MV^EVMoFTHEAnEILICAN INDIAN: i-i ' "'" """ illilililliliiilililllMi lIMjIl" MARSHALL -H.SAVILLE COLLECTION 3 1924 103 987 412 Cornell University Library The original of this book is in the Cornell University Library. There are no known copyright restrictions in the United States on the use of the text. http://www.archive.org/details/cu31924103987412 CERTAIN ANTIQUITIES OF EASTERN MEXICO JESSE ~V^ ALTER, FEAVKES '-1 221 o RMC i=i CERTAIN ANTIQUITIES OF EASTERN MEXICO By Jesse Walter Fewkes PREFACE The explorations and studies embodied in the present paper were undertaken by Doctor Fewkes, Ethnologist in the Bureau of American Ethnology, at the instance of the Secretary of the Smithsonian Insti- tution, and funds for the field work were furnished by the Institution. The particular field of operations was chosen with the view of deter- mining, if possible, by means of a reconnpissance of the eastern states of Mexico, whether or not any definite connection or relationship existed between the ancient peoples north of the Rio Grande and within the area of the United States and those to the south in Mexico, especially the semicivilized tribes of middle Mexico and Yucatan. The present study must be regarded as preliminary only, the field work not yet having progressed sufficiently to furnish data for definitive results; but it is, nevertheless, a very important contribution to our knowledge of the ancient culture of the Gulf states of Mexico, and it is hoped that the researches thus initiated maj' be continued and completed in the near future. W. H. Holmes, Chief. 223 O O N" T E I^ T S Page Introduction 231 Ruins of Cempoalan 233 Construction of buildings 236 Nomenclature and position of buildings ' 237 Building A 238 Building B 240 Building 240 Building D 242 Building E 243 Mounds near Antigua '. 243 Ruins of Xicochimalco 244 Texolo mounds 245 Xico Viejo ■ 246 Stone idols near Xico 247 Modem Xicochimalco '. 248 Papantla 249 Castillo de Teayo : 250 Objects from Cempoalan and Xico 250 Classification 251 Stone ring 251 Closed stone yokes 253 Open stone yokes 253 Curved stones 258 Theories di the use of stone yokes , 259 Paddle-shaped stones 261 Padlock stones 263 Stone heads 264 Sling stones 266 Stone idols 266 Clay objects from Barra Chachalicas 268 Mounds near Tampico 271 General remarks 271 Shell heaps 275 Ruins near Altamira 276 Ruins near Champayan lagoon 277 Archeological objects 277 Stone idols at Altamira * 278 Tampico stone idols 280 Huaxtec pottery 280 Conclusions - 284 25 ETH— 07 15 225 ILLITSTRATTOI^S Plate XCIV. Relative positions of the main buildings at Cempoalan. XCV. INFodern building on foundation of Cempoalan pyramid. XCVI. Templp del Aire. XCVII. Building A, Cempoalan. XCVIII. Building B,.Cempoalan. XCIX. Building C, Cempoalan (front view). C. Building C, Cempoalan (lateral rear view). CI. Building D, Cempoalan (front view). CII. Building D, Cempoalan (lateral rear view). cm. Building E, Cempoalan. CIV. Mounds at Texolo. CV. Mounds at Texolo. CVI. Stone idol at Xico Viejo. CVII. Stone serpent at Fuente. CVIII. Stone idol at Texolo. CIX. View of Xico Viejo: a, the pyramid. b, view of the old houses, ex. View of el Tajin, Papantla. CXI. Castillo de Teayo. CXII. Stone yokes of the first group (Dehesa collection): • a, front view. b, side view. c, top view. CXIII. Stone yokes of the first group (Dehesa cGllection): a, front view. b, side view of a. c, front view. d, side view of c. CXIV. Stone yoke of the first group (Sonora XewH Company coUfction); a, front view. b, side view. c, base view. CXV. Stone yokes of the first group (Dehesa collection) : i(, a', front and lateral views of same yoke. b, c, d, side views of other yokes. CXVI. Paddle-shapect stones (Dehesa collection) : a, front view. b, front view. CXVII. Paddle-shaped stones (Dehesa collection): a, front view. 6, front view. 227 228 ILLUSTRATIONS Plate CXVIII. Paddle-shaped stones (Dehesa collection): u, front view. h, reverse view of a, plate cxvi. OXIX. Fan-shaped stones (Dehesa collection): a, b, with human face. r, with bird head and body. CXX. Stone heads, masks, and idols ( Dehesa coUeption) : a, small yoke. b, fiatiron-shaped specimen. c, perforated specimen. d, e,f, heads. g, h, padlock-shaped specimens. )', seated figure. j, mask. OXXI. Stone heads (Dehesa collection): a, b, heads of clowns, c, d, heads of old men. e, f, g, masks. h, flattened head. i, bird-shaped specimen. OXXII. Clay images from Cempoalan and vicinity (Dehesa collection) : a, effigy. b, head of Eain god. c, Eain god. d, e, bowls in shape of death's head. /, head of Flower goddess. , g, small painted effigy. CXXIII. Pottery images from Barra Chachalicas: a, b, front and side view of figure without arms or legs. c, painted effigy of female figure. CXXIV. Clay objects from Oempoalan: . a-f, small heads. g, well-made large head. h, painted head. I, }, heads from panel of temple. ■CXXV. Pottery from Otates (Estefania collection): a, b, decorated bowls, from exterior. c, painted bowl, showing spiral ornament. d, deep bowl, silie view. e, bowl with interior decorated with picture of monkey. /, bowl with exterior decorated with death's head. CXXVI. Pottery objects from near Jalapa and Tampico (Estefania and Pressley collections) : a, classic pitcher with graceful handle. 