You are on page 1of 5

Journal of Archaeological Science 39 (2012) 3475e3479

Contents lists available at SciVerse ScienceDirect

Journal of Archaeological Science

journal homepage:

Applying GIS tools to define prehistoric megalith transport route corridors: Olmec
megalith transport routes: a case study
Leslie C. Hazell a, *, Graham Brodie b
FILAS, La Trobe University, Melbourne, Victoria 3086, Australia
Melbourne School of Land and Environment, Melbourne University, Dookie, Victoria 3647, Australia

a r t i c l e i n f o a b s t r a c t

Article history: Applying spatial data management and analytical tools is an accepted method for investigating potential
Received 7 March 2012 settlement patterns. This study applied Geographic Information System (GIS) technology to the study of
Received in revised form Preclassic Olmec megalith transport in Mesoamerica. The GIS slope gradient tool was used to identify
16 May 2012
potential transport corridors with gradients of less that 1:10, which defines the limit of human effort
Accepted 17 May 2012
needed to control these stones as demonstrated in ethnographic and replication experiments by various
researchers. These outcomes were also used to refine transportation methods and likely start points that
would be used to cross lowlands near to the San Lorenzo centre of Olmec society.
Megalith transport
Ó 2012 Elsevier Ltd. All rights reserved.
GIS analysis
Slope gradient

1. Introduction aspects of this complex undertaking. In Mesoamerica transporting

megaliths involved technological knowledge, logistical expertise
Stone use is enduring evidence of the industry, enterprise and (Diehl, 2007) and significant resources to overcome terrestrial
resource of prehistoric cultures. Transporting large stones or features and e according to some e oceanographic hurdles (Coe,
megaliths was a pivotal part of this prehistoric industry. The 2000; Stuart, 1993; Velson and Clark, 1975). While some aspects of
methods and routes that were chosen for this undertaking continue replication experimentation are inappropriate it is still possible to
to be the subject of debate and have given rise to considerable use human physical capabilities, identified in replication experi-
speculation. The sources for stones were often distant from their ments and corroborated by ethnographic observations. These sour-
final position and the routes that were used to retrieve the stones ces have been adopted as part of Geographic Information System
traversed difficult terrain or open seas (Hazell, 2011). The sources (GIS) analyses in order to identify viable and sustainable land route
and routes often bypassed, what to modern eyes, may seem to be options and transportation methods (Ayres and Scheller, 2001;
suitable stones and more convenient sources; so adding significant Erasmus, 1965; Heyerdahl, 1958; Horvath and Finney, 1976; Irwin
logistical issues to the problem of stone retrieval. et al., 1990; Richards and Whitby, 1997; Schiffer and Skibbo, 1987).
A limited material and often fragmented, archaeological record, This Mesoamerican investigation of Preclassic (1800 BCe
limits any endeavours to determine the methods and routes that 400 BC) Olmec society megalith transportation includes these
were used to undertake these monumental activities. To overcome and other challenges including limited field survey access to the
this, ethnographic examples are often cited (Ayres and Scheller, large areas involved and securing sufficient funds to study such an
2001; Dillon, 2004; Heyerdahl, 1958; Souden, 1997) and replica- extensive site; therefore another technique was required to study
tion experiments (Atkinson, 1960; Ayres and Scheller, 2001; this geographically extensive transport problem. The focus of this
Richards and Whitby, 1997; Van Tilburg, 1995) are also used as study was on land transport of megaliths used in sculptures known
a basis for research. as the Colossal Heads, ten of which were found at San Lorenzo.
Experiments have limitations because the conditions under Large numbers of stones, most weighing between 6 and 26 tonnes,
which they were conducted may differ substantially from those were moved to this area and also to La Venta, which was another
encountered by ancient cultures and are unlikely to replicate all major and later Olmec complex. For this research a mean mass of
twenty tonnes has been adopted.
This paper documents our use of GIS technology to determine
* Corresponding author. 105 Lewis Road, Tatong, Victoria 3673, Australia.
Tel.: þ61 3 5767 2353. whether viable land routes and methods, which comply with
E-mail address: (L.C. Hazell). human power capability calculations and terrain conditions, can be

