The southern African tick shell, Nassarius kraussianus (Dunker, 1846), has been identified as being the earliest known ornamental object used by human beings. Shell beads dated from ~75,000 years ago (Pleistocene era) were found in a cave located on South Africa's south coast. Beads made from N. kraussianus shells have also been found in deposits in this region dating from the beginning of the Holocene era (<10,000 years ago). These younger shells were significantly smaller, a phenomenon that has been attributed to a change in human preference.
We investigated two alternative hypotheses explaining the difference in shell size: a) N. kraussianus comprises at least two genetic lineages that differ in size; b) the difference in shell size is due to phenotypic plasticity and is a function of environmental conditions. To test these hypotheses, we first reconstructed the species' phylogeographic history, and second, we measured the shell sizes of extant individuals throughout South Africa. Although two genetic lineages were identified, the sharing of haplotypes between these suggests that there is no genetic basis for the size differences. Extant individuals from the cool temperate west coast had significantly larger shells than populations in the remainder of the country, suggesting that N. kraussianus grows to a larger size in colder water.
The decrease in fossil shell size from Pleistocene to Holocene was likely due to increased temperatures as a result of climate change at the beginning of the present interglacial period. We hypothesise that the sizes of N. kraussianus fossil shells can therefore serve as indicators of the climatic conditions that were prevalent in a particular region at the time when they were deposited. Moreover, N. kraussianus could serve as a biomonitor to study the impacts of future climate change on coastal biota in southern Africa.
Citation: Teske PR, Papadopoulos I, McQuaid CD, Newman BK, Barker NP (2007) Climate Change, Genetics or Human Choice: Why Were the Shells of Mankind's Earliest Ornament Larger in the Pleistocene Than in the Holocene? PLoS ONE 2(7): e614. doi:10.1371/journal.pone.0000614
Academic Editor: Michael Hofreiter, Max Planck Institute for Evolutionary Anthropology, Germany
Received: March 12, 2007; Accepted: June 5, 2007; Published: July 18, 2007
Copyright: © 2007 Teske et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: PRT was supported by a Postdoctoral Research Fellowship from the Claude Harris Leon Foundation. This work was funded by a National Research Foundation Grant (GUN 2069119) to NPB.
Competing interests: The authors have declared that no competing interests exist.
The southern African tick shell, Nassarius kraussianus (Dunker, 1846), is an estuarine and shallow water marine snail that has recently been identified as being the earliest known object to be used for ornamental purposes by human beings; shells found in Blombos Cave on the south coast of South Africa (Fig. 1) were estimated to have been worn as a necklace during the Middle Stone Age (MSA; ~75 000 years ago , ). Fossils of N. kraussianus are absent from south-west African deposits of the Pliocene (1.8–5.3 million years ago; ), but by the late Pleistocene (~120 000 years ago), the species had become an important component of the coastal fauna throughout much of southern Africa , . Tick shells are also common in deposits from the Late Stone Age (LSA; beginning of the present interglacial, <10 000 years ago , ) and today are among the numerically dominant coastal gastropods in southern Africa . Tick shells from Blombos Cave dated from the MSA were significantly larger than ornamental LSA shells from the same locality and from another fossil site in the region (Die Kelders), as well as from modern specimens collected in estuaries in the region (Goukou and Duiwenhoks) . From this, D'Errico et al.  concluded that MSA humans had selected only exceptionally large shells for their necklaces, whereas during the LSA, shells were collected randomly. We question this interpretation for two reasons: Firstly, while several of the 29 MSA shells were more than 10 mm in length, none of the 2 098 LSA shells and none of the similarly sized 2 587 modern shells that were measured exceeded 10 mm. The effort of finding shells that exceeded 10 mm would thus have been exceptional during the LSA. Secondly, not all of the MSA shells were exceptionally large, but they rather comprised a size range from 6.83 to 10.42 mm. This could indicate that instead of a change in human preference, the sizes of shells in the region decreased. We tested two alternative hypotheses, both of which assume that shell sizes of N. kraussianus on South Africa's south coast were larger in the MSA than in the LSA: a) the size difference has a genetic basis and N. kraussianus comprises at least two distinct genetic units and b) the difference in shell size is due to phenotypic plasticity responding to changing environmental conditions.