6, food bowl with three legs, c, d, bowls with two handles. e, bowl with legs. /, g, h, clay heads. i, clay effigy of a human being. j, section of a bowl. k, rude effigy of human being. I, m, dippers. ILLUSTKATIONS 229 Plate OXXVII. Pottery objects from Panuco valley (Pressley collection): a, melon-shaped specimen. 6, melon-shaped specimen with handle. c, double-handled vase with human face. d, globular dipper with human face. e-h, clay heads. /, j, unknown objecta. k, I, figurines. m, seated figure. n, paint mortar. CXX VIII. Pottery images from the Panuco valley (Presaley collection) : a, h, t , d, figurines. e-h, clay heads. 4, seated figure. j, figurine smoking (?). h, unknown quadruped. I, bowl with legs. CXXIX. Pottery from Ohampayan lagoon, Tampico: a, vase with meridional swelling. b, vase with painted curved designs. c, undecorated vase. d, e, painted vases. Page Figure 44. Battlements of building A, Cempoalan 240 45. Stone idol at Xico Viejo (front view) 247 46. Stone idol at Xico Viejo (back view) 247 47. Stone idol at Texolo (side view) 248 48. Stone idol at Texolo (back view) 248 49. Stone ring (Deheaa collection) 252 50. End view of arms of stone yokes (Museo Nacional, Mexico) 254 51. Side view of stone yoke, second group (Dehesa collection) 256 52. Side view of stone yoke, second group (Dehesa collection) 257 53. Convex side of curved stone (Museo Nacional, Mexico) 258 54. Concave side of curved stone (Museo Nacional, Mexico) 259 55. Paddle-shaped stone representing lizard (Dehesa collection) 262 56. Reverse of paddle-shaped stone 263 57. Paddle-shaped stone with dancing figure (Dehesa collection) 263 58. Reverse of a, plate ex vii 263 59 Bird-shaped stone with notched base (Dehesa collection) 264 60. Sling stones (Dehesa collection) 265 61. Sling stones (Dehesa collection) 266 62. Snake idol from Jalapa Viejo 266 63. Maize goddess (Dehesa collection) 267 64. Rain goddess (Dehesa collection) 268 65. Bowl frorn Cempoalan 269 66. Clay image (Dehesa collection) 270 67. Serpent-god idol 278 68. Stone idol at Altamira , 279 69. God with staff (Pressley collection) 280 70. Melon-shaped vase with handle (Pressley collection) 283 INTRODUCTION The geographical position of the Gulf states of Mexico gives them a special significance in comparative studies of the prehistoric culture of the mound builders of the lower Mississippi valley and that of the Maya and other tribes of the far south. Notwithstanding this fact very little has been contributed in recent years " to our knowledge of the archeology of this interesting region, and comparativel^y little is known of the culture of the prehistoric races that inhabited it. With hopes of increasing this knowledge the author was directed in the winter of 1905 to visit portions of these states for field work under the auspices of the Smithsonian Institution. Some of the general results of this visit are published in the following pages. When Hernando Cortes disembarked his little army of invasion in what is now the state of Vera Cruz he found it inhabited by aborigines of comparatively high culture. The inhabitants called themselves Totonac, and their territory was known as Totonacapan. The con- queror was not long in discovering that the Totonac were subjects of Moctezuma, a great ruler in the mountains to whom they unwill- ingly paid tribute, and that they chafed under his yoke. Shortly after landing Cortes visited their settlements at Quiauistlan and Cem- poalan, near the former of which he laid the foundation of a city that he called Villa Rica de la Vera Cruz, the Rich Town of the True Cross. He was well received by the inhabitants of these cities, making friends with those above mentioned and thirty other dependent pueblos whose aid greatly facilitated his march to the interior pf Mexico. But this friendship of the natives of Cempoalan and their settlements for Cortes was not shared by all the Indians of the Mexican Gulf coast. In the valleys of the Panuco and Tamesi rivers, that is, in what is now northern Vera Cruz and