HomeBiographyWritingsPhotosContactAncient Maya Settlement Patterns and Environment at Tikal, Guatemala: Implications for Subsistence Models Dennis Edward Puleston A Dissertation in Anthropology, University of Pennsylvania, 1973

Chapter Seven: Environmental Aspects of the Cultural Ecology of the Lowland Maya


Geology and Topography


Don Callendar preparing for jungle camping during mapping of the survey strips


One of the great problems with respect to the Classic Maya has been the search for an explanation as to how this complex and civilized people were able to achieve such success in the face of a tropical forest environment. In comparison to other "cradles of civilization" in the Near East, Asia, North Africa, and the Highlands of Mexico and Peru where semi-arid and comparatively open conditions prevailed, the Lowland Maya situation seems to represent a real exception. Now with an understanding of the indicated settlement and population densities the entire situation is even more incredible. Only in Cambodia, where the Mon Khmer, who have been compared to the Lowland Maya by M. Coe (1957), offset many of the difficulties through the utilization of steel tools and an immediate and abundant river protein resource (Briggs 1951:37), is there even average similarity. Since neither of the latter advantages were available to the Classic Maya of Tikal, their evident success is even more puzzling. In my master's thesis (Puleston 1968) the hypothesis was presented that the fruit of the ramon tree (Brosimum alicastrum) helped the Maya convert the liability of dense and exuberant vegetation into an asset. With the exploitation of a food-producing forest, the necessity of removing forest vegetation in order to plant domesticated annuals would have been largely eliminated.

Thus, in view of the potential and the uniqueness of the Lowland Maya environmental situation, it is clear that the physical and biological aspects of their environment deserve careful attention. One should note here that the interrelationship between a culture and its environment may be viewed from two sides: the effect of the environment on the culture and the effect of the culture on the environment.

From the standpoint of the influence of the environment on human and culture, one of the principal areas for consideration lies in the realm of subsistence and technology. The effect of a particular environment on a particular society and its culture will largely determined by the subsistence strategies and technology of the group concerned. Here we touch upon the hypothesis of techno-environmental and technoeconomic determinism recently championed by Marvin Harris as a nomothetic principle in an emerging strategy of "cultural materialism" within anthropological theory.

This principle holds that similar technologies applied to similar environments tend to produce similar arrangements of labor in production and distribution, and that these in turn call forth similar kinds of social grouping, which justify and coordinate their activities by means of similar systems of values and belief (Harris 1968:4).

Assuming that there is some basis for such a principle, we are left with little as to specifics. What aspects, for example, of environment and technology are to be measured for similarity? In the case of the Lowland Maya, we have an environment that is outwardly very different from that in which we find civilizations evolving in other parts of the New World as well as the Old World. Are there differences more apparent than real, or is it perhaps that "arrangements of labor in production and distribution . . .  social groupings . . . systems of values and beliefs" of the ancient Maya were in some sense very different from those of other civilizations? These are questions that remain to be answered by archaeologists working in many parts of the Maya Lowlands. I will begin here, however, with a review of the physical and biological aspects of the larger southern Maya Lowland eco-system with the hope of revealing those aspects which were of particular significance in the Tikal area.

Geology and Topography

The Maya Lowlands comprise a total area of nearly 100,000 square miles. Geographically this includes Quintana Roo, Yucatan, Campeche, parts of Tabasco and Chiapas, Belize (British Honduras), northern Guatemala, and the western most portions of Honduras. North of Lake Peten the surface geology is characterized by karsted limestone dating back to the early Cenozoic period. The western most portions of the Maya Lowlands on the Gulf Coast, however, are overlain by comparatively recent alluvial deposits (West 1964, fig. 18). The limestone bedrock was an important natural resource for the ancient Maya. Where it was hard enough to be cut into blocks, it was used extensively for masonry. The softer limestone (cal) could be burned to make quicklime which, when mixed with calcareous sand (sascab) and water, produced a mortar that was used extensively as a cement for construction as well as for plasters and stucco. Chert, in the form of hard siliceous nodules, occurs with great frequency in the limestone and was used to make a broad range of tools. Chert nodules are generally found on ridge tops where they are exposed by weathering down of the limestone bedrock and are not more abundant or of better quality at Tikal than elsewhere, in my experience.

Local variations in surface geology and topography were evidently of considerable importance to the Maya all over the Lowlands. In northern Yucatan, the fact that water could be obtained from sinkholes and caves must have often played a part in determining site locations. In contrast to more northern portions of the peninsula, the region around Tikal cannot be considered to be typically karsted in the way the north is. Nevertheless, shallow funnel-shaped sinkholes, locally called resumideros, occur fairly frequently. However, these sinks of doline type have been distinguished from cenotes by Robles Ramos (1950). A number were found on the west and south survey strips of the present survey (see Plates 3c, 2d, 2h). All these features, designated as sinkholes on the survey maps presented here, apparently serve as drains when the water is high during the rainy season and undoubtedly connect to underground caverns and streams. One of the most impressive of these in the Tikal area is located 4.3 km. ENE of the center of Tikal on the edge of what appears to be a fault. This sinkhole is located only ten meters or so north of where the fault is crossed by the trail shown on the large scale map of the National Park. At various times members the Tikal Project, including myself, have visited this sinkhole during the rainy season to see the large whirlpool that is formed as the water enters it. During the dry season it is filled with leaves and rotting logs and no entrance or water is visible in this or any of the other resumideros in this part of the Peten.

West of Uaxactun, up on the El Palmar ridge, I found sinkholes of the cenote type, the only difference being that they do not reach the water table. They are in the form of conical pits up to ten meters deep. Stalagtites were visible on the walls of one of these. These features are mapped on the Ricketson Survey as chultuns, including some, perhaps all, of the series 41 through 49 on the end of the west arm of the Ricketson Housemound survey (Ricketson and Ricketson 1937: fig.2)! Sinkholes of this type have not turned up in the vicinity of Tikal, and perhaps are found only on the higher El Palmar ridge.

Another geological feature unique to the El Palmar ridge is small caves of alabaster. These vary from five to ten feet in height and three to five feet in depth. The location of one is indicated on the Uaxactun Survey Strip at 3 km (Plate 6c). Though there was no obvious evidence of exploitation of these caves by the Maya, alabaster objects have been found in Tikal burials. On this basis of these discoveries, it would be useful to correct the assumption of Tourtellot and Sabloff (1972:126) that alabaster was an exotic substance that played a role in the external exchange relationships of the Classic elite. These El Palmar caves do not represent the first reports of alabaster in the southern lowlands. In fact Cortés, in his fifth letter to Carlos V, describes alabaster deposits north of Lake of Yasuncabil, he writes, his route ". . . led over high and rocky mountains, and I had to cross a dangerous pass of which al the rocks were of very find alabaster, hence I named it Puerto del alabastro" (Cortés 1908:269).

The surface configuration, which played a major part in settlement patterns in this region, has been determined by two principal factors, folding and erosion. The southwest-northeast orientation of hills and drainage, particularly noticeable to the south and southeast of Tikal, appears to be the result of folding. The long ridges are often broken up into roughly conical hills called "hums" (West 1964:73). The ridges and hums appear to have been preferred for settlement. The El Palmar ridge, northwest of Tikal, with exposed limestone escarpments, visible from Temple IV, also appears to be the product of folding and faulting. As one follows the edge of the ridge counter-clockwise from southwest to northeast it curves around until it runs north and south. This curve is concentric to the more gradual curves to the southeast and indicates that the El Palmar Ridge, in the Tikal National Park, falls on or a little to the east of the axis of the U-shaped series of folded limestone ranges that cross the base of the Yucatan peninsula (Wet 1964:71 fig. 18).

