International Biodeterioration & Biodegradation 46 (2000) 335–341 www.elsevier.com/locate/ibiod Biodeterioration ofMayan archaeological sites in the Yucatan Peninsula,Mexico H.A. Videla∗;1 , P.S. Guiamet, S. Gomez de Saravia INIFTA, Department of Chemistry, Faculty of Pure Sciences, University of La Plata, Argentina Accepted 23 October 2000 Abstract Many of the monuments of the Mayan civilization are su�ering deterioration caused by environmental factors (high temperatures and relative humidities), increasing contamination from natural and anthropogenic sources, and by the action of micro- and macro-biological communities. Archaeological sites and historical monuments in the Mayan area were constructed with di�erent limestones which o�er di�erent resistances to degradation by the various types of contamination. Two di�erent sampling sites were chosen at the archaeological site of Uxmal in the Yucatan Peninsula, Mexico. Heterotrophic bacteria, cyanobacteria and di�erent fungi were isolated and classi�ed taxonomically. The other archaeological site chosen for this study was the fortress of Tulum, located at the side of the Caribbean Sea and exposed to chloride of marine spray and sand erosion. In this case, heterotrophic aerobic and anaerobic bacteria, cyanobacteria and fungi were isolated from the four sampling areas selected. In both archaeological sites crust deposits were observed by using light microscopy, SEM and ESEM. Surface analyses were made by means of EDAX and electron microprobe. Possible mechanisms of stone decay, based on the type of microorganisms isolated, the physico-chemical characteristics of the constructional materials and environmental factors are discussed. c© 2001 Elsevier Science Ltd. All rights reserved. Keywords: Biodeterioration; Cultural heritage; Mayan; Limestone 1. Introduction Mayan cultural heritage covers a wide variety of archae- ological monuments spread over several Mexican states located in the Yucatan peninsula (Yucatan, Campeche and Quintana Roo) as well as in the states of Tabasco and Chiapas together with the countries of Belize, Guatemala and Honduras in Central America. Many of the monuments relating to Mayan civilization are su�ering deterioration caused by the environmental factors (high temperatures and relative humidities), by increasing contamination due to natural and anthropogenic sources and by the action of micro- and macro-biological communities. Among the few references available in the literature on the deterioration of the Mayan ruins, reports have included the loss of painted stucco sculptures in Dzibilchaltun, Yuca- tan (Robertson, 1989), the microbial attack on the structural limestone of several buildings in the city of Uxmal, also in Yucatan, (Guiamet et al., 1998) and the weathering of ∗ Corresponding author. Tel.=Fax: +54-221-425-7945. E-mail address: hvidela@infovia.com.ar (H.A. Videla). 1 Postal address: Ave. 51-337 (1900) La Plata, Argentina. the ruins of Tulum by marine spray and sand erosion (Mal- donado et al., 1998). Archaeological sites and historical monuments in the Mayan area were constructed with di�erent limestones that o�er di�erent resistances to degradation by the vari- ous types of contamination. Limestones from the Yucatan peninsula contain basically two types of rocks: red and white limestone, both consisting of calcium carbonate ac- cording to X-ray di�raction analyses. A major di�erence is their porosity: white rocks are 2.48 times more porous than red rocks (Maldonado et al., 1998). Accelerated tests for limestone dissolution by sulfate=chloride show that red limestone is more resistant than white limestone. In all cases the main degradation products are gypsum, formed in the presence of sulfate, and calcium chloride, formed in the presence of chloride from marine spray. Both compounds are either retained in the rock pores, or washed o�, depend- ing on the available water. One of the archaeological sites chosen for this study, corresponds to the city of Uxmal, lo- cated in a rural environment in the north-western part of the Yucatan peninsula, in the state of Yucatan, Mexico. Uxmal is one of the main city centers representing the Puuc style of the Mayan classic period, prevalent between the 8th and 0964-8305/01/$ - see front matter c© 2001 Elsevier Science Ltd. All rights reserved. PII: S 0964 -8305(00)00106 -2 336 H.A. Videla et al. / International Biodeterioration & Biodegradation 46 (2000) 335–341 Fig. 1. Uxmal map. Sampling sites A and B have been marked on the map. Fig. 2. Fa�cade of the Governor’s Palace at Uxmal. In the foreground is the building’s monumental stairway. 