid is square, instead of long and narrow as at Uxmal. The inner rooms of the edifice have floors two feet higher than the floors of the outer rooms, and are entered by two stone steps. In one instance these were cut from a single block with the lower step in the form of a scroll. At Kabah we meet with an entirely new feature in Maya architecture, and the reader’s acquaintance with the terraced casas, of the New Mexican region, will supply the lack of an illustration at this point. In the style of building referred to, the pyramid instead of serving as a foundation for the building, serves as a central support around which the house with its receding stories, one above another, is built. The first story of the building referred to is built upon the ground, with the perpendicular sides of a mound for its rear wall. Just above, on a level with the roof of the first story on the platform of the first terrace of the mound, stands the second story, with the roof of the first serving as a promenade in front of it, while the third story rests upon the second platform of the mound. The platforms or roofs of the first and second stories are reached by means of a stone stairway supported upon a half arch. The first story is accessible from the ground by doorways. The interior apartments are constructed on the model of the Yucatec arch. Here, however, lintels of stone are met with, supported in the centre by rude stone columns surmounted by square capitals. These buildings are of large proportions, equalling any we have thus far described. The decorations of the edifices were considered by Mr. Stephens equal to those of any known era, even when tried by the severest rules of art.[508] At Zayi, one of the finest illustrations of this style of architecture is to be seen in what is known as the Casa Grande. The dimensions of the Casa Grande are as follows: lower story, 120 by 265 feet; the second story, 60 by 220 feet; and the third, resting on the summit platform of the mound, 18 by 150 feet; a stairway thirty-two feet wide furnishes a means of ascent to the third story on the front, while a narrow stairway leads to the second story at the rear. Round columns both in doorways and the façade constitute the chief variation from the styles already observed. An “elephant trunk” ornament protruding from the cornice (also found on Casa del Gobernador and the Casa de Monjas at Uxmal) is a marked feature of decoration. It is unnecessary for us to say that its presence has given rise to much speculation as to its origin. M. Waldeck has given the figure the name which we have applied to it, and perhaps with some reason.[509]

At Labná ruins of a curious and extraordinary nature exist, though far gone in decay. The accompanying cut, employed in Stephens’, Baldwin’s and Bancroft’s works, will serve to show the extravagant decoration lavished upon the cornices of the edifices. At Chichen-Itza, the so-called “Nunnery” is supported by a solid mass of masonry, with perpendicular walls. The dimensions of this base are one hundred and twelve by one hundred and sixty feet and forty-two feet high. This was crowned by a building having two receding stories. The great pyramid of Chichen is celebrated for the solid stone balustrade which guards its northern stairway of ninety steps, forty-four feet wide. These balustrades terminate in colossal serpent heads, ten feet long.[510] Both at Chichen and at Mayapan circular structures are met with and are figured by Stephens.[511] The same author has described the rectangular watch-towers of Tuloom, which rise majestically amid the extensive ruins of the ancient city of the same name, situated upon the eastern coast in latitude 20° 10´. At Tuloom, Mr. Stephens (its only describer), found the first walled city in Yucatan. He believes it to have been occupied long after the conquest, and probably was one of the cities whose many towers met the gaze of the wondering Spaniards, who beheld them as they coasted along the shore.[512]

