Eleusine indica (L.)
Goosegrass, Yardgrass,
Wiregrass, Silver Crabgrass,
Crowfoot Grass
Poaceae

       Eleusine indica is a cespitose annual grass introduced from the Old World (Radford et al.1964; Steyermark 1963). Plant genetic studies have revealed it to be one of the wild progenitors for domesticated finger millet (Eleusine coracana) in Africa (Werth et al. 1994). Reported in several New World archaeobotanical assemblages, E. indica is an intrusive non-native.

Description
       Because E. indica is an intrusive contaminant in components that antedate European contact, it is likely to occur in an uncarbonized condition in archaeobotanical samples. Uncarbonized caryopses of E. indica are red-black to black in color and ovate in outline. Caryopsis length ranges from 1.0-1.5 mm. A sample of caryopses from a single plant had an average length of 1.33 mm, and fairly uniform widths of 0.9 mm (Delorit 1979).
       Although the seed component of E. indica usually is found as a caryopsis or grain (Delorit 1979), remains may be preserved in one of various states of disarticulation depicted in Figure 1. It is easy to separate the glumes from the floret, and fairly easy to remove the lemma. Disarticulation of the palea is slightly more difficult; recovery of the caryopsis with the palea still attached is quite possible. After disarticulation of the lemma and palea, a coating may be visible on the caryopsis. On extremely fresh specimens that are harvested slightly green, it may appear under the light microscope as a fuzzy greenish tint, and is often more pronounced toward the top of the seed, away from the hilum. This coating follows the topographic undulations of the seed below, but blurs them somewhat. On less green seeds, the coat takes on a more papery aspect. Areas of weakened adherence or absence of the coat may be evident. The description in Gray’s Manual of Botany (Fernald 1970) suggests that this coating is the pericarp:

“… lemmas broader, with a thickened 5-ribbed keel; palea shorter, acute, the narrowly winged keel distant; grain black, the loose pericarp marked with comb-like lines, free within the subrigid lemma and palea” (Fernald 1970, p. 183).

Although the coat is not visible in Figure 1, it is evident in all of the SEM views of the caryopsis (Figures 2, 3, and 4). Fernald’s (1970) description of the pericarp corresponds to this structure in most respects. Unlike his description of the pericarp as loose, however, this coating can be quite difficult to remove. SEM micrographs (50X) show that attempts to remove it from the seed were not entirely successful. It clings to the lower extremities of the seed in Figure 2, covers the entire seed in Figure 3, and is only slightly damaged in Figure 4. The tenaceious adherance of the coat is probably due to the green condition of the plant at the time of collection, but the correct designation for this coat is not yet certain.
       In transverse cross-section, caryopses are Y-shaped or heart-shaped. The ventral side contains two lobes separated by a longitudinal trough roughly ¼ the width of the seed (Figure 2; right side; Figure 5: caryposis on extreme right). The dorsal side has a marked longitudinal ridge extending up from the embryo area and narrowing toward the top of the caryopsis. The embryo is not clear on any of the figures due to insufficient magnification (Figure 5) and adherence of the seed coat (Figure 4).
       Textural details on the surface of the caryopsis are quite distinctive. About a dozen fine, wavy ridges extend up from the embryo area and out onto the lateral face. From there, they reach to the ventral face and descend into its longitudinal groove. These wavy ridges are visible with the seed coat intact and under the light microscope. Finer features include the delicately corrugated surface between the ridges. The corrugations are roughly longitudinal in orientation. They may or may not be visible depending on the magnification and the condition of the seed coat.
       All of the above description concerns uncharred speciemns, and intrusive material is likely to be recovered in this state. It is possible, however, that a recent fire could carbonize contaminants as well as archaeological material. Charring experiments do not indicate detectable shrinkage. Charred florets may be difficult to identify, however, because attempts to remove the lema and the palea usually cause the entire floret to disintegrate.

Archaeological Distribution
       Eleusine indica has been reported at several archaeological sites throughout the Eastern Woodlands. Asch et al. (1972) record the presence of six E. indica seeds in Middle Archaic contexts at the Koster site in the Illinois river valley. At Koster, the uncharred state of these and other seeds facilitated the differentiation of intrusive and primary materials during analysis. E. indica has also been found amongst material from the Late Woodland Mason Phase at the Parks site near the Elk and Duck river valleys in Tennessee (McMahan 1983). At the Carrier Mills site in the Saline valley in Illinois, carbonized specimens of E. indica have been recovered from two separate, shallow components of the site; most material from these contexts was attributed to the Middle Woodland (Lopinot 1982). E. indica has also been identified in archaeobotanical samples from Middle Woodland contexts at the Old Monroe site, a Middle/Late Woodland site in Missouri northwest of Saint Louis (Pulliam 1987). In all but the last case, the status of E. indica as a probable contaminant was not noted or discussed by the authors.

The Modern Plant and its Distribution
       The genus Eleusine has nine species. E. tristachya is the only member of the genus that is native to the New World. It grows in South America and north to the southwestern United States (Werth et al. 1995), but not in Eastern North America. Any Eleusine in pre-contact Eastern Woodland archeobotanical assemblages is therefore both intrusive and non-native.
       Ususally growing to a height of 1 m or less, Eleusine indica is a tufted annual easily recognized for its narrow, dense spikes, about four of which emanate from the terminus of each culm (Steyermark 1963). Its seeds are available in late summer and early fall (Steyermark 1963). In recent times, E. indica has extended all over much of North America, ranging from Québec and New Hampshire west to South Dakot, and south to Oklahoma and Florida (Steyermark 1963). It also grows in several western states (Hitchcock and Chase 1971). Now a common weed in both the Old World and the New World, E. indica grows in fields, open ground, waste places, lawns, gardens, and along roadsides and railroad tracks (Salimanth et al. 1995; Hitchcock and Chase 1971; Radford et al. 1964).