9th centuries AD (Sablo�, 1997). Ten di�erent groups of quadrangular buildings can be distinguished in the Uxmal urban nucleus: (1) North group, (2) Cemetery group, (3) the Nunnery Quadrangle, (4) the House of the Magician (El Adivino), (5) Ball court, (6) House of the Turtles, (7) the Governor’s Palace, (8) the Great Pyramid, (9) Doves-Cotes building, and (10) South West group (Figs. 1 and 2). The other archaeological site studied in this work is the ancient fortress of Tulum located in the state of Quintana Fig. 3. Tulum map. Sampling sites A, B, C1 and C2 have been marked on the map. Roo, in the north-eastern part of the Yucatan peninsula. It is perched on a cli� 50 m from the Caribbean sea and sur- rounded on three sides by a wall. This site belongs to the late post classic style of the 12th and the 13th century AD (Kelly, 1993). The main buildings at the fortress of Tulum are: the Castle (El Castillo), located high on a cli� over the Caribbean, occupying one of the most spectacular sites in the Mayan world; the Temple of the Descendent God, decorated by a large stucco statue of this god; the Temple of the Frescoes, built over a building and surrounded by a porch, with the gods Chaac and Ixchel performing rituals painted on the temple fa�c ade; Temple of the Wind, built on a natural rise on the north cli�, having a single room with an entrance on its north side and dedicated to Ehecatl, the god of the wind (one of the aspects of Quetzalcoatl); and �nally the House of the Cenote, built over an old cenote (natural well) of the Mayan area and having three chambers and a staircase that runs down into the cenote to provide access to the water (Figs. 3 and 4). Several of the external walls of the Uxmal buildings and some of the Tulum structures possess a black crust that turns into copious green patinas when the walls belong to inner compartments that are not directly exposed to sunlight. The aim of this work is to assess the kind and variety of biodeteriogenic microorganisms a�ecting the integrity H.A. Videla et al. / International Biodeterioration & Biodegradation 46 (2000) 335–341 337 Fig. 4. East side of The Castle at the fortress of Tulum, high on a cli� over the Caribbean sea. of two of the archaeological buildings of Uxmal and four di�erent constructions of Tulum. Di�erent isolation techniques and biochemical assays complemented with microscopic observations were used to identify the micro- bial components of the natural bio�lms. Chemical and sur- face analyses combined with light and electron microscopy were also used in an attempt to characterize the mechanisms of limestone biodeterioration. 2. Materials and methods 2.1. Sampling procedures Samples were taken aseptically from two di�erent sites in the city of Uxmal. One of the sites was an external rock wall exposed to open air, rain and direct sunlight, located on the left side of the House of the Magician (sampling site B). Here, the rocks were partially covered by a thick black crust. In the other sampling site (a wall located in one of the inner chambers of the Governor’s palace), the rock surface was protected from sunlight, rain and outer air, and was heavily fouled by a dark greenish bio�lm (sampling site A). Non-destructive sampling techniques were used to scrape the bio�lms from the surfaces with a sterile plastic scalpel. Samples were immediately placed into a sterile plastic ves- sel. A similar procedure was followed with the Tulum sam- ples, that were taken from four di�erent locations (Fig. 3): from the ceiling of a small cubicle on the left side of the Castle (sampling site A), from one of the inner walls of the Temple of the Wind (sampling site B), from a wall of a small corridor near the House of the Cenote (sampling site C1) and from an external wall of this building (sampling site C2). Site C2 was the only sampling site directly ex- posed to the external atmosphere. The Uxmal samples were collected during the Winter period (February) whereas Tu- lum sampling was accomplished during the late Summer (September). 