Quiché Architecture.—The propriety of classifying the great ruins of Honduras and Guatemala as Quiché in their origin and style, may be questioned by some of our readers. It must be admitted that great contrasts in style are found in this region, which was occupied by the powerful kingdom of the Quichés and Cakchiquels, at the time of the conquest. However, it is probable that the ancient Quichés (who, as we have already seen, at an early day developed a religion and literature), were the authors of the more ancient cities, like Copan and Quirigua. The Quiché-Cakchiquels of more modern times were quite another people, whose institutions, language, and no doubt their architecture, had been largely influenced by Nahua people from the Mexican plateau. Utatlan, the magnificent capital of this modern and mixed people, was in the height of its glory just before the blighting power of the conquerors laid it in ruins. As ours is not an attempt at the history of discovery, we omit entirely that interesting feature in the treatment of antiquities, and call attention at once to the features conspicuous in Quiché architecture. The ancient city known as Copan, on the eastern bank of a river of the same name, in latitude 14° 45´ and longitude 90° 52´ in Honduras, and four leagues from the Guatemala line, is interesting in furnishing material for study in this department. It is probably the most ancient city on the continent. Copan no doubt could successfully contend with Palenque for the palm of antiquity. It is again to the indefatigable Stephens and the skillful Catherwood that we are most indebted for our knowledge of these ruins.[513] The period of the abandonment of Copan is a question with reference to which we possess too few data to render an intelligent decision concerning it. Following the example of Stephens and Bancroft, we first introduce the account of Fuentes contained in Juarros.[514] “In the year 1700, the great circus of Copan still remained entire. This was a circular space, surrounded by stone pyramids about six yards high and very well constructed; at the base of these pyramids were figures, both male and female, of very excellent sculpture, which then retained the colors they had been enameled with; and what was not less remarkable, the whole of them were habited in the Castilian costume. In the middle of this area, elevated above a flight of steps, was the place of sacrifice. The same author (Fuentes) relates that, a short distance from the circus, there was a portal constructed of stone, on the columns of which were the figures of men, likewise represented in Spanish habits, with hose, ruff round the neck, sword, cap, and short cloak. On entering the gateway there are two fine stone pyramids, moderately large and lofty, from which is suspended a hammock that contains two human figures, one of each sex, clothed in the Indian style. Astonishment is forcibly excited in viewing this structure, because, large as it is, there is no appearance of the component parts being joined together; and although entirely of stone and of an enormous weight, it may be put in motion by the slightest impulse of the hand. Not far from this hammock is the cave of Tibulca; this appears like a temple of great size hollowed out of the base of a hill, and adorned with columns having bases, pedestals, capitals and crowns, all accurately adjusted according to architectural principles; at the sides are numerous windows faced with stone exquisitely wrought. All these circumstances lead to a belief that there must have been some intercourse between the inhabitants of the old and new world at very remote periods.” The swinging stone hammock is probably a work of the fancy rather than that of the artist’s hand, though the padre at Gualan told Stephens that he had seen it, and an Indian remembered to have heard his grandfather speak of it. None of these remarkable remains have been identified with certainty, though it is not improbable that they might be discovered if the heavy growth of vegetation were removed by a conflagration and explorers to extend their observations farther from the banks of the Rio Copan. According to Stephens’ survey, a wall encloses a rectangular area measuring about nine hundred by sixteen hundred feet. The principal group of buildings is designated as the temple. It is built of heavy blocks of cut stone, with walls of about twenty-five feet in thickness, and when examined they were between sixty and ninety feet high on the river’s bank. The temple measured six hundred and twenty-four feet north and south by eight hundred and nine feet east and west. The general feature of the ruin is that of an immense pyramidal terrace, with a platform elevated about seventy feet above the ground. The river side of the terrace is perpendicular, while the remaining sides are sloping; viewing the ruin from this general platform seventy feet high, depressions such as amphitheatre-like courts descend from it in some instances thirty or forty feet, or about half way to the level of the ground, while above the level of the general platform pyramidal structures rise to a considerable height, in one instance one hundred and twenty-two feet. It is difficult to conceive of what might have been the nature of the superstructure, if any surmounted the general platform. It is probable that for the purposes of assembly the amphitheatres with their sloping sides may have answered every purpose, while the pyramids may have been surmounted by temples now in ruins. Of the sculptured columns of this locality we will speak farther on. Utatlan, the former capital of the modern Quiché kingdom, would naturally be selected as a point at which to seek for remains of the newer Quiché styles of architecture. The conquerors, however, left little that can serve as the basis for architectural study. The city was surrounded by a deep ravine or barranca, which can be crossed at only one point, and there long lines of stone fortifications still guard the passage. A fortress, called El Resguardo, is among these works. It rises one hundred and twenty feet high in the form of a terraced pyramid, with a stone wall plastered with cement enclosing its summit platform, on which a circular tower provided with a stairway was built. Only fragmentary walls of the Quiché palaces remain; their dimensions were eleven hundred by twenty-two hundred feet, and nothing but their cement covered floors have survived the vandalism of the conquerors and the architects of the modern town; the latter having carried away the upper portions for building purposes. A pyramidal structure near by, known as El Sacrificatorio, presents no architectural contrasts to pyramids already described. Its stairway, composed of nineteen steps each eight inches broad and seventeen inches high, is characteristically Central American.[515] In the province of Vera Paz, especially in the Rabinal Valley, Brasseur de Bourbourg observed numbers of tumuli, resembling those of the Mississippi Valley both in material and structure. These were especially prevalent in the neighborhood of the villages, and sometimes were associated with pyramidal structures equal in finish to any we have described. The name cakhay, “red houses,” is generally applied to these tumuli.[516]

