by Andrew F. Malof


The study and interpretation of snail assemblages from Texas archaeological sites is intermittent in implementation and inconsistent in application.A fuller understanding of snail species and their corresponding environments should enable Texas archaeologists to more fully develop problem-driven research designs capable of answering a variety of questions by utilizing a data set that is often overlooked or under used.


Snails are a commonly encountered portion of the faunal assemblage in many archaeological sites. In somecases the density of snail remains demand attention.The result has been a mixture of approaches to understanding and interpreting land mollusks in archaeological contexts.These include questions regarding dating methods, climate and environmental reconstruction, and aboriginal subsistence.

These broad questions entail specific methodologies.Snails are relatively short-lived, so if appropriate methods are used they should be useful as a means of dating prehistoric occupations.Snails are very niche specific and so should allow conceptualizing paleoclimatic conditions.The density and ubiquity of Rabdotus species has led to a largely unquantified, and variously contested, assumption of snail gathering and consumption.Proper levels of analysis allows lines of evidence to be constructed for all of these issues. 

Unfortunately snails are often glossed over in reports, and their potential as an aid in interpreting archaeological sites has remained largely unrealized.This paper serves to summarize uses to which snail data can be applied.An a priori understanding of these uses combined with research specific questions can be of great utility while designing a date recovery project.If mixed deposits are suspected, or charcoal preservation is poor, snail recovery may help determine the integrity of the site or the age of deposits within it.If it is known that other methods of climate reconstruction will be affected by differential preservation then recovery of microscopic snails may be of tremendous help.If testing indicates large numbers of Rabdotus snails will be present then careful research design will help determine if they were a portion of the subsistence base or scavengers feeding off of human detritus.

Careful consideration of these issues during either the inductive or deductive portion of the research design should allow a valuable source of data to be used as effectively as possible.The lowly snail holds great promise for a better understanding of the archaeological record in Texas. 



There is no one single source that summarizes all snail species within Texas, or within a single region within Texas.In the 1940s Henry Pilsbry compiled four volumes of data on North American land snails (Pilsbry 1940, 1946, 1948), but the volumes are difficult to access, and contain much outdated taxonomy, and are therefore most useful for the illustrations (Ken Brown 2000: pers. comm.).Between 1971 and 1973 E. P. Cheatum and R. W. Fullington produced a series of bulletins for the Dallas Museum of Natural History describing the aquatic and land mollusca of Texas (Cheatum and Fullington 1971a, 1971b, 1973).Cheatum died before the project was completed (Fullington and Pratt 1974:ii) and Fullington , assisted by W. L. Pratt, produced one more volume in 1974 (Fullington and Pratt 1974).The series was never finished, and although very useful at times, it contains omissions that can prove frustrating.Further, it describes no aquatic species, other than in the supplemental key (Cheatum and Fullington 1971b).Perhaps a precursor to the Dallas Museum bulletins, D. C. Allen and E. P Cheatum published a short but very informative summary of land and aquatic snails in the 1960 Bulletin of the Texas Archeological Society (Allen and Cheatum 1960).Of interest was an emphasis on snails found within archaeological sites. 

It is apparent that there is no single source for comprehensive references on snails.The major sources require supplementation from scattered reports from various sources, and even these often lack information that might be of interest or relevance, both from a natural history and an archaeological perspective.Very little is known about the most basic aspects of snail ecology; reproduction, life spans, environmental preferences, etc.In fact, it is not known what the most commonly recognized snail species in archaeological assemblages, Rabdotus, eats (Brown 1999:251, although see Randolph 1973, for strong evidence of plant consumption). 

Species, Morphology, and Distributions 

By combining information from the Dallas bulletins and Allen and Cheatum’s 1960 

BTAS publication it is possible to begin to inventory Texas snail species.Aquatic snails are represented by at least 10 families containing 18 species (Cheatum and Fullington 1971b).Six of these families and 11 associated species are commonly found in archaeological contexts (Allen and Cheatum 1960).Land snails exhibit a much richer diversity.Twenty-one familes with 35 genera containing over 140 species are documented (Cheatum and Fullington 1971a, 1973, Fullington and Pratt 1974).Of these, 14 families containing 32 species are reported to be common in archaeological sites (Allen and Cheatum 1960).Another source (Hubricht 1985) indicates 127 land snail species are found in Texas.This discrepancy in numbers is probably largely due to exclusion of species found west of the Pecos River, as well as species considered non-native 

The range in species is reflected in range of phenotype.Fourteen major shell forms are described as globose, depressed, discoidal, domed, bulimoid, pupilliform, conical,and turbinate, among others.Other diagnostic features of shells include number and direction of body whorls, presence or absence of a lip at the shell aperture, or opening, presence or absence of lamella, also called teeth, also at the shell opening, and an open or closed termination of the whorls on the underside of the shell, described as either perforate or imperforate.Shell size ranges from around 30-35 mm to as little as 1-2 mm, or smaller.The animal which inhabits the shell, the gastropod, presents a whole new range of morphologies, which will not be covered here (but see Purchon 1977, for details of anatomy). 