2.2. Isolation and identi�cation of microorganisms For both Uxmal and Tulum samples, light microscopical observations of samples from di�erent sampling areas at 40× magni�cation showed abundant motile bacteria and scarce �lamentous cells. Small amounts of each type of sample (1 g weight) were aseptically suspended in 10 ml of sterile saline solution and then a 1ml aliquot was cultured on di�erent nutrient media in Petri dishes. Nutrient agar was used to grow bacteria and yeast extract=glucose=chloramphenicol (YGC) agar for fungi. The incubation time was 48 h for bacteria and 72 h for fungi, both at 28◦C. In the case of bacteria, the colonies isolated from nutrient agar plates were Gram stained and observed by light microscopy. EMB and Cetrimide agar were used to isolate Gram-negative bacteria while Mosel agar was used for isolating spore forming bacteria and �nally, Postgate B medium was used for the detection of anaerobic bacteria. Laboratory tests to con�rm acidi�cation of the medium caused by the growth of bacteria and dark pigmented mi- tosporic fungi were done (Guiamet et al., 1998). Using 24-h cultures of isolates, batch cultures were made in a simpli- �ed mineral medium (Salvarezza et al., 1983) containing 1% glucose for bacteria and 2% for fungi, as the sole car- bon source. In both cases, the initial pH value (7.0) was de- creased to ca. 4.5–4.6, showing a capacity for acidi�cation of both types of isolates. 2.3. Electron microscopy and surface analysis Small pieces of crust samples from sampling sites A and B from Uxmal and sites B and C from Tulum were ob- served using a stereoscopic light microscope, a scanning electron microscope (SEM) and an environmental scanning microscope (ESEM). Samples for SEM observations which had been �xed initially in 2.5% glutaraldehyde bu�er solu- tion, were later washed with phosphate bu�er, dehydrated through an acetone series and �nally critical point dried. Fi- nally, the samples were gold coated at 10−3 mm Hg (30 Torr overpressure) in a Balzers Model SCD 030 sputtering apparatus. SEM observations and energy-dispersive X-ray analysis (EDAX) of samples were made using a Philips 500 microscope. To avoid distortion of samples and ar- tifacts due to dehydration procedures, the samples were also observed with a ESEM microscope (Philips Electro- scan 2010). Occasionally, to observe bacterial and fungal mor- phology on a clean surface (without the interference of the particles present in the samples), microbial bio�lms of the isolates were formed on an inert smooth sur- face (AISI type 316 stainless steel). These bio�lms were �xed and dehydrated according to the procedures used for SEM observations. Some surface analyses of Tulum samples were done by means of an electron microprobe (CAMECA SX 50). 338 H.A. Videla et al. / International Biodeterioration & Biodegradation 46 (2000) 335–341 3. Results 3.1. Microbiological results 3.1.1. Uxmal samples Cultures of both samples revealed Gram-negative rods and also spore-forming bacteria. Subsequent cultures in EMB and Cetrimide agar identi�ed bacteria of the genus Pseudomonas. From the inoculation of spore-forming bac- teria into Mosel agar it was possible to identify Bacillus cereus. This presumptive taxonomic identi�cation was later con�rmed by means of biochemical tests. The growth on YGC agar allowed us to obtain three di�erent isolates: (i) sterile fungi (Mycelia sterilia); (ii) hyphomycetes of the genus Monilia sp. and (iii) a dark-pigmented mitosporic fungus. Although microscopic observations of cyanobacteria in samples in this work only showed the presence of genus Cyanocystis, in a recent investigation (Gaylarde and Gaylarde, 1998) a higher diversity of coccoid forms of cyanobacteria were found. The acidi�cation test performed with bacteria or fungi showed a marked decrease of pH (ca. 3.0 units) due to microbial growth. 