Nahua Architecture.—It would be quite impossible for us to devote that space to this subject which the number of remains would justify, and the presentation of the typal features of the architecture of that interesting family of nations will be all that we shall here attempt; of geographical and detailed treatments there are several on the different departments of the subject.[517] In the pages which follow we will select a few examples of Nahua architecture in order to illustrate our subject, but we would state that many equally important works, though perhaps presenting no new features, have been purposely passed by unnoticed. In a preceding chapter we referred to those intermediate nations which occupied the transition position between the Mayas and Nahuas. The Miztecs, Zapotecs and others, were probably a mixed people, related in different degrees to both of the great families on the north and south of them. Oajaca and Guerrero were the homes of these peoples, where they developed their own civilization and styles of art in channels distinct from those of their neighbors. The isthmus of Tehuantepec presents some interesting remains, chief among which we may cite two stone pyramids situated three leagues west of the city of Tehuantepec. One of these measures fifty-five by one hundred and twenty feet at the base and thirty by sixty-six feet on the summit. A grand stairway composed of forty steps and thirty feet in width leads up the western slope. The summit is also made accessible by smaller stairways on the north and south sides. The lower of the four terraces composing the structure, is perpendicular; the others have inclined walls. On the face of the second terrace were four ranges of flat stones, one above another, extending entirely around the pyramid and furnishing a series of shelves, devoted no doubt to some sacred or sacrificial use. The whole structure was plastered with a cement, colored brilliantly by red ochre. The adjoining pyramid presents an architectural novelty in its gracefully curved sides. Castañeda has sketched and Dupaix described it. The height of the pyramid is over fifty feet while its general dimensions are about the same as those of its neighbor. In close proximity to the pyramids, altar-like structures were observed, one of which was composed of eight circular stones, like mill-stones, placed one above another. The base measured ten and a half feet, but the summit only four and a half feet; the height measures twelve feet.[518] Numerous earthen tumuli resembling those of the Mississippi Valley were observed by the German traveler Müller, scattered over the region, especially to the south-east.[519] The most important group of ruins in Oajoca is that at Mitla, situated about thirty miles south-east of the capital of the State. This is probably the finest group of remains north of the isthmus of Tehauntepec. Still they are not purely Nahua in their style, being, according to tradition, the work of the Zapotecs. This group has been described several times by explorers, whose accounts have differed considerably in value. The most important of these are the descriptions and drawings by Dupaix and Castañeda, made in 1806, and the description and valuable photographs by Charnay, the latest explorer of this group, whose work was performed in 1859.[520]

The mitla ruins are distributed into four groups of buildings (generally called palaces or temples) and two pyramids. The principal edifice is described as follows: three low oblong mounds only six or eight feet high but surmounted by stone buildings, enclose a court. The court measures 130 by 120 feet. The eastern and western buildings are in a fallen and ruined condition. The northern building, however, presents a singular example of ancient grandeur. The southern portion measures 36 by 130 feet, and the northern 61 feet square. The edifice is about eighteen feet high, having walls varying from four to nine feet in thickness. The accompanying cut, a photographic reduction of Charnay’s photograph, gives a correct idea of the western façade of the northern building.[521]

The walls of this edifice are constructed in a somewhat novel manner, their interior portions being nothing more than clay intermixed with stones, thus furnishing a poor substitute for the cement and stone filling in the inner parts of Yucatanic walls. However, the exterior facing of the walls is of hewn stone blocks cut in different forms and sizes, and so set in relation to each other as to present examples of perhaps the finest variety of grecques found in any structure in the world.[522] Two layers of large stone blocks form the base of the palace, from which rises buttresses and a framework of stone, filled in with panels of mosaic, in patterns as described. We pronounce these grecque patterns mosaics, because of the manner of their structure. They are not of the nature of sculpture, since each pattern, with all its regularity, is composed of small brick-shaped blocks of stone built into the wall, mosaic-like, thus forming the graceful patterns shown in the cut. No trace of mortar has been found at Mitla. The inner surface of the wall in the northern building was smoothly plastered without any ornament. Six round stone columns standing in line occupy the centre of the apartment, and no doubt supported a roof of wood or stone, but more probably of the former.[523] The cut in Baldwin’s work, copied by Bancroft showing the interior of the apartment and the six columns, conveys an incorrect impression as to the form of the columns and the character of the walls, as is proven by Charnay’s photograph.[524] The façades of the inner court of the northern wing of the palace are finished with mosaics of great beauty. Four or five feet of the wall is plain at the bottom except that the plastering was evidently frescoed in various colors. The remainder of the wall is decorated with bands of mosaic grecques, as shown in the cut, which is a fac-simile of Charnay’s photograph engraven for Mr. Bancroft’s work. We should not fail to note the use of immense stones in the base, framework and lintels of the southern wing of the building. One of these is of granite, sixteen or nineteen feet long, with the pattern of the adjacent grecques sculptured on its face. None of the other buildings at Mitla present any architectural contrasts to the one already described, and require no special attention. Under a temple on the south-west side of the one we have just referred to, is a subterranean gallery, constructed in the form of a cross. The opening is at the base of the mound upon which the temple stands. The arms of the cross pointing toward the East, North and West, are each twelve feet long, five and a half feet wide, and six and a half feet high. The southern arm is, however, about twenty feet long, and not more than four feet high throughout most of its length. Near the centre of the cross (which lies directly under the centre of the temple above) a flight of four steps descends in the southern arm of the cross to a lower level, so that the southern arm of the passage is somewhat lower than the others. The entire subterranean chamber was roofed with large flat stones reaching from side to side. The walls, besides being painted red, were ornamented with panels of mosaic, but of a ruder style than that of the superstructure, which is suggestive of an earlier period in the growth of the art. A circular pillar resting on a square base, and called by the natives “the pillar of death,” because of the belief entertained among them that whoever embraced it would immediately die, supports the large flagstone which covers the intersection of the galleries. An immense fortification over a mile in circumference and with stone walls six feet thick and eighteen feet high crowns the summit of a hill, which stands three-fourths of a league south-west of Mitla. The place was inaccessible except on the side toward the village where the wall was double. Castañeda has delineated and Bancroft copied the plan of this fortress.[525]