0305-4403/$ e see front matter Ó 2012 Elsevier Ltd. All rights reserved.
3476 L.C. Hazell, G. Brodie / Journal of Archaeological Science 39 (2012) 3475e3479

established. The outcomes derived from this work were able to Any direct land route would encounter swamps and perennial
show that ancient Olmec could have transported stones by land floodplains (INEGI, 1985, 1986, 1999). The existence of these natural
in contrast to the more favoured open water and river routes. obstacles endorsed argument for water transport of these mega-
However, it went further by defining some transport methods liths (Cyphers and Ortiz, 2000; Cyphers and Zurita-Noguera, 2006,
because ground conditions also precluded the use log rollers, sleds Tamayo and West, 1964).
and wheeled vehicles, even though the wheel was known to Olmec It is known that the Olmec used canoes for trade, whether these
society, as evidence of wheeled toys demonstrate (Charnay, 1974). would be suitable for megalith transport is uncertain (Anawalt,
The distances involved are considerable, and the terrain is 1992; Callaghan, 2003; Dillon, 1975; Edwards, 1978). Analyses of
challenging, with swamps, floodplains and rivers, many of which waterborne options highlighted serious limitations of the most
are seasonal. From their source near Cerro Cintepec in the Tuxtla likely watercraft known at the time these stones were moved. A
Mountains to their final destination at the Preclassic centre of San study of rafts and multi-hull dugout canoes (Hazell, 2003), based on
Lorenzo is a straight line distance of around 80 km across this performance observations, structural integrity, and stability under
difficult landscape (Coe, 2000: 68). Importantly gradient analysis significant loads, combined with human capability studies
using the GIS software allowed us to establish a potential transport (Baumeister, 1967; Hazell, 2011; Henderson and Haggard, 1925;
corridor from which the Olmec could begin to cross the lowlands. Horvath and Finney, 1976; Severin, 1988) of maximum human
Transportation methods were also defined by gradient and related power and sustained effort (Doran, 1971, 1978; Edwards, 1965;
road structures needed to support this activity. For this reason, our Kent, 1958; Ling, 1970; Muckle, 1975; Walton, 1906; West, 1961) has
GIS matrix included soil types and their distribution, as existing questioned whether waterborne options for moving these mega-
archaeological evidence shows significant earth structures associ- liths would be viable or reliable. Even if the Olmec did use water
ated with the San Lorenzo complex. transport for moving these stones from their source to final desti-
nation, establishing land based pathways from their source points
1.1. Preclassic Olmec in the foothills of the Tuxtla Mountains to suitable water staging
sites would still require analysis of large tracts of land.
Olmec society had a significant sphere of influence and is often With the considerable area involved and limited access, the use
described as the “Mother culture” (Benson, 1981; Benson and de la of GIS allowed a viable, cost effective research methodology for this
Fuente, 1996; Coe, 1965, 1981; Coe and Diehl, 1980; Cyphers, 1996; aspect of the research.
Puche et al., 1996). Archaeological surveys and excavation projects
have shown that Olmec society built complex centres at different 1.3. Land based megalithic transport
sites during their florescence. They rose in prominence in about
1500 BC and then declined again by 600 AD (Coe, 1965, 1970, 2000; The region between the source in the Tuxtla Mountains and final
Coe et al., 1967; Coe and Diehl, 1980; Stirling, 1943, 1965). Their destination on the San Lorenzo Plateau is characterized by many
“heartland” dominated much of what is now known as Veracruz rivers, seasonal swamps and floodplains which are subject to
State in Mexico. inundation during wet seasons or are permanently under water
The “Olmec heartland”, which is accepted as the Olmec sphere (Cyphers, 2008; Cyphers and Ortiz, 2000). The use of land routes to
of social, political and economic influence, was an area of some avoid these constraints during drier months, or using strategies to
16,000 km2. The Olmec traded in obsidian and jade, neither of which overcome inundated areas, has already been suggested (Cyphers,
occur naturally around Olmec centres. Trade pathways, while suit- 2006). Besides these landform challenges, manpower needs must
able for precious mineral trade, would not necessarily suit megalith be considered for transportation methods.
transport; however trade in precious materials suggests that regu- Gradient is an important constraint when moving large stones.
larly used pathways may have existed during Olmec times. This constraint will limit the ability of humans to move the Colossal
Olmec society moved many large stones for stelae, altars or Heads across the landscape, irrespective of whether these mega-
thrones and sculptures. Amongst the hundreds of stones that were liths are being hauled down hill, where there is a need to control
used for different purposes, the largest known stone was La Venta 1, their descent, or when they are being hauled uphill to isolated
which weighed about 40 tonnes (Clewlow et al., 1967; Clewlow, pockets of higher ground to avoid uncertain conditions. Gradient
1970). At San Lorenzo near Tenochtitlan extensive raised plat- also plays an important part in the final stages when hauling them
forms, numerous mounds, stelae, altars and stone sculptures, known onto the San Lorenzo Plateau itself. Therefore slope gradient anal-
as Colossal Heads, have been found (Benson and de la Fuente, yses were used to link pathways from the higher ground to
1996; Clewlow et al., 1967; Clewlow, 1970, 1974; Pohorilenko, potential routes across the low country.
1996; Stuart, 1993).
2. GIS analysis strategy
1.2. Colossal Heads
Although swamps, rivers and floodplains were recognised
Most of the seventeen Colossal Heads weigh between 6 tonnes obstacles, the GIS analysis began by examining gradients from
and 26 tonnes. They vary in height from between one and a half sources at Cerro Cintepec (Williams and Heizer, 1965) or nearby
metres to nearly 3 m; their circumference varies from just over 3 m (Grove et al., 1993) in the Tuxtla Mountain foothills. Controlling the
up to nearly 6 m (Clewlow et al., 1967) and each stone has a human stones, while descending these slopes, is difficult with the area
face carved into it. The distinctively individual features of each characterised by many steep gullies. Conversely providing adequate
stone’s carved face, arguably links each of the Colossal Heads to an motive power to drag the stones uphill also defines acceptable
Olmec ruler (Clewlow et al., 1967; Stirling, 1965). The stones are routes. To establish manpower parameters on routes, ethnographic
generally accepted as being sourced near the foothills of the Tuxtla (ArchaeoNews, 2004; Harmon, 2005; Mladjov and Mladjov, 1999),
Mountains (Williams and Heizer, 1965). Ten of these stones have historical and experimentation examples of human power capa-
been found on or in gullies on the San Lorenzo Plateau. Some heads bilities and sustained effort expectations (Richards and Whitby,
are believed to have been reused altar stones, perhaps pointing to 1997; Van Tilburg, 1995) were adopted. These analyses suggested
their value and linked to difficulties in acquiring these stones a gradient of 1:10 defined the limit of human effort needed to
(Cyphers, 2006; Porter, 1990). control these stones so that the combined power of a work party
L.C. Hazell, G. Brodie / Journal of Archaeological Science 39 (2012) 3475e3479 3477