These species are widely distributed across the state.Distributional maps in Hubricht (1985), and Cheatum and Fullington (1971, 1973) and Fullington and Pratt (1974) show that some species are widespread, indicating a broad range of environmental tolerance, while others are seen to cluster in various regions of the state, denoting narrower environmental preferences.Still others are represented in only a very few counties.

The environmental preferences suggested by regional distributions are also seen on a more local scale.These descriptions, however, are widely variable, depending, no doubt, on available literature and observations of the various authors.The species Holospira goldfussi, for instance, is distributed largely across Central Texas, but its more specific habitat is described merely as “associated with limestone rock and humus” Cheatum and Fullington (1973:37-38). Descriptions for other species’ habitat preferences are more useful, illustrating such variables as moisture preferences and zonation in grass or woodlands.

Other information is also inconsistent.Some comments help clarify distributions and morphology, while others indicate that very little is known about a particular species.More species specific data requires searching through literature sources from the natural sciences or specialized journals (e.g. Hubricht 1982, Randolph 1973).In general, the quality and quantity of available knowledge is highly variable. 



Archaeological investigators have been including snails within their reports for many years, but there has been little constancy in level of detail or analysis.Some reports mention snails largely in passing, under sections such as “Non-Vertebrate Faunal Remains”, although others attempt, with various degrees of success, to incorporate snails in some aspect of site interpretation.A sample of some of the more involved studies are summarized briefly below.

Raymond Neck’s(1994) analysis of non-vertebrate remains at Mustang Branch in Hays County was a succinct summary of species collected and implications of species diversity, with an emphasis on Rabdotus.Change in species ratios at this site were inconclusive in reconstructing paleoenvironments.The discussion of holes found in Rabdotus shells, (from roots and screen damage), and habits of the snail were especially useful. 

Henry (1995) compares seven sites in the Hog Creek basin of North Central Texas and compared their frequencies with those of pollen and vertebrate fauna to argue for an increasing woodland replacing grasslands.Species of Oligyra and Rabdotus varied inversely, with the woodland adapted Oligyra replacing Rabdotus over time.

An exception to the general disregard of snails by some authors is LeRoy Johnson, who makes an effort to explain, or at least describe, the presence of snails in sites he has reported on.In 1962 he stated that snails and shellfish were likely gathered by women as a regular part of the subsistence regime (Johnson et al. 1962:47).In a series of reports produced for the Texas Department of Transportation (e.g. 1991, 1995, 1997) snails and molluscan remains received generous attention as a means of environmental reconstruction (1991), patterns of site use (1995), and general descriptive data (Neck 1997). 

Ken Brown (1999) has recently produced what is probably the definitive work on snails in regard to methods of environmental reconstruction and subsistence questions.Based on data from the Smith Creek Bridge Site in DeWitt County, he has produced an analysis that summarizes the current state of knowledge of snail use in paleoenvironmental studies and subsistence models.This report will be a valuable asset to anyone interested in these aspects of snail studies. 

Ellis et al. (1996) and Lintz and Abbot (1997), meanwhile, have been active in using snails to obtain dates for archaeological deposits using amino acid racemization and carbon dating techniques.These methods are proving useful for absolute dating of archaeological remains, and for helping to determine integrity of buried deposits. 

Many other examples could be cited.What becomes clear though, in this very brief literature review, is that the use of snails to understand archaeological sites in Texas has been building for a number of years.Early studies recognized the potential for snail analysis, and later works have been developing a framework from which future studies can proceed.The following section examine three areas of snail studies:use of snails for dating of archaeological deposits, for reconstructing past environments, and for understanding prehistoric subsistence patterns. 


