Palenque stands alone. Architecturally it is by far the most advanced of the cities of the central area, while the artistic qualities of the stucco reliefs, and the low-relief stone carvings (Figs. 49 and 52), the technique of which appears to be based on stucco-work, place them in a category by themselves. It was, in fact, only at Copan and Quirigua that the Maya showed anything like mastery over stone. The deficiency lay no doubt not so much in the artistic faculties of the mason, as in the lack of suitable tools, but it is a fact that only at the southern sites he displayed definite signs of modifying the technique borrowed from wood-carving to suit a less tractable material. Even the fine reliefs of Menché are wood-carvings translated into stone, and the same applies to the reliefs at other sites, except perhaps to the figures in niches at Piedras Negras, which in other respects noted above display a certain similarity to the Copan sculptures. When the Maya artist began to develop the art of modelling in stucco he showed that the quality of his work depended to a great extent upon the plasticity of his material, while he had gained much in technique from his experience in stone. A peculiar feature of Palenque is the comparative restraint exhibited by the artist. That he still loved complexity of detail is shown by such reliefs as Fig. 61 (p. 297), but he no longer feared the vacant space, and appreciated its value as a background. The dates on the monuments do not assist us in any way to determine the epoch which saw the rise of Palenque. The initial series in the temples give dates relating to the first cycle or even before, that is to say more than 3000 years B.C., and as such they must be regarded as purely legendary. It is impossible to believe that the site was of early foundation compared with the others. It would be against all experience to suggest that the people who built Palenque could at a subsequent period have adopted the clumsy and unnecessarily laborious architecture of Tikal. On the other hand, the site of Palenque bears a certain similarity to that of Menché, including the presence at both of buildings with underground passages and chambers, though the “palace” at Palenque is a far more elaborate construction than the analogous building at Menché, being in fact a complex of associated buildings. All the evidence seems to point to the conclusion that Palenque is the latest of the central Maya sites, and that it is most closely associated with Menché. In further proof of the latter supposition I might mention the so-called “crosses,” surmounted by birds, held by the figures in more than one of the Menché reliefs, which have their counterparts in the “crosses” of Palenque alone. The significance of these I have already discussed on p. 257.

I think, then, that the various dates found on the monuments may be taken as a fair indication of the relative periods at which the various cities flourished, except of course those at Palenque. Arguments based upon the style of decoration alone are apt to be misleading, especially as far as the lesser sites are concerned, since it is only natural that at these the workmanship should be of an inferior and “provincial” character. Allowance too must be made for the varying capacity of local artists, and above all for the relative tractability of the local materials. The evidence of architecture is more valuable, and, to speak generally, supports the dates as given by the glyphs, though the possibility of greater conservatism at some sites, as compared with others, must not be overlooked. On the whole the best results must necessarily accrue from a careful consideration of all three, with due allowance for a natural tendency to quote dates relating to past and often mythical history.

With respect to certain similarities presented by the attributes of the figures depicted at certain sites, allusion has been made above to the hand appearing as part of the head-dress at Copan, Piedras Negras and Palenque (p. 227), to the jaguar-man at Tikal, Seibal, Piedras Negras and Quirigua (p. 343), and to the fish and flower, also shown as a head-dress, at Naranjo, Palenque, Chichen Itza and on the Nebaj vase (p. 310). Attention will be called later to a peculiar interlaced ornament, also borne on the head, by figures at Copan, Piedras Negras, Menché and Naranjo, and found again at Xochicalco, Teayo and in the Mexican valley (p. 355). One more may be mentioned, the head-dress representing a heron with a fish or frog in its beak, seen at Palenque (Fig. 61; p. 297) and twice at Seibal, an ornament which recalls the Quiché legend of how they sent eastward to a country, most probably to be identified with the Usumacinta valley, to obtain royal insignia, and received amongst other objects the plumes of the heron.

