The Work

Specialist Report 2004 - Land Snails

Posted by steven on 11/04/2005 at 01:31 PM

(NG 5943 1971)

Claire L. Pannell - University of Glasgow


Fourteen species of gastropod in the form of shell remains were recovered from the excavations at High Pasture Cave.  None of the shells appeared to have been predated upon.  All snail shells are native to Britain and are terrestrial, with one exception, Littorina saxatilis, the rough periwinkle.  Thirteen species were identified from the four samples received; one species remains unidentified.  All of the terrestrial gastropod species prefer moist, shady habitats and no species were found in this preliminary study to infer that very different conditions prevailed in the past environment.  Inferences from the gastropod assemblages are complicated by the apparent mixed age of all of the samples which prevents the recognition of any clear trends.  Separation of the shells into chronologies by an appropriate dating methodology may yet yield more information about past habitat change if it occurred and exhibited an effect upon the snail species distributions in the area.  Further analyses of the shells, as long as not diagenesced may yield information about the snails’ past dietary intake, the local vegetation and the environmental moisture at the time of shell formation.


High pasture cave is a limestone cave situated 1.25 km south east of the village of Torrin on the Isle of Skye.  The cave entrance is situated in a shallow valley on the northern slopes of Beinn an Dubhaich, and is one of the longest on Skye with a total length of 320 m and a maximum depth of 12.8 m.  Disturbed archaeological deposits were initially found in May 2002 and subsequently excavated between April and June 2004 (see report by Birch et al 2003).  Four samples of land snail shells from the excavations at High pasture cave, Isle of Skye, were received for identification.  The aims were to identify the shells from each sample to species level and to evaluate their use as potential palaeo-habitat indicators.


Shells were processed prior to being received for identification and had been extracted from the sedimentary cave material by wet sieving to a minimum mesh size of 1.5 mm and subsequently air dried.  Shell totals were compiled following the convention of Sparks (Sparks 1959) in which gastropod apices count as single individuals.  Gastropods were identified by comparison with type specimens housed in the National Museum of Scotland, Edinburgh.  Nomenclature follows that of Kerney (Kerney & Cameron 1979). 


The total number of species found in the four samples numbered fourteen; thirteen of those have been positively identified.
The unknown species is represented by seven fragments of individual shells, comprising solely the protoconch and have proved to be unidentifiable at present until larger individuals are found.

The snail species found are listed below and summarised in Table 1 (p. 4).

Species A:  Vitrea crystallina (Müller 1774)
Family Zonitidae; a large family, occurring throughout the Northern hemisphere, favouring damp habitats, such as under stones and among ground litter in woods.  Some species are subterranean and many are partly carnivorous.
A universal species, but commonest in damp places, such as marshes and moist grassland.
Widespread in Europe.

Species B: Clausilia bidentata (Ström 1765)
Synonyms C. nigricans, C. rugosa.
Family Clausilidae, Subfamily Clausiliinae; occurring in the western Palaearctic region, South America and South East Asia.  They live mainly in woods and among rocks, normally hiding in crevices or under ground litter and emerging in damp weather and at night and climbing high up on bare surfaces to graze on algae and lichens.  A few species are ovo-viviparous.
Found in moderately moist places, among rocks, old walls, woods, hedgebanks: rare above 1000m.  Widespread in the British Isles and the Northern hemisphere.

Species D: Discus rotundatus (Müller 1774)
Synonym Goniodiscus rotundatus
Family Endodontidae; a very large and ancient family, occurring in all parts of the world and characteristic of moist, shaded habitats.
Found in moist sheltered places of all kinds: woods, under ground litter and stones and damp herbage; often among rubbish in gardens.
Common throughout western and central Europe, as far as southernmost Scandinavia where it has been anthropogenically spread (in Finland in gardens and greenhouses only).

Species E: Oxychilus alliarus (Müller 1774)
Family Zonitidae; a large family, occurring throughout the Northern hemisphere, favouring damp habitats, such as under stones and among ground litter in woods.  Some species are subterranean and many are partly carnivorous.
A universal species; found in woods, fields, on rocks, occasionally in gardens and greenhouses.  Tolerant of poor acidic places such as conifer plantations.
Common in Western Europe; increasingly local to the east.

Species F
Juvenile specimens; protoconch plus ¼ whorl only.  Unidentifiable at present.