Sampling localities (A–J) from which specimens of Nassarius kraussianus were collected. The sample sizes were: A = 350, B = 104, C = 132, D = 112, E = 47, F = 183, G = 83, H = 54, I = 71, J = 94. Blue circles indicate localities dominated by haplotypes present in the western range of the species' distribution, and red circles indicate sites where haplotypes of the eastern lineage were mostly found. The location of Blombos Cave (where N. kraussianus fossils were found that were worn as beads by human beings during the Middle and Late Stone Age) is indicated.
The present-day geographical range of N. kraussianus spans three marine biogeographical provinces in southern Africa : the cool-temperate west coast, the warm-temperate south coast, and the subtropical east coast (Fig. 1). Many marine organisms in the region are associated with more than one of these biogeographic provinces, but each province has its own combination of species –. Genetic data show that some coastal invertebrates with planktonic larvae or direct development that occur in more than one province comprise two or more genetically distinct regional lineages, with boundaries that coincide with those between the biogeographic provinces –. Given the wide distribution range of N. kraussianus and the fact that it disperses by means of planktonic larvae , it is possible that this species may comprise several regional phylogeographic units that differ in shell size, e.g. a large-bodied lineage adapted to cooler water temperature and a small-bodied lineage present in warmer water. Their distributions may have shifted as a result of climate change, with the small-bodied lineage having replaced the large-bodied lineage on the south coast. Alternatively, the larger lineage may have become extinct (e.g. during the last glacial maximum ~15–25 000 years ago) and been replaced by the smaller lineage throughout its range (e.g. at the beginning of the present interglacial, <10 000 years ago). The two lineage-hypothesis is rejected if a) no genetically distinct lineages of N. kraussianus are found that differ in shell size and b) expansion events from south-eastern to south-western Africa evident in the genetic data do not significantly postdate the MSA.
Morphological differentiation in molluscs is often the result of phenotypic plasticity rather than genetics . The fact that both homeotherms and poikilotherms grow larger at colder temperatures ,  may account for a change in shell size as a result of an increase in water temperatures on the south coast from the MSA to the LSA. This hypothesis is rejected if tick shells in different biogeographic provinces do not differ in shell size.
A total of 1230 specimens of Nassarius kraussianus were collected from 10 South African estuaries/lagoons throughout the species' distribution range (Fig. 1). Ten specimens from each sampling locality were randomly selected for genetic analyses. Genomic DNA was isolated following the Chelex® extraction protocol . Partial mitochondrial cytochrome oxidase c subunit I gene (mtDNA COI) and 16S rDNA sequences were obtained by means of the polymerase chain reaction (PCR) using universal primers , . PCR reactions and sequencing followed previously published protocols , .
A minimum spanning network of mtDNA haplotypes was constructed using statistical parsimony  as implemented in the program tcs version 1.21 . Gaps were treated as missing data. Relationships between groups of haplotypes (nested clades) and geography were established with the program geodis version 2.5 , and the latest version of the inference key for nested clade analyis  was used to infer the most likely evolutionary scenario that may have resulted in the observed phylogeographic patterns. As gene flow along the coast is essentially one-dimensional, along-coast distances between sampling localities were specified rather than the geographic coordinates.
To determine whether N. kraussianus underwent a range expansion, the mismatch distribution  of the mtDNA sequences was estimated under the spatial expansion model  in arlequin version 3.1 . The time at which this event had taken place was determined by converting the expansion time parameter τ to time in years using the formula τ = 2ut, where u is the mutation rate per nucleotide per year multiplied by sequence length (i.e. number of nucleotides), and t is the time since population expansion in years. Ninety-five percent bootstrap confidence intervals of τ were calculated using 100 000 coalescent simulations in arlequin. The procedure was repeated five times to check for consistency of results. We also estimated the divergence time between two regional lineages of N. kraussianus identified by the nested clade analysis using the program im . This program simultaneously estimates divergence time, time to most recent common ancestor, effective population sizes of the present-day lineages and their ancestor, the proportion of the ancestral population that has founded one of the descendant populations (to account for changes in population size), as well as pairwise migration rates, under the coalescent model . We specified the HKY model  with an inheritance scalar of 0.25 for mitochondrial DNA. After a number of exploratory runs to determine suitable upper bounds for each model parameter, we used the following search strategy: -b500000–q1100–q2200–qa50–qu1–t2.5–m15–m210–fg–n20–g10.01–g22–k20–j8(population1: western; population 2: eastern). To ensure consistency of results, 10 independent runs with random starting seeds and at least 2 million genealogical steps were performed. Final divergence time estimates were calculated based on the means of the five runs with the highest effective sample sizes (ESS). The value obtained for divergence time was converted to time in years using the formula t = t/u, where t is time in years, t is scaled divergence time and u is the mutation rate per site multiplied by sequence length.