The largest concentration of major Classic Maya sites in the southern Lowlands seems to occur to the east of the same axis where the ridges are interspersed with large bajos or swamps composed of hydromorphic soils. Apart from general reduction of the landscape, erosion has cut many relatively steep arroyos in the hills. Good examples of these can be seen on the map of Tikal Report No. 11. One of the largest goes right through the center of Tikal, beginning with the Palace and Hidden Reservoirs, before continuing northeast on the other side of the Mendez Causeway towards the Tikal Reservoir. The short unnamed causeway that crosses the east end of the Palace Reservoir must be entirely artificial. Other less altered examples may be seen west and north of Temple IV.

The material removed by this erosion has been converted to clays and deposited in the broader depressions to form the bajos mentioned above, which are one of the geographical hallmarks of the northeast Peten. The largest of these bajos include the Bajo de Santa Fe, East of Tikal, and the Bajo de Joventud which stretches east and north of Uaxactun. Many smaller patches of bajo exist; irregularities in the hills act as dams behind which the clays have accumulated as in the small bajos north of H-Group (Square 2D on the Tikal maps). Another bajo lies south of the group which includes Str. 6B030 (the "Barringer Group"). Steps or platforms on the hills themselves sometimes allow such accumulations as in the area of the Aguada Subin in Square 4B. All these features have had a profound effect on settlement patterns in the northeast Peten, for it was in the clay soils of the bajos that water-retaining aguadas could be constructed.

Though fairly detailed studies of the subsurface geology have been undertaken for the Esso and Argus petroleum companies, unpublished portions of these data are not easily accessible and little can be reported here. One of the most recent available geological descriptions of the area by Vinson (1962) indicates that surface material is underlain by Lower Eocene limestone of the Sanata Amelia formation, consisting of cream-colored microgranular dolomite interbedded with evaporatic clays. Tikal, however, lies very close to contact with the slightly younger Buena vista formation described as being of a more massive limestone with intebedded gypsum. Samples of Tikal bedrock examined by Cowgill (Cowgill and Hutchinson 1963:5) were more calcitic than dolomitic, though large and apparently natural dolomitic boulders occur at Tikal; some of the best examples may be found at the head of the ravine in the northwest corner of Square 5E on the Tikal site map (Carr and Hazard 1961). It was this stone that was apparently used for the sculpture of Early Classic monuments at the site.

Informal communications from drillers who contracted for the petroleum companies reveal that the limestone deposits in the Peten are up to 610 meters (2,000 feet) thick. Though limestone caverns appear to be rare on the surface in the Tikal region (the nearest known are the caves at Jobitzinaj just south of Lake Peten), they apparently exist beneath the surface. Otto Dauber, one of the contractors referred to above, reported that large caverns, in which water could be heard running, were encountered far beneath the surface during drilling at the Gato Salvaje camp 20 km south of Tikal. The presence of water in the above case was confirmed by the fact that several plane-loads of cement poured into the hole to fill the cavern had been carried away by the time drilling could be resumed.

The great depth and porosity of the limestone bedrock in the region is certainly one of the reasons there is so little surface drainage of water. Subsurface drainage is very deep and played a major role in ancient Maya dependence on the development of reservoirs or aguadas. Wells drilled at Tikal in the 1950's in an attempt to reach water went to a depth of about 180 meters (600 feet) without success, despite the assurances of the water witcher, Henry Gross. Water is not so deep in all parts of the Peten. Cunningham (1948:24, see Hester 1954:52) reports ground water within 50 feet of the surface in a "natural well" 12 km southwest of La Libertad, south of Lake Peten. Other wells had water within 20 feet of the surface. The southern edge of the Tayasal Peninsula is pocked with sinkholes characteristic of karst topography some of which contain water. I am not sure whether or not this water derives from contact with a water table. While it may represent a perched water supply maintained by local rainfall, this alternative seems unlikely in view of general porosity of the limestone where it is not sealed by sediments.

In the process of earlier attempts in 1956 to reach water at Tikal, a deep pit was excavated by hand in front of Antonio Oritiz's Hotel. I am unaware of the exact location though it is now used as a septic tank (W. R. Coe, personal communication). While it is unfortunate that geologic samples were not taken from this pit, personal reports from those who saw and worked in the pit are that alternating layers of very hard and soft limestone were found. Small open pockets containing calcitic crystals were occasionally found. Needless to say, water was not reached.

Concern about the availability of water has ed students of Maya studies into an interesting but untenable hypothesis--namely that abundant water resources once occurred in these regions but have since disappeared. C. W. Cooke (1931) was the first to suggest that the logwood bajos which so characterize the Peten today are all that remain of extensive lakes which were filled in by catastrophic erosion of upland soils as a result of removal of the foret vegetation to make milpa. At least three attempts have been made to test this theory by examination of bajo soils. Wauchope (1965:13) describes taking a fourteen-foot core with a pipe, apparently in the Bajo de Joventud in 1932. In the same year, A. L. Smith (Ricketson and Ricketson 1937:11) excavated a 5.5-eter-deep pit in the same bajo in whichevidence for lake deposits was supposedly found. The contradictory nature of the evidence, however, makes the conclusions highly suspect. Ricketson (Ricketson and Ricketson 1937:11) says that "the bottom 2.05 m. [of the pit] were composed of highly calcareous clay or marl which evidently accumulated very slowly in water. The sample from above the 2.05 m. level is black clay . . . . Between the 4.46 and 4.81 m. levels the deposit contains water-worn pebbles. . . "(he seems to e measuring from the bottom of the pit upward). However, A. L. Smith, in a personal communication about the excavation to U. Cowgill (Cowgill and Hutchinson 1963:7), has the gravel layer near the bottom of the pit, and Cowgill interprets Cok's information as if the 2.05 m. was at the top. ":Cooke states that the upper 2.05 meters of material resembles the soil of the uplands."

The third attempt to deal with the bajo-lake hypothesis may be found in Ursula Cowgill's own excavation in the Bajo de Santa Fe nearer Tikal. On the basis of her exhaustive studies of sediments from a 5-meter pit, she concludes that the despoits are the result of "sedimentation in a swamp rather than in a lake" (Cowgill and Hutchinson 1963:41).

Other geological features which may or may not have had significance for the Maya include salt domes that occur north of Uaxactun. In 1967 an abortive attempt was made by the Ghetti Company to drill for sulfur in that region.


Don Callendar reconstructing pottery in Tikal laboratory


Rainfall and temperature variation were of great importance to the Maya. Following the hypothesis of Holdridge (1947), one can say that natural terrestrial ecosystems are regulated by and are thereby subordinate to climatic factors. Balance in this scheme is achieved when the calculated potential evapotranspiration ratio is equal to 1.0. The latter is achieved when potential evapotranspiration by vegetation and soil is equal to precipitation. In this sense, Holdridge has provided us with a definition of climax vegetation that is independent of species composition. He has further provided us with an indirect mens of measuring the efficiency and stability of agricultural systems imposed by man on a particular environment. Evapotranspiration rates are extremely difficult to measure, but because a direct relation between temperature and potential evapotranspiration can be expressed as a logrithmic constant, the ratio can be calculated on the basis of annual figures for rainfall and temperature. Data from Tikal for the years 1960, 1961, and 1962 has been summarized and used to calculate the potential evapotranspiration ratio for Tikal on the basis of the most recent version of Holdridge's three-dimensional model (Tosi 1964, Fig. 1).

                        Mean Monthly                                       Potential Evapo-
Year                Biotemperature           Total Precipitation             Transpiration Ratop
1960                    24.90o C                        1622 mm.                              0.9
1961                    25.34o C                        1514 mm.                              1.0
1962                    24.78o C                        1304 mm.                              1.1

This summary reveals that in Tikal, today, precipitated moisture is essentially equal to potential evapotranspiration. Recent studies suggest that this ratio is highly correlated to:

. . . the ratio of effectively dry to effectively wet months in the temperature growing season, to open water evaporation, and to actual evapotranspiration, and runoff. Moreover, because the moisture balance rather than total precipitation is most influential in soil genesis and morphology, further research may well show that this ratio is also correlated significantly with several important characteristic of soils . . . (Tosi 1964:179).

