Fall/Winter 1997
Chenopodium quinoa are present in the southwestern Titicaca Basin by at least 3500 BP.	LAKE TITICACA Studying Chenopobium quinoa Evidence is found for an Andean domesticate by Phyllisa Eisentraut IN THE SPRING 1995 ISSUE OF BACKDIRT, I presented the preliminary results of a macrobotanical analysis for Quelcatani, a rock shelter located in the southwestern Titicaca Basin of southern Peru. Because the rock shelter contains materials from the little-understood Late Archaic/Early Formative transition (roughly 6000 to 3000 bp)-the time period during which many Andean peoples adopted agriculture-I hoped to document the domestication process of Chenopodium quinoa, a high-altitude cultivar native to the area. Unfortunately, I was unable to identify domesticated Chenopodium at Quelcatani based on measurements of seed diameter, or relative abundance. Since the spring of 1995, however, seed coat or testa thickness has been determined to be a more accurate measure of domesticated status and I have had the opportunity to submit a sample of Chenopodium seeds from Quelcatani and three additional Titicaca Basin sites believed to date from the Late Archaic/Early Formative to scanning electron microscope (SEM) measurement. These results indicate that domesticated Chenopodium quinoa are present in the southwestern Titicaca Basin by at least 3500 bp (see photo in upper right). In January, I did some work with Dr. Bruce Smith, the curator of ethnobotany at the Natural History Museum of the Smithsonian and an authority on eastern North American Chenopodium domestication. Dr. Smith's work has shown that seed-bearing plants, such as Chenopodium, undergo five interrelated morphological changes as the result of human intervention in the life cycle of the plant: 1. infructescence compaction and terminalization; 2. loss of natural shatter mechanisms; 3. uniform maturation of fruit; 4. loss or reduction of germination dormancy; and 5. increased perisperm-endosperm food reserves for germination. The first three changes occur to the plant itself, and are unlikely to be represented in the archaeological record. The latter two changes are preserved in the Chenopodium seeds from the Titicaca Basin. Seed dormancy is primarily controlled through a variety of chemical and/or physiological inhibitors but also through the presence of a thick seed coat; a thick seed coat prevents premature germination of naturally dispersed, mature seeds. For wild and weedy plants, a thick seed coat is advantageous. It restricts embryonic growth and development to periods of optimal survival by blocking water, gases, and other external elements necessary for germination. For domesticated crops, unless dormancy is of short duration, a thick testa is nonadaptive as human selection favors seeds that germinate soon after planting. Domesticated Chenopodium generally have testa thickness under 20 µm (see photo), although some types lack an outer seed coat entirely. Wild Chenopodium have measurements over 40 µm. Weedy types, a third variety of Chenopodium that oftentimes grow in tandem with domesticated plants, range in size from 25 to 30 µ. Increased food reserves, the second change observable in fruit and seed specimens from the Titicaca Basin, is indirectly linked to testa loss or reduction in seed coat thickness, especially for domesticated varieties of Chenopodium. Researchers have observed that testa reduction is accompanied by a change in overall fruit shape. The biconvex, rounded, and equatorial-banded shapes characteristic of wild Chenopodium, are replaced with flattened or truncated seed coats, rectanguloid in cross-section (see drawing). The form is the result of a shift in the cross section of the embryonic leaves adjacent to the fruit wall. Rectangular leaves increase the internal volume of the seed, without necessarily increasing overall seed size, and allow for an increased food supply for the emerging plant. The archaeological Chenopodium from the Titicaca Basin demonstrate a range of seed-coat measurements and cross-section shapes. The Chenopodium from Palermo and Sillumoccu-Huaquina, two open-air sites situated on the western shore of Lake Titicaca near the town of Juli, include very thin specimens (>10 µm), weedy specimens (about 25 µm), wild specimens (40+ µm), and margin configurations characteristic of all three types. Unfortunately, the archaeological context for these seeds has yet to be confirmed, and I am unable to examine the order in which they occur. I would generally expect wild varieties to precede domesticated and weedy varieties, although weedy types may continue to be used after domestication has occurred. The Chenopodium from Camata, also an open-air site on the western shore of Lake Titicaca, range from domesticated (4-18 µm) to weedy (21-26 µm) to wild (40+ µm), and support domesticated, weedy, and wild designations Several of the domesticated specimens have margins slightly less truncate on the embryonic leaf portion of the seed, but surface morphology is relatively smooth in cross-section and supportive of a domesticated identification. Thin testa Chenopodium are present at Camata from the earliest identified occupation (3400 bp) onward. Because excavations at Camata were relatively limited, it is unknown whether settlement during this time period (a phase prior to the first evidence of ceramics at the site) was year-round, only during the growing season, or temporary. A small sample accelerator mass spectrometer (AMS) age for a Chenopodium seed recovered slightly above the earliest occupation is consistent with other radiocarbon dates from the site, suggesting that disturbance is not responsible for the relocation of archaeobotanical remains within the site deposit. The Chenopodium from Quelcatani, the rock shelter I discussed in the Spring 1995 issue, include only domesticated (4-20 µm) and weedy (21-24 µm) varieties. The domesticated specimens have truncate margins, resulting in the characteristic rectanguloid cross-section. The weedy specimens unfortunately lack margins because of fragmentation or distortion caused by carbonization. The earliest, undisturbed thin testa Chenopodium at Quelcatani occur about 3500 bp when settlement at the site appears to be transitioning from a relatively mobile band of hunter-gatherers to a sedentary or near-sedentary group of agropastoralists. Thin testa are present even earlier (potentially back to 5000 bp) but from deposits badly damaged by roof collapse. A small sample AMS age for a Chenopodium from the oldest portion of the site is significantly younger than expected, confirming the disturbed nature of the deposit. The macrobotanical analysis of the these archaeobotanical assemblages provides valuable insight into issues of Late Archaic-Early Formative subsistence, while the SEM measurements address questions of early plant domestication in the Titicaca Basin. Dosmesticated C. quinoa are confirmed in the southwestern Titicaca Basin by at least 3500 years ago. Since the Camata deposits date back to only 3400 years ago and the oldest Quelcatani levels are badly disturbed, we have no idea how much earlier domesticated C. quinoa may appear. I hope to begin addressing this problem with additional work at Late Archaic-Early Formative sites soon. Phyllisa Eisentraut is Coordinator of the South Central Coastal Archaeological Information Center.
	Top | About Backdirt | Contents | Next