Of the Yucatec sites the most important is that of Chichen Itza. The remains here are considerable, though they conform to no definite plan save that the same arrangement of building round rectangular courts prevails. The buildings themselves are typically Maya, though structurally they exhibit a certain advance upon those of the central area, and Maya glyphs are found throughout, with certain important exceptions. Of the peculiarities which distinguish the architecture of Yucatan from that of the last region, many have already been mentioned, including the mosaic ornament with which the façades of buildings are decorated. But another important feature remains to be recorded. Throughout the whole of Yucatan only two dates in the long count have been discovered, and one of these is at Chichen Itza. The site is interesting as providing evidence of more than one stage of construction. One of the finest buildings, the so-called Monjas, shows plainly that a considerable addition was made to the foundation at a time later than the original building (Fig. 83). At the same period an additional storey was added, and one of the chambers in the earlier structure was filled solid with masonry to support the weight. Certain annexes were also built (Pl. XXIX, 2; p. 350), in the decoration of which ornamental fragments from the façades of previous “palaces” were included, but the inscriptions are in regular Maya characters, though no date in the long count is included. Other buildings in the immediate neighbourhood are in similar style and belong presumably to the same epoch. But further north is a large group of structures of an entirely different character of ornamentation. These include the famous ball-court and attached temples (Pl. XXVIII; p. 348), the Castillo (also seen in Pl. XXVIII), a large structure of which only the numerous square sculptured columns remain, and the so-called “Temple of the Tables.” On these buildings neither Maya glyphs nor mosaic masks are found, while the Castillo and upper ball-court temple are furnished with serpent-columns similar to those of Tulan. In the lower temple attached to the ball-court is an elaborate relief showing figures, exhibiting no signs of cranial deformation, armed with spears and spear-throwers, assisting, so it would seem, at the obsequies of some personage in the centre who is distinguished by a huge feathered snake which overshadows him. Many of the figures are accompanied by glyphs which are distinctly Nahua in type, while their ornaments and dress combine both Nahua and Maya characters; further, a sun-disc in Nahua style occurs at the top of the relief. In the Castillo are atlantean figures carved in relief in similar style, and in the Temple of the Tables were found a number of slabs supported by small caryatides (Fig. 84), carved in the round, exactly similar to some which have been found at Tlaxcala. Finally, several stone recumbent figures supporting vases, of the type shown on Pl. VIII, 2; p. 74, have been discovered buried in the neighbourhood.[8] The buildings themselves, as far as preserved, present, architecturally speaking, Maya characteristics, except the large site with the numerous square columns, which probably supported a flat roof laid on transverse beams. To the west and south of the buildings first described are many unexplored mounds, probably the remains of the earliest settlement of all, and it is in one of these that a slab bearing a date in the long count, but later than any in the central area, has been discovered. Thus we appear to have three, or possibly four, periods represented at Chichen Itza, the first by the mounds where the initial date was found, the second and possible third by the Monjas group with its reconstructions in typical Yucatec style, the last by the northern group with its definite “Toltec” affinities. It is to be noted that this corresponds in a most remarkable manner with the Tutul Xiu tradition. The first settlement would be that of which they heard when they were yet at Balcalar, the second and doubtful third periods would represent their own occupation and reoccupation, while the last would illustrate the result of the Nahua influx which brought about the fall of the league of Mayapan. The other date in the long count which has been found in the Yucatec area is at Tulum. This is a very early date, in the sixth katun of the ninth cycle, and may possibly relate to some event in past history. At any rate it supports the evidence furnished by Chichen Itza that the Maya peoples who reckoned by the long count and built the cities of the central area had penetrated into Yucatan before the arrival of the Tutul Xiu. Tulum has not yet been satisfactorily explored, but the buildings there are neither imposing nor richly decorated; the site however presents one feature of interest in so far as it is surrounded on three sides by a wall, the fourth being protected by the sea.

Uxmal, which, according to the Tutul Xiu account, was founded from Chichen Itza, shows, as regards the majority of its buildings, a very close correspondence with the Monjas group of the latter site. Most of the buildings exhibit the true Yucatec character, including the mosaic masks with projecting “trunks,” and pilastered ornament is also found. A few however do not show these characteristics to the same degree, and present abnormal varieties of cornice, while others display signs of alteration and modification which imply at least that the site was inhabited for a considerable period. The most striking buildings, perhaps the finest in the true Yucatec style which exist, are the so-called “House of the Governor” (Pl. XXX; p. 358), the “Temple of the Magician,” peculiar in its oblong pyramid with rounded corners (Fig. 73, h; p. 321),[9] and the “House of the Doves,” distinguished by a vandyked roof-comb ornamented with open-work tracery. Lack of space forbids allusion to the many other Yucatec sites, such as Labna, Sayil, Kabah, and so forth, but it may be said that they embody similar forms of decoration such as the grotesque mosaic mask, and geometrical and pilastered ornament (Pl. XXV, 1; p. 332), though the structures can hardly be said to compare with those of Chichen Itza and Uxmal.

As regards British Honduras, mention has already been made of the interesting building at Santa Rita, decorated with frescoes somewhat akin in style to those at Mitla, though rather nearer the true Maya in type. It is difficult not to see in this art an extension of the “Toltec” influence so evident at Chichen Itza, combined with a stronger element of the local Maya art. But buildings in this region are few, though the substructures of former edifices are common. On the Colombia branch of the Rio Grande is found an extensive two-tiered foundation-mound, supporting a number of stepped pyramids, which recall to some extent the site of Uxmal, in so far as the corners of most are rounded. Both mound and pyramids are faced with cut stone, and quantities of brick are also found which appear to have been used as hearting and paving material.

