Family
  Shock
The Shock Family Web Page

Archaeobotanical insights into diet on the Brazilian savanna, 500-2000 BP

Myrtle P. Shock1 and Renato Kipnis2

Poster presented at the 73rd Annual Meeting, Society for American Archaeology, March 26-30, Vancouver, BC, Canada.

Larger versions of most images are available by clicking on the image.

vegetation map of modern Brazil

Introduction

Located in central Brazil in the modern day savanna, Lapa dos Bichos is a rock shelter with amazing preservation of botanical remains. The limestone has created protected living surfaces where preservation could occur. The preservation of botanical remains brought into the shelter is enhanced by both the dry nature of the rock shelter and the six month dry season of the savanna. While the site dates from 11,500 to 500 BP, analysis began with the most recent levels of occupation. Domesticated plants are thought to have been introduced here between 2500 and 2000 BP. Collected botanical remains show that native savanna plants were retained as food sources with the introduction of domesticates.

picture of rock shelter entrance

Lapa dos Bichos entrance

picture of rock shelter stratigraphy

The stratigraphy of Lapa dos Bichos is visible in a reopened 1994 test pit just over a meter deep.

picture comparing mutamba preservation

Archaeological (left) and modern (right) examples of mutamba (Guazuma ulmifolia)

Plant preservation

Botanical remains were preserved through both carbonization and desiccation. Most pictures presented here are of desiccated remains as the degree to which the seed structures are visible is greater. The botanical remains found include both domesticated species such as corn, beans, and cassava and native plant foods such as palm kernels, custard apple, and passion fruit.

picture comparing umbu preservation

Archaeological (left) and modern (right) examples of umbu (Spondias tuberosa)

picture of rock shelter excavation

Excavation at Lapa dos Bichos

Dating

Dating of the agricultural material is limited at present, and so while there might be temporal distinctions between structures or levels 2 and 3, there is no basis upon which to analyze diachronic change.

picture of rock shelter excavation

The center of this unit has a small pit structure visible as a brownish circle.

Samples and data

All of the material used in this analysis come from the 2006 excavations. All of the sediment from the shallow pit structures was processed through nested geological screens (down to 0.5 mm) and sediment was collected. Around the structures the sediment varies in hues from red, natural to the soils of the area, to whites where the inclusion of ash is very high. Two liters of sediment from each excavation unit was collected for flotation and the remainder was screened in the field.

Botanical samples come from both the structures, standard excavation, and flotation samples, each of which have distinct limitations. However for prolific species with relatively large seeds, greater than 3 mm diameter, all of the procedures recover the remains. All structure material was sorted in the lab. Meanwhile the botanical remains from the fourteen excavation units were bagged by level and unit in the field.

picture of domesticated plants

Domesticated plants

Foods Plant parts found
A. Maize cobs
C. Bean seeds
D. Squash seeds
F. Cassava seed shell
H. Peanut seed pod
Industrial Plants
B. Cotton seeds and fibers
E. & G. Bottle gourd rinds and seeds

Comparing domesticates with native plants

The native plant species occur in great abundance. There are 50 different native species which have been identified and an additional 200 unknown seed and fruit rind morphologies. The identified plants include species considered to be folk medicine and industrial fiber sources.

Agriculture was adopted by the inhabitants of this site and concurrent local changes in material culture include the introduction of pottery. A local survey documented many additional archaeological sites along the river where soil may have been farmed.

picture of native plants

Some native savanna plants

Foods Plant parts found
1-3 Palm nuts shell
Syagrus spp., Attalea sp.
A. Custard apple seed
Annona sp.
B. Pequi seed
Caryocar brasilensis
C. Cansanção seed and shell
Cnidoscolous pubescens
D. Passion fruit seed
Passiflora sp.
E. Grass seed
Graminae - large seeded specie
F. Umbu seed
Spondias tuberosa
G. Jatobá seed and pod
Hymenaea sp.
Cajá seed
Spondias mombin
Industrial Plants
Native cotton seed
Gossipium barbadense
picture of palms

A variety of palm species were used. The most predominant are Attalea sp. (1) and two Syagrus spp. (2 and 3).

picture of palm quantities

Palm nut shells, approximately 700 grams

Plant quantities

Botanical remains are very common at Lapa dos Bichos. There are both structures and units with two kilograms of palm nut shells. These samples number in the thousands of fragments. Palms are not the only plant where the number of remains is high. There are large quantities of jatobá pods, cansanção shells and corn cobs.

Plant Ubiquity

graph of selected plant ubiquities

Ubiquity, the percentage of sample locations where a specie is found, gives a measure of specie frequency. Immediately striking is that some native plant species are nearly ubiquitous, being found in most of the structures and excavation units. The relatively even distribution of species between structures and excavation units lends credence to the perspective that the fill of structures are midden materials from the site's inhabitants. The ubiquity of common native plant species and maize is high. Many of the rarer species occur infrequently and have ubiquities similar to cotton and wild cotton.

Maize variability

picture of maize from one structure

Maize from one structure

The maize ears from Lapa dos Bichos are extremely variable. The number of rows on an ear varies from four to 12 rows. Cob diameters vary from three to eleven millimeters. The cob is often very fragile, however on some specimens this layer is extremely durable and thick. Maize ears have various arrangements of kernels: spirals, ladder, side by side, or random organizations. The maize from both structures and the excavation units is variable with no patterns that would indicate the development of homogeneous races. The variability within corn ears is curious considering the location of the site within central Brazil far from the Mesoamerican centers of maize domestication. A homogeneous structural type would be expected from a small introduced population.


Many thanks to the Museum of Archaeology and Ethnography at the University of São Paulo and to Mike Jochim.

1 Department of Anthropology, University of California Santa Barbara

2 Department of Biology, Universidade de São Paulo, Brazil


If you would like more information about the ongoing research, please contact the author.

Myrtle Shock: myrtleshock gmail.com



Return to Myrtle's Web Page

Return to Shock Family Web Page