can be effectively coordinated. This was also demonstrated in and joined in the GIS. This process became a potential source of
ethnographic and replication experiments (Heyerdahl, 1958; accumulative errors.
Richards and Whitby, 1997). Therefore potential routes identified All scans were enhanced using Adobe Photoshop 7 to improve
by applying the GIS slope analysis, were based on gradients that did legibility and associated accuracy during digitisation. Points on
not exceed 1:10 at any point. Subsequent cross checking showed contour lines were digitised with changes in direction and inter-
that the routes identified in the GIS study mirrored modern roads mediate points as frequently as possible to minimize error (Douglas
or tracks. and Peucker, 1973). Nevertheless this stage added another potential
source of error. Our interpretation of specific features such as
2.1. Establishing a spatial data set for GIS analysis contour position and value imposed limitations on final accuracy of
the spatial data set.
GIS have been used for many spatial analyses (Longley et al., A Triangular Irregular Network (TIN) surface model of the
2005; Romero-Calcerrada et al., 2008; Zhijun et al., 2009). In this landscape was generated from the contour data. Assembling this
instance though, by using an appropriate spatial database, GIS can data as a surface model was a necessary precursor to applying
provide a valuable insight into different aspects of any prehistoric a slope gradient analysis of the landscape to identify pathways of
society, particularly when archaeological evidence is limited and suitable gradient for land transport.
terrestrial features are likely to have played a part in the activity.
We used ArcView 3.2 GIS software, and its associated 3D Analyst 2.2. Results of the GIS analysis
extension, to create and digitise geographic themes.
The data set was primarily established from survey mapping at From the spatial analysis, based on slope analysis and circum-
scales of both 1:250,000 (INEGI, 1984, 1986, 1997) and 1:50,000 navigation of major obstacles in the flood plain, two possible
(INEGI, 1984, 1985, 1987, 1988, 1990, 1996, 1999, 2000, 2003). These overland routes for transport of the Colossal Heads were identified
maps were scanned, georeferenced and used to create themes of (Fig. 1). To adopt other less direct routes would require crossing
contours, rivers, swamps, soil types and vegetation. Lowland gullies and associated rivers with constant changes of direction
country, intersecting with possible routes and encompassing the while controlling the stones. Within the projected pathway
region surrounding the San Lorenzo/Tenochtitlan plateau, was corridor, the Olmec Llano del Jicaro workshop site described by
included. The GIS database also included La Venta, thus allowing Gillespie is situated (Gillespie, 1994).
future analysis of transport routes to that site.
The initial digitising procedure was undertaken using a HP 2.3. Some comments on applying GIS technology to determine
5100c flatbed scanner with a default scan speed for high quality possible transport routes
scans. Resolution was set at 150 dots per inch (DPI). Even with the
larger scale of the survey maps, each map required at least four and The management advantages of a GIS database are well known
sometimes six A4 scanned sheets, which needed to be geolocated (Longley et al., 2005) but its application for this research should be