southern Tamaulipas, dwelt the so-called Huaxtec, a people linguisticajly allied to the Ma^a and culturally similar to the Totonac. They had populous towns, having reached a high degree of culture, and thej' had never been conquered by the Aztec. At first they resisted the Spaniards, but subsequently were subdued by Cortes and their main city, called Chila, situated on the Panuco rivei-- about 15 miles from its mouth, and certain other settlements on « A valuable summary of what is known of the ruins in these states may be found in Bancroft, The Native Races, iv (Antiquities), San Francisco, 1882. Mr Hugo Fink, in Smithsimian Report for 1870, p. 873-375, refers to the abundance of antiquities in Vera Cruz, 231 232 INTRODUCTION lagoons of the Tamesi near the present pueblo, Altamira, were destroyed. The survivors of these villages who escaped slavery or massacre fled to the mountains, where their descendants, bereft of ancestral arts, lost much of their culture and settled in new localities. Let us begin our account with the Totonac ruin, Cempoalan, and follow with a brief description of prehistoric earth mounds near Xico," a Nahuatl pueblo not far from Jalapa, closing with a brief mention of that near Antigua, the modern name of the second Villa Eica de la Vera Cruz founded bj' Cortes. (1 A contracted form of Xicochimalco; a better known pueblo of the name Xico is situated on an island of Lake Chalco near Mexico City. RUINS OF CEMPOALAN The earliest historical references to Cempoalan occur in the accounts of the Conquest by Bernal Diaz del Castillo," Francisco Lopez de Gomara,* and other contemporaries. At the time of the Conquest Cempoalan was so striking a metropolis that it excited the admiration of the Europeans, and from its many temples ("towers") and large buildings was called Se villa. Its streets and plazas are said to have swarmed with people, one author estimating the population at 30,000 souls. Whether this statement was exaggerated or not we may never know, but the size and number of the temples prove that the city had a considerable population. After the Conquest' Cempoalan rapidly declined in power and its population so dwindled that in 1580, accord- ing to Patino,'' it had shrunk to 30 inhabited houses; it is stated that in the year 1600 only one or two Cempoalanos lived on the old site, the most of the sui'vivors having been moved to the jurisdiction of Jalapa, where they were distributed in new "congregations" by the then Viceroy of Mexico, the Count of Monterey. The adjacent forests and an exuberant tropical vegetation rapidly grew over the deserted build- ings of the once populous city, so that in a few generations its site was practically forgotten by students. Regarding the position of the ruins, Bancroft writes as follows :** About the location of Cempoalan, a famous city in the time of the Conquest, there has been much discussion. Lorenzana says that the place "still retains the same name; it is situated 4 leagues from Vera Cruz and the extent of its ruins indicates its former greatness." Rivera tells us, however, that "to-day not even the ruins of this capital of the Totonac power remain, although some human bones have been dug up about its site." All the old authors agree that the people who inhabited Cempoalan belonged to the Totonac stock. This identification gives the study of this ruin both an archeologic and ethnologic importance. A student of the antiquities of Cempoalan need not doubt the kinship of its inhabit- ants; but regarding the affinity of the inhabitants of many other Vera a Historia Verdaclera de la Conqvista de la Nueva-Espaiia, Mexico, 1632. f> Cronica de la Nueva-Espana. In chapter xxxii of this work -the author deacribea a plaza of Cem- poalan with rooms on one side and towers on the other, the walls of the latter shining in the sun like silver. Manuel Rivera, Historia Antigua y Moderna de Jalapa y de las Revoluciones del Estado de Vera Cruz, Jalapa, 1869. The -author had an opportunity to examine this work in Jalapa and from it obtained the above statement ascribed to Patino, whose writings were not seen. According to Rivera, Alonzo Patino presented a "piano" of Cempoalan in 1580 to Martin Enriquez, but much to his regret the author has not been able to see this plan. d Bancroft, The Native Races, iv (Antiquities), 436, San Francisco, 1882^ 233 234 CERTAIN ANTIQUITIES OF EASTERN MEXICO [eth. ann. 25 Cruz ruins there is not the sume certainty. Some are Nahuatl, manj are Totonac, and still others were once inhabited by people of unknown stock. Although lost and forgotten by the outside world, the name of the Totonac metropolis clung to a