A mutation rate to convert both τ and t into time in years was obtained as follows: for COI, we used an evolutionary rate of 1% per million years based on a marine gastropod group for which a good fossil record is available . A rate for 16S rDNA was determined using two sister species of marine gastropods for which published sequences of both COI and 16S rDNA are available, namely Conus brunneus and C. regius . COI and 16S rDNA sequences of these differ by 13% and 6%, respectively. This corresponds to an evolutionary rate of 0.4% per million years for 16S rDNA, or 0.7% per million years for the combined fragment, taking into account differences in sequence length.
Shell lengths of specimens from all 10 estuaries/lagoons were measured as described previously . Measurements were made using digital callipers and rounded to the nearest 0.05 mm. In most cases, the samples included a small number of juveniles and sub-adults (identified using the criteria in d'Errico et al. ). As the differences between adults and large sub-adults were not always obvious, we removed the smallest 25% of individuals from the data set of each population (following d'Errico et al. ) to eliminate the impacts of differences in recruitment between populations when comparing shell sizes of extant populations with the MSA shells. Ninety-five percent confidence intervals of the means were estimated based on 10 000 bootstrap pseudoreplications as implemented in PopTools version 2.6.3 .
Consensus sequences of 600 and 483 nucleotides were obtained for COI and 16S rDNA, respectively. All unique sequences generated in this study were submitted to GenBank (accession numbers DQ456981–DQ456995 and EF636006–EF636023). A total of 28 mitochondrial haplotypes were identified. A haplotype network constructed from these comprises two regionally confined units that are both characterised by a star-like pattern (Fig. 2), indicating population expansion . Sixteen of the haplotypes were mostly present on the south-east coast (shown in red), including the basal haplotype of the network. Twelve haplotypes occurred primarily on the south-west coast (shown in blue), with only the basal haplotype of clade 2-3 being present but rare on the south-east coast. Significant genetic structure was found between clade 2-1 (which includes the basal haplotype of N. kraussianus) and clade 2-3 (which comprises younger haplotypes). Using nested clade analysis , the evolutionary scenario inferred for these clades was a contiguous range expansion from south-east to south-west (inference chain: 1-2-11-12-No).
A statistical parsimony haplotype network constructed from combined partial mtDNA COI and 16S rDNA sequences of N. kraussianus. Haplotypes shown in blue were mostly found in the western portion of the species' geographical range (localities A–F; Fig. 1) and those in red were mostly found in the eastern portion of its range (localities G–J). Haplotypes are represented as ovals, with sizes being proportional to a haplotype's frequency. Letters within ovals indicate in which sampling localities a particular haplotype was found and correspond to the letters in Fig. 1. Small white circles are interior node haplotypes not present in the samples. Grey areas around groups of haplotypes depict two-step nested clades.
The mismatch distribution of the N. kraussianus sequences did not depart from the expectations of the spatial expansion model (SSD = 0.003, P = 0.2), thus supporting the result from the nested clade analysis. A spatial expansion time parameter τ of 1.35 (95% C.I. = 0.94–2.72) was found, and the scaled divergence time t calculated for the divergence of the south-eastern and south-western lineages was 0.47 (95% C.I. = 0.26–1.21). Using a rate of 0.7% per million years for the combined COI and 16S rDNA fragments, it was estimated that the range expansion from south-east to south-west took place ~89 000 years ago (95% C.I. = 62–179 000 years ago), and that the two regional lineages then diverged ~62 000 years ago (95% C.I. = 34–160 000 years ago).
Shells of extant populations from the west coast (Olifants Estuary and Langebaan Lagoon) were significantly larger than those from the south and east coasts, and individuals whose shells exceeded 10 mm in length were only found in these two populations. The largest 75% of the shells from the Olifants Estuary population were not significantly different in size from the MSA fossil shells (Fig. 3).
Shell sizes of the Middle Stone Age (MSA) specimens of N. kraussianus from Blombos Cave  and the largest 75% of the shells of each of ten extant South African populations (A–J). Vertical bars represent means and whiskers are upper 95% confidence limits. Horizontal lines represent the shell size of the largest individual from each population. The insert shows how shell size was measured.