Dating:The dating of snails has been refined to the point where there is a real possibility of gathering useful information from snail shells in archaeological contexts.It was recognized quite early that snail shells might be useful for providing dates for archaeological assemblages (Allen and Cheatum 1960).Snails precipitate carbonates that are ingested from local limestone or soils, and the shell therefore dates much older than the snail itself (Goodfriend 1992:665).It is necessary to test a modern, preferably pre-atomic bomb sample to enable use of a corrective factor to adjust for an up to 3000 year discrepancy in carbon dates (Ellis et al. 1996:192).This age anomaly can be corrected for on a regional basis (Lintz and Abbot 1997:15), and by extension, a database could conceivably be initiated that would allow for local correlations.

Amino acid racemization and epimerization is perhaps more directly applicable to snail studies.The difference between the two methods is not immediately clear, and may be minor, as both are based on the fact that an L-form of a particular amino acid converts (racemizes [Ellis et al. 1996:192] or epimerizes [Ellis and Goodfriend 1994:184-185]) to a D-form upon death of an organism.Once the rate is determined it can be calibrated with other absolute dating methods resulting in a quick, cost-efficient dating method (Ellis and Goodfriend 1994:185).

These methods may be utilized for directly dating archaeological deposits.They can also be used for determining site integrity.Snails and other carbon-bearing materials can be extracted from excavation levels and used to determine potential for mixed deposits.If the dates are not comparable, the site is likely mixed; if the sample indicates clustering of ages, the deposits are likely intact.At 41ZP39 and 41ZP176 snail and charcoal samples were dated by AMS methods, and by statistical comparison it was determined that the site maintained enough vertical integrity to warrant further testing (Lintz and Abbot 1997).Racemization rates of snails were used, somewhat more rigorously, by Ellis et al. (1996) indicating that a number of sites at Fort Hood contained mixed deposits. 

Paleoenvironments:Recreation of paleoenvironments is another direction snail research has followed.Snails are capable of informing about past environments in much the same way as other fauna, and it is the smaller, microscopic snails that are more subject to selective pressure, and so are more reliable indicators of surrounding environmental variables (Brown 1999:213).There are two methods of gaining information on past environments from snails.One is through survey level data using local and regional collections, and the other is through experimental studies (Goodfriend 1992:666).Survey data has been used for determining such variables as biome abundance or frequency, elevation, rainfall amounts and frequency, temperature, and moisture (Goodfriend 1992: Tables 1 and 3).An assumption that environmental requirements do not change is necessary to support a good fit with known ecological parameters of modern populations (Allen and Cheatum 1960:292).This assumption of evolutionary stasis is expressed through relative scarcity or abundance of snail species based on fluctuations in favorable climates, ignoring the implications of selection embedded in this comparative process (Goodfriend 1992:669).

An equally complex issue is determining relationships between different aspects of the environment.Variables that interact to produce today’s climatic regimes may not have been in place in the early Holocene, when the factors controlling the Earth’s orbit were different enough to produce increased seasonality in the northern hemisphere relative to present conditions (Goodfriend 1992:669).Careful measurement of snails from datable contexts may help resolve some of these issues, as shell size is dependent partially on available moisture and temperature, as well as population density (Brown 1999:217).Comparative studies utilizing other lines of evidence may allow environmental factors to be teased apart from micro-evolutionary changes. 

Another means of recreating past climates is through experimental research on present populations (Goodfriend 1992:666).Controlling primary factors of light, temperature and moisture can provide information on mortality rates, reproduction, feeding habits, and mobility (Goodfriend 1992:667).Randolph (1972) conducted experiments on two species native to Central Texas, Mesodon roemeri and Bulimulus dealbatus (Rabdotus).Randolph was able to show that a more fluctuating environment results in more generalized species behavior, and conversely, less species diversity (Randolph 1972:934).Using a standard biological diversity index (Shannon-Weaver), it became apparent that a less diverse environment (woods), contained a richer diversity of snail species than did the more diverse grasslands, where snail diversity dropped (Randolph 1972:936-937).The seeming discrepancy of grasslands being described as more environmentally diverse than woods is due to a greater range in temperature and relative humidity in the grasslands (Randolph 1972:936) and may also be the result of a number of micro-habitats (downed wood, small bushes, etc) within the broader grass zone (K. Brown 2000: pers. com.).

Subsistence:A long standing issue revolves around the question of whether snails formed a portion of the subsistence base or whether their presence is the result of enriched human midden deposits attracting existing populations.Although it is generally accepted that Rabdotus species were a food source for prehistoric populations (e.g. Simmons 1956, Allen and Cheatum 1960, Jelks 1960, Johnson 1964), others maintain that even the extremely dense populations seen at some archaeological sites are the result of “commensal scavengers” (Brown 1999:243).A method that may help resolve this issue,the use of off-site control samples, is little utilized (Brown 199:244).If on-site snail deposits are richer than those found off-site, there will be strong evidence for a correlation between human and snail populations, although what kind of relationship will still require clarification.