Species G: Cochlicopa lubrica (Müller 1774)
Family Cochlicopidae; occurring in North America and the western Palaearctic region.  A catholic species; found in moderately damp places of all kinds; marshes, grasslands, woods. 
Common throughout North America, Europe and Asia (the Holarctic).

Species I: Lauria cylindracea (da Costa 1778)
Synonyms Pupa umbilicata, P. anconostoma.
Family Pupillidae, Subfamily Truncatellininae
This is a variable species: the angular tooth may be absent, the reflected lip may be poorly developed, and the shape may vary from tall cylindrical to stumpy conical.
Found in woods, on rocks, grassland; not usually in very wet places.  Often abundant under ivy on stone walls.  Many species are ovo-viviparous.
A world-wide family.  Range western European and Mediterranean.  Commonest near the Atlantic.

Species J: Littorina saxatilis (Olivi 1792)
Subclass: Prosobranchia, Order: Neotaenioglossa, Family Littorinoide.
Known as the rough periwinkle, it retains its young within a series of brood chambers until their shells are sufficiently developed for the young to lead an independent existence (Barnes & Hughes 1999).  It is thus able to reproduce in harsher environments (high shore levels, salt marshes, estuaries, mobile pebble beaches and silty shores) than related species that do not brood their young.  Littorina saxatilis occurs from the upper eulittoral zone down to the littoral fringe of the intertidal and is typically found in crevices of bedrock, empty barnacle shells and under stones.  L. saxatilis occurs on salt marshes on the base of Spartina species and on firm mud banks.  It also occurs submerged in sheltered, brackish lagoons generally attached to Zostera, Fucus, Ruppia and Potamogeton species.
Common around the coasts of Britain and Ireland. 

Species K: Cochlicopa lubricella (Porto 1838)
Synonyms C. lubrica var. lubricella, C. minima.
Order Stylommatophora, Family Cochlicopidae.
A catholic species; characteristically of drier places than Cochlicopa lubrica, i.e. limestone grassland, calcareous sand dunes, screes, but the two species are often associated.
Common throughout the Holarctic, but more local than C. lubrica; also occurring in the Western Palaearctic region.

Species L: Arianta arbustorum (Linné 1758)
Synonym Helicigona arbustorum.
Family Helicidae, Subfamily Ariantinae. 
Widespread amongst meadows, herbage, woods and hedgerows, but always in damp places, and very restricted in areas with dry climate and good drainage.
Found throughout Central and North West Europe, but becoming rare in the extreme west and south.

Species M: Euconulus fulvus (Müller 1774)
Family Euconulidae; found in damp situations in most parts of the world.  Represented in Europe by one genus only.
A catholic species: widespread in deciduous and coniferous woods, grasslands, marshes; usually in fairly most places.  Distributed throughout the Holarctic.

Species N: Vitrina pellucida (Müller 1774)
Synonyms Helicolimax pellucidus, Phenocolimax pellucidus.
Family Vitrinidae; snails of mostly cool, moist habitats; some partly carnivorous.  The shell is very small relative to the body size, and the body is generally not fully retractable, although this is possible in this species.
Common species in a wide variety of moderately humid places: woods, grassland, amongst rocks; often abundant in the grassy hollows of coastal sand-dunes and the banks of streams.
Distributed throughout the Holarctic.

Species O: Zonitoides excavatus (Alder 1830)
Family Zonitidae, Sub family Gastrodontinae.
Found amongst ground litter in woods; occasionally in marshes.  Restricted to non-calcareous soils.
Widespread in the British Isles.

Species Q: Ashfordia granulata (Alder 1830)
Synonym Monacha granulata (Alder).
Family Helicidae, Subfamily Monachinae.
Found in woods, hedges, marshes and damp, shady places generally. 
British Isles only; widespread but local in Britain.

Physical Description of Shells

The following information details the condition of the shells recovered from each trench.

Trench 1.  Context C002, S.21.

Vitrea crystallina - white and opaque; Clausilia bidentata – still brown but not shiny;
Discus rotundatus – white, opaque, very faint markings present; Oxychilus alliarus – white; grey in places, dull sheen; Species F (unidentifiable at present) – opaque, dull sheen; Cochlicopa lubrica – cream coloured, opaque, dull sheen; Lauria cylindracea – faintly translucent, pale brown, dull sheen; Arianta arbustorum – pale brown, spiral line present, faintly translucent, dull sheen.

