Printer Friendly

Prehistoric shellfish-harvesting strategies: implications from the growth patterns of soft-shell clams (Mya arenaria).

Shellfish-gathering is the stuff of many a hunter-gatherer economy. It is technically hunting -- the beasties are animals -- but they conveniently sit in the mud and on the rocks ready to be gathered. A new means of studying growth-rings in clam shells gives insight into shellfish-gathering and the seasonal pattern of life-ways in southern New England.

This paper presents an archaeological application of a new method to study the growth patterns of Mya arenaria, the soft-shell clam common along North Atlantic and North Pacific coasts, by cross-sectioning the chondrophore, a small appendage on the hinge region of the left valve that preserves well in archaeological contexts. It complements recent advances in sectioning whole valves of another Atlantic bivalve, the hard-shell clam (Mercenaria mercenaria) (Bernstein 1987; Cerrato 1987; Claassen 1986; Hancock 1984; Kennish 1980; Kerber 1985; McManamon 1984a; 1984b; Quitmyer et al. 1985). Sectioning can now be used to compare and contrast harvesting patterns, age profiles, and growth rates of Mercenaria mercenaria and Mya arenaria that are common in shell deposits along the Atlantic coast of North America.

Clam-harvesting strategies

Atlantic clams represent an expedient, low-risk and relatively predictable food source during most of the year. People of all ages can readily harvest Mercenaria mercenaria and Mya arenaria with minimal tools and facilities. Clam collecting may neatly 'fit' into a diverse range of coastal subsistence-settlement systems, with harvests being scheduled around other environmental, economic and sociopolitical considerations. The relative ease and flexibility of clam collecting has interesting implications for the organization and composition of shellfish gathering parties who may use encounter and targeting strategies.

In the encounter strategy, frequent or even daily collecting forays are made by a segment of the group living in coastal residential bases near productive estuaries. Estuary resources may provide accessible food to 'range-limited' gatherers kept near camp by illness, age, gender, adverse weather conditions, or other reasons (Meehan 1982; 1983; Jones 1991: 435). Specifically, brackish-water shellfish, such as clams, can provide a ready source of fresh meat to young children playing around camp, to mothers nursing infants, to elderly and injured people whose movements are limited, and to everyone when bad weather keeps people close to home.

In the targeting strategy, organized work parties from residential bases harvest specific species of molluscs in great volume. Among contemporary Australian shellfish-gatherers, Meehan (1983: 4-10) notes that organized work parties often 'target' a particular mollusc species found in a specific estuarine habitat for exploitation, with a 'carefully planned strategy'. Estuarine locations are scouted to monitor specific mollusc beds, work parties are recruited and organized, and the movements of people and food back and forth are arranged.

Shelter Island shell deposits

We recently excavated two shell-bearing deposits, the Sungic Midden and Laspia sites, on the Mashomack Preserve, Shelter Island, New York that may have been produced by prehistoric shellfish-gatherers using encounter and targeting strategies (Lightfoot et al. 1987).

Sungic Midden site

Situated along a former estuary (the Sungic Pond), the site consists of a discrete shell-bearing deposit (280 sq. m) surrounded on three sides by an extensive scatter of lithics and ceramics (520 sq. m). A total of seven 1 x 1-m squares were excavated. Three in the shell-bearing deposit yielded marine molluscs, mammal, fish and reptile remains; charred and uncharred seeds and nuts from flotation; and various classes of artefact. Two uncalibrated radiocarbon determinations (1200|+ or -~20 b.p. (Beta-19911), 1240|+ or -~100 b.p. (Beta-19909)) and the presence of Levanna projectile points, scallop-brushed and cordmarked pottery indicate an occupation during the late Middle and Late Woodland periods in coastal New York. From the stratigraphic profiles, we believe the site was occupied at least three times (Lightfoot et al. 1987: 102-18).

We interpret Sungic Midden as a coastal residential base where a cluster of households conducted domestic, subsistence, and ceremonial activities for much of the year. We suspect that 'range-limited' foragers may have harvested large numbers of molluscs from the nearby Sungic Pond; the molluscs were brought back, processed, cooked and shared among the group in residence, and the shell refuse discarded. 'Range-limited' foragers are expected to produce mollusc assemblages that contain a diverse range of shellfish species collected throughout the occupation of the coastal residential base. Meehan's (1983: 10-11) ethnographic study of contemporary shellfish-gatherers in Australia indicates that children and elderly people tend to be eclectic in the kinds of mollusc species harvested near camp, their selection determined largely by the species encountered. We also expect that frequent collecting at easily accessible mollusc beds should yield evidence of an intensively harvested population. Younger and smaller specimens should be found in the archaeological assemblage.