Brown (1999) argues that an active campsite is a very poor environment for active Rabdotus populations; foot traffic and soil compaction resulting in lessened soil moisture would not support viable populations.It may be they were attracted to peripheral garbage middens, but as it is unknown what these snails eat, that contention cannot be supported.Not until the site has been abandoned for a long enough period to revegetate will it be favorable for recolonization.Therefore, in a well stratified site, snails should be found in zones free of cultural material, if they are indeed recolonizing previous human occupation zones.It remains possible that enriched soils would support rich plant growth that potentially might reach a level supporting snails quicker than surrounding areas.This factor, combined with mixing of deposits could account for the dense concentrations of snails seen in many sites (Brown 1999:243-248).A possible objection to this line of thought, if subsistence is being argued for, is that other snail species should follow similar patterns can be countered by a lack of snail concentrations by species other than Rabdotus (Brown 1999:249).

Ethnographic accounts indicate that snails were used as a food resource.Apparently Cabeza de Vaca witnessed consumption of snails by Native Americans (Clark 1969:43, 1976, Hester and Hill 1975).North African groups, both recently and prehistorically, conduct activities that result in large piles of burned rock, called rammadyat, that are quite similar to the burned rock middens of Central Texas (Honea 1962:317), and one of the primary purposes seems to be the preparation of snails and other food resources (Honea 1962:318).Modern Maya collect snails for subsistence purposes and evidence strongly suggests prehistoric populations did as well (Healy et al. 1990).

Archaeologically snail clusters accompanied by cultural material is good evidence for snail consumption, and Brown lists a number such cases, some where snails numbered in the thousands (Brown 1999:248-249).Looking at ratios of adults to juveniles may also be of use.If adults are present in numbers exceeding those found in general populations it is reasonable to suggest that they were being selectively gathered (Neck 1994:496).Clark calls for an analysis of coprolites (Clark 1973) and although snail shell fragments have been found in some samples, diagnostic body portions such as radulae, part of the feeding mechanism (Clark 1969) have not been recovered (Clark 1973, Brown 1999:250), which means that the recovered shell may have been introduced accidentally into the diet. 

Although it appears that snails (Rabdotus sp.) could have been, and probably were consumed on a fairly regular basis, another line of thought is to ask if they should have been eaten; i.e., following on optimal foraging theory (Kelly 1995) can it be shown that snail consumption would be a rational and inevitable subsistence choice. 

Optimal foraging theory includes the concepts of diet-breadth and patch choice (Kelly 1995:97).Diet-breadth predicts whether a resource will be taken during a foraging excursion and is based on total acquisition costs (search and handling) compared to return rates (Kelly 1995:79).The result is ranked resources from which choices of acquisition can be made.Higher ranked resources will be utilized as primary food sources and diet-breadth will be low.As the availability of higher ranked resources decreases acquisition costs rise, and to maintain efficiency in return rates lower ranked resources become incorporated,and diet-breadth increases (Kelly 1995:84).

Incorporating resource “patches” into the model more accurately depicts distributions of resources across any particular section of terrain.Searching a resource patch becomes part of acquisition costs, and a patch remains productive as long as it can provide a return rate higher than one that would encountered at another patch, once the costs of moving to that patch are factored in (Kelly 1995:91).

Combining these two models allows prediction of which area will be chosen and what will be the focus of the acquisition activity (Kelly 1995:97).Optimal foraging theory thus becomes a method from which predictive or explanatory models from a wide range of situations can be tested (Smith and Winterhalder 1992:51).Optimization can also be placed within a broader perspective of selection-based archaeology, which examines how potential reproductive fitness determines human choice-making, and, ideally, why some choices are made over others (O’Brien and Holland 1995:194).

Differing return rates are dependent on several factors.For instance, in the Ache diet, the paca has a higher number of kilocalories per unit than does deer, but also has a much longer handling time (encounter plus acquisition), and so has a substantially lower return rate, and correspondingly is ranked lower than deer (Kelly 1995:Table 3-5, 85).If only acquisition and processing costs are calculated food sources can be ranked in a more general sense (Kelly 1995:79-80, Table 3-3).It follows that if processing costs are unavailable, then indices of nutrition, such as kilocalories, protein, fats and carbohydrates must be used as a baseline to which experimental or ethnographic data can be added. 