Briefly then, the potential evapotranspiration ratio may well be linked to the decline in the productivity of soils with the removal of vegetation for agriculture, the rate of water loss from aguadas by evaporation, and the rate of soil erosion by runoff. The possible implications of these relationships for the stability of food production under different types of agriculture are enormous and will be discussed later.

Moving now to the basic factors in this approach, temperature and rainfall, we find that they are influenced by a series of factors including latitude, the cycles caused by the earth's movement around the sun, winds, nearness to the sea, topography, and, to a significant degree, the activities of man. First of all, latitude places this region south of the Tropic of Cancer, where the difference between winter and summer temperatures begins to decrease as the equator is approached. Further, it places the Lowlands in the path of the northeast Trade Winds which sweep across the Caribbean from the Atlantic Ocean. As they cross the warm surface of the Caribbean, they absorb great quantities of water vapor. Though these moisture-laden winds cross the Maya Lowlands all year, they only produce rainfall when they are forced to rise and cool. This occurs when the thermal equator, along which air is ascending, shifts northward in the months of April, May, and June (Vivo Escoto 1964:192); and, as can be seen on the daily precipitation graphs (Gifs. 23 and 24), it is at this time that the summer rainy season begins. On a diurnal scale a similar relationship between local convection currents and rainfall seem to be what causes it to rain so regularly in the early afternoon during this period.

Figure 23. Daily precipitation for June 1959 through May 1961, Tikal, Guatemala. (This chart is continued in Figure 24.

Figure 23. Daily precipitation for June 1959 through May 1961, Tikal, Guatemala. (This chart is continued in Figure 24

Figure 24. Daily precipitation for June 1961 through May 1963, Tikal, Guatemala. (This chart represents a continuation of Figure 23.)


In  the graph on which monthly averages for rainfall have been projected, another phenomenon of great significance to agriculture is evident. This is the canicula of Yucatan or "Little Dry" or "Meager Season" of Belize (Wright et al. 1959:21), a short but intense drought which occurs at the height of the rainy season during the crucial growing period for maize.  Beginning in the latter half of July the canicula appears to last from a week to more than month.

The regularity and timing of this phenomenon, which follows on the heels of a short period during July and August when the sun's rays fall vertically on latitudes north of the Maya Lowlands, suggest that these events are related. According to Hester (1954:23), the canicular drought starts during the first week of July in northern Yucatan, though in a section of his dissertation entitled the "Statistical Obscurity of the Canicula," he reports that "the abruptness and impact of the canicula . . . is not obvious from a typical (rainfall) chart." Such is clearly not the case of Tikal. Though the canicula is not so obvious on the charts of daily precipitation (Figs. 23 and 24), it shows up very well on the monthly one (Fig. 25). This is logical, as the shift northward of the thermal equator would take the rain-causing convection currents with it. These would not return until after the summer solstice, and so would be north of Tikal longer than they would be north of northern Yucatan where the canicula would be shorter and thus less visible on the charts. There seems to be a lag in this process, as heavy rains following the canicula start in August and September at Tikal. As compared to northern Yucatan, where the true dry season begins in October or November, it does not seem to really get started at Tikal on the basis of the data presented here, until November or December. The canicula is of considerable significance to maize agriculturalists, for unusual dryness before this period or an abnormal extension of it of a week or two can cause an entire maize crop to be lost to drought, particularly in the north where total precipitation is less. When Hester (1954:28) was working in Yucatan, a canicular drought wiped out 70% of the entire maize crop for the village of Telchaquillo in 1953. While the canicula does pose such a thereat to farmers further south in the Peten, it is still longer, and local milperos testify that major portions of a year's crop are occasionally lost to this cause. Such losses appear to have occurred in the central Peten in 1971 (S. Loten 1972, personal communication).

Figure 25.  A chart for monthly precipitation and temperature for June 1959 through July 1963. For the temperature data the absolute minimum and maximum for each month are indicated by the dotted lines. The monthly average of daily minimums and maximums are indicated by the dashed lines. The solid line connects values for monthly precipitation totals. The canicular drought is more evident here than in the charts of daily precipitation presented in Figures 23 and 24.


Too heavy rains can be as destructive to maize as a drought; excessive humidity in the soil can ruin the crops (Reina 1967:6) and is one of the reasons why water retaining bajos are not preferred for milpa agriculture. Many root crops are equally sensitive to poor drainage; as little as 10 day's submergence kills the roots of manioc, resulting 100% crop loss (Bolian 1971).

Individual storms may come as quick intense showers, or as violent thunderstorms. In the case of the latter, the tempest is often preceded by rapid darkening of the sky and an ominous calm in which the breathless air seem almost electric. The silence is broken only by the strident flight calls of parrots as they fly to roost. Then from the southeast comes the sound of the wind as it tosses the branches of the forest. Small twigs, leaves, and swallows (which seem to be catching insects blown up out of the forest by the wind) fly before the powerful gusts. Close behind them driving sheets of rain can be heard approaching through the forest like a waterfall. Large portions of a field of standing maize can be leveled by such storms.

Truly shattering thunderclaps occasionally accompany these storms. In one case, in 1967, we experienced a hailstorm with hailstones up to half an inch in diameter. These occasionally do local damage to crops and gardens and are "most frequent during the first weeks of the rainy season" (Hester 1954:22). Single hailstorms are also reported for the summer of 1961 at about 4:00 p.m. (W. Coe, personal communication) and in February 1961 (A. V. Kidder II, personal communication) so they may not be infrequent. Though rain may continue for an entire afternoon, it is more often the case that after a short rainstorm the sky clears, and the sun is shining through the glistening leaves and branches within an hour or two after the commencement of the storm. Topical hurricanes cross the Yucatan peninsula once every few years. The hurricane of 1961 which crossed the southeast Peten virtually leveled those portions of the forest that lay in its path, according to the chicleros who were in the region at the time.

While I have no information on how wide this swath was, an idea of the destructive potential of theses storms can be gained from a glance at the 1945 hurricane which devastated the forest on a 25-mile front, from the coast to a point nearly 40 miles inside Guatemala. Around Tikal, the 1961 storm only knocked down large trees. There was a notable bias for large sapote trees (Manikara zapota) weakened by excessive bleeding for chicle sap. The latter is a fact which will have to be taken into account in future surveys of sapote stands and their possible relationship to ancient settlement.

Landa (Tozzer 1941:40) describes the effects of a hurricane which passed over Mayapan:

This wind overthrew all the large trees, causing a great destruction of every kind of game; and it destroyed also all the tall houses which, since they were covered with straw and contained fire on account of the cold, were set on fire, and they burned up a large part of the people. If any escaped, they were crippled by the blows which they received from the (flying) wood.

The many ways in which rainfall and storms intruded upon the lives of the Maya lead one to suspect that their Chaacs and Mams had considerably more significance for them than simply the timely arrival of rain for crops.


As mentioned previously, there is very little surface drainage in the central portion of the southern Maya Lowlands. This may be attributed to a number of factors, including: 1) the near equality of precipitation and potential evaportranspiration (i.e. there would be more run off if precipitation exceeded potential evaportranspiration); 2) the reduction of the impact of precipitation by vegetation and humus which permits water to "soak in"; 3) the high porosity of the limestone bedrock on the uplands which draws off a significant percentage of excess water that cannot be absorbed by the soil.