Discussion
       Eleusine indica is probably originally native to Africa. It is now considered a Pantropical weed (Salimanth et al. 1995). Genetic analyses have confirmed that it and another, as yet unidentifies, diploid species of Eleusine were the progenitors of the tetraploid Eleusine coracana (L.) Gaertn ssp. africana (Hilu 1988; Werth et al. 1994). E. coracana ssp. africana is the wild ancestor of modern finger millet, E. coracana ssp. coracana (Hilu 1995; Salimanth et al.1995). Finger millet has long been an important crop throughout much of sub-Saharan and eastern highland Africa (Portères 1976), where it is used primarily for beer (Harlan 1993).
       Non-domesticated Eleusine continues to be used in various parts of the Old World. E. indica is harvested regularly in the vicinity of Lake Chad, and in some areas of South Africa (Portères 1976). Steyermark (1963) mentions the collection and grinding of E. indica by Arabs and nomadic peoples of North Africa, to produce flour and gruel during times of food shortage. Various species of wild Eleusine are gathered and eaten in India (Portères 1976).
       The significance of E. indica for Eastern Woodland paleoethnobotanical research is its status as a contaminant, possibly indicating the intrusion of plant material from non-archaeological sources into prehistoric components. Because of the small width and thickness of its caryopsis, E. indica may be missed in flotation using 0.7 mm or larger mesh sizes, but it should be recovered in 0.5 mm mesh. Once recovered, this intrusive grass species is readily identifiable according to its distinctive shape and textural details visible at high magnification.

References
Asch, N. B., R. I. Ford, and D. L. Asch
    1972   Paleoethnobotany of the Koster Site: The Archaic Horizons. Illinois State Museum Reports
            of Investigations 24. Springfield.
Delorit, R. J.
    1979   An Illustrated Taxonomy Manual of Weed Seeds. Agronomy Publications, River Falls,
            Wisconsin.
Fernald, M. L.
    1970   Gray’s Manual of Botany. Illinois Statue Museum Scientific Papers, Vol. 20, Illinois State
            Museum, Springfield.
Harlan, J. R.
    1993   The tropical African cereals. In The Archaeology of Africa: Food, Metals, and Towns,
            edited by T. Shaw, et al. Routledge Press, London, pp 53-60.
Hilu, K. W.
    1995   Evolution of Finger Millet: Evidence from Random Amplified Polymorphic DNA. Genome
            38:232-238.
    1988   Identification of the “A” Donor of Finger Millet Using Chloroplast DNA. Genetics 118:163-
            167.
Hitchcock, A. S., and A. Chase
    1971   Manual of Grasses of the United States. United States Department of Agriculture,
            Miscellaneous Publications 200.
Lopinot, N. H.
    1982   Plant Remains and Paleoethnobotanical Implications. In The Carrier Mills Archaeological
            Project: Human Adaptation in the Saline Valley, Illinois, edited by R. W. Jeffries, and B. M.
            Butler. Center for Archaeological Investigations, Southern Illinois University, Carbondale.
MacMahan, J. D.
    1983   Paleoethnobotany of the Late Woodland Mason Phase in the Elk and Duck River Valleys,
            Tennessee. Unpublished M.A. thesis, University of Tennessee, Knoxville.
Pulliam, C. B.
    1987   Middle and Late Woodland Mason Phase in the Western Margin of the Mississippi River
            Valley. In Emergent Horticultural Economies of the Eastern Woodlands, edited by W. F.
            Keegan. Center for Archaeological Investigations, Carbondale.
Portères, Roland
    1976   African Cereals. In The Origins of African Plant Domestication, edited by J. R. Harlan, J. M.
            J. DeWet, and A. B. L. Stemler. Mouton, The Hague, pp. 409-452.
Radford, A. E., H. E. Ahles, and C. R. Bell
    1964   Manual of the Vascular Flora of the Carolinas. University of North Carolina Press, Chapel
            Hill.
Salimanth, S. S., A. C. de Oliviera, I. P. Goden, and J. L. Bennetzen
    1995   Assessment of Genome Origins and Genetic Diversity in the Genus Eleusine with DNA
            Markers. Genome 38(4):757-763.
Steyermark, J. A.
    1963   Flora of Missouri. Iowa State University Press, Ames.
Werth, C. R., K. W. Hilu, and C. L. Langner
    1994   Isozymes of Eleusine (Graminae) and the Origin of Finger Millet. American Journal of
            Botany 81(9):1196-1197.

Written by: Elisabeth Hildebrand

Figure 1. Florets of Eleusine indica, in various states of disarticulation.
Figure 1. Florets of Eleusine indica, in various states of disarticulation.

Figure 2. Micrograph of ventral aspect of E. indica (toward the left side); 50X.
Figure 2. Micrograph of ventral aspect of E. indica (toward the left side); 50X.

Figure 3. Lateral view of E. indica caryopsis; 50X.
Figure 3. Lateral view of E. indica caryopsis; 50X.

Figure 4. Dorsal view of E. indica caryopsis; 50X.
Figure 4. Dorsal view of E. indica caryopsis; 50X.

Figure 5. Dorsal, lateral, and ventral views of Eleusine indica.
Figure 5. Dorsal, lateral, and ventral views of Eleusine indica.

Figure 6. Line drawing of Eleusine indica (after Steyermark 1963).
Figure 6. Line drawing of Eleusine indica (after Steyermark 1963).