Trench 1.  Context C003, S.59.

Vitrea crystallina – white, most opaque, a few translucent; Clausilia bidentata – pale brown, mat, opaque; Discus rotundatus – pale brown, matt, some with faint markings and ribs; Oxychilus alliarus – grey or white, most opaque and mat, a couple are translucent and possess a dull sheen; Species F (unidentifiable at present) – one translucent, other opaque;
Cochlicopa lubricella – off-white, opaque, mat; Arianta arbustorum – pale brown, spiral markings visible, dull sheen on one; others opaque.

Feature F.001.  Context C014, S.118.

Clausilia bidentata – pale brown, opaque and dull; Discus rotundatus – white with faint markings, opaque and dull; Oxychilus alliarus – some opaque, white and dull; two have dull sheen and semi-translucent; Littorina saxatilis – grey, opaque and dull; Arianta arbustorum - pale brown to white, translucent with pale sheen, others opaque.

Much secondary carbonate deposits on the shells from this context were noticed, possibly indicating contamination by ground water.

Context C001.  Disturbed deposits, S.14.

Vitrea crystallina – some white, opaque and dull, others translucent and glossy;
Clausilia bidentata – some cream, dull and opaque, others brown and with a slight sheen; Discus rotundatus – some cream, faintly marked, opaque and dull, others strongly marked and translucent with a mild sheen; Oxychilus alliarus – some white, opaque and dull, others semi-translucent and with a slight sheen; Species F (unidentifiable at present) – grey, dull sheen, semi-translucent; Cochlicopa lubrica – grey to off-white, dull sheen, semi-translucent;
Lauria cylindracea – cream, opaque and dull to brown, shiny and transparent;
Littorina saxatilis – grey and opaque; Cochlicopa lubricella – one brown and aestivating or recently dead; Arianta arbustorum – pale brown with spiral markings, dull sheen;
Euconulus fulvus – some white and opaque, others pale brown and semi-translucent;
Vitrina pellucida – grey, translucent; Zonitoides excavatus – cream coloured, slight sheen, faintly translucent; Ashfordia granulata – white to grey, semi-translucent to opaque.

Some of the land snails recovered from High Pasture Cave


All snail shells found are native to Britain and belong to the gastropod subclass Pulmonata, order Stylommatophora and are terrestrial snails, with one exception, Littorina saxatilis, the rough periwinkle; a marine species that belongs to the subclass Prosobranchia, order Neotaenioglossa.

Snail shells rapidly lose the periostracum (the outer proteinaceous covering), post-mortem and hence colour and other markings will fade with time.  Shells also become opaque with time and the shell’s outer surface structures such as ribbing can become worn due to erosion.  Factors such as these that affect the shells’ appearance can be considered a very rough guide to age.  Where there is wide variation in the appearance of the shells it is possible that the deposit represents a mixed-age deposit and/or disturbance post burial that would serve to reassemble the deposits.
It can be seen from the detailed descriptions of the shells (p.5) that all the excavated samples appear to exhibit a degree of sediment mixing and thus cannot be considered as time stratified or time specific.  The sample labelled Context C001, S.14. is known to be from a disturbed deposit and this is confirmed by the condition of the shells and also by the presence of one snail, a Cochlicopa lubrica, that was found to have very recently died.  It was initially thought that this snail was aestivating, but after being placed in a humid environment with food and water for many weeks which would usually bring about the end of aestivation, the snail did not resume activity.

In mixed age deposits, the presence of a species in a particular dated level is not a reliable indicator of its age – ages cannot be determined by stratigraphic association.  Brecciation of the sediments and secondary carbonate deposits on the shells indicate that water has penetrated the sediments frequently.  Shells could also reach the cave floor from crevice and solution hole fills as well as being washed in from the surface above the cave (Goodfriend & Mitterer 1993).  Periodic flooding of the cave could cause flotation of the shells and subsequent mixing of levels.  Biasing of species abundances may occur as a result of the accumulation of shell middens or by differential destruction by certain predatory species such as birds or rats.  The transport of shells by humans or by predators can also result in the occurrence of a mixed-habitat assemblage.  Taphonomic changes could also occur as a result the sorting action of earthworms.  Differential destruction of the shells of different species of snails can also be expected to take place depending upon the acidity of the sedimentary environment and the resistance of the shells, which will vary interspecifically.

All of these influences mean that the sub-fossilised assemblages are not an accurate representation of the former live populations.