geographical locality near the left bank of the Actopan river, where certain mounds and ruined pyramids are still known to the people of the neighborhood as the remains of ancient Cempoalan. In modern times the attention of archeologists was first called to this site by Sra Estefania Salas, a lady of Totonac extrac tion, still living in Jalapa, who was then a zealous collector of land shells. In 1883 Dr H. Strebel, led by information furnished bj' Sra Estebania and others, published an illustrated account of six of the temples of Oempoalan" that represented for several years all that was known of the ruins.' Descriptions of objects from Cempoalan appeared also shortly after in Strebel's work, Alt-Mexico,"- which has long been the authoritj' on the antiquities of Vera Cruz. Strebel apparently had not visited Cempoalan when his articles were written, and he makes no attempt to locate the geographical or relative positions of the buildings he describes. In 1891, eight years after the publication of Strebel's work, in commemoration of the fourth centenary of the discovery of America by Columbus the Mexican Government made a survey of Cempoalan and neighboring ruins, under direction of the well-known Mexicanist, Sr Paso y Troncoso. At that time the dense, almost impen- etrable jungle covering the mounds was thoroughly cleared away and the walls of several large buildings, including those described l)y Strebel, were laid open to view. The whole ruin was then surveyed by an engineer, Pedro Pablo Romero, and a model prepared of the central buildings adjoining a court identified as tjie Plaza Mayor. In the course of the work here and in the adjoining Totonac region more than two hundred photographs were taken and much valuable material was collected. The models and plans were exhibited in the Columbian Exposition at Madrid in 1892, where they attracted considerable atten- tion, and an account of the material as well as of the different temples was published in a catalogue '' of the exhibits that appeared at that time. The above-mentioned model and plans, with crayon copies of some of the photographs enlarged bj^ Sr Jos6 M. Velasco, are now on exhibition in the Museo Nacional in Mexico cit}^ and the collection of photographs preserved in the library of the same institution is open to inspection. With the exception of a visit of Senor Batres, official "As so often happens in Mexico, the same name is applied to several places. The Cempoalan near Paso del Ovejas from ita position can not be the historic city conspicuous in the conquest of Mexico. ''Die Ruinen von Cempoallan in Staate Vera Cruz (Mexico) and Mitteilungen iiber die Toto- imken der Jctzeits, Ablmndlmigen des Naturwisaenschaft Vereins lu Hamburg, vii, Teil 7. 1883. c Alt-Mexico, Archeologischer Beitriige zur Kulturgeschichte seiner Bewohner, Hamburg, 1P8'>. dCatalogo de la Seccion de Mexico en la Expo8ici6n Hist6rico- Americano de Madrid, 189'2, tomos, i-ii, 1892-3. PBWKES] ETJINS OF CEMPOALAN 2B5 inspector of archeological monuments and one or two others, the ruins of Cempoalan passed the next decade without being disturbed or even visited, and a new jungle spi'ead itself over the stately pyra- mids. The author made two excursions to Cempoalan in February-, 1905, I'emaining there a week on his second visit. The limitation of time prevented extended work at the ruins, but photographs of the main buildings were made and data regarding them collected. Even this limited work was attended with somediiEculty, since the clearings made by Troncoso in 1891 had already disappeared, the trees and underbrush having grown to so great an extent as to obscure the build- ings, making it difficult to secure satisfactory photographs. Although much of this vegetable growth was cut away by the owner, the jungl<» is still dense over the greater part of the ruins." A visit to the ruin Cempoalan can be readily made from Jalapa or Vera Cruz in a single day. It lies not far from the left bank of the Chalchalicas or Actopan river, a short distance from the coast and two hours' ride on horseback from a station on the railroad between Jalapa and Vera Cruz, called San Francisco. The roads (plate xoiv) from this station to the ruins are fairly good, passing through a com- paratively level country, lined in part with groves of tropical trees to which cling beautiful air plants, and in the branches of which live many parrots and other brilliantly colored birds. The shortest road passes through hamlets called Gloria and Bobo, and near the latter is a ford of the Rio Actopan. Although the road from San Francisco to Bobo at first is uninteresting, distant mountains are always visible, and as the traveler approaches the river trees are more numeroxis and the country becomes more attractive. On the right side of this road before fording the Actopan * there are several artificial moUnds belonging to the Cempoalan group, the first being passed a mile from what was once the central plaza of the city. After fording the river the traveler crosses several irrigating ditches and the cultivated fields increase in number, showing evidences that much of the plain on the left bank of the river is fertile and once may have been extensively farmed.'