When Chocolate Was Young

1. Into the Garden
Born on the slopes of the Amazon Basin, the first chocolate trees (Theobroma cacao) shed their fruit pods to the ground and could barely capture the attention of a passing squirrel or macaw. The Amazon by this time was a paradise of intertwined plant and animal life in greens, browns, all colors, in a quantity and variety not seen anywhere else on earth. Immense ceiba and laurel trees rose to protect the cacao from the burning rays of the sun, and grasses and leaves provided a moist and fertile blanket below. It would be an unknown many thousands of years but eventually the first man and woman walked barefoot down the slopes of the Andes and found the cacao tree, broke open a pod on a rock, and ate and drank of the sweet pulp inside. This was the beginning of the end of cacao in its truly pure and natural state; so began thousands of years of history as men became inextricably linked to the bitter seeds of the plant and made the first chocolate, and the seeds helped fuel war campaigns of the Olmec, Maya, Aztec and others, and meanwhile the seeds became the engine of trade, and literally the money that made the ancient Mesoamerican economies go round. Later they were the seeds of enslavement and the forced relocation of many millions of African men and women to cacao plantations in the Americas. In this article we will focus on the first cultivation of the cacao tree and the first making of chocolate; we are interested in who made chocolate, when, where, and for what purpose. We concentrate on the very beginning of the story. It is a story we know from the chemists and archaeologists whose artifact testing is writing the history of the earliest chocolate in the world.

As demonstrated above the cacao tree is exotic in appearance, with colored seed pods ranging from green to red to yellow to brown, and grow from the trunk instead of the branches (“cauliflory”). As the tree will only grow in warm and humid conditions of plus or minus 20 degrees of latitude, the surrounding plants are also exotic. Inside the thick rind of the fruit is a white or purplish pulp that can be made into a beverage–alcoholic or non-alcoholic–and also almond sized seeds which can be dried in the sun, roasted on a heated stone, de-shelled, and finally ground into a semi-liquid paste (raw chocolate). At some point, by man or by nature, the trees spread to upper central America and southern Mexico (the region known as Mesoamerica). From this vast garden setting the early peoples would add maize and water to chocolate to make gruel, and add also relishes like chili peppers, petals of “ear flower,” honey, maguey sap, vanilla, and achiote (Coe, 2013, pp. 62, 87, 94; Presilla, 2009, p. 9). It is important to know that both the pulp of cacao and the seed contain 1-2 % by weight caffeine and related stimulants (Coe, 2013, p. 29; Henderson, 2007). The Aztecs were known to make disks of solid chocolate, and amplified by the equivalent of 10 cups of coffee in 50 grams of chocolate their warriors would take to the trails, hunting for enemies and looking for new sources of slaves.

2. Into the Pot
Most of what we know or can guess about the earliest chocolate use in Mesoamerica is projected backwards from chocolate use today in rural regions, first-hand accounts of the Spanish colonizers, and Mesoamerican texts (codices), languages, and artifacts. As far as prehistoric use of chocolate goes, almost all of what we know has been gleaned from close examination of cooking and serving vessels, which may have illustrations, hieroglyphics, or even ancient residues that can be analyzed in the laboratory. 

Mesoamericans have made and used ceramic cookware for at least 3,500 years (Soleri, 2013). The forms can be remarkably evolved, as for example vessels with handles, double spouts, bridged spouts, lids with locking bayonets, colored illustrations, relief sculptures, or tripod legs (Hall, 1990; Powis, 2002). The vessels were made well enough to last several thousand years in some cases. Illustrations of cacao on prehistoric cookware are not so common and the textbook example is the Princeton Vase, a Mayan vessel from 670-750 AD (Late Classic); this illustration shows an elegantly dressed woman pouring what is undoubtedly a chocolate drink from a height to make a froth, with a wild assortment of other women and Mayan gods in attendance. The scene suggests gender aspects and class aspects of the early Mayan consumption of chocolate. The Maya were interested in the metaphysical aspects of frothy drinks, and the cylindrical jar and method of preparing chocolate shown is also described in Conquest era literature; perhaps this pouring from a height originally resembled the froth from waterfalls in the Yucatan. On another Mayan vessel a smart-looking monkey contemplates a cacao tree; of course monkeys were cracking open cacao pods on the peninsula long before man arrived in the region.