Further Work and Recommendations

Quantitative samples are the most informative in order to analyse a deposit with the goal of inferring the palaeo-habitat.  Sampling and processing of sediments through a series of sieves, either wet or dry, to a minimum size mesh of 0.5 mm (Goodfriend 1992) (Evans 1972) facilitates this and also ensures the complete retrieval of all species present and individuals.  This in turn allows counts of whole shells and apices in addition to species identification.  Most species of land snail present in the Quaternary and Holocene fossil records are still extant, so it is possible to glean palaeo-environmental information from the study of their modern populations (Goodfriend 1992).

The dating of shells from cave deposits can be highly problematic due to the mixed age of the sediments.  Snail shells from limestone areas are not suitable for radio-carbon dating alone due to the ingestion of old carbonates from limestone or sediments during the snail’s lifecycle which releases old CO2 into the bicarbonate pool from which the shell carbonate is precipitated.  This leads to variation in dates of as much as 3,120 yrs in some instances (Goodfriend & Stipp 1983).  Conversely, snails from non-limestone areas have been shown to produce no variation, implying their suitability for radiocarbon dating.  As land snail shell carbonate is subject to such an anomaly, an adjustment has to be made for apparent radiocarbon ages reported by the laboratory.  Values have firstly to be corrected for the anomaly and then corrected for fractionation (Pigati 2002).  The amount of material recovered from cave deposits is also often insufficient for radiocarbon dating.  Radiocarbon dating requires relatively large sample sizes (around 25g) so that only very large shells can be analysed individually.  AMS radiocarbon analyses require much smaller sample sizes (~ 30 mg) and thus allow radiocarbon dating of even relatively small individual shells.  AMS analyses are expensive, however.

A combination of radiocarbon dating and amino acid epimerization analysis could be used to determine the chronology of species in a deposit.  An integrated approach combines 14C dating with amino acid epimer (alloisoleucine/isoleucine A/I) dating (Goodfriend & Mitterer 1988).  Amino acid epimer ratios have been shown to be good predictors of age in land snails and large numbers can be analysed.  They require only small amounts of material (~ 10mg).  The A/I values are calibrated against radiocarbon in order to determine the epimerization rate and thus permit the calculation of age from the A/I values.  A study involving 38 radiocarbon-dated samples of a land snail showed that the D/L ratios of each of the amino acids showed a strong correlation with age (r2 = 0.84) and thus good age predictive ability (Goodfriend 1991) (Engel, Goodfriend, Qian et al 1994).

Mollusc shells appear to be one of the best materials for amino acid dating due to their non-porous structure which retains the organic matrix well.  The formation of the terrestrial mollusc shell has only ever been under terrestrial influence and has not been complicated by the repeated submergence and exposure that characterise aquatic species in the littoral zone, which may complicate the hydrolysis of shell proteins and the racemization/epimerization of amino acids. Environmental conditions that can produce variation in epimerization/racemization rates (especially temperature) are considered to be uniform within a deposit, thus giving greater precision to estimates of age differences among individuals.

Another method that may yield information about the past environment is analysis of the stable isotopes of carbon and oxygen contained in the shell carbonate.  The shells of terrestrial gastropods are comprised of two parts; a calcium carbonate layer overlain by a protein layer; the periostracum.  Land snail shell carbonate is derived from plants, atmospheric CO2, and ingested limestone (Goodfriend & Hood 1983).  Relatively direct information about the rainfall 18O values of an area may be obtained from the 18O of land snail shell carbonate (Lécolle 1985, Goodfriend, Magaritz & Gat 1989).  The carbon isotopic composition of plant material depends upon the plant’s photosynthetic pathway (C3 or C4).  The 13C of shell carbonate may be a useful indicator of the presence or absence of C4 plants, although when the organic component of the shell matrix is measured, derived only from the snail’s diet, there is less ambiguity (Goodfriend & Magaritz 1987a).  However, the method may still be of use in older fossils were the preservation of the organic component may be poor (Magaritz & Heller 1983).

The organic component of the shell material can also provide information about the past rainfall of the area, either mean annual rainfall or the mean number of rainy days per annum.  The δ13C of land snails shell organic matter has the advantage that it reflects strictly local environmental conditions and thus permits reconstruction of palaeoclimate on a fine geographical scale (Magaritz & Heller 1980).  Although shells only contain a small amount of organic matter (~ 0.03‰ by weight), they are not prone to contamination by soil organics due to their non-porous structure, unlike that of bone (Goodfriend 1988).