Our initial analysis of Sungic Midden provided some support for the encounter strategy. The shell deposits are a mix of mollusc species that vary greatly in frequency (as measured by shell weight) by unit and depth (Lightfoot et al. 1987: table 20). Mya arenaria dominates, followed by Mercenaria mercenaria, Crassostrea virginica (American oyster), Aequipecten irradians (bay scallop), Busycon sp. (whelk), and other snails. The sectioning of 52 Mercenaria mercenaria valves suggested two trends (Cerrato 1987; Lightfoot & Cerrato 1988). First, clam harvests occurred throughout the year, with the exception of late winter. The majority were harvested in the colder seasons of autumn (45%) and early winter (24%), in contrast to the summer (19%) and spring (12%). The four seasons of harvest are represented in each of the three occupation episodes. Second, the majority (76%) were collected at a relatively young age, between the ages of 4 and 7 years old.

We previously noted that the age profile of the Mercenaria specimens from Sungic Midden may indicate an intensively harvested population (Lightfoot & Cerrato 1988: 145):

the age distribution of Sungic Midden clams resembles very closely that of a modern, intensively-harvested Mercenaria population in Great South Bay, Long Island. This heavily harvested population, as reported in Buckner (1984), is characterized by low proportions of older individuals in contrast to an unharvested population experiencing only natural mortality. Given that the life-span of Mercenaria is greater than 20 years (Walker & Tenore 1984), and assuming that natural mortality averaged about 10% per age class per year as it does in modern day Great South Bay populations (Buckner 1984), one would expect a greater percentage of 8-15-year-old clams than that exhibited in the Sungic Midden sample.

Laspia site

We interpret the Laspia site as a shellfish-processing location at the edge of the Fan Creek estuary where organized work parties harvested and processed Mya arenaria in quantity. It consists of a discrete stratum of soft-shell clams, 12.56 sq. m in area and 10-20 cm in depth, with a very light scatter of lithics and ceramics. Nine 1x1-m squares were excavated -- five in the midden proper, two on the periphery and two outside the shell deposit. 99% of the faunal remains are Mya arenaria, calculated either by weight or MNI counts. No mammal, bird, fish or reptile remains were recovered. A Late Woodland occupation is suggested by two Levanna projectile points, ceramic sherds representing six vessels and an uncalibrated radiocarbon determination (460|+ or -~80 b.p. (Beta-19913)). We observed no stratigraphic evidence for multiple occupation episodes. Charcoal, sharp-edged flakes and ceramic sherds led us to suggest that the clam meat was extracted by roasting or boiling at the site (Lightfoot et al. 1987: 85-92).

If the Laspia site was made by organized work parties, we would expect a mollusc assemblage that differs significantly from that produced by 'range-limited' gatherers. It should reflect a population with a greater proportion of older and larger specimens, because organized work parties would target distant, uninhabited estuaries that contain fully mature clam specimens. Also, the seasons of death of the clams should reflect bulk harvesting over a limited period. Even if several harvesting episodes are represented, season of death should exhibit discrete seasonal modes rather than episodic collecting throughout the year.

We would also expect that clam harvests occurred in the warmer seasons of the year, when organized work parties travelled by foot to estuaries on Shelter Island, or by canoe to prime mollusc beds in other locales in eastern Long Island. Watercraft would increase the distances over which organized work parties could range and transport shellfish back to residential bases. The late spring, summer and early fall represent the most favourable water conditions for canoe travel in eastern Long Island. Winter travel may be hindered by strong winds and very rough or even frozen waters.

These expectations remained untested at the Laspia site until the chondrophore sectioning method provided information on the season-of-death, age structure, and growth rates of harvested Mya arenaria.

The sectioning method for Mya arenaria

True to its name, the valves of the soft-shell clam are thin-walled and prone to damage and erosion, especially along the outer margin. Growth patterns, poorly preserved in the valve, are unreliable for estimating age (MacDonald & Thomas 1980; Malione 1987). As an alternative, investigators (Brousseau & Baglivo 1987; MacDonald & Thomas 1980; Malione 1987; Newell 1982) have observed annual patterns by cross-sectioning the chondrophore, a small, internal, spoon-shaped shelf projecting from the hinge region of the left valve. Prior archaeological studies have inferred seasonal harvesting practices from annual patterns in the chondrophore and hinge region of Mya arenaria (Barber 1982; Hancock 1982; Spiess & Heddon 1983), comparing the amount of the final year's growth completed at death with prior years' annual growth or predictions of the final year's growth. These studies have not accounted for seasonal variations in growth. The sectioning method below details inter-annual variation in the growth of an individual animal.