Fig. 1. Spatial model of the Olmec sites showing the overland routes identified during the GIS study.
3478 L.C. Hazell, G. Brodie / Journal of Archaeological Science 39 (2012) 3475e3479

explained. The regional nature of the Olmec stone transport matrix encompassing geographical features such as vegetation,
imposed more extensive use of preparatory scanning and data swamps and floodplains are common themes a linked to settle-
processing than might otherwise be the case for a less extensive ment patterns for example. In our research though, these features
project. In addition, the resolution or interpretation quality were barriers and their complexity and their area, indicated
required to generate the terrain model of the Olmec heartland a matrix was ideally suited for looking at transport routes. In this
became a compromise between practicalities of managing file sizes mode the tools could also be applied to investigating trade routes
and the need for reasonably accurate data. Larger files allowed clear for example.
image processing but required extensive handling to exchange and
interpret the data.
The type of analysis that has been proposed in this paper could
be undertaken empirically or by simply reading the contours on
LCH thanks La Trobe University library staff for access to
the map to trace likely routes that showed minimal grades;
resources and his wife Dianne for her continued patience, support
however this would have relied on visual assessment of the
and constructive comments.
contour spacing or measuring and calculating the grade along
a multitude of potential pathways. This is further complicated
because the foothills near the stone source are complex land- References
forms, reflecting their volcanic origins and access permission from
landholders is uncertain (Chase, 1981; Nelson, 1990; Williams and Anawalt, P.R., 1992. Ancient cultural contacts between Ecuador, West Mexico and
the American Southwest: clothing similarities. Latin American Antiquity 3,
Heizer, 1965). 114e129.
The surface model was developed using current topographical ArchaeoNews, 2004. Megalithic Tradition Is Preserved in Indonesia. Stone Pages,
maps of the area. Without ground-truthing, our interpretation and 5th June 2004 ed.
Atkinson, R.J.C., 1960. Stonehenge. Penguin Books, Mitcham, Victoria, Australia.
digitising is subject to the tolerances and accuracy of the source Ayres, W.S., Scheller, C.J., 2001. Experiments in Stone Transport Status Architecture
survey maps. This potential source for error was recognised in this and Tool Stone on Pohnpei, Micronesia. To appear: Pacific Archaeology: the
assemblage process and in subsequent analyses of the Digital Legacy of Richard Shutler, Jr.
Baumeister, T. (Ed.), 1967. Standard Handbook for Mechanical Engineers, seven ed.
Elevation Model (DEM) associated with the slope analysis tools Mc Graw-Hill Book Company, New York, London, Sydney.
(Hageman and Bennett, 2000). Benson, E. (Ed.), 1981. The Olmec & Their Neighbors. Dumbarton Oaks Research
Library, Washington, DC, USA.
Benson, E., de la Fuente, B., 1996. Olmec Art of Ancient Mexico. National Gallery Of
2.4. Limitations associated with GIS application to this research
Art, Washington, USA.
Callaghan, R.T., 2003. Prehistoric trade between Ecuador and West Mexico. Antiq-
A major limitation in this study was the process of data acqui- uity 77, 796e804.
sition. Digitising such extensive landscapes is always a challenging Charnay, D., 1974. Wheeled "toys". In: Deuel, L. (Ed.), Conquistadors without
Swords: Archaeologists in the Americas. Schocken Books, New York,
undertaking. Without access to large scale remote sensing tech- pp. 175e186.
nologies, this study relied on the digitisation and interpretation of Chase, J.E., 1981. The sky is falling: the San Martin Tuxtla volcanic eruption and its
topographical land survey maps. Errors in interpretation of contour effects on the Olmec at Tres Zapotes, Verucruz. Vinculos 7, 53e59.
Clewlow, C.W., Cowan, R.A., O’Connell, J.F., Benemann, C., 1967. Colossal Heads of the
lines occur and may have influenced the gradient analysis; however Olmec Culture. Department of Anthropology, University of California, Berkeley.