The nutritional value of Rabdotus species is not well known, but they seem to be high in protein and low in carbohydrates and fats (Brown 1999:250).Brown presents figures of generic cultured snails as containing approximately 80 kcal, 16 grams of protein, 2 grams of carbohydrates, and 1 gram of fat per 100 grams of snails.This compares well with snails known ethnographically to be consumed by lowland Maya populations (Healy et al. 1990), although the carbohydrate values of the aquatic snail they consume is considerable higher, at 12 grams per 100 grams of meat.Kilocalories are comparable to rabbit (73 kcal/100gm) but considerably less than deer (126 kcal/100gm), and both animals contain higher amounts of protein and fats, but no carbohydrates (Healy et al. 1990:Table 2, 178).The post encounter return rate for deer in the Great Basin averages to 24,700 kcal/hour, while that of jackrabbit averages to 14,400 kcal/hour, but that of snails is not given (Kelly 1995:Table 3-3).Based simply on return rates deer would be ranked higher than rabbit.Because return rates for snails are not known it is necessary to use a simple comparison based on total kilocalories.Snails would be ranked higher than rabbit, but lower than deer.

Diet breadth has now been addressed, but the question of patch choice remains to be clarified.Rabdotus is colonial in its habits, maintaining localized populations that become active during periods of high humidity (Fullington and Pratt 1974:14).The species escapes high temperatures by climbing above super heated ground and estivating on upright vegetation(Randolph 1972:934) or seeking shelter underground (Hubricht 1960:69) or under rocks and logs (Fullington and Pratt 1974:14).Colonies may be as large as a city block (Hubricht 1960:69).Densities of up to 18 individuals per m2 could therefore be found in prickly pear (tuna) patches along with rabbits and woodrats (Brown 1999:250), and presumably in other locations as well.

Rabdotus, then, is a prime candidate for testing with an optimal foraging model.Its colonial, or “patchy” nature combined with periods of high availability, and thus low acquisition costs, would tend to boost its ranking relative to otherwise higher ranked resources having higher associated costs.Under the right conditions, a dense patch of snails could well become highly ranked within a diet breadth widened by even a relatively minor fluctuation in larger game animals.Food energy is maximized through proper heat treatment (Wandsnider 1997:10), so once Rabdotus species are quantified as to nutritional value, and experimental gathering and processing activities are conducted, a clearer picture of when Rabdotus should enter a particular subsistence strategy should emerge. 


Site 41BL116 is located in southern Bell County on a high interfluvial terrace lying between the Lampasas River and a small intermittent tributary.The site has been looted for decades, which has resulted in large pits exposing cultural material to a depth of 2 meters.Physical stratigraphy is largely undifferentiated within the cultural zone, and exhibits a sharp disjunction below it, where gray silty clays are replaced by tan silty sand which appears to completely devoid of cultural material.Profiling of the “looter pit” has resulted in recovery of Pedernales projectile points, among others.These points, indicative of the early half of the Late Archaic Period (Collins 1995: Table 2), appear to be the earliest evidence of in situ occupations.

Test excavations were initiated in 1995 by John Voltin of the University of Texas at Austin.These consisted of a number of 1 x 1 meter squares placed selectively over the ca. 100 x 100 meter site.Test Unit 8 was excavated by Michael Williams of the University of Texas at San Antonio.It reached a depth of 50 cm before it was decided to place another unit along side it to more fully expose a possible rock feature noted at about 30 cm below the surface.This apparent feature resulted in initiation of two additional units resulting in a 2 x 2 m block.

Additional labor became available and it was decided to open another two units to determine if an intact surface could be uncovered.These excavations indicated a fairly distinct patterning of cultural material, with Late Prehistoric ceramics and arrow points being followed by Late Archaic dart points in a sequence relatively consistent, most notably with large numbers of Darl and Fairland type points, which seem to overlay a fairly discrete Montell component.Once the 2 x 3 m block was excavated to the level of possible feature it became apparent that it was actually a fortuitous arrangement of fire-fractured limestone, but by this time, recovery of large amounts of snail and mussel shell, combined with a relative lack of ground stone implements, caused a partial redirection in research design.Snail, which had previously been noted and discarded, was collected at a 100% level.In addition, another 1x1 m unit has been added to the block, and designated the “snail unit”.This unit is being screened through nested 1/4 in and 1/8 in nested hardware cloth, and 1 gal (2-3 kg) matrix samples are being bagged for future fine sieving and possible additional analyses.