Nevertheless, during the height of the rainy season, before or after the canicula, a heavy series of cloudbursts can cause arroyos to turn into tumbling streams for several days. Bajos, filled with impermeable clays, and generally provided with only narrow outlets, may stay flooded for many weeks. Three arroyos are shown on the map of the National Park: the Arroyo Holmul; the Arroyo La Pava, both of which cross the South Brecha; and an unnamed arroyo which follows the North Brecha survey strip into the Bajo de Joventud east of Uaxactun. I have only seen the Arroyo La Pava and the arroyo on the north survey strip actually flowing, and in each case that was for only two or three days after a heavy rain near the height of the rainy season.

It was this lack of surface drainage and the inaccessibility of subsurface water discussed above that led the ancient Maya of the Tikal region to depend on largely constricted clay-lined aguadas. Stands of water, however, sometimes occur in apparently naturally impermeable stretches of certain arroyos, but such instances seem to be rare. One example may be the Naranjal Aguada, the location of which may be noted in the southeasternmost corner of the 16 km2 site map in Tikal Report No. 11. In fact, this feature may be the result of an artificial dam built across its lower end. A second example may be found in portions of the Arroyo Holmul near the center of the Santa Fe Bajo and in the same arroyo where it passes south of the site of Nakum. Here long ponds are formed during the dry season which have been permanent enough to allow populations of crocodiles (prob. Crocodylus) to become established.


The significance of soil studies to archaeology would be difficult to overestimate. As the very matrix for the bulk of archaeological data, they often provide important clues to the source and history of a particular deposit. More important than this, however, is how they affected the lives of the people who formerly lived on them.

Houses and temples are built on soils; roads and pathways passover them. soils have a great influence on local fauna and flora, and perhaps most important is their crucial role in food production. In his excellent summary of the present status of soil studies in Middle America, however, Rayfred L. Stevens (1964:265) points out that "soil is perhaps the least adequately studied of all the major features of the habitat of the Middle American Indians . . . .

Two major surveys, based for the most part on local testing and aerial photography, summarize data on the southern Maya Lowlands. A Guatemalan survey by Simmons, Tarano and Pinto (1959) covers the Department of El Peten. The report of the British Honduras Land Use Survey Team (Wright et al. 1959) has provided excellent data for the eastern portions of the lowlands in Belize. Apart from these broad based surveys, a number of small-scale studies have been carried out in direct association with anthropological work. For the most part, these have been oriented to a specific problem--soil fertility in relation to milpa agriculture.

One of the earliest of these was Steggerda's (1941:135-139) rather inconclusive study at Pisté in Yucatan in which he analyzed samples taken from four experimental agricultural plots over a period of six years. Re-analysis of the data by Cowgill (1961:7) reveals that the percentage content of the various elements studied is a function of the year they were collected, rather than the plot they came from. This suggests that variation can be attributed to differences in analytical technique from year to year, rather than to real variation in soil constituents. Morley (1956:135) refers to a similar ten-year study carried out at Chichen Itza. Hester (1954:145), on the basis of his own analyses, demonstrated the critically short supply of available phosphates as a nutrient element in the soils of Yucatan. More recently, Ursula Cowgill (1961;962) carried out a study of milpa soils near Lake Peten. Chemical analysis of the soil samples showed that potassium and magnesium increased with burning, while all other soil nutrients declined. Cultivation also brought about an absolute decline in all items measured, including total nitrogen, organic matter, plant-available phosphorus, exchangeable sodium, potassium, calcium, magnesium, and pH (U. Cowgill 1961:54).

In 1968 one of the first detailed, areal surveys in the southern Maya Lowlands was carried out at Tikal under the direction of Gerald W. Olson. The survey covered the entire nine square kilometers of central Tikal, as well as sites on milpas beyond the edges of the map, and on the North and south Brecha Survey Strips. Techniques were more comprehensive than those used in previous surveys. Preliminary analyses of the results have been presented in various reports and papers (Olson 1969, Olson and Puleston 1970, Puleston and Olson 1971, Olson and Puleston 1972).

Mapping was done in two stages. The first stage involved characterizing the local soil types by excavating pits to bedrock or as deep as was feasible at a total of twelve different sites. At these sites details were recorded on local vegetation, physiography, relief, slope, drainage and the like. Then, entering the pit, horizons were defined and described. Information on soil color, texture, structure, consistency, temperature, pH, and boundary definition were recorded. Samples were then collected from each horizon for laboratory investigation. As more pits were examined in this way, the characteristics of typical soil profiles in different situations became recognizable. Representative profiles were then selected and used to characterize the soils observed in the second stage.

For the second stage, auger cornings were made at 20 to 100 meter intervals over most of the area. Depth to limestone bedrock was determined where possible and the solid brought up by the auger were compared with those of the representative profiles so that soil areas could be defined.  Finally, an attempt was made to characterize soil under cultivation. At ten sites surface samples from cropped soils in milpas and adjacent forest areas were collected along with relevant cropping history.

Information resulting from the survey was relevant to many questions. Most surface layers of soils around Tikal are very dark colored and contain up to 30% organic matter.  Despite their shallowness, they contain a comparatively tremendous supply of readily available plant nutrients that apparently have accumulated over a period of thousands of years through vegetative cycling of nutrients in the leaf and forest litter. The evidence suggests, however, that when the tropical forest is cleared, burned, and suggests, however, that when the tropical forest is cleared, burned, and cultivated for milpa, there is a dramatic decline in the percentage content of organic matter, nitrates, and soluble salts as revealed by the analysis of four pairs of adjacent milpa and forest soil samples presented in Table VII. These data in part confirm Ursula Cowgill's (1961:53) research which showed an absolute decrease over the long run in all soil components measured as a result of cultivation. Burning apparently brought about a temporary increase in potassium and magnesium but had no effect on sodium of pH. The indicated losses undoubtedly may be attributed to a combination of factors including erosion, leaching, and oxidation. During soil survey and field sampling it was observed that milpa surface soils, upland silty and loamy soils were much lighter in color than forested areas with correspondingly lower contents of organic matter. It was further observed that temperatures of surface soil horizons in milpas reach much higher levels than surface soil horizons of the same soils under forest cover. Such great temperature differentials and exposure differences cannot help but have serious consequences on soil fertility by increasing the oxidation and volatilization of nutrients.

TABLE VII: A comparison of soils from adjacent forest and milpa sampling sites. Samples 107/108 come from the area of milpas at the east end of the Tikal airstrip 2 1/2 years after clearing. Samples 112/113 come from the area of a milpa near the Tikal Project Village (4F:S370,E70) cleared within half a year before the samples were taken. Samples 115/116 come from milpas NW of the Hotel (3F:S210,E170) which had been cleared and cultivated for nine years. Samples 118/117 come from a recently abandoned milpa also NW of the Hotel (perhaps SE of 3F:S210,E10). The most significant declines are in organic matter, nitrates and soluble salts. The increases in iron and aluminum in samples 107 and 117 may be attributed to the initiation of laterization.


The increase in temperature would also upset the water balance by causing the transpiration ratio to increase. A micro-environmental shift towards aridity would be unavoidable. Better management through the use of tree crops, mulching, fertilization, rotation, and son on, could reduce many of the detrimental effects of slash-and-burn agriculture on soil fertility.

Much of the soil area studied, within the central nine km2 of Tikal, had to be considered "disturbed." It was interesting to note that these soils have not recovered their original content of organic matter after more than a thousand years. They are generally lighter in color as well. Practically every mound group, plaza, causeway, and other place where the ancient Maya disturbed soils in the process of construction is characterized by soils with lighter-colored surface horizons than can be found in adjacent undisturbed soils. Where new soils have formed in soft limestone fill, as on the Tikal earthworks, or on the collapsed redeposited soil fills of house platforms, soil horizonation over the last 1000 years or so can be measured and sampled. The importance of maintaining levels o soil organic matter during cultivation is emphasized by the long recover time needed to reestablish this important aspect of soil fertility. It is worth investigating the possibility that even with fallowing periods of ten years or longer, soil fertility still suffers a continual, if gradual, decline under slash-and-burn agriculture.