One of the most interesting ruins of British Honduras consists of a huge mound near Benque Viejo, sixty feet high and built of limestone, which supports a three-storeyed stone building. This is surrounded by three others of similar type, at the foot of each of which is a plain stone monolith, recalling the plain stelæ mentioned above, which doubtless were once ornamented with painted designs. The ruins of this site probably belong to the same series as Tikal, Naranjo, Seibal, etc., and though more excavation, combined with accurate surveying, is necessary before we can speak with confidence regarding the monuments of British Honduras as a whole, we may assume that they fall into two classes, viz. ruins connected with the early central Maya area, those to the north of British Honduras at Tulum being in part a “provincial” extension of the early culture,[10] and later edifices which owed their birth to Toltec influence filtered southward through Yucatan.

In the west of the Maya area, the remains, though plentiful, are inferior in quality, and have not attracted the attention of explorers to the extent which their importance, as indicating the spread of Maya culture, and its relation to that of Oaxaca, deserves. Sacchana and its neighbourhood is a site of great interest, since Seler has discovered there two stone slabs bearing initial dates in the Maya style, falling just each side of the date at Chichen Itza, which is otherwise the latest known. The style of these slabs is rude and decadent, but they are of the highest importance as indicating that the early method of reckoning time had spread as far west at least as the department of Huehuetenango. Otherwise the remains of this district are rude and coarsely built, of unsquared blocks usually without mortar; but mortar is found in some cases, notably in the tlaxtli-courts, which, as well as the pyramidal mounds, bear a very close resemblance to those of Quiengola in Oaxaca (p. 175). A feature of this district is constituted by the number of cave-temples, which again recall the last-named province of Mexico. It is perhaps worthy of mention that a tendency appears in this neighbourhood to arrange temple-mounds in groups of three, in a straight line from north to south, with the stairways facing west. The western position of pyramid stairways, which implies that the worshippers faced east, has already been observed in Oaxaca and Mexico. An account of certain ruins at Utatlan, Iximché and Rabinal is given in the works of Maudslay, but they cannot compare with those of the central Maya area, and seem rather to resemble those of the district of Huehuetenango.

The Maya appear to have resorted occasionally to caves as dwelling-places, and certain investigations have been made in caverns at Copan and Loltun (immediately south of Uxmal), in Yucatan. The remains however do not imply that their occupation was of early date, and it is probable that the caves served as places of temporary refuge in times of war. This chapter would be incomplete without some mention of the peculiar bottle-shaped subterranean structures found at certain sites in Yucatan. These are known as “chultunes,” and, since they invariably occur at places remote from rivers, may almost certainly be regarded as cisterns for preserving a supply of water, though in some cases they have served the secondary purpose of burial chambers. They are lined with dressed stone or a coat of stucco, and the mouth is furnished with a stone ring which is closed by a slab. Careful exploration of the chultunes at Labna has been made by Thompson and reported in the “Memoirs” of the Peabody Museum.

It will be unnecessary to enter into a discussion of Maya art, since the subject has been fully treated by Spinden in a monograph which was published only a week or so before this chapter was written, and, moreover, much may be gathered from the preceding pages supplemented by the illustrations. The kinship of the stone reliefs with wood-carving, except at Copan and Quirigua, has already been indicated, and the symbolic nature of Maya ornament is too obvious to need especial mention. Attention may be called to the use of glyphs, extremely ornamental in themselves, to give balance to a design, and to the horror vacui which appears in all reliefs except those of Palenque. Though bound by convention, the result in a large measure of the symbolic nature of his designs, the Maya artist occupies a relatively high plane amongst barbarous peoples. In particular the quality of his line is excellent, even in stone, and in certain respects, notably foreshortening (especially in the treatment of feathers, for which see Figs. 61 and 87; pp. 297 and 367), he was superior to the sculptor of Egypt or Assyria. One peculiarity of Maya art lies in the fact that the artist was obsessed by the motive of the serpent (rattle-snake) head. In nearly all its essentials this design appears in the remarkable gargoyle from Copan shown in Fig. 85. We have here the exaggerated upper jaw, the front fang, and the curl at the corner of the mouth. One detail is lacking, viz. the nose-ornaments which were usually added to the reptile. Maudslay has shown how derivatives of this design, often modified almost beyond identification, are constantly applied to every form of ornament, such as the edges of girdle-flaps and so forth, and the point has been elaborated by Spinden. Most interesting is the almost invariable occurrence of the serpent-motive on the wings of birds (e.g. Figs. 48 and 49; pp. 225 and 230), the combination no doubt expressing the high god Kukulkan, who, among the early Maya, seems to have been represented only in this indirect way (see p. 226). In the later art, especially that exhibiting Toltec affinities, we have a more direct delineation of the deity in the shape of a human figure whose face is surrounded by the jaws of a serpent, but who is provided with wings carved in the best Maya style (Fig. 87). The foreshortening of feathers has already been mentioned, but it may be added that perhaps no people has shown in its art such an appreciation of this form of ornament as the Maya, and for sheer beauty the feather-motive shares the honours with the plant-design (see Fig. 60; p. 289), often shown in combination with fishes and grotesque masks, which appears not only on the sculptures, but also in the frescoes.