Very little analysis has been conducted on any of the materials excavated from the block, and so any conclusions are at this point highly subjective.Small fragments of bone are common throughout the units, and many exhibit burning, and some show evidence of cutting.Formal processing tools, other than projectile points, are rare.Most lithic tools are flakes showing evidence of retouch or use.Burned rock is common, but seems to fluctuate in density, with some levels containing dense clusters and scatters, while others contain little.Flake counts are high, with 2-300 per 5 cm level being common.It is presently unknown if flake counts will vary with depth.

Mussel shell is also very common.It is found scattered through out all levels that contain burned rock.No features have been recognized, but it is often found nested, in small groups of 2-4 individuals.Snail counts are mostly subjective from earlier levels and units.The 100% collection strategy recently implemented has resulted in snail counts well over 300 in many levels.The majority of these seem to be Rabdotus species.The snail unit is of interest in that the upper levels, which produced ceramics and distal arrow tips, contain relatively few snails, and few mussels as well.This unit is now at the point where Archaic materials are expected, although only one thin biface, unidentifiable as a projectile point, has been recovered.The last two levels excavated showed a distinct increase in molluscan remains.This allows postulation that later populations at the site had a lesser reliance on this type of resource. 

Eleven species of snail have been recognized so far (Table 1).The majority of snails are represented as preferring wooded, near-stream environments.This describes fairly well the present-day setting.One of the exceptions, Pupoides albilabris, is “microscopic”, and so far is represented by only one specimen.Rabdotus and Helicina species are both found in large numbers within some contexts.Additional quantification of all snails should provide useful information for interpreting the general environment through out the Late Archaic and Late Prehistoric periods at this site. 


Texas Distribution
Angiuspira strongylodes
Central and East
Widely tolerant
Gastrocopta pentodon
Scattered statewide
Dry upland woods, occasionally in moist low areas
Helicina orbiculata
Along and east of escarpment
Semi-arboreal, prefers sun
Mesodon roemeri
Central and North-Central
Wooded slopes near streams
Mesomphix friables
Central and East
Floodplains, river bluffs
Planorbidae sp
Polygyra texasiana
All but plains and West
Woodland or prairie
Praticollela berlandieriana
Woods near streams, and open areas
Pupoides albilabris
Bare ground
Rabdotus sp 
South Central and West
Semi arboreal, grass and shrubs
Retinella umbilicata
Central and East
Woods near streams

Table 1.List of snails recovered to date at 41BL116.Identification of land snails is confident to genus level, some species designations are based on distributions rather than morphological characteristics.Land snail distributions and habitat data are from Hubricht 1985.The single aquatic snail is recognized only by Family at this time.


The analysis of snails at 41BL116 holds great promise for determining subsistence strategies and recreating of past environments.Once snail species are identified and quantified it is probable that information on environmental zones will emerge.This will allow moisture and temperature regimes to be hypothesized, which additional lines of evidence (faunal, palynological), if available, can be used to support. 

Large numbers of Rabdotus snails are the result of either intentional human gathering or innate natural processes of the species.Rabdotus can be segregated from other species being utilized for dating or environmental reconstruction;it is the smaller snails that are more relevant for these purposes, which allows subsistence questions to be pursued separately from other issues.

At 41BL116 Rabdotus are found in direct association with mussel shells and concentrations of burned rock, as well as lithic and bone material.As yet it is too early to determine if these snails are a part of the subsistence base or are naturally occurring.Comparison of adult to juvenile ratios within natural populations may present the strongest evidence for either hypothesis.

The close association of snails and mussel bears closer examination.Aquatic resources in the coastal plains were a low ranked resource that became more prevalent when the general resource base was stressed, although fish, at least, appear to have been a decisive factor in Late Archaic settlement (Hall 1998:4,5).Natural mussel distributions are non-random, and they can occur within beds (Howells et al. 1996:21).It seems likely that the association of the two molluscan species and their similarities in habits may be more than coincidental.The people at 41BL116 may well have been exploiting similarly structured resources available at a single location, perhaps within a seasonally dependent context.Additional research and correlations of species and artifact counts should help determine the answers to these and similar questions.

Snail analysis provides an avenue of data that is often overlooked.With careful research design information on past environments, subsistence strategies, relative and absolute ages of occupations, and site integrity can be gathered and consolidated.If such methods become commonplace, potential exists for substantial additions to the regional archaeological database.