Definite evidences of erosion were found during field investigations of soils. In areas which drained the site epicenter large alluvial flats with more than half a meter of silty, loamy, and gravelly sediments were found deposited over the original clayey soil. In the Causeway Reservoir north of the Great Plaza, a four-meter column of stratified layers of fine sandy loam, loamy sand, and fine gravels were found, indicating a period of fairly intense erosional deposition. Around old quarry areas shallow silt loam, clay, and occasionally silty clay loan, have been deposited directly on hare bedrock. These areas must have suffered severe soil erosion during quarry operations and periods of construction. One of the sampling sites included a trench through artificially constructed feeder ditches leading into the Tikal Reservoir. Neglect of these channels, presumably beginning at the time of collapse, was indicated by silting of these ditches.

Typical upland soils at Tikal were described and sampled at five sites. Soils over large areas, particularly around quarries, were quite shallow, with as little as .28 m to hard limestone bedrock. In other areas the soil was more in the range of half a meter in thickness. It was these upland soils that the Maya apparently preferred for construction and habitation as well as for horticulture. Typically, structures occur on well-drained ridges and the convex hills and knolls known as hums. The soils are of silt loam texture with increasing amounts of fine gravel closer to bedrock. A six-inch layer of soft limestone, or weathered gravelly hard limestone, is generally encountered just above the limestone bedrock. A hard surface layer of bedrock, one-half to two meters thick, usually overlies a softer limestone bedrock or marl. This soft limestone has pockets of red, brown, grey and yellow clay a swell as flint. Though in many areas the hard limestone lies directly beneath the soil cover, in other areas only the softer marl is found. In arroyos, the ravine commonly cuts through a hard limestone cap into soft limestone beneath. While the cap, to some degree, appears to be formed by the surface action of rain water, there also seem to be layers within the limestone of varying hardness as discussed above in the section on geology. Exactly how rain water or exposure might bring about this formation has been in question. professor H. E. Wright of the University of Minnesota Geology Department has suggested to Webster (1972:316) that a process similar to that responsible for caliche formation may be involved. Unfortunately the mechanics of caliche crust formation are not well understood, though the process is apparently not restricted to any particular climatic zone (Reeves 1970:353).

Webster has attempted to determine which of two possible mechanisms is responsible for the formation of the bedrock cap. These are:

1)         dissolved carbonates carried upward to the surface by capillary
            action during rapid evaporation and . . . concentrated to form a
            hard crust.
2)         soluble carbonates near the surface carried downward by
            penetrating rain water, leaving behind accumulations of non-
            soluble materials, especially clays. Capillary action would
            reprecipitate some of the dissolved carbonates which would then
            cement the non-soluble residue into a crustal layer (Webster


Weighing of Becan limestone samples from depths of 0.5, 1.0, and 2.0 meters, after bathing them in 25% hydrochloric acid until all soluble  carbonates had been removed, revealed non-soluble residues of 25%, 5% and 2.5% respectively. These results, as determined by Webster, support the second alternative which is actually an elaboration of the first. The larger non-soluble residues in the samples nearer the surface were fine and light in color, probably representing a mixture of kaolin clay and ash, carried down into the limestone with carbonates by rain water.

The possibility that the crust was caused by laterization brought about by deforestation by man is rejected by Webster on the basis of the observation that the hard bedrock cap frequently underlies Preclassic structures where he intimates exposure time would have been relatively short. It should be pointed out, however, that from Middle Preclassic to terminal Preclassic covers a span of 500 to 1000 years which would seem to be time enough. Consideration of the various possible causes for the formation of this crust is thus intimately linked to discussion of whether or not the lowlands were ever heavily utilized for slash-and-burn agriculture.

Bajo soils in swampy areas are remarkably similar to each other. Typically, they have thick (up to half a meter) reddish-lack to black clay surface  horizons overlying gray subsoils. Theses clays are highly mottled with manganese stains, manganese concretions, iron stains, secondary lime deposits, clayskins, slickensides, and other characteristics of hydromorphic glei soils (Stevens 1964:287). Generally, these features increase with depth in the gray subsoils. Manganese concretions in these subsoils seem to identify them as the source of clay used for local pottery (Fry 1969:209,232). As reported by Cowgill and Hutchinson (1963), these soils  are high in montmorillonite clay minerals and exhibit deep cracking when exposed to the sun during the dry season. A principal factor in their formation is the alternation between wetting and drying. Some places in the bajos have well-defined gilgai topography and black clay extends down some distance into the gray subsoils. These properties, including the cracking of the soil which tears the roots of plants  not adapted to this environment, pose serious limitations to bajo agriculture.

When wet, the soils make trails almost impossible to travel on if there is much traffic. Apart from simply sinking into them the traveler is impeded by the black clays which are sticky and accumulate on feet and footgear in increasing number of layers with each increasingly difficult step. These soils certainly presented problems for travel, foundation support, and food production in the past. In spite of these difficulties, escoba bajo (this will be described in the section on vegetation) soils are used for slash-and-burn agriculture today when upland locations are not available. Apart from purpling of the stems and leaves, which can probably be attributed to phosphorus and zinc deficiencies, maize grown in such locations at Tikal seems to do well enough to be worth cultivation in a situation where available land is scarce.

A management system involving ridged fields, such as we have recently discovered on the Candelaria River (Siemens and Puleston 1972) and northern Belize might have made these soils productive. Apart from a very small area six kilometers east of Tikal (see Plates, block 36), where we found a series of stone ridges in 1966, there is no evidence for the use of such techniques in the region of Tikal. If the bajos were not used for cultivation the way river levees along the Candelaria appear to have been, there probably was a good reason for it. Further research is needed, but difficulties associated with bajo cultivation may be related to poor drainage and the comparatively slow movement of water into and from their surface, which has resulted in the deposition of clays comparatively finer than those deposited on the river levees.

The high acidity is also a factor which prevents the use of these soils for certain varieties of maize and other crops. The accompanying long periods of saturation of soils during the rainy season are very destructive to the roots of many crops, including maize and manioc. Stevens (1964:300) writes of these soils:  "Where flooding does not entirely preclude agricultural land use, the productive capacity of the hydromorphic soils is limited by the intensely leached, highly acidic, and frustratingly sticky soil." With respect to acidity, Hester (1954:48) records that in bajos pH may drop as low as 5.0, "becoming much too acid for maize and most of the other cultivated  plants of the peninsula as well as for most of the trees." Cowgill and Hutchinson (1963:6) who have carried out one of the few intensive sediment studies in tintal bajo (see the section on vegetation) report on an unpublished MS of Simmons (n.d.) in which it is stated that the pH of the first 5 cm. of Yaloch clay "Oscillates around 5." For 5-30 cm. the pH "is about 5.0" though Table I on the same graph gives the figure 4.0. It suffices to say that Yaloch bajo soils are too acid for the cultivation of maize and most other crops. The findings of Cowgill in the Bajo de Sante Fe apparently corresponded to those of Simmons (Cowgill and Hutchinson 1963:15), though inexplicably the data are not presented.

A surface soil sample taken from tintal bajo at the end of the South Brecha Survey Strip in 1967 had a pH of 6.6 (Table 8). A pH reading of 6.2 was taken at surface of the Santa Fe Bajo with a dagger-probe pH meter on June 6, 1966. Unfortunately both the sample and the probe reading were taken after and during rainfall which might tend to reduce acidity, since the pH of rainwater in the Peten may vary from 6.8 to 7.4 (Cowgill and Hutchinson 1963:19). Bajo soils seem to have had another function, perhaps even more important than food production--the conservation of drinking water. Around the edges of the bajos the soil pattern is complex. Typically, clayey soils at the bajo edges are increasingly shallow to soft or hare limestone as one moves up slope. Clayey soils extend up slope for considerable distances in some places. These areas, shallow and moderately deep to limestone, have thinner dark brown (not so black) surface horizons over coffee-colored or chocolate-colored subsoils which overlie limestone.