the points of shallowest gradient in the landscape correspond to Clewlow Jr., C.W., 1970. Comparison of Two Unusual Olmec Monuments. June 1970
those parts of the digitised map, which are easiest to interpret ed.. Department of Anthropology, University of California, Berkeley.
Clewlow Jr., C.W., 1974. A Stylistic and Chronological Study of Olmec Monumental
accurately. To control a twenty tonne stone on steep downward Sculpture. Department of Anthropology, University of California, Berkeley.
slopes would have been impractical, as replication and historical Coe, M., 1965. The Olmec style and its distribution. In: Willey, G. (Ed.), Archaeology
observation have illustrated (Dillon, 2004; Heyerdahl, 1958; of Southern Mesoamerica. University Of Texas Press, Austin Texas, USA,
pp. 739e775.
Richards and Whitby, 1997). Additionally there is strong correlation
Coe, M., 1970. The Archaeological Sequence at San Lorenzo, Tenochtitlan, Veracruz,
between corridors, identified by this study, and modern roads in Mexico. June 1970 ed.. University of California Department of Anthropology,
the general area. Therefore the choice of potential transport paths, Berkeley.
Coe, M., 1981. Gift of the river: ecology of the San Lorenzo Olmec. In: Benson, E.
based on gradient analyses has some validity and could be used
(Ed.), The Olmec and Their Neighbours. Dumbarton Oaks Research Library and
with confidence as part of ongoing investigation. Collections, Washington DC.
Coe, M., 2000. Mexico. Thames & Hudson, London, UK.
3. Conclusion Coe, M., Diehl, R., Stuiver, M., 1967. Olmec Civilization, Veracruz, Mexico: dating of
the San Lorenzo phase. Science 155, 1399e1401.
Coe, M.D., Diehl, R.A., 1980. In the Land of the Olmec: the Archaeology of San
The outcomes presented in this paper are not final solutions to Lorenzo Tenochtitlán. University of Texas Press, Austin USA.
the stone transport problem. While our use of GIS was able to Cyphers, A., 1996. Reconstructing Olmec life at San Lorenzo. In: Benson, E.B.d.L.F.
(Ed.), Olmec Art of Ancient Mexico. National Gallery of Art, Washington,
highlight the value of using GIS software and using its extensions Washington USA.
for studying prehistoric cultural activity, the corridor outcomes Cyphers, A., 2006. Re: Olmec stone transport. In: (Ed.), Mel-
linked to gradient viability established other important results. bourne, Australia.
Cyphers, A., 2008. Prehistoric floodplains. In: LCH (Ed.), San Lorenzo, Tenochtitlan.
Firstly this corridor could be linked to the known Olmec workshop Cyphers, A., Ortiz, M.A., 2000. Geomorphology and ancient cultural landscapes of
at Llano Del Jicaro and secondly, the viable gradients indicated such Southern Veracruz. In: Ahler, S.R. (Ed.), Mounds, Modoc, and Mesoamerica:
gradients could only be traversed by dragging stones in contact Papers in Honor of Melvin L. Fowler. Illinois State Museum Scientific Papers,
Springfield, Illinois, pp. 99e110.
with the ground. In this regard the slope gradient tool has provided
Cyphers, A., Zurita-Noguera, J., 2006. A land that tastes water. In: Lucero, L.J.,
a useful insight into potential pathways and viable transport Fash, B.W. (Eds.), Precolumbian Water Management. University of Arizona
methods that may have been used by the ancient Olmec to move Press, Tucson, pp. 33e50.
Diehl, R., 2007. De cómo los reyes olmecas obtenían sus cabezas colosales, XXIX
their Colossal Heads from their source at or near Cerro Cintepec to
Coloquio de Antropología e Historia Regionales Colegio de Michoacán. Zamora,
their final destination on the San Lorenzo Plateau. Michoacán, México.
We have noted limitations in its use and demonstrated how this Dillon, B.D., 1975. Notes on Trade in Mesoamerica. Department of Anthropology
technology contributed to this research by defining areas of interest, University of California, Berkeley.
Dillon, P., 2004. Indonesia: Megalithic Arts Struggle On.
and establishing a starting point for investigating routes across the Doran, E., 1978. Seaworthiness of sailing rafts. Anthropological Journal of Canada 16,
lowlands between San Lorenzo and the Tuxtla Mountains. A GIS 17e22.
L.C. Hazell, G. Brodie / Journal of Archaeological Science 39 (2012) 3475e3479 3479