^ a My investigations in the state of Vera Cruz were greatly facilitated by Governor Dehesa, who not only gave me permission to study his valuable collection, but also directed the owner of the hacienda, Agostadero, and the alcalde of San Carlos, to aid me in every way. I take this opportunity to express my gratitude to Governor Dehesa for this and many other kindnesses. Don Firmen Zarete, owner of the property on which Cempoalan stands, Sr Alejandro Viu, alcalde of San Carlos, and Mr Qaw, of Jalapa, also rendered valuable assistance, for which I wish to thank them. I was accom- panied to Cempoalan and Xioochimaloo by Senor Ximenes, photographer of the governor of Vera Cruz. i G6mara says the river crossed by Cortes was about a mile from the court of Cempoalan, which is approximately the distance of the Bobo ford (Paso del Bobo). He also mentions the irrigated gar- dens (Huertas de regado), which were evidently north of the Actopan, through which the road passes. From Gbmara's account it appears that Cempoalan was not a compact city with buildings crowded together, but composed of many clusters of buildings, each s\]rrounded by gardens, and groves of trees so tall that the buildings were not visible from a distance. Actopan, according to Alonzo de Molina, means land, rich and fertile. Its Aztec rebus is a maize plant growing out of an irregular circle filled with black dots. Note, however, the difference in spelling the name of the river, Actopan and Atocpan. 236 CERTAIN ANTIQUITIES OF EASTERN MEXICO [eth. ann. 25 The visitor may obtain shelter and food near the ruins at the hospi- table hacienda, Agostadero, owned by Don Firmen Zarete. This set- tlement consists of a collection of primitive cabins of the simplest construction characteristic of the Tierra Caliente, clustering about the house of the owner. Evidences of the older population crop out everywhere in this region, and well-defined rows of rubblestones mark the remains of foundation walls of old temples that have been appropriated for the same purpose in modern cabins (plate xcv). It would appear that practically the majority of the houses of Agos- tadero are built on walls of the older settlement, and that the present inhabitants cultivate the same fields as their prehistoric predecessors. There are indications that these fields are still irrigated by water drawn from the Rio Actopan, as in prehistoric times. The ruins of Cempoalan are quite extensive, covering a large extent of territory, but^ as a majority of the mounds are inaccessible except by cutting one's way through the underbrush with a machete, the locations of their sites can be only surmised. Mounds belonging to this metropolis were found extending over a territory a mile square, but the main buildings are crowded into a limited area. Wherever one turns in this neighborhood, if vegetation permits he encounters evidences of former human occupation. Not only mounds and pyra- mids rise on all sides but also plastered walls, and fragments of con- creted road-beds lined with rows of stones set in cement (not unlike curbs) are seen on all sides. It does not take long to discover that Cempoalan was constructed alrbost entirelj"- of .plaster and rubble- stones ; " none of its walls were made of adobe or of cut stones. Construction of Buildings So far as can be determined, the four buildings of old Cempoalan now standing are pyramids, the bases of former temples. The}' are constructed of a concrete core made of water-worn stones laid in lines one above another and faced with concrete. Wherever thi^i super- ficial covering has fallen, especially on the stairways, rows of stones are clearly seen. ' The surfaces of these buildings were originally highly polished, so smoothly that it was supposed hy one of the sol- diers of Cortez that the walls were covered with plates of silver. These walls were decorated with yellow and red paititings, traces of which are still visible, especially in places not exposed to the weather. Two typical forms of buildings are represented at Cempoalan, one circular, the other rectangular. Both types have stairways with mas- sive balustrades on one side. Examples of the circular type are not as well preserved as those of the rectangular, but their form is similar ■ to that of the temple now in ruins at Calera near Puente Nacional.