Allen, D. C. and E. P Cheatum 

1960Ecological Implications of Fresh-Water and Land Gastropods in Texas Archeological Studies.Bulletin of the Texas Archeological Society 31:291-316 

Brown, Kenneth M. 

1999Snails From the Quarter-Inch and Eighth-Inch Screens.Appendix F (pp. 213-275) in Dale Hudler, Keith Prilliman, and Thomas Gustavson, The Smith Creek Bridge Site (41DW270) A Terrace Site in DeWitt County, Texas.Texas Archeological Research Laboratory, Studies in Archeology 35 and TxDot Environmental Affairs Division, Archeology Studies Program, Report 17. 

Cheatum, E. P. and R. W. Fullington 

1971aThe Aquatic and Land Mollusca of Texas.Part One:The Recent and Pleistocene Members of the Gastropod Family Polygyridae in Texas.Bulletin 1.Dallas Museum of Natural History, Dallas. 

1971bThe Aquatic and Land Mollusca of Texas:Supplement:Keys to the Families of the Recent Land and Fresh-Water Snails of Texas. Bulletin 1.Dallas Museum of Natural History, Dallas. 

1973The Aquatic and Land Mollusca of Texas.Part Two:The Recent and Pleistocene Members of the Puplidae and Urocoptidae (Gastropoda) in Texas.Bulletin 1.Dallas Museum of Natural History, Dallas. 

Clark, John W. Jr. 

1969Implications of Land and Fresh-Water Gastropods in Archaeological 

Sites.Arkansas Academy of Science Proceedings 23:38-52. 

1973The Problem of the Land Snail Genus Rabdotus in Texas Archeological Sites.The Nautilus87(1):24 

1976Alvar Nunez and the Snail Rabdotus in Texas.The Nautilus 90(1):13-14. 

Collins, M. B. 

1995Forty Years of Archeology in Central Texas.Bulletin of the Texas Archeological Society. 66:361-400. 

Ellis, G. Lain, and James T. Goodfriend 

1994Chronometric and Site-Formation Studies Using Land Snail Shell:Preliminary Results.in Archeological Investigations on 571 Prehistoric Sites at Fort Hood, Bell and Coryell Counties, Texas.edited by W. Nicholas Trierweiler, pp. 183-201.United States Army Fort Hood Archeological Resource Management Series Research ReportNo. 31.Mariah Associates, Austin. 

Ellis, G. Lain, Glenn A, Goodfriend, James T. Abbott, P. E. Hare, and David W. Von Endt 

1996Assessment of Integrity and Geochronology of Archaeological Sites Using Amino Acid Racemization in Land Snail Shells:Examples from Central Texas.Geoarchaeology:An International Journal.11(3):189-213. 

Fullington, Richard W., and Wiliam Lloyd Pratt 

1974The Aquatic and Land Mollusca of TexasPart Three:The Helicinidae, Carychiidae, Achatinidae, Bradybaenidae, Bulimidae, Cionellidae, Haplotrematidae, Helicidae, Oreohelicidae, Spiraxidae, Streptaxidae, Strobilopsidae, Thysanophoridae,,Valloniidae (Gastropoda) in Texas.Bulletin 1.Dallas Museum of Natural History, Dallas. 

Goodfriend, Glen A. 

1992The Use of Land Snail Shells in Paleoenvironmental Reconstruction.Quaternary Science Reviews.11:665-685 

Hall, G. D. 

1998Prehistoric Food Resource Patches on the Texas Coastal Plain.Bulletin of the Texas Archeological Society69:1-10. 

Henry, Donald O. 

1995Cultural and Paleoenvironmental Successions Revealed by the Hog Creek Archeological Investigations, Central Texas.in Advances in Texas Archeology:Contributions From Cultural Resource Management, Volume 1.edited byJ. E. Bruseth and T. K. Perttula.Cultural Resource Management Report 5.Texas Historical Commission Department of Antiquities Protection. 

Healy, Paul F., Kitty Emery, and Lori E. Wright 

1990Ancient and Modern Maya Exploitation of the Jute Snail Pachychilus).Latin American Antiquity1(2):170-183. 

Hester, Thomas R., and Hill, T. C., Jr. 

1975Eating Land Snails in Prehistoric Southern Texas:Ethnohistoric 

and Experimental Data.The Nautilus89(2):37-78. 

Honea, Kenneth H. 