We have found that the best method for observing growth patterns in the chondrophore of Mya arenaria is to prepare thin sections ground to between 80 and 250 microns in thickness (using procedures recommended by Clark 1980). Microgrowth analysis of Mya arenaria does not require complete valves, which are rarely recovered in the Shelter Island deposits. Intact chondrophores with the hinge still attached provide all the growth record that is required for this analysis. Specimens are sectioned from the umbo to the chondrophore edge, mounted on to petrographic slides, sectioned a second time, and ground and polished by hand. Thin sections are examined under transmitted light using a compound microscope at very low magnifications (12.5-100x). Both the thickness of the section and the illumination provided by a compound microscope are critical to observing growth patterns.

For comparative purposes, samples from a modern Mya arenaria population were collected bimonthly in 1986-7 and monthly during 1988-89 at an intertidal location in Stony Brook Harbor, Long Island, New York. Sectioning revealed semidiurnal growth increments in the warmer months of the year (complete description in Cerrato et al. 1991). Criteria for estimating season-of-death involve the optical density and growth features primarily at the edge of the chondrophore. The season assigned to a specimen is not an exact chronological time, but 'season' as defined by local environmental conditions that influence development of the animal. Spring, for example, begins with the transition from irregular to regular growth increments and the production of a spawning band. It ends when the opaque, spring layer changes to a translucent, summer pattern. In modern populations, the timing of these changes is related primarily to temperature and food supply, and will vary from one year to the next. From the modern collections, individuals spawn from April to early June, and the summer pattern is initiated sometime in July.

Age is estimated by counting the number of spring regions in the chondrophore. Growth of the shell can be inferred from the chondrophore; Malione (1987) shows that an allometric relationship exists between chondrophore length and shell length or height. Chondrophore length at a particular age is taken to be the distance from the umbo to the beginning of the appropriate spring region.


The archaeological application

Representative samples of 86 and 31 chondrophores were selected from the Sungic Midden and Laspia sites, respectively. Those from Sungic Midden were from the same archaeological contexts as the sectioned Mercenaria mercenaria valves.

All of the features observed in modern specimens of Mya arenaria were seen, with little difference in the degree of preservation between sites. As the chondrophores were weathered, and some exhibited evidence of damage, features readily discernible in modern specimens were less apparent in the archaeological samples. The spawning bands in 91% of the modern specimens (ages 2-5 years old) we have examined (Cerrato et al. 1991) were found in only one-third of the archaeological samples.

Sungic Midden site

Seasons of harvest

The timing of death, estimated for 57 of the Mya arenaria specimens, indicates a very similar harvesting pattern to that determined for Mercenaria mercenaria. A log-likelihood ratio test (Sokal & Rohlf 1981) comparing the harvesting schedule of soft-shell and hard-shell clams was not significant (G=1.47 with 3 df, p |is greater than~ 0.5). The majority of Mya arenaria were collected during the fall (51%) and winter months (25%), with some collecting in the warmer months (spring (14%), summer (10%)). The four different seasons of harvest are represented in each of the three occupation episodes.

Age distribution

Mya arenaria specimens ranged from 1 to 13 years. The great majority were young, a finding that correlates closely to the age profile of the Mercenaria mercenaria from Sungic Midden. 67% were only 2 and 3 years old, while 96% were 6 years or less. Few very young (less than 2 years) or very old (greater than 6 years) clams were collected.

Growth rate

Chondrophore size at age measurements could be obtained from 69 of the Mya arenaria specimens.

Laspia site

Seasons of harvest

Collection at Laspia appears to have been a warm-season activity, the vast majority harvested in summer (64%) and spring (32%). A log-likelihood ratio test comparing the Sungic Midden and Laspia scheduling patterns was highly significant (G=44.1 with 3 df, p |is less than~ 0.001).

Age distribution

Mya arenaria ranged from 3 to 9 years. To test for differences in the ages of soft-shell clams from the Sungic Midden and Laspia sites, data were grouped into four age-classes (1-2, 3-4, 5-6 and 7 or more years old) to avoid the bias associated with low frequencies (Sokal & Rohlf 1981). The resulting log-likelihood ratio test between the two sites was highly significant (G=31.44 with 3 df, p |is less than~ 0.001). Fewer young clams were harvested at Laspia; only 11% were 2 and 3 years old, while 51% ranged between 7 and 9 years. Older, and larger, molluscs were collected, processed and then deposited at Laspia.