* II In this respect unlike the Tptonac ruin Tajin, near Papantla. 6 See Bancroft, The Native Races, iv (Antiquities), San Francisco, 1S82. FBWKES] BUINS OF CEMPOALAN 237 The rectangular type' (plate xciv) may be still further classified into two groups, one of which (plates xcix, ci) has two stories, forming a basal and a second terrace on the latter of which stood the temple. The other group (plates xcvii, xcviii) has more than two stories or ter- races, dirhinishing gradually in size from base to apex. The top or upper platform of the latter group is reached by a continuous stairway " on one side, but in the former there are two flights of stairs, one above the other, the lower mounting to a landing or the platform of the basal story, from which the second flight of stairs takes its rise. As a rule the foundations of these Cempoalan pyramids are con- cealed by a luxuriant growth of vegetation, but it is apparent from the clearings of the foundations made here and there that all were built on slightly raised artificial bases, somewhat larger than those of the pyramids. In some instances smaib buildings or annexes of the temple were erected on the same foundation platforms as the pyramids. Nomenclature and Position of Buildings One observes the first group of temple mounds of Cempoalan on the left-hand side of . the road from Agostadero to San Isidro, just after leaving the hacienda. The largest belongs to the round type and lies in a cultivated field much overgrown with bushes and vines. Venturing into this field, wliiqh is full of ; troublesome insects, the ob- server discovers that near these two mounds are others forming a group. One of the largest of these two mounds (plate xcvi) was called by Troncoso Templo del Aire ("Temple of the Air"*), and like allround temples is supposed, ^p have been dedicated to Quetzalcoatl, Plumed Serpent, or the God of the Air. The many smaller mounds are crowded together, indicating houses once possibly inhabited by priests. On the right of the road crop out fragments of walls, some of which, extending into the adjacent forests, are Ipst in the jungle while others continue pai-allel with the road for some distance and fartlier on dis- appear. Slabs of plastering or rows of rubble stones extending in all direc- tions indicate the crowded arrangement of houses in this immediate locality, which must have been not far from the center of the city. Just beyond the second of the two mounds identified as temples of the Plumed Serpent, there enters from the left the road to San Isidro, a little-traveleid pathway (plate xciv) that follows the barbed-wire fence of the field in which lie the circular ruins. Making one's way with some difiiculty through a dense forest along this pathwEiy a short quarter of a These stairways are not unilorm in their orientation; that of building A faces west, that of B, south, while those of C and D face east. 666mara,(CroniGa, p. 83), in speaking of the temples' of the City of Mexico, says: "I entre ellos' (teocalli) havia vno redondo dedioado al Diosdel Aire, dicho Quezalcovatlh; porque asi como el Aire anda al rededor del Cielo, asi hacien el Templo redondo.'V There are many other references to the round temples of the Air god, Quetzalcoatl. 238 CERTAIN ANTIQUITIES OF EASTERN MEXICO [bth. ANN. 25 a mile north of the Templo del Aire, one suddenly sees rising before him, in fact, blocking the way, a pair of massive pyramids (plates xcvii, xcviii) that evidently formed parts of two sides of an inclosed court. An observer facing the larger of these, with the smaller on the left hand, probably stands in the great court of Cempoalan, where, perhaps, Cortes marched his' soldiers on his memorable visit to this city almost four centuries ago. The larger of the two massive pyramids (plate xcvii) is locally known as Templo del Pozo (" Temple of the Fountain "), or Chinimeos (" Chim- neys ") ; the other is nameless. Peering into the jungle that surrounds these buildings, we get glimpses of other mounds hidden for the greater part in the dense forest. Passing onward between the two great pyramids (a, b) already men- tioned, leaving the larger on the right, following a fallen wall one descends by a few artificial steps (plate xciv) to a plastered pathway par- alleling to the eastward a barbed-wire fence. This trail brings one in a short time to one of the best preserved buildings (plate xcix) in Cem- poalan, locally