In southern Africa, most of the examined coastal invertebrate species that, like Nassarius kraussianus, disperse by means of planktonic larvae, show considerable genetic differentiation across a transition area between the warm temperate and subtropical coastal regions in south-eastern Africa , , , . Phylogeographic discontinuity in this region is also evident in N. kraussianus, but in contrast to other invertebrates studied to date, this differentiation is supported by comparatively few nucleotide differences, indicating that this species became established along the south-west coast relatively recently. The time estimates of demographic events in the evolutionary history of N. kraussianus suggest that these are likely to have been linked to climatic changes in the region. Fossil data indicate that warmer conditions during the last interglacial (~80–130 000 years ago) enabled estuarine-lagoonal molluscs that are presently confined to the tropical east coast to extend their ranges to south-western Africa . As N. kraussianus is also commonly found in deposits from this time and the range expansion estimate falls into this time (~89 000 years ago), it is likely that this event took place as a result of elevated sea temperatures. In contrast to other gastropods, cooler temperatures after the last interglacial did not eliminate the species from south-western Africa. Confidence intervals of time estimates for the range expansion and subsequent divergence of the two lineages are wide. Nonetheless, the estimates do not significantly postdate the time when MSA shells were deposited at Blombos Cave (~75 000 years ago), and they significantly predate the LSA. This suggests that there is little support for the hypothesis that a now extinct larger-bodied lineage may have been present on the south coast during the MSA and that it was subsequently replaced by a smaller-bodied lineage from the east. This conclusion is further strengthened by the fact that snails present on the cool-temperate west coast can attain sizes similar to those of the MSA individuals from the south coast, and that their mtDNA sequences are identical to individuals found as far east as the subtropical Mzingazi Estuary (Fig. 1).
The rejection of the hypotheses that Nassarius kraussianus comprises or formerly comprised multiple regional genetic lineages that differ morphologically, suggests that shell size instead depends on environmental conditions. Body size in ectotherms has often been reported to be affected by temperature, with many species growing larger at lower temperatures –. The observed decrease in shell size in N. kraussianus between the two west coast populations and the remaining populations may thus be linked to sea surface temperatures, although microhabitat conditions are also likely to be important (the samples with the smallest shell sizes [localities F and I] were collected in shallow, vegetation-rich side arms of estuaries, where temperatures were higher than in the main channel). A decrease in shell size from west to east has also been reported in the South African rocky shore limpet Patella granularis . Fossil shell sizes of marine gastropods in Chile showed marked increases during periods of intensified cold water upwelling throughout the Pleistocene and Holocene . In South Africa, temperatures on the south and east coasts were cooler between ~50–80 000 years ago than at present . The fact that the shell sizes of N. kraussianus from the vicinity of the Goukou Estuary were similar to those of the present-day populations on the west coast suggests that similarly cooler environmental conditions were on the south coast during the MSA. We consider this the most parsimonious solution explaining the decrease in fossil shell size, and hypothesise that the sizes of N. kraussianus fossil shells can thus provide information on climatic conditions prevalent at the time during which they were deposited. The hypothesis that tick shells grow to a larger size in colder water could be further investigated by raising snails, collected at the same locality, at different temperatures in the laboratory. A confirmation of this would suggest that N. kraussianus could be a suitable biomonitor to study the effects of global climate change on the coastal biota of southern Africa, particularly because unlike most other southern African coastal invertebrates for which phylogeographic data are available, there is no indication of cryptic speciation in this species.
Plots of IM posterior probability distributions. Marginal posterior probability distributions for three parameter estimates from one of five IM runs scaled by the neutral mutation rate, including the population size parameter θ calculated for the southwestern lineage (θ1), southeastern lineage (θ2), and the ancestral lineage prior to divergence (θA), time since population divergence (t) and migration rates (m1 = southwestern to southeastern; m2 = southeastern to southwestern). Posterior probabilities are shown on the y-axes.
(0.60 MB TIF)
This is a contribution from the African Coelacanth Ecosystem Programme. We are grateful to Henning Winker and three anonymous reviewers for comments on earlier versions of the manuscript.
Conceived and designed the experiments: PT. Performed the experiments: PT IP. Analyzed the data: PT. Contributed reagents/materials/analysis tools: NB PT IP CM BN. Wrote the paper: NB PT IP CM BN.