In the case of hieroglyphic writing on vessels, this shows the vivid and imaginative writing system created by the Mayans, the fine detail the artists were capable of, and an interesting Mayan version of the monogram. The various serving vessels, for example, might be labeled to identify the owner (usually male), the function of the vessel, and the recipe of the contents; in fact one of these personalized jars used for chocolate (vessel 15 of Hurst, 1989) was key to begin decoding the Mayan hieroglyphic system. (See David Stuart’s article in McNeil, 2006.)
Of course, careful examination of the excavation site around an artifact is essential to estimate age and function. 90% of Mesoamerican vessels shown to be associated with chocolate consumption are discovered in burial locations (Powis, 2002), and by context these are invariably burials of high-status individuals (see Powis for diagrams of graves containing approximately 20 vessels each, dubbed a “complete table setting”). These vessels are thought to have been filled with chocolate drinks and other foodstuffs for nourishment on the journey to the underworld (Hall, 1990); by the fill lines still visible we know they contained liquids, and sometimes residues remain that can be scraped off and tested. Intact chocolate vessels are likely to be found in burial sites because of the natural protection, however pieces of vessels (“potsherds”) are also found in ancient refuse heaps and ancient construction fill. Intact vessels removed from protected locations are preferred for residue testing because they are thought less likely to be contaminated.

(A detail of a line drawing by John Montgomery of the Princeton Vase is shown below. The original vase can be viewed at the Princeton University website, http://artmuseum.princeton.edu/collections/objects/32221 .)

3. Into the Laboratory

The sections above have hinted at some interesting questions without providing all of the answers. By the end of the 1980s, research into the natural history of chocolate had generated more questions than answers and new methods of attacking problems were needed; in 1989 Jeffrey Hurst, a chemist at the Hershey company, performed for the first time chemical testing on ceramic vessels recovered from digs in Mesoamerica (Hurst, 1989). There was a new expectation of discovery regarding who had made the first chocolate, what were the prehistoric recipes, which social classes consumed chocolate, and so forth.

In that first study Hurst tested an unspecified number of Mayan vessels from the Rio Azul site in Guatemala by High Performance Liquid Chromatography (HPLC), the usual laboratory method for separating and identifying compounds (Fig. 1 below). The Hurst paper determined that several vessels dating to 460-480 AD had contained cacao; this pushed the use of cacao back to approximately one millenium before the first Spanish records and was therefore a landmark study. The paper also introduced two important notions that would be used (and sometimes questioned) in subsequent studies: 1. Theobroma cacao is the only Mesoamerican commodity containing theobromine, so theobromine can be used as a marker for cacao, and 2. Unglazed ceramics absorb theobromine, which is stable indefinitely and can be sampled without contamination by removing some of the ceramic (page 279, and see also Soleri, 2013, pages 349 and 357.) We remarked above that cacao contains 1-2% by weight theobromine and caffeine.

In 2002 Terry Powis sampled Mayan spouted pots from the Colha site in Belize on the Gulf Coast and found two of the pots had used for cacao, pushing the use of cacao back to 600-400 BC (Powis, 2002). Powis entertained the notion that the spout itself could be used as a marker for frothing and chocolate consumption (pp. 94, 96), although a spout may just be easier for pouring in general.

In 2007 John Henderson pushed consumption back to 1150 BC when he found 11 of 13 vessels from the Puerto Escondido site, Honduras, had been used for cacao (Henderson, 2007). Recall that pulp of cacao as well as the seeds contains theobromine and caffeine; Henderson concluded that the most ancient cacao drinks were not chocolate but made from the fermented cacao pulp. Unfortunately, both the volatile alcohol in pulp drinks and the lipids in chocolate drinks would disappear quickly from containers (Henderson p. 18939; Hurst, 1989, p. 286), and there is no test to distinguish between chocolate drinks and pulp drinks. Like Powis, Henderson suggested that spouted pots appearing approximately 1000 BC may be associated with frothing and chocolate drinks, versus pulp drinks. The vessels in this study were also tested for beeswax and capsain (markers for honey and chili pepper, respectively); the findings were negative.

In an effort to pin down the prehistoric use of cacao geographically, Powis next sampled over 100 vessels from both the Pacific and Gulf Coast regions (Powis, 2008) and found two very old vessels testing positive, one from each coastal region. It turns out a Mokaya vessel from the Paso de la Amada site in the state of Chiapas, Mexico is the oldest vessel with cacao residue ever found–dated to 1900-1500 BC by the construction fill where it was discovered. It is actually a broken piece from a small brown gourd-shaped pot with vertical ridges. The Mokaya were some of the very earliest settlers in Mesoamerica (1900-1700 BC; p. 36), and the find may settle the question of which coast first spread the cultivation of cacao, unless cultivation arose independently in both regions. Powis concluded that the Mokaya brought chocolate drink consumption to the Gulf Coast; in any case the Mokaya and pre-Olmecs had cacao drinks long before the Maya and the Aztecs had them.