Growth rate

Growth results for the harvested population were very similar to the Sungic Midden clams.


The sectioning results support the interpretation that the clams in Sungic Midden were from an intensively harvested population. Both hard-shell and soft-shell clams were harvested throughout much of the year, and the clams were younger (and therefore smaller) than in an unharvested population experiencing only natural mortality. This finding is consistent with a mollusc assemblage that was harvested frequently by 'range-limited' gatherers employing an encounter strategy. The frequency of both harvested soft-shell and hard-shell clams at the Sungic Midden site increased in the colder months, the leanest and most inclement time of the year. Adverse weather may have forced the general population to become 'range-limited' gatherers, practising an encounter strategy for brief periods in the winter and early spring when fresh meat was scarce. McManamon & Bradley (1986: 35-6; 1988: 102) also note that shellfish collection at this time would not interfere with warm weather horticultural practices that may have involved women, children and other 'range-limited' workers in nearby fields.

In reality, the mollusc assemblage at Sungic Midden was probably produced by both encounter and targeting harvesting strategies. Some clams may have derived from organized work parties operating in the nearby estuary during the initial occupation; the lowest (and earliest) stratigraphic feature at the site is a pit containing large, whole Mercenaria mercenaria specimens (Lightfoot et al. 1987: 102-109). Other clams may have been transported by organized work-parties from estuaries some distance away. However, the current data, based on the sectioning of both Mercenaria mercenaria and Mya arenaria samples, suggest that many of the clams were harvested in frequent foraging forays, probably in the adjacent estuary.

In contrast, both the seasons of harvest and the age structure of Mya arenaria specimens from Laspia suggest a less intensively harvested population. More mature and larger clams were collected primarily during warm weather -- a finding consistent with organized work-parties who targeted specific mollusc species from less intensively harvested waters for bulk procurement. A warm-weather harvesting pattern is also consistent with canoe transport. While a more discrete seasonal mode of collecting is represented at the site, the presence of clams that were harvested during more than one season of their developmental cycle is more ambiguous. The Laspia site may have been deposited during one harvesting episode, but the seasonal indicators may not be discrete enough to identify a single event. Alternatively, organized work parties may have allowed enough time between harvests for the mollusc beds to regenerate. Mya arenaria are capable of repopulating themselves in large numbers in only a few years. Repeated harvests by organized work parties at intervals of 3 to 7 years would likely yield little stratigraphic evidence of distinct events in shell deposits.


In this paper we have applied a method for detailing growth patterns in Mya arenaria to archaeological investigations. Sectioning studies can now compare and contrast the harvesting patterns, demographic profiles, and growth rates of soft-shell and hard-shell clams commonly found in coastal Atlantic archaeological assemblages. We have shown the value of comparing the sectioning results of two or more mollusc species from archaeological assemblages to evaluate whether encounter or targeting strategies were employed by prehistoric coastal peoples.

Acknowledgements. We thank The Nature Conservancy (especially Michael and Susan Laspia), James Moore and Robert Kalin for their participation in the archaeological investigation of the Mashomack Preserve. We are grateful to Jonathan Salerno, Richard Muller and Mark Wiggins for their assistance in field collecting modern Mya arenaria samples. Art Spiess and Russell Barber were very helpful in discussing previous archaeological research on Mya arenaria. We appreciate greatly the constructive comments provided by Roberta Jewett, Roger Luebbers and Francis McManamon.


BARBER, R.J. 1982. The Wheeler's Site: a specialized shellfish processing station on the Merrimack River. Cambridge (MA): Harvard University, Peabody Museum. Monograph 7.

BERNSTEIN, D.J. 1987. Prehistoric subsistence at Greenwich Cove, Rhode Island. Ph.D dissertation, Department of Anthropology, State University of New York, Binghamton.

BROUSSEAU, D.J. & J.A. BAGLIVO. 1987. A comparative study of age and growth in Mya arenaria (soft-shell clam) from three populations in Long Island Sound, Journal of Shellfish Research 6: 17-24.

BUCKNER, S.C. 1984. Aspects of the population dynamics of the hard clam, Mercenaria mercenaria L., in Great South Bay, New York. Ph.D dissertation, Marine Sciences Research Center, State University of New York, Stony Brook.

CERRATO, R.M. 1987. Microgrowth line analysis of hard clams from the Sungic Midden Site (2N3E 1-0), Shelter Island, New York, in Lightfoot et al.: 175-96.

CERRATO, R.M., V.E. WALLACE & K.G. LIGHTFOOT 1991. Tidal and seasonal patterns in the chondrophore of the soft-shell clam Mya arenaria, Biological Bulletin 181: 307-11.