1962The Rammadyat of Northwest Africa and the Burned Rock Middens of Texas.Bulletin of the Texas Archeologcial Society.32:316-320. 

Hubricht, Leslie 

1960The Genus Bulimulus in Southern Texas.The Nautilus.74(2)68-70. 

1985The Distributions of the Native Land Mollusks of the Eastern United States.Fieldiana:New Series 24.Publication 1359.Field Museum of Natural History, Chicago. 

Jelks, E. B. 

1962The Kyle Site:A Stratified Central Texas Aspect Site in Hill County, Texas.Archeology Series 5.Department of Anthropology, The University of Texas at Austin. 

Johnson, L. 

1964The Devils Mouth Site:A Stratified Campsite at Amistad Reservoir, Val Verde County, Texas.Archeology Series 6.Department of Anthropology, The University of Texas at Austin. 

1991Early Archaic Life at the Sleeper Archeological Site, 41BC65 of the Texas Hill Country, Blanco County, Texas.Report No.39.Publications in Archaeology,Texas State Department of Highways and Public Transportation, Highway Design Division, Austin. 

1995Past Cultures and Climates at Jonas Terrace, 41ME29, Medina County, Texas.Report No. 40.Office of the State Archeologist, Texas Department of Transportation and Texas Historical Commission, Austin. 

1997The Lion Creek Site (41BT105), Aboriginal Houses and Other Remains at a Prehistoric Rancheria in the Texas Hill Country (Burnet County).Report 1, Environmental Affairs Division, Archeology Studies Program, Texas Department Transportation, Austin, and Report 41, Office of the State Archeologist, Texas Historical Commission, Austin. 

Johnson, LeRoy Jr.D. A. Suhm, and C. D. Tunnell 

1962Salvage Archeology of Canyon Reservoir:The Wunderlich, Footbridge, and Oblate Sites.Bulletin No. 5. Texas Memorial Museum, The University of Texas at Austin. 

Kelly, R. L. 

1995The Foraging Spectrum:Diversity in Hunter-Gatherer Lifeways.Smithsonian Institution Press.Washington. 

Lintz, Christopher and James T. Abbot 

1997Chronometric Dating of Snail Shells and Charcoal as an Assessment of Sediment Integrity at Archeological Sites 41ZP39 and 41ZP176, Porcion 18 Tract, Near Falcon Reservoir, Zapata County, Texas.Project No. 21116.TRCMariah and Associates Inc, Austin. 

Neck, Raymond 

1994Interpretation of Molluscan Remains from the Mustang Branch Site (41HY209).in Archaic and Late Prehistoric Human Ecology in the Middle Onion Creek Valley, Hays County, Texas.by R. A. Ricklis and M. B. Collins, pp. 491-497.Studies in Archeology 19.Texas Archeological Research Laboratory, TheUniversity of Texas at Austin.

1997Appendix:Environmental Assessment of the Immediate Vicinity of the Lion Creek Site (41BT105).in The Lion Creek Site (41BT105), Aboriginal Houses and Other Remains at a Prehistoric Rancheria in the Texas Hill Country (Burnet County).by Leroy Johnson, pp. 193-186.Report 1, Environmental Affairs Division, Archeology Studies Program, Texas Department of Transportation, Austin, and Report 41, Office of the State Archeologist, Texas Historical Commission, Austin. 

O’Brien, Michael J , and Thomas D. Holland 

1995The Nature and Premise of a Selection-Based Archaeology.in Evolutionary Archaeology Methodological Issues.edited by P.A. Teltser.pp 175-200.The University of Arizona Press, Tucson.

Purchon R. D. 

1977The Biology of the Mollusca, Second Edition,Pergamon Press, Oxford. 

Randolph, Polley Ann 

1973Influence of Environmental Variability on Land Snail Population Properties.Ecology 54(4):933-955.

Simmons, Frank 

1956Snails of the Burnt Rock Middens.in Central Texas Archeologist No. 7.edited by Frank Watt.pp.48-51.Central Texas Archeologist, Waco, Texas. 

Smith, E. A. and B. Winterhalder 

1992Natural Selection and Decision-Making: Some Fundamental Principles.in Evolutionary Ecology and Human Behavior. edited by E. A. Smith and B. Winterhalder, pp, 25-60.Aldine de Gruter, New York. 

Wandsnider, LuAnnn 

1997The Roasted and the Boiled:Food Composition and Heat treatment with Special Emphasis on Pit-Hearth Cooking.Journal of Anthropological Archaeology.16(1):1-48.