Doran, E.J., 1971. The sailing raft as a great tradition. In: Riley, C.L., Kelley Charles, J., Ling, S.-S., 1970. A Study of the Raft, Outrigger, Double and Deck Canoes of Ancient
Pennington Campbell, W., Rands Robert, L. (Eds.), Man across the Sea. Univer- China, the Pacific and the Indian Ocean. Institute of Ethnology Academia Sinica,
sity of Texas Press, Austin Texas, pp. 115e138. Republic of China, Nankang, Taipei, Taiwan.
Douglas, D.H., Peucker, T.K., 1973. Algorithms for the reduction of the number of Longley, P.A., Goodchild, M.F., Maguire, D.J., Rhind, D.W., 2005. Geographical
points required to represent a digitized line or its caricature. The Canadian Information Systems and Science. John Wiley & Sons.
Cartographer 10, 112e122. Mladjov, R.V., Mladjov, I.R., 1999. This side up. Discovering Archaeology, 90e93.
Edwards, C.R., 1965. Aboriginal Watercraft on the Pacific Coast of South America. Muckle, K., 1975. Naval Architecture for Marine Engineers. Butterworth & Co,
University of California Press, Berkerley, Los Angeles. London, England.
Edwards, C.R., 1978. Pre-Columbian maritime trade in Mesoamerica. In: Nelson, S.A., 1990. Volcanic Hazards in Mexico e a Summary. In: Univ. Nai. Auton.
Navarette, L. (Ed.), Papers of the New World Foundation, pp. 199e209. Mexico, Inst. Geologia 9, pp. 71e81.
Erasmus, C., 1965. Monument building: some field experiments. Southwestern Pohorilenko, A., 1996. Portable carvings in the Olmec style. In: Benson, E.B.d.L.F.
Journal of Anthropology 21, 277e301. (Ed.), Olmec Art of Ancient Mexico. National Gallery of Art, Washington USA,
Gillespie, S.D., 1994. LLano del Jicaro: an Olmec monument workshop. Ancient pp. 119e131.
Mesoamerica 5, 231e242. Porter, J.B., 1990. Las cabezas colosales olmecas como altares resulpidoes: "multi-
Grove, D.C., Gillespie, S.D., Ponciano, O.C., Hayton, M., 1993. Five Olmec monuments lación," revolución y reesculpido. Arqueología 3, 91e97.
from the Laguna de los Cerros hinterland. Mexicon 15, 91e95. Puche, M.C.S., Gonález De La Vara, Fernán, Karina, R., Durand, V., 1996. Daily life in
Hageman, J.B., Bennett, D.A., 2000. Construction of digital elevation models for Olmec times. In: Benson, E.B.d.L.F. (Ed.), Olmec Art in Ancient Mexico. National
archaeological applications. In: Westcott, K.L., Brandon, J.R. (Eds.), Practical Gallery of Art, Washington USA, pp. 35e39.
Applications of GIS for Archaeologists. Taylor & Francis, London UK, pp. 113e127. Richards, J., Whitby, M., 1997. The engineering of stonehenge. In: Cunliffe, B.R.C.
Harmon, D.b.P., 2005. Qala Yampu Project: Tiwanaku Experimental Archaeology. (Ed.), Science and Stonehenge. Oxford University Press, Oxford.
Hazell, L.C., 2003. Using environmental constraints, human power capability and Romero-Calcerrada, R., Novillo, C., Millington, J., Gomez-Jimenez, I., 2008. GIS
technological performance as parameters to investigate transport of megaliths analysis of spatial patterns of human-caused wildfire ignition risk in the SW of
using canoe rafts by Olmec society in Pre Classic Mesoamerica. In: Madrid (Central Spain). Landscape Ecology 23, 341e354.