3.1.2. Tulum samples Microbiological results for samples from sites A, B and C, showed the major biomass to be composed of cyanobacteria (coccoid types) and chlorophyta. In the early observations, no �lamentous types of cyanobacteria were found. A description of the microorganisms found in each sam- pling site can be summarized as follows: Site (A)Major biomass:Dermocarpa. In lesser amounts: Gloeocapsa, Pleurocapsa, Gloeothece, Synechococcus, Chlorella. Fungi, actinomycetes and bacteria were also found. Site (B) Major biomass: Myxosarcina, Dermocarpa, Gloeocapsa. In lesser amounts Synechocystis, Synechococ- cus, Chlorogloepsis, Chroococcidiosis, Chlorella were found. The fungus Aspergillus ochraceus, aerobic bacteria (Pseudomonas sp.) and anaerobic bacteria (sulphate-reducing bacteria (SRB)) were also isolated. Site (C) Major biomass: Gloeocapsa. In lesser amounts Xenococcus, Gloeothece, Gongrosira were found. Others isolated were Aspergillus niger among the fungi, some protozoa, aerobic bacteria (Pseudomonas sp., Bacillus circulans) and SRB. 3.2. EDAX and SEM, ESEM results 3.2.1. Uxmal samples EDAX pro�les of round particles present in the black crust obtained from sampling site A in Uxmal (Fig. 5) re- vealed a major peak of calcium and minor peaks of chlo- ride, potassium and aluminium. If an overlapping analysis Fig. 5. SEM micrograph of particles of black crust from sampling site A from Uxmal (scale 10 �m). of EDAX pro�les obtained in similar areas from samples of sites A and B is made (Fig. 6), similar peak heights for sulfur, chloride, potassium and calcium can be observed for both samples. Conversely, samples from site B show higher peaks for magnesium, sodium and aluminium. 3.2.2. Tulum samples In some of the Tulum samples from site B (Temple of the Wind), which is the nearest place to the sea, major peaks of chloride and minor peaks of sulfur (probably associated with the presence of SRB, detected in these bio�lms) and cal- cium can be seen (Fig. 7, bottom). Conversely, in di�erent samples taken from other areas of the same site, the Cl=Ca relationship is inverted (Fig. 7, top). Samples from site C1 (a partially closed corridor near the House of the Cenote) show high peaks of calcium accompanied by smaller peaks for sulfur and chloride (Fig. 8, top) whereas the EDAX pro- �le from site C2 (a wall exposed to the atmosphere) shows higher peaks of chloride and minor peaks of sulfur and cal- cium (Fig. 8, bottom). 4. Discussion Almost all surfaces in natural environments may be col- onized by microorganisms, including buildings and monu- ments belonging to the cultural property (McCormack et al., 1996). Limestone is one such material that can be readily colonized by algae (Gaylarde and Gaylarde, 1998), lichens (Garcia Rowe and Saiz-Jimenez, 1991), bacteria and fungi, as was found in this work. Among the few references available in the literature on the microorganisms causing biodeterioration of Mayan ar- chaeological monuments, several types of cyanobacteria has H.A. Videla et al. / International Biodeterioration & Biodegradation 46 (2000) 335–341 339 Fig. 6. Overlapping of EDAX pro�les of samples from sites A and B from Uxmal. Fig. 7. EDAX pro�les from sampling site B (Temple of the Wind) of Tulum. Top: EDAX pro�le of a bio�lm grown on one inner wall of the Temple of the Wind. Bottom: EDAX pro�le of a bio�lm grown on other inner wall of the Temple of the Wind. Fig. 8. EDAX pro�les from sampling sites C1 and C2 (House of the Cenote) of Tulum. Top: EDAX pro�le of a bio�lm grown on a wall of sampling site C1. Bottom: EDAX pro�le of black crust formed on an external wall of the house of the Cenote. 340 H.A. Videla et al. / International Biodeterioration & Biodegradation 46 (2000) 335–341 been reported recently (Gaylarde and Gaylarde, 1998). Mi- croorganisms are able to obtain several elements that they need for their metabolism (e.g. calcium, aluminium, silicon, iron and potassium) from the limestone by biosolubiliza- tion. This biosolubilization process generally involves the production of various organic and inorganic acids by al- gae, lichens, fungi and bacteria (Warscheid et al., 1991). The release of aggressive acids is one of the best known biogeochemical destructive mechanisms at rock surfaces (Resende et al., 1996) and it occurs