Prior to any analyses shells must firstly be checked for diagenesis by X-ray analysis (XRD) to check the composition remains that of the original aragonitic structure.  The presence of calcite in the matrix indicates that the original aragonite has undergone re-precipitation as calcite, thus entering the possibility of re-fractionation of the isotopes and therefore they are not representative of those that were originally laid down during shell formation.


Thanks are extended to Sankurie Pye, Assistant Curator of Mollusca at the NMS, and to Dr Barry Colville for their help with identification.


Barnes RSK, and Hughes RN. 1999. An Introduction to Marine Ecology. London: Blackwell Science. 254 pp.

Birch, S. A., Wildgoose, M., and Kozikowska, G. 2003. Archaeological deposits from a limestone cave on the Island of Skye. Unpublished report.

Engel M, Goodfriend GA, Qian Y, and Macko S. 1994. Indigeneity of organic matter in fossils: A test using stable isotope analysis of amino acid enantiomers in Quaternary mollusk shells. Proceedings of the Natural Academy of Science, USA. 91: 10475-10478

Evans JG. 1972. Land snails in Archaeology. London: Seminar Press, London

Goodfriend GA. 1988. Mid-Holocene rainfall in the Negev Desert from 13C of land snail shell organic matter. Nature 333: 757-760

Goodfriend GA. 1991. Patterns of racemization and epimerization of amino acids in land snail shells over the course of the Holocene. Geochimica et Cosmochimica Acta 55: 293-302

Goodfriend GA. 1992. The use of land snail shells in paleoenvironmental reconstruction. Quaternary Science Reviews 11: 665-685

Goodfriend GA, and Hood DG. 1983. Carbon Isotope Analysis of Land snail shells: implications for carbon sources and radiocarbon dating. Radiocarbon 25: 810-830

Goodfriend GA, and Magaritz M. 1987a. Carbon and oxygen isotope composition of shell carbonate of desert land snails. Earth and Planetary Science Letters 86: 377-388

Goodfriend GA, Magaritz M, and Gat JR. 1989. Stable isotope composition of land snail body water and its relation to environmental waters and shell carbonate. Geochimica et Cosmochimica Acta 53: 3215-3221

Goodfriend GA, and Mitterer RM. 1988. Late Quaternary land snails from the north coast of Jamaica: local extinctions and climate change. Palaeogeography, Palaeoclimatology, Palaeoecology 63: 293-311

Goodfriend GA, and Mitterer RM. 1993. A 45,000-yr record of a tropical lowland biota: The land snail fauna from cave sediments at Coco Ree, Jamaica. Geological Society of America Bulletin 105: 18-29

Goodfriend GA, and Stipp JJ. 1983. Limestone and the problem of radiocarbon dating of land-snail shell carbonate. Geology 11: 575-577

Kerney MP, and Cameron RAD. 1979. A field guide to the land snails of Britain and North-West Europe. London: Collins, London

Lécolle P. 1985. The oxygen isotope composition of landsnail shells as a climatic indicator: applications to hydrogeology and paleoclimatology. Chemical Geology (Isotope Geoscience Section) 58: 157-181

Magaritz M, and Heller J. 1980. A desert migration indicator- Oxygen isotopic composition of land snail shells. Palaeogeography, Palaeoclimatology, Palaeoecology 32: 153-162

Magaritz M, and Heller J. 1983. Annual cycle of 18O/16O and 13C/12C isotope ratios in land snail shells. Isotope Geoscience 1: 243-255

Pigati JS. 2002. On correcting 14C ages of gastropod shell carbonate for fractionation. Radiocarbon 44: 755-760

Sparks BW. 1959. The Ecological Interpretation of Quaternary Non-marine Mollusca. Proceedings of the Linnean Society of London Symposium on Quaternary Ecology: 71 - 80

Species No of ID's Sample Weight (g)
Alder 1 0.4
Birch 91 40.7
Hazel 360 286.9
Pine 5 3.7
Cherry/blackthorn type 1 0.3
Oak 5 7.4
Willow 4 1.30
Heather 25 5
Total 492 345.7

Next entry: Specialist Report 2004 - Charred Plant Remains

Previous entry: Specialist Report 2004 - Amphibian Bone