Steinbrenner, L., Cripps, B., Georgopoulos, M., Carr, J. (Eds.), 36th Chacmool Schiffer, M.B., Skibbo, J.M., 1987. Theory and experiment in the study of techno-
Conference: Flowing through Time. The Archaeological Association of the logical change. Current Anthropology 28, 595e622.
University of Calgary University of Calgary, pp. 56e67. Severin, T., 1988. The Ulysses Voyage. Arrow Books Ltd., London, UK.
Hazell, L.C., 2011. An Investigation of Olmec Water and Land Megalith Transport Souden, D., 1997. Stonehenge. Collins & Brown.
Methods and Route Options, Historical and European Studies e Archaeology. La Stirling, M.W., 1943. Stone Monuments of Southern Mexico. Smithsonian Institution
Trobe, Melbourne, p. 320. Bureau of American Ethnology, Washington DC USA.
Henderson, Y., Haggard, H.W., 1925. The maximum of human power and its fuel. Stirling, M.W., 1965. Monumental sculpture of Southern Verucruz and Tabasco. In:
American Journal of Physiology 72, 264e282. Willey, G. (Ed.), Archaeology of Southern Mesoamerica. University of Texas
Heyerdahl, T., 1958. Aku-Aku. George Allen & Unwin Ltd., London, England. Press, Austin, Texas, USA, pp. 716e738.
Horvath, S.M., Finney, B., 1976. Paddling experiments and the question of Polynesian Stuart, G.E., 1993. New light on the Olmec. National Geographic, 88e114.
voyaging. In: Finney, B. (Ed.), Pacific Navigation and Voyaging. The Polynesian Tamayo, J.L., West, R.C., 1964. The hydrography of Middle America. In: West, R. (Ed.),
Society Incorporated, Wellington, pp. 47e54. Handbook of Middle American Indians. University of Texas Press, Austin, USA,
INEGI, 1984. Coatzacoalcos: Carta Edafologica, first ed. pp. 84e121.
INEGI, 1985. El Progreso, first ed. Van Tilburg, J.A., 1995. Moving the Moai: transporting the megaliths of Easter
INEGI, 1986. Coatzacoalcos: Carta Hidrologica de Aguas Subterraneas, first ed. Island. Archaeology 47, 34e43.
INEGI, 1987. Agua Dulce, 2a ed. Velson, J.S., Clark, T.C., 1975. Transport of Stone Monuments to the La Venta and San
INEGI, 1988. Chinameca, second ed. Lorenzo Sites. Department of Anthropology, University of California, Berkeley
INEGI, 1990. Juan Diaz Covarrubias, second ed. USA.
INEGI, 1996. San Juan Volador. Walton, T., 1906. Know Your Own Ship, tenth ed. Charles Griffin and Company
INEGI, 1997. Coatzacoalcos: Carta Topográfica, second ed. Limited, London, UK.
INEGI, 1999. Coatzacoalcos, first ed. West, R.C., 1961. Aboriginal sea navigation between Middle and South America.
INEGI, 2000. Las Chopas, first ed. American Anthropologist 63, 133e135.
INEGI, 2003. San Andrés Tuxtla, first ed. Williams, H., Heizer, R.F., 1965. Sources of Stones Used in Prehistoric Mesoamerican
Irwin, G., Bickler, S., Quirke, P., 1990. Voyaging by canoe and computer: experiments Sites. University of California Archaeological Research Facility.
in the settlement of the Pacific Ocean. Antiquity 64, 34e50. Zhijun, T., Jiquan, Z., Xingpeng, L., 2009. GIS-based risk assessment of grassland fire
Kent, J.L., 1958. Ships in Rough Water. Thomas Nelson and Sons. Ltd., London, disaster in western Jilin province, China. Stochastic Environmental Research
Edinburgh, UK. and Risk Assessment 23, 463e471.