through the leaching of rock-binding materials with the consequent weakening of the crystal structure (Lewis et al., 1988). The �nal result of this type of biodeterioration is the physical and mechanical breakdown of the stone. Results shown in the EDAX pro�les for sites A and B (Fig. 6) are consistent with a previous re- port on the biodeterioration of concrete (McCormack et al., 1996) where the high levels of calcium are an indication of the alteration of the base material. According to this work, calcium can be leached from the calcite crystal lattice by an ion exchange mechanism in which ammonium and mag- nesium ions are involved. The liberation of ionic sodium, potassium, calcium, magnesium and aluminium could be used as an indicator of deterioration. In our SEM and ESEM observations it was also possible to verify the existence of bacterial and fungal bio�lms on the crystalline structure of the limestone. It has been reported (Videla and Characklis, 1992) that dramatic changes in pH, ionic concentrations and redox con- ditions can occur within the matrix of the bio�lms, leading to highly aggressive conditions, that are not usually found in uncontaminated areas. In addition, the environmental pa- rameters like light intensity and relative humidity can be important factors in controlling the microalgal communities (Ariño and Saiz-Jimenez, 1996). The stone colonization by lichens and bryophytes is one of the early stages leading to the later invasion by vascular plants, an important cause of mechanical deterioration. This kind of attack would not be prevalent in the moderate climate found at Uxmal, but it could be one of the most relevant causes of biodeterio- ration at the Mayan archaeological sites located in tropical jungles like those found at Palenque (state of Chapas, Mex- ico), Tikal in Guatemala (Garc��a de Miguel et al., 1995) or Copan (Honduras). According to a recent publication (Maldonado, 1997) de- scribing a corrosion map for the Yucatan peninsula, Uxmal corresponds to a rural area relatively free of airborne salinity and pollutant contamination, typical of marine and urban at- mospheres, respectively. The signi�cant acidi�cation in our laboratory tests, caused by the growth of microorganisms isolated from both Uxmal sampling sites, suggests that acidic attack of the stone structures may be one of the main mech- anisms of biodeterioration. This assumption is supported by the EDAX pro�les of Uxmal samples that show a major peak of calcium and minor peaks of chloride and aluminium for both sampling sites. Conversely, in Tulum samples, there is a synergistic action of the marine environment with the microbial bio�lms, according to the sampling sites (Figs. 7 and 8). Atmospheric corrosion data giving monthly deposi- tion rates for the Yucatan peninsula (Maldonado et al., 1998; Maldonado and Veleva, 1999), show that airborne contami- nants (airborne salinity and sulfur dioxide) are much higher at Tulum (50m from the sea) than at Uxmal (60 km from the sea), due to marine spray. Thus, according to our sur- face analysis results, the atmospheric contribution to stone decay will be considerably greater in Tulum than in the Uxmal monuments where microbial biodeterioration would predominate in the weathering process. 5. Conclusions Samples obtained at two di�erent sites in the city of Ux- mal and at four di�erent sites in the fortress of Tulum were studied in order to isolate and identify the main microor- ganisms and their metabolic characteristics. Microbiological procedures and microscopic observations (light microscope, SEM and ESEM) allow us to conclude that a heterotrophic micro ora, composed of bacteria, fungi and cyanobacteria, colonizes the structural materials in the buildings at both archaeological sites. Chemical analyses, surface analysis by EDAX and elec- tron microprobe complemented with electron microscopy (SEM and ESEM), suggest that biodeterioration may be performed through a biosolubilization mechanism involv- ing the production of metabolic acids by bacteria and fungi. In Tulum, rock decay would be also markedly a�ected by the aggressive marine atmosphere, evident in the high percentages of chloride and calcium found in the EDAX