- 1. Henshilwood C, d'Errico F, Vanhaeren M, van Nieker K, Jacobs Z (2004) Middle Stone Age beads from South Africa. Science 304: 404.
- 2. d'Errico F, Henshilwood C, Vanhaeren M, van Niekerk K (2005) Nassarius kraussianus shell beads from Blombos Cave: evidence for symbolic behaviour in the Middle Stone Age. J Human Evol 48: 3–24.
- 3. Kensley B (1977) A second assemblage of Pliocene invertebrate fossils from Langebaanweg, Cape. Ann S Afr Mus 72: 189–210.
- 4. Tankard AJ (1975) Thermally anomalous late Pleistocene mollusks from the south-western Cape Province, South Africa. Ann S Afr Mus 69: 17–45.
- 5. Cooper JAG, Kilburn RN, Kyle R (1989) A late Pleistocene molluscan assemblage from Lake Nhlange, Zululand, and its palaeoenvironmental implications. S Afr J Geol 92: 73–83.
- 6. Compton JS (2001) Holocene sea-level fluctuations inferred from the evolution of depositional environments from the southern Langebaan Lagoon salt marsh, South Africa. The Holocene 11: 395–405.
- 7. Kilburn R, Rippey E (1982) Sea Shells of Southern Africa. Macmillan, Johannesburg, South Africa.
- 8. Bolton JJ, Anderson RJ (1990) Correlation between intertidal seaweed community composition and sea water temperature patterns on a geographic scale. Bot Mar 33: 447–457.
- 9. Emanuel BP, Bustamante RH, Branch GH, Eekhout S, Odendaal FJ (1992) A zoogeographic and functional approach to the selection of marine reserves on the west coast of South Africa. S Afr J Mar Sci 12: 341–354.
- 10. Harrison TD (2002) Preliminary assessment of the biogeography of fishes in South African estuaries. Mar Freshwater Res 53: 479–490.
- 11. Ridgway TM, Stewart BA, Branch GM, Hodgson AN (1998) Morphological and genetic differentiation of Patella granularis (Gastropoda: Patellidae): recognition of two sibling species along the coast of southern Africa. J Zool 245: 317–333.
- 12. Evans BS, Sweijd NA, Bowie RCK, Cook PA, Elliott NG (2004) Population genetic structure of the perlemoen, Haliotis midae in South Africa: evidence of range expansion and founder events. Mar Ecol Prog Ser 270: 163–172.
- 13. Teske PR, McQuaid CD, Froneman PW, Barker NP (2006) Impacts of marine biogeographic boundaries on phylogeographic patterns of three South African estuarine crustaceans. Mar Ecol Prog Ser 314: 283–293.
- 14. Zardi GI, McQuaid CD, Teske PR, Barker NP (2007) Unexpected population structure of mussel populations in South Africa: indigenous Perna perna and invasive Mytilus galloprovincialis. Mar Ecol Prog Ser 337: 135–144.
- 15. Teske PR, Papadopoulos I, Zardi GI, McQuaid CD, Griffiths CL, et al. Implications of life history for genetic structure and migration rates of southern African coastal invertebrates: planktonic, abbreviated and direct development. Mar Biol. In press.
- 16. Teske PR, McQuaid CD, Barker NPLack of genetic differentiation among four southeast African intertidal limpets (Siphonariidae): phenotypic plasticity in a single species? J Molluscan Stud.. In press.
- 17. Teske PR, Froneman PW, McQuaid CD, Barker NPPhylogeographic structure of the caridean shrimp Palaemon peringueyi in South Africa: further evidence for intraspecific genetic units associated with marine biogeographic provinces. Afr J Mar Sci.. In press.
- 18. Palmer CG (1980) Some aspects of the biology of Nassarius kraussianus (Dunker) (Gastropoda: Prosobranchia: Nasseridae) in the Bushmans River estuary with particular reference to recolonization after floods. Grahamstown: Rhodes University. M.Sc. thesis.
- 19. Knowlton N (2000) Molecular genetic analyses of species boundaries in the sea. Hydrobiologia 420: 73–90.
- 20. Bullock TH (1955) Compensation for temperature in the metabolism and activity of poikilotherms. Biol Rev 30: 311–342.
- 21. Ray C (1960) The application of Bergmann's and Allen's rules to the poikilotherms. J Morphol 106: 85–108.