called Las Caritas ("Small Heads"), from the many small pottery heads that have been found at its base, apparently having fallen out of the walls. A fourth pyramid (plate o), sometimes called Casa de Moctezuma, lies in the forest about due east from that last mentioned and is approached by a circuitous trail through the woods. This structure is likewise the pyramidal base of a temple but is less shut in by the forests than those already considered. In an open field north of the temple Las Caritas, and to the left as one passes to it from the main plaza, there will be noticed a large structure (plate cm) overgrown with shrubbery, from which project smooth polished faces of cement walls. This is one of a group of mounds designated by Trancoso, Sistema de los Paredones, and is fig- ured in the accompanying illustration. The several buildings above referred to are designated by the letters a, b, c, d, e. The name Templo del Aire is retained for the round ruin. BUILDING a This building (plate xcvxi) is one of the large pyramids in the main court and, judging from its present size and annexes, must have been one of the most important structures in Cempoalan. Its ruins are still impressive and, considering the material used in construction, in a fair state of preservation. The pyramid is simple" and has several adja- cent minor buildings evidently belonging to it, forming a cluster. The front of the pyramid is indicated by a staijway, before which is a sec- ond building, longer'than broad, the roof of which was supported by n In the author's description the term " temple " is applied to the room on the upper terrace, and "pyramid" to the solid terraced base upon which this sanctuary stood. FEWKES] EUINS OF CEMPOALAN 239 columns, two of which are still visible at the base of the pyramid. This building was possibly an antechamber or gateway, a waiting room for those who' took part in the ceremonies in the temple on the pyramid, A third structure on the same base as the others is a roof- less inclosure, situated in the rear side of the pyra^nid and extending the whole width of the basal platform." All these buildings stood on a common platform that was slightly raised above the surface of the court or plaza. The steps by which one mounted to the platform are still visible. The accompanying illustration (plate xcvii) shows this pyramid as seen by one facing the stairway, which is continuous from the base to the apex. At the foot of the stairs are seen the broken remains of hollow, chimney -like plaster columns that once supported a roof, for- merly decorated on their flat sides with stucco figures. Adjoining them are fragments of the foundations of old walls of the room. Three of the four pillars appear in the illustration, the missing one having been broken off at its base and covered with rank vegetation and other debris. The round pedestal of solid concrete seen a little to the right in the foreground of the plate resembles a pillar, but is in reality an altar, the remains of which stood in front of the pyramid. An exami- nation of the structure of the rear and sides of the pyramid shows that it had six terraces, the size of which gradually diminishes from the base to apex, the upright walls being inclined slightly inward from the perpendicular. The plaster covering the surface of the lowest story is somewhat more broken than that of the upper, revealing the rows of rubble stones laid in the concrete. which forms the interior. On the left face of the pyramid, about midway from each corner, a row of pits, one in the surface of each terrace, forms a continuous series of foot holes, by means of which one could ascend to the top of the pyramid without making use of the main stair, a feature not found in the other Cempoalan temples. The stairway is continuous from base to top and has a massive bal- ustrade on each side, following the angle of inclination of the steps except at the top, where it ends in a cubical block, the sides of which are practically perpendicular. This struct^ure, like all other parts of the pyramid, is made of plaster applied to a core of water-worn stones laid in concrete. On ascending to the spacious top or upper platform of the pyramid, remnants of the temple walls are found somewhat back from the land- ing of the stairs. These walls, now fallen or broken, once formed three sides of a chamber, the fourth side being occupied by a door- way. It would appear that formerly there were two idols in this temple, the pedestal on which one of these stood being still visible to n Possibly the bodies of tliose sacrificed in the temple were thrown down the pyramid into this inclosure. 24