pro�les. The presence of cyanobacteria would serve as a primary support for heterotrophic bio�lms, providing organic mat- ter for growth through photosynthesis. This organic matter would be supplemented by the excreta of birds and bats in the inner chambers of buildings of Uxmal where copious dark greenish bio�lms were found. Acknowledgements The authors wish to thank Dr. Christine Gaylarde for the isolation and identi�cation of cyanobacteria in the Tulum samples and to Dr. Ricardo Echenique for the Uxmal sam- ples. The University of La Plata, CONICET and CICPBA of Argentina are gratefully acknowledged for research fund- ing. This work is part of the development in progress in the Research Network XV.E within the CYTED program. References Ariño, X., Saiz-Jimenez, C., 1996. Colonization and deterioration process in Roman mortars by cyanobacteria, algae and lichens. Aerobiologia 12, 9–18. H.A. Videla et al. / International Biodeterioration & Biodegradation 46 (2000) 335–341 341 Garc��a de Miguel, J.M., S�anchez-Castillo, L., Ortega-Calvo, J.J., Gil, J.A., Saiz-Jim�enez, C., 1995. Deterioration of building materials from the Great Jaguar pyramid at Tikal, Guatemala. Building and Environment 30, 591–598. Garcia Rowe, J., Saiz-Jimenez, C., 1991. Lichens and bryophythes as agents of deterioration of building materials in Spanish cathedrals. International Biodeterioration 28, 151–163. Gaylarde, C.C., Gaylarde, P.M., 1998. Phototrophic bio�lms on monuments of cultural heritage in Latin America. Latincorr ’98 Proceedings, Paper S11-03. NACE International, Houston, TX. Guiamet, P.S., G�omez de Saravia, S.G., Videla, H.A., 1998. Biodeteriorating microorganisms of two archaeological buildings at the site of Uxmal, Mexico. Latincorr ’98 Proceedings, Paper S11-01. NACE International, Houston, TX. Kelly, J., 1993. Tulum. In: An Archaeological Guide to Mexico’s Yucatan Peninsula. University of Oklahoma Press, Norman, pp. 330–336. Lewis, F.J., May, E., Bravery, A.F., 1988. Metabolic activities of bacteria isolated from building stone and their relationship to stone decay. In: Houghton, D.R., Smith, R.N., Eggins, H.O.W. Biodeterioration 7. Elsevier Applied Science, London, pp. 107–112. Maldonado, L., 1997. Corrosion map of the Yucatan Peninsula. International Materials Research Congress (Book of Abstracts), Cancun, Quintana Roo, Mexico, p. 40. Maldonado, L., Diaz, L., Cabrera Sensores, A., Pacheco Avila, J., Halsey, D., 1998. Paper No. 350: natural and accelerated dissolution of limestones used in monuments under tropical atmospheres. Corrosion ’98. NACE International, Houston, TX. Maldonado, L., Veleva, L., 1999. Corrosivity category maps of a humid tropical atmosphere: the Yucatan Peninsula, Mexico. Materials and Corrosion 50, 261–266. McCormack, K., Morton, L.H.G., Benson, J., Osborne, B.N., McCabe, R.W., 1996. A preliminary assessment of concrete biodeterioration by microorganisms. In: Gaylarde, C.C., S�a, E.L.S., Gaylarde, P.M. (Eds.) Proceedings LABS 2, Biodegradation & Biodeterioration in Latin America. UNEP=UNESCO=ICRO-FEPAGRO=UFRGS, Porto Alegre, Brazil, pp. 68–70. Resende, M.A., Rezende, G.C., Viana, E.M., Becker, T.W., Warscheid, T., 1996. Acid production by fungi isolated from historic monuments in the Brazilian state of Minas Gerais. In: Gaylarde, C.C., S�a, E.L.S., Gaylarde, P.M. (Eds.), Proceedings of LABS 2, Biodegradation & Biodeterioration in Latin America, UNEP=UNESCO=ICRO- FEPAGRO=UFRGS, Porto Alegre, Brazil, pp. 65–67. Robertson, M.G., 1989. Mesoamerica. The Journal of Middle America, Merida, Yucatan, 19. Sablo�, J.A., 1997. Uxmal and the Northern Maya Lowlands. In: The Cities of Ancient Mexico. Reconstructing a Lost World. Thames and Hudson Inc., New York, pp. 89–104. Salvarezza, R.C., de Mele, M.F.L., Videla, H.A., 1983. Mechanisms of the microbial corrosion of aluminum alloys. Corrosion 39, 26–32. Videla, H.A., Characklis, W.G., 1992. Biofouling and microbiologically in uenced corrosion. International Biodeterioration & Biodegradation 29, 195–212. Warscheid, T., Oelting, M., Krumbein, W.E., 1991. Physicochemical aspects of biodeterioration processes on rocks with special regard to organic pollutants. International Biodeterioration 28, 37–48.