By the time of a study by Daniela Soleri in 2013, researchers had gained confidence in their knowledge of Mesoamerican cacao consumption; in this study vessels from the Valley of Oaxaca, Mexico were tested for maize as well as cacao, with the hopes of obtaining information on ancient cacao recipes (Soleri, 2013). As mentioned above, it is known that the Aztecs and Mayas prepared drinks and gruels made from cacao with maize; in fact this practice continues among some of the indigenous peoples of Mexico. The study found 3 of 8 artifact vessels had been used for cacao and 2 of these had also been used for maize; the conclusion was these very old vessels had been used for a recipe containing both cacao and maize. Of course, the positive test for cacao and maize does not prove simultaneous use beyond a doubt; vessels can be used across time for different things. The Soleri study is noteworthy as the first effort to provide any kind of a control: a present-day ceramic vessel used for five years to prepare cacao was also sampled.

Finally we report on a study of Daniela Washburn, which came under immediate fire because of its sampling method and conclusions. The 2014 study is notable for introduction of a proper control and testing for statistical significance. By sampling dust on the shelves at six museums (including the Peabody at Harvard) Washburn showed there is always some background methylxanthine contamination (theobromine and caffeine), but the theobromine levels found in vessels used for cacao are significantly greater and this is easily tested by an analysis of variance. The new non-destructive sampling method involves simply rinsing the ceramic vessel followed by centrifugation of the rinse water. The paper of Washburn was interested in whether cacao from Central America was traded to indigenous peoples of the American Southwest, Midwest, and Southeast; the study found that cacao was being used by populations as far away as Florida (1000 AD).

We caution the reader that we have not summarized here every study on the subject of artifact testing for chocolate/cacao; undoubtedly the number of studies will continue to multiply due to simpler testing procedures, the popularity of chocolate, and interest in the Aztecs and Maya and the history of chocolate. We hope we have given, however, some idea of what the archaeological interest in ancient chocolate consumption is all about.

Figure 1. A typical High Performance Liquid Chromatography (HPLC) setup is shown below. A solvent or solvents in (1) that are appropriate to the substance (7) to be separated out and identified are pumped by (5) though a “column” (9) that separates by adsorption. (10) is a detector for the wavelengths of e.g. UV light absorbed by the sample, and this information or “chromatogram” is recorded by the computer (11). The substance (in solution) is collected in (12). 


Coe, M. (2013). The true history of chocolate. London, England: Thames & Hudson, Ltd.
Hall, G. (1990). Cacao residues in ancient Maya vessels from Rio Azul, Guatemala. American Antiquity, 55(1), 138-143.
Henderson, J. (2007). Chemical and archaeological evidence for the earliest cacao beverages. Proceedings of the National Academy of Sciences, 104(48), 18937-18940.
Hurst, W. (1989). Authentication of cocoa in Maya vessels using High Performance Liquid Chromatographic techniques. Journal of Chromatography, 466, 279-289.
Lo Coco, F. (2007). Determination of theobromine, theophylline, and caffeine in by-products of cupuacu and cacao seeds by High Performance Liquid Chromatography. Journal of Chromatographic Science, 45, 273-275.
Loudon, G. (2016). Organic Chemistry. New York, NY: W. H. Freeman.
McNeil, C. 2006. Chocolate in Mesoamerica: a cultural history of cacao. Gainesville, FL: University Press of Gainesville.
Powis, T. (2002). Spouted vessels and cacao use among the Preclassic Maya. Latin American Antiquity, 13(1), 85-106.
Powis, T. (2008). The origins of cacao use in Mesoamerica. Mexicon [sic], 30, 35-38.
Presilla, M. (2009). The new taste of chocolate. Berkeley, CA: Ten Speed Press.
Snyder, L. (2010). Introduction to modern liquid chromatography. Hoboken, NJ: Wiley.
Soleri, D. (2013). Archaeological residues and recipes: exploratory testing for evidence of maize and cacao beverages in Postclassic vessels from the Valley of Oaxaca, Mexico. Latin American Antiquity, 24(3), 345-362.
Washburn D. (2014). Chemical analysis of cacao residues in archaeological ceramics from North America: considerations of contamination, sample size and systematic controls. Journal of Archaeological Science, 50, 2014.
* The red pod cacao tree illustration is (C) 2013; we obtained kind permission to use it from Lake Champlain Chocolates. The Princeton Vase line drawing is (C) 2000 by John Montgomery. The HPLC schematic drawing is licensed under Creative Commons.


Leave a Reply

Please log in using one of these methods to post your comment:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s