It was in these soils near where bajos and uplands met that the ancient Maya constructed reservoirs or aguadas. The impermeability of the fine clays prevented rainwater from draining through to the porous limestone bedrock. By constructing encircling dikes or clay and earth, reservoirs capable of holding millions of gallons of water were created. Many reservoirs have been found in the vicinity of Tikal, (Carr and Hazard 1961). In the process of mapping surveyors recognized that an artificial drainage system was associated in the Peridido Reservoir (Carr and Hazard 1961:13). Since that time excavations by Bennet Bronson and Donald Callendar have revealed that the Tikal Reservoir was fed by an elaborate system of clay-lined drainage ditches which came from upland areas, including the large paved plaza areas at the center of the site. These water collection systems seem to have been constructed in association with sites throughout the central southern Lowlands. In many areas, including Tikal, they must have provided the only permanent and local source of drinking water.

Present Tikal soils provide a record of past land conditions and us eextending back to Maya times and beyond. With careful site selection soils can be selected for study that show evidences of past conditions. During Maya occupation, for example, lare areas were paved and devoid of soil as such. When these paved areas were abandoned, desert conditions persisted for a short time as a result of rapid runoff, intense solar heating, and lack of tree cover. Soils at this stage of development wer collected from the tops of Temples V and VI. These soils are like true desert soils, consisting of fine gravelly loams, light brown in color, zero to three inches thick, that reach temperatures of 100o F or more only under grass, shrub, maguey, century plant, or cactus vegetation. When tree cover becomes established the soil environment changes considerably with the addition of leaf litter and cooler temperatues due to tree shading. Soils at this stage of development were found on the top of buildings on which tree cover had become established.

Another way of examining past soil conditions is by searching out and sampling soil horizons which were buried in Maya times by construction. It is hard to find good examples of undisturbed soils buried in this way since the Maya often stripped off soils to bedrock before beginning construction of even house platforms. An exception was found beneath the Tikal Earthworks, however, where the rapidly excavated limestone was simply piled up on existing soils. The buried surface soil is easily recognized by its dark color and characteristic concentration of tiny white fungal mycelia which were also found in buried topsoils beside the artificially built-up reservoir drainage ditches. Laboratory analysis of these soils, however, revealed that in spite of the dark color, the content of organic matter had apparently dropped from levels of near 25% to only 3%. Apart from their lack of organic matter these prehistoric topsoils proved to be remarkably fertile, as are modern surface soils in these forests.

The analysis of soils has proved valuable for investigating the possibility that variation in the availability of soil nutrients may have played a significant part in the distribution of ramon trees. The analysis of 22 surface samples taken at 500-meter intervals along the center of the South Brecha Survey Strip are presented in table 8. The data are also presented graphically and compared with information on elevation, settlement density, ramon tree counts, and zapote tree counts in Fig. 26. An idea of the significance of these data for the various soil constituents may be obtained from Table 9 where correlation coefficients along with information on their statistical significance are presented.

In Fig. 26 it is fairly clear that none of the measured constituents of soil match the correlation to settlement and ramon trees. This situation suggests that soil differences and particularly higher concentrations of phosphorus resulting from human habitation, are not responsible for the differential distribution of ramon trees.

At first glance it might seem surprising that phosphorus levels are not generally higher in habitation areas between zero and 6.5 km on the strip. The association of higher phosphorus, calcium, carbon, and nitrogen levels with habitation areas of archaeological sites has been demonstrated and investigated a s a site finding technique by Cook and Heizer (1956).

TABLE VIII.  Soil fertility analyses of surface horizon samples collected at 500-meter intervals along the center of the South Brecha Survey Strip. Sample No. 1 was taken at 1.5 km., sample No. 2 from 2.0 km., and so forth. These data are presented graphically in Figure 4. (Analyses by Cornell Soil Test Lab.)


Figure 26.       Chart comparing data from the South Brecha Survey Strip on elevation, structure counts, tree counts, and soil fertility (Table VII). Correlation coefficients for the total of 118 soil samples collected at Tikal. These data make it clear that there is not a close relationship between any measured constituent of the soil and the high correlation between ancient Maya structures and ramon trees.


TABLE IX.    Simple correlation coefficients and statistical significance of coefficients from 118 soil samples from Tikal. Phosphorus, manganese, and aluminum appear to be the least related to other soil constituents. (Date prepared by Gerald W. Olson)


With respect to organic matter, we have already seen how this soil component decreased from 25% to 3% under the North Earthworks. For this reason, along with the probability that concentrations of organic matter are redistributed fairly rapidly in the shallow soil, it does not seem surprising that there is no correlation between areas of former settlement and organic content. The high degree of relationship between soil constituents (Table IX) suggests that any inequities that might have resulted from former occupation or clearing have largely disappeared. Only phosphorus, manganese, and aluminum seem to be uncorrelated with the others and they do not seem to follow the variation in ramon tree and settlement densities either.

Only with calcium does there appear to be even a slight chance of relationship to settlement distribution. I suspect that it is not due so much to residues from concentrations of this element in animal tissues and excreta (Cook and Heizer 1965:19) as much as to the greater quantities of slope soils where calcium seems to be more concentrated (U. Cowgill 1963:23). In confirmation of this possibility, we may note that the terrain nearer to Tikal on the South Brecha Survey Strip is more irregular than in the outer half. There is also the possibility that the greater degree of construction activity in areas of heavy settlement has bought about an increase in the mixing of nearly pure limestone with surface soils.

With respect to phosphorus, the lack of correlation is inconsistent with data from a very limited sample of five sites tested by Ursula Cowgill (1963: Table 5) at the western edge of the Santa Fe bajo. She found that "highest phosphorus is associated with ancient sites as might be expected" (U. Cowgill 1963:23). Only one of her five sites was in the vicinity of ancient habitations, however, the lack of correlation in the Tikal data may be due simply to redistribution of any differential phosphorus distribution in these soils, where turnover must be much faster than in the semi-arid areas and deeply buried soils examined by Cook and Heizer. It may also be due to the special nature of kitchen garden nutrient cycles. In a situation where kitchen sweepings and human excreta, derived ultimately from the products of such a garden, are all returned to it, the expected build-up may never occur.

In view of the evident vulnerability of these soils, and the large populations they supported in the past, it is safe to say that the ancient Maya could not have sustained at least 1500 years of growth without finding ways to cope with problems of erosion, soil depletion, and water supply. In spite of the eventual failure of Classic Maya civilization, it seems clear that the ancient Maya understood well and were able to manipulate to their advantage the potentials and limitations of the "soil resource base."


Following Holdrige's scheme for classifying world plant formations on the basis of annual precipitation and potential evapotranspiration through its relationship to biotemperature (Tosi 1964), the Peten forest in the vicinity of Tikal falls in the zone of transition between tropical dry forest and low subtropical moist forest. The data for Tikal are summarized in the section on climate. This terminology seems equivalent to Lundell's (1937) classification of the vegetation as being "quasi-rainforest."

Formal studies of the Peten forests began with collections made by O. F. Cook and R. D. Martin in 1922. The first extensive collections, however, were made by H. H. Bartlett in conjunction with the Carnegie Institution of Washington's archaeological excavations at Uaxactun (Bartlett 1932, 1935). This was followed by Lundell's classic survey, The Vegetation of Peten (1937), based on field work carried out in 1933. Since that time comparatively little has been done, though for the three-year period from 1959-1961, Lundell directed a project financed by the Rockefeller Foundation aimed at taking an inventory of plants of possible economic significance to the ancient Maya (Lundell 1961).