- 22. Walsh PS, Metzger DA, Higuchi R (1991) Chelex 100 as a medium for simple extraction of DNA for PCR-based typing from forensic material. Biotechniques 10: 506–513.
- 23. Folmer O, Black M, Hoeh W, Lutz R, Vrijenhoek R (1994) DNA primers for amplification of mitochondrial cytochrome c oxidase subunit I from diverse metazoan invertebrates. Mol Mar Biol Biotechnol 3: 294–299.
- 24. Palumbi SR (1996) Nucleic acids II: The polymerase chain reaction. In: Hillis DM, Moritz C, Mable BK, editors. Molecular Systematics. 2nd edition. Edited by. Sinauer Associates, Sunderland: pp. 205–247.
- 25. Teske PR, Cherry MI, Matthee CA (2004) The evolutionary history of seahorses (Syngnathidae: Hippocampus): molecular data suggest a West Pacific origin and two invasions of the Atlantic Ocean. Mol Phylogenet Evol 30: 273–286.
- 26. Templeton AR, Crandall KA, Sing CF (1992) A cladistic analysis of phenotypic associations with haplotypes inferred from restriction endonuclease mapping and DNA sequence data. III. Cladogram estimation. Genetics 132: 619–633.
- 27. Clement M, Posada D, Crandall KA (2000) tcs: a computer program to estimate gene genealogies. Mol Ecol 9: 1657–1659.
- 28. Posada D, Crandall KA, Templeton AR (2000) geodis: a program for the cladistic nested analysis of the geographical distribution of genetic haplotypes. Mol Ecol 9: 487–488.
- 29. Templeton AR (2004) Statistical phylogeography: methods of evaluating and minimizing inference errors. Mol Ecol 13: 789–809.
- 30. Rogers AR, Harpending H (1992) Population growth makes waves in the distribution of pairwise genetic differences. Mol Biol Evol 9: 552–569.
- 31. Excoffier L (2004) Patterns of DNA sequence diversity and genetic structure after a range expansion: lessons from the infinite-island model. Mol Ecol 13: 853–864.
- 32. Excoffier L, Laval G, Schneider S (2005) arlequin ver. 3.0: An integrated software package for population genetics data analysis. Evol Bioinform Online 1: 47–50.
- 33. Hey J, Nielsen R (2004) Multilocus methods for estimating population sizes, migration rates and divergence time, with applications to the divergence of Drosophila pseudoobscura and D. persimilis. Genetics 167: 747–760.
- 34. Kingman JFC (1982) The coalescent. Stochastic Process Appl 13: 235–248.
- 35. Hasegawa M, Kishino K, Yano T (1985) Dating the human-ape splitting by a molecular clock of mitochondrial DNA. J Mol Evol 22: 160–174.
- 36. Meyer CP, Geller JB, Paulay G (2005) Fine scale endemism on coral reefs: archipelagic differentiation in turbinid gastropods. Evolution 59: 13–125.
- 37. Duda TF Jr, Kohn AJ, Palumbi SR (2001) Origins of diverse feeding ecologies within Conus, a genus of venomous marine gastropods. Biol J Linn Soc Lond 73: 391–409.
- 38. Hood GM (2004) PopTools version 2.6.2. Available on the internet, URL http://www.cse.csiro.au/poptools.
- 39. Slatkin M, Hudson RR (1991) Pairwise comparisons of mitochondrial DNA sequences in stable and exponentially growing populations. Genetics 129: 555–562.
- 40. Atkinson D (1994) Temperature and organism size–a biological law for ectotherms? Adv Ecol Res 25: 1–58.
- 41. Partridge L, Barrie B, Fowler K, French V (1994) Evolution and development of body size and cell size in Drosophila melanogaster in response to temperature. Evolution 48: 1269–1276.
- 42. Atkinson D, Silby RM (1997) Why are organisms usually bigger in colder environments? Making sense of a life history puzzle. Trends Ecol Evol 12: 235–239.
- 43. Trussell GC (2000) Phenotypic clines, plasticity, and morphological trade-offs in an intertidal snail. Evolution 54: 151–166.
- 44. Teusch KP, Douglas SJ, Allmon WD (2002) Morphological variation in turritellid gastropods from the Pleistocene to recent of Chile: association with upwelling intensity. Palaios 17: 366–377.
- 45. Deacon J, Lancaster N (1988) Late Quaternary Palaeoenvironments of Southern Africa. Cape Town: OUP.