CLAASSEN, C. 1986. Shellfishing seasons in the prehistoric southeastern United States, American Antiquity 51: 21-37.

CLARK, G.R. 1980. Study of molluscan shell structure and growth lines using thin sections, in Rhoads & Lutz (ed.): 603-6.

HANCOCK, M. 1982. The determination of archaeological site seasonality using the remains of Mya arenaria (L.): examples from the Central Maine coast. MA thesis, Institute for Quaternary Studies, University of Maine, Orono.

1984. Analysis of shellfish remains: seasonality information, in McManamon (ed.): 121-56.

JONES, T.L. 1991. Marine-resource value and the priority of coastal settlement: a California perspective, American Antiquity 56: 419-43.

KENNISH, M.J. 1980. Shell microgrowth analysis: Mercenaria mercenaria as a type example for research in population dynamics, in Rhoads & Lutz (ed.): 255-94.

KERBER, J.E. 1985. Digging for clams: Shell midden analysis in New England, North American Archaeologist 6: 97-113.

LIGHTFOOT, K.G., R. KALIN & J. MOORE. 1987. Prehistoric hunter-gatherers of Shelter Island, New York: an archaeological study of the Mashomack Preserve. Berkeley (CA). Contribution of the University of California Archaeological Research Facility 46.

LIGHTFOOT, K.G. & R.M. CERRATO. 1988. Prehistoric shellfish exploitation in coastal New York, Journal of Field Archaeology 15: 141-9.

MACDONALD, B.A. & M.L. THOMAS. 1980. Age determination of the soft-shell clam Mya arenaria using internal growth lines, Marine Biology 58: 105-9.

MCMANAMON, F.P. (ed.). 1984a. Chapters in the archaeology of Cape Cod I 2. Boston (MA): National Park Service.

1984b. Prehistoric cultural adaptations and their evolution on Outer Cape Cod, in McManamon (ed.): 339-417.

MCMANAMON, F.P. & J.M. BRADLEY 1986. The Indian Neck ossuary and Late Woodland prehistory in southern New England, in F. McManamon et al. (ed.), The Indian Neck ossuary, chapters in the archaeology of Cape Cod V: 3-47. Boston (MA): National Park Service.

1988. The Indian Neck ossuary, Scientific American 256: 98-104.

MALIONE, B.R. 1987. Analysis of growth in Mya arenaria using internal growth lines. MS thesis, Marine Sciences Research Center, State University of New York, Stony Brook.

MEEHAN, B. 1982. Shell bed to shell midden. Canberra: Australian Institute of Aboriginal Studies.

1983. A matter of choice? Some thoughts on shell gathering strategies in northern Australia, in C. Grigson & J. Clutton-Brock (ed.), Animals and archaeology 2: Shell middens, fishes, and birds: 3-17. Oxford: British Archaeological Reports. International series S183.

NEWELL, C.R. 1982. The soft-shell clam Mya arenaria L.: growth rates, growth allometry, and annual growth line formation. MS thesis. Department of Zoology, University of Maine, Orono.

QUITMYER, I.R., H.S. HALE & D.S. JONES 1985. Paleoseasonality determination based on incremental shell growth in the hard clam, Mercenaria mercenaria, and its implications for the analysis of three Southeast Georgia coastal shell middens, Southeastern Archaeology 4: 27-40.

RHOADS, D.C. & R.A. LUTZ (ed.). 1980. Skeletal growth of aquatic organisms. New York (NY): Plenum.

SOKAL, R.R. & F.J. ROHLF 1981. Biometry. New York (NY): Freeman

SPIESS, A.E. & M.H. HEDDON. 1983. Kidder Point and Sears Island in prehistory. Augusta (MA): Maine Historic Preservation Commission. Occasional Publication in Maine Archaeology 3.

WALKER, R.L. & K.R. TENORE. 1984. The distribution and production of the hard clam, Mercenaria mercenaria, in Wassow Sound, Georgia, Estuaries 7: 19-27.
COPYRIGHT 1993 Antiquity Publications, Ltd.
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1993 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Lightfoot, Kent G.; Cerrato, Robert M.; Wallace, Heather V.E.
Date:Jun 1, 1993
Previous Article:Accelerator radiocarbon dating of plant-fibre binders in rock paintings from northeastern Australia.
Next Article:Wooden churches and their paintings in the Maramures region of Romania: a preliminary study.

Terms of use | Copyright © 2018 Farlex, Inc. | Feedback | For webmasters