Food Plants

Many plants are, and undoubtedly were, of significance to the Maya. Lundell (1938) lists almost 200 species on the basis of modern ethnohistorical and archaeological data that he feels were " probably utilized" by the ancient Maya. These range from food sources to timber trees and dye plants. Of particular importance to us here are those which were used for food. Theses are listed in the accompanying Table X.

The total of 90 species  is certainly minimal. Numerous fruit trees, shrubs and herbs, such as Casimiroa edulis and Puteria hypoglauca, that are reported only from gardens in the northern lowlands but which quite logically may have been cultivated further south, are not include. Many other ornamentals, fiber and dye plants, such as Bixa orellana L. (achiote), were expectably also planted in kitchen gardens. The many cultivated, semi-cultivated, and wild plant species that were probably used for construction timber, cordage, thatching materials, dugout canoes, decorations, shade and other miscellaneous uses, will not be treated here. Lists of plants which could have served these needs may be found elsewhere (Lundell 1938: Mangelsdorf, MacNeish and Willey 1964).

TABLE X. HUMAN FOOD PLANT PROBABLY UTILIZED BY THE ANCIENT MAYA (derived principally from Lundell, 1938:46, see also Mangelsdorf, MacNeish, and Willey 1964). All of these were or could have been grown at Tikal.


What is particularly impressive about the list presented here is the large number of tree crops. More than 60 of those listed can be considered trees, and another 17 or so are shrubs and vines. Only 13 are representative of the more standard category of "vegetables," which apart from certain root crops are annuals: one cereal, three curcurbits, six root crops, and three other vegetables (beans, tomato, and Physalis). The majority of these food plants, trees as well as vegetables, are amenable to cultivation around the home in kitchen gardens. Exceptions to this cannot be specifically indicated, for a species which grows practically everywhere in some areas, such as the corozo (Orbignya cohune), apparently grows well only in certain bajos around Tikal.

The great majority of the tree crops produce soft fruits which could only have had significance when in season as they are not particularly amenable to storage. Of special interest are the seed tree crops producing hard seeds which can be cooked, stored, and eaten in various forms, including ground flour in tortillas. Of the six seed tree crops listed, Brosimum alicastrum has received the most attention with respect to it role as a food resource in my master's thesis (Puleston 1968). Anacardium occidentale (cashew) is not a common tree in the southern lowlands outside of cultivated gardens, though it does seem to grow wild in the pine ridges of British Honduras. The "pulpy receptacle and the roasted seed are both eaten." A fermented beverage can be made from the fruits (Lundell 1938:44). Bactris spp. comprises several species of low spinney palms with edible kernels. Bactris major is an extensively utilized food resource in South America. Pachira aquatica (zapotebobo), after which the site of Bobal on the South Brecha survey strip was named, grows around aguadas and ancient reservoirs where it may be a relic. The fruit is not considered palatable, but the large seen inside can be eaten after it is roasted. Crescentia cujete (calabash tree) produces a large gourd-shaped fruit which contains a white pulp used a as medicine. The seeds can be cooked and eaten, but the pulp is not considered edible. Cattle eat it, but it apparently causes abortion (Standley 1920-26:1324). The hard shell of the fruit was much used in former times for the manufacture of drinking cups. It is this tree which grows in dense groves on the ridged fields of the Candelaria River in Campeche.

Of all these, only the ramon appears to have real significance as a food crop. The tree is easily cultivated and produces large quantities of seed, which can be successfully stored in chultuns (Puleston 1971). After being cooked in water it has traditionally been prepared by grinding with a mano and metate (Puleston 1968:74) and cooked as small loaves or tortillas. As a food it appears to be highly nutritious (Puleston 1968:102-108).

Lundell's (1937:10) claim that there is a high correlation between ruin areas and existing populations of the tree has been borne out by a survey that extended the entire length of the South Brecha Survey Strip (Figures 26 and 27). A product moment correlation of 0.86, significant to the 0.01 level was revealed (Puleston 1968:51). The survey was particularly significant for the fact that the association between ruins and trees was not limited to areas of large collapsed buildings, but extended out into residential districts where a general association with house platform areas was indicated. Higbee (1948:461), following the suggestion of O. F. Cook (1921), seems to favor the idea that this association is the result of selective cutting for milpa. Later, when the fields were abandoned, the ramon presumably would have propagated itself so effectively that it would have become a dominant in areas of secondary forest. This possibility is unacceptable to me for three reasons: 1) the available land areas around residential groups was too small for long-term milpa agriculture even without considering the space that might have been taken up by ramon trees left standing after clearing; 2) partial shading is highly detrimental to maize yields so it is unlikely that in a situation where a) space was severely limited and b) dependence on maize was high, that such trees would have been left standing; 3) it is hard to believe that in a situation of high population density a resource  such as ramon would not have been used to its full extent.

The magnitude of the potential role of Brosimum alicastrum in southern Maya Lowlands subsistence is revealed in Table X, where the results of the FAO-FYDEP timber survey are presented. The 100 most common species of the Peten survey are listed in the order of their abundance as revealed by spot checks all over the Peten. This list is reproduced here as it has not been published elsewhere and represents the only attempt to statistically quantify the frequency of Peten tree species that the present author is aware of. Of particular interest here is the discovery that out of 200 or more upland forest tree species, the ramon (Brosimum sp.) is the most frequently encountered, comprising, on the basis of the sample, nearly 10% of the standing forest. Calculations in terms of volume indicate the Brosimum sp. accounts for as much as 25% of the total timber volume in the southwest Peten. Other facts to come out of the survey include an apparently natural association, at least in the southwest Peten, of Brosimum sp. and 1) Dialium guianense and 2) Terminalia amazonia. In view of the fact that at least 60 trees in the Peten produce ood, it is perhaps surprising that only 17 show up on the FAO-FYDEP list.

Figure 27.  A graphic representation of the relationship between ancient Maya structures and living ramon trees in contrast to elevation and slope on the South Brecha Survey Strip. Note the lack of correlation between elevation and the density of structures or ramons beyond 6.5 km.


TABLE XI. A table of tree species of the Peten listed in the order of the frequency of their occurrence. Access and permission to use these data was granted by the FAO-FYDEP offices in Guatemala.

They represent a portion of the results of a timber resources survey carried out from 1964 through 1966 by means of spot checks in the upland forest of many pats of the Peten, though mot were concentrated in the west and southwest regions. A total of 4,464 trees were included in the sample. Asterisked items are food-producing trees listed in Table X.


Major Plant Associations

The definition of the extent of major plant associations was included as one of the objectives of the survey. The collection of this information was oriented to the problem of defining and quantifying areas of greater and lesser agricultural potential. Plant associations are excellent indicators of the cultivability of different soils, and they are used today by soil scientists and planners as well as traditional Maya agriculturalists. The associations identified on the survey maps include tintal (logwood bajo), escobal (escoba bajo), corozal, jimbal, pital and, in a very general way, the upland mesophitic or "climax" broad-leaved forest.

The wettest of these associations, apart from aguadas, in tintal bajo, which, is inundated by virtually stagnant water for long periods during the rainy season. For purposes of the survey, Haematoxylum campechianum (palo tinto) was used as an indicator species for this association. The vegetation tends to be low (five-eleven meters). Trees and shrubs "are gnarled, twisted, and thorny, with interlocking knotty branches." Distorted trunks lie at all angles as if larger trees are soon pushed over by the winds. The leaves of the species found here are "thick, hard, glossy or glaucous, generally small or pinnately compound. All protective devices against intensive insolation and excessive evaporation are present" (Lundell 1937:29). The ground in wetter areas is often densely overgrown with coarse saw-toothed cutting sedges "forming an intertwined growth that can be penetrated only with pain and difficulty" (Lundell 1937). The extreme difficulty involved in getting through tintal, as a result of the vegetation and the often very rough hummocky surface, suggests that it could have served as an important defense perimeter. The North Earthworks at Tikal, which stretch from one large bajo of this kind on the east to one on the west, sustain such an hypothesis. Though Wolf (1959:78) suggests that the bajos may have been used for chinampa-like cultivation, the possible defensive function, the poor drainage, and the lack of evidence for ridging all seem to argue against such a possibility as has been discussed in the previous section on soils. Even the concentration of settlement along the edges of bajos can perhaps best be attributed to the fact that water reservoirs are generally built in the clay soils at the edges of bajos.

Escobal, as an intermediate form of bajo between tintal and the upland forests, is characterized by the presence of the spiney escoba palm (Crysophila argentea). The soils in these areas are flooded for correspondingly less time during the rainy season. The forest, while considerably denser in terms of cover, is only slightly higher than tintal.

This form of bajo may have been used marginally for cultivation, but settlement densities do not suggest it was very important. An illustrative example may be seen on the East Survey Strip at 8.5 km. At this location a small rise is surrounded by escoba bajo but does not seem to be associated with an increase in settlement density; in fact, density seems to be lower. Other demonstrations of this lack of association of settlement with escobal may be observed elsewhere on the strips.

Corozal, characterized by dense groves of Orbignya cohune, and possibly Scheelea lundellii, occur only occasionally in the Tikal-Uaxactun area. The corozo, a magnificent palm with a towering straight trunk and huge pinnate leaves, was probably an important oil source for the ancient Maya. Oil content of the seed kernels is high, and oil was otherwise probably low in aboriginal Maya diets (O. Puleston 1971). Two large stands are located on the survey strips, one at 11 km on the East Strip and the other at 5.7 km on the South Strip. A single palm was found on the Uaxactun Strip at 2.5 km. It seems possible that the groves mentioned above represent relic populations of groves tended by the Maya Lundell (1937:153) notes that the "corozal association is quite permanent, and where it occurs it represents one of the most stable phases of the vegetation of the limestone valleys." It is difficult to imagine that this important oil resource would have been neglected if it were available. Since these trees seem to grow better in groves than as single trees in kitchen gardens, it is possible that the corozo was a restricted resource in this area, with control in the hands of only certain segments of the society.

Jimbal, characterized by small but dense stands of a thorny bamboo (Guadua sp.),were noted more for their impressive impenetrability to us as surveyors rather than any supposed significance to ancient Maya economy. The possibility that this plant may have been planted at bajo edges to serve as a defensive screen might be considered, however. Stands are located at 3.0 km. on the south strip at 5.0 km. and 5.3 km. on the west strip, and in and around the aguada west of the site of Jimbal. It seems to prefer deep moist soils but for some reason is of limited distribution.

Pital, characterized by the large terrestrial bromeliad Aechmea magdalenae, tend to be concentrated around aguadas, again in relatively small colonies. The plant was almost certainly of economic significance to the Maya as a source of cordage (Lundell 1938:52). Stands are located at 3.2 km., 5.8 km., and 10.6 km. on the South Strip, and 11.3 km. on the North Strip. The broad-leaved upland forests in all their diversity of mixed-forest, and the major associations, including ramonal, zapotal and caobal (Lundell 1937:35), are simply indicated by lack of symbol.


Information on the modern fauna of the Tikal area is scattered. The mammals have been stuied by Anne Meachum Rick (1968), whose thesis, based on a collection of 252 specimens (44 species), is quite comprehensive. Another small collection of 19 specimens has been made by Peter Puleston.The birds of Tikal have received equal if not more attention in a series of publications. The original checklist (Smithe and Paynter 1963) has recently been expanded into a handsome field guide (Smithe 1966) entitled The Birds of Tikal. An excellent supplementary source with information o habits, calls, species distribution maps,. and habitat preferences in Land's (1970) Birds of Guatemala. Blake (1953) provides an excellent source for adjacent portions of Mexico. Puleston (1966) records a number of species at Tikal not listed by Smithe.

Reptiles and amphibians have been studied by L. C. Stuart (1958). Anne M. Rick (personal communication) also reports that a manuscript was prepared on the 295 specimens she collected in Tikal. The collections reside at the University of Florida in Gainesville.

Fish from Tikal aguadas and intermittent streams, the Aguada Naranjal of the Arroyo Holmul, have been collected sporadically by icthyologists, including D. E. Rosen. Information on these collections has not been published. A collection of cichlids, cyprinodontids, and two eels made by Donald Callendar and the present author has been identified by Rosen and includes the following species from the following locations.

Tikal Reservoir; caught with seine along edges, mainly along east edge of southwest quadrant, July 3, 1968.
1            Belonesox belizanus (eel)
4            Astyanax fasciatus
5          Cichlasoma octofasciatum

Tikal Reservoir; caught with hook and line; July 16, 1968.
1          Belonesox belizanus (eel)

Dimick Reservoir (on south edge of east end of airport runway); caught with seine; July 7, 1968.
4            Astyanax fasciatus
1            Cichlasoma octofasciatum
8          Peocilia mexicana

Naranjal Reservoir; caught with seine at 6 p.m., July 16, 1968.
11          Astyanax fasciatus
2            Heterandria bimaculata
4            Xiphophorus helleri
4          Gambusia yucatans

Little work has been done with insects, apart from a brief study of dragonflies (Kormondy 1959), psocoptera (Mockford 1957), and certain species of diptera (Hays 1960).Stored food pests, including mites as well as insects, were identified in chultun storage experiments (Puleston 1971:329-330). Other invertebrates, apart from land mollusca (Basch 1959; Moholy-Nagy 1963), have not received nearly the attention they deserve.

The potential importance of the large mammal and bird fauna, fish, and certain crustaceans to ancient Maya subsistence is obvious. Faunal remains collected from archaeological contexts at Tikal are currently being studied by Kent Flannery at the University of Michigan. Identifications of many items including birds, reptiles, amphibians, and fish have been provided by other specialists.

Historical accounts as well as modern ethnographic data further suggest that a number of species of birds and mammals were frequently raised as pets and kept around the house in a semidomesticated state for eventual use as food (Tozzer 1941:127). If we may judge from ethnohistoric sources, a good deal of this meat resource was passed on to the elite in the form of tribute (Tozzer 1941:97, Relaciones de Yucatan 1898-1900 Vol. 1:78). Bones are not abundant in middens at Tikal, but this may in part be attributed to poor preservation and the ways in which organic garbage was disposed of. Skin, bone, sinew, and feathers undoubtedly served as important natural resources for the manufacture of tools, clothing, and ornaments. Bees and a few honey-making wasps were certainly an important source of sweets (Stuart 1964:337).

The fauna were not important only as a resource for exploitation. Insects as vectors of disease transmitted leishamaniasis and trypanosomiasis. Thompson (1966:24) has outlined the evidence for the idea that the introduced, insect-transmitted diseases, malaria and yellow fever, played a major role in the dramatic depopulation of certain parts of the Maya area in colonial times. Demitri Shimkin (Willey and Shimkin 1971:7) has recently argued that Chagas' disease or trypanosomiasis " is likely to have contributed heavily to infant mortality and to adult debilitation" and thus the decline of Classic Maya civilization. It should be pointed out that, while serious, this disease is not epidemic in nature and "cases are not numerically important in Mexico" (Johnson 1972:20). Transmission is not by the bite of the bug (Triatoma) which caries the trypanosomes, but by inoculation of contaminated excreta into bites, other open wounds, mucous membrane, or the conjuctiva. The bug is nocturnal, and most prevalent in areas where hygienic standards are low (Leclerq 1969:11). It is doubtful to me that it had the significance Shimkin attaches to it. In 1957 only one death by Chagas' disease was recorded for all of Mexico (World Health Organization 1960).


Go Back to Top of Page

PagesNotesPrefaceIntroductionCh 1Ch 2Ch 3Ch 4Ch 5Ch 6Ch 7Ch 8Ch 9ContactVoivode Multimedia