University of Johannesburg eResesearch repository Article in press in the Journal of Molluscan Studies

Molecular insights into species recognition within southern Africa’s endemic Tricolia radiation (Vetigastropoda: Phasianellidae) Tshifhiwa C. Nangammbi1,2, Peter R. Teske3 and David G. Herbert 2 1,2

Department of Nature Conservation, Faculty of Science, Tshwane University of Technology, Private Bag X680, Pretoria 0001, South Africa;

2

Department of Mollusca, KwaZulu-Natal Museum, Private Bag 9070, Pietermaritzburg 3200 & School of Life Sciences, University of KwaZulu-Natal, Private Bag X01, Scottsville 3209, South Africa; and 3

Molecular Zoology Laboratory (Aquatic Division), Department of Zoology, University of Johannesburg, Auckland Park 2006, South Africa.

ABSTRACT The validity of morphology-based species boundaries between the southern African representatives of the genus Tricolia Risso, 1826 was assessed using mitochondrial COI and 16S rRNA sequence data. Most phylogenies obtained from individual and combined genetic datasets recovered 10 of the southern African members of the genus as a monophyletic clade. No COI sequences of the 11th species (T. adusta) were available, but this species clustered among the other African species in the 16S rRNA phylogeny. Discrepancies between morphology and genetics were identified in two clades within which there was limited genetic variation and no differentiation between nominal species, comprising respectively T. africana (Bartsch, 1915) and T. capensis (Dunker, 1846), and T. bicarinata (Dunker, 1846), T. insignis (Turton, 1932) and T. kraussi (Smith, 1911). In both cases the distributions of the nominal taxa coincide with well-known biogeographic disjunctions, and there is evidence of overlapping and intergrading shell characters. We propose that both of these unresolved clades be recognized as single, phenotypically plastic species, for which the oldest available names are respectively T. capensis and T. bicarinata. Despite the resultant loss of species due to synonymy, the phasianellid fauna of southern Africa remains the most diverse in the world, with 10 endemic species and three tropical species extending south into KwaZulu-Natal.

INTRODUCTION The Tricoliinae, a subfamily of the Phasianellidae (pheasant shells), is presently considered to include two genera, Tricolia Risso, 1826 and Eulithidium Pilsbry, 1898 (Hickman & McLean, 1990). Eulithidium is a New World taxon and includes all the Caribbean and Western Pacific species formerly included in Tricolia. In contrast, Tricolia is an Old World and Australasian genus with varying levels of species diversity and endemicity across this region (Gofas, 1982; 1986; 1993; Robertson, 1985; Bogi & Campani, 2007; Nangammbi, 2010): Eastern Atlantic/Mediterranean (12

endemic taxa); tropical Indo-West Pacific (3); south-western Australia (3); Amsterdam and St. Paul Islands (1); northern Japan (1); tropical and subtropical East Africa (1) and southern Africa (25 unrevised nominal endemic taxa). Previous taxonomic revisions of the genus have mainly focused on the tropical Indo-West Pacific and the Eastern Atlantic/Mediterranean regions (Gofas, 1982; 1986; 1993; Robertson, 1985), and a thorough revision of the nominally speciose southern African Tricolia fauna has never been published. In the absence of such a revision, the taxonomy of the genus in this region remains poorly resolved,

1 Correspondence: Tshifhiwa C. Nangammbi; e-mail: [email protected]

having suffered from profligate species description (Turton, 1932) and subsequently from rather vague and indecisive discussion of synonymy (Barnard, 1963). However, through more detailed comparative morphological study some clarity is beginning to emerge, and the 31 nominal species of Phasianellidae recorded from the region are now thought to comprise only 16 entities that can be recognized as morphologically distinct taxonomic units (Nangammbi, 2010). Thirteen of these are regionally endemic taxa belonging to Tricolia, a revision of which is in preparation. With such diversity, the southern African Tricolia fauna may be the richest in the world. The present paper aims to test the validity of these morphologically determined entities using molecular data, so as to establish whether these taxonomic units represent reciprocally monophyletic lineages that can be considered distinct species.

South Africa (NMSA). See Table 1 for locality data, museum accession numbers and GenBank accession numbers, as well as details regarding samples of the outgroup taxa.

For the most part our confidence in the morphology-based species hypotheses is high, but in some cases, where distributions are allopatric or parapatric and where strong environmental gradients are at play, it is unclear whether the observed morphological differences result from ecological rather than genetic phenomena.

PCR reactions were carried out in 50 µl volumes with the following reagents: distilled water (36.8 µl), 10x amplification buffer with 15 mM MgCl2 (5 µl), 10 mM dNTP (1 µl), 2 units of Super Therm Gold Taq (Southern Cross Biotechnology, Cape Town), 5 µM solution of each primer (3 µl) and DNA template (1 µl). Thermal cycling conditions were: 11 minutes at 95ºC, followed by 35 cycles of 1 minute at 94ºC, 1 minute 30 seconds at markerspecific annealing temperatures, and 1 minute 30 seconds at 72ºC. The PCR was terminated with 5 minutes at 72ºC. Annealing temperatures were between 47ºC and 50ºC for COI, and 50ºC for 16S rRNA. Amplifications were performed on a Perkin Elmer GeneAmp PCR System 9600 (Applied Biosystems). PCR success was evaluated by electrophoresis of 2.5 µl PCR products on a 1.3% agarose gel stained with ethidium bromide and visualized with UV fluorescence.

MATERIAL AND METHODS Taxon sampling Samples were obtained throughout southern Africa (Table 1) and included 11 of the 13 morphologically diagnosable Tricolia taxa endemic to southern Africa, the distributions of which are illustrated in the Supplementary material (Fig. S1). The two species not included in this study were T. striolata (Turton, 1932) and T. retrolineata Nangammbi & Herbert, 2008. The reasons for not including these species were that no live collected material was available for the former, and that DNA extraction was not successful for the one available preserved specimen of the latter. The type species of Tricolia, T. pullus (Linnaeus, 1758), was used as the outgroup taxon in phylogenetic analyses, in addition to representatives of closely related genera, including Tricolia (Hiloa) variabilis (Pease, 1861), Phasianella solida (Born, 1778) and Eulithidium perforatum (Philippi, 1848). These species were selected on the basis of an unpublished phylogeny using combined 18S and 28S sequences, which recovered them basal to the African species of Tricolia (Nangammbi, 2010). Vouchers of the southern African species included in this study are deposited in the KwaZulu-Natal Museum, Pietermaritzburg,

Laboratory protocols Genomic DNA was isolated from the foot (large specimens) or the entire body (small specimens) using one of three equally successful DNA isolation protocols: DNeasy Blood and Tissue Kit (Qiagen), SV Total RNA Isolation System (Promega), following manufacturers’ protocols, and the ammonium acetate protocol of Nicholls et al. (2000). Amplification of the 5'-half of mitochondrial cytochrome-c oxidase subunit I (COI) was achieved using standard LCO1490 (forward) and HCO2198 (reverse) primers (Folmer et al., 1994). Amplifications of the mitochondrial 16S rRNA marker were achieved with universal forward primer 16Sar and reverse primer 16Sbr (Palumbi et al., 1991).

Amplified DNA was purified for cycle-sequencing using either the QIAquick PCR purification kit (Qiagen) or the 1:4 ammonium acetate protocol (Moussalli et al., 2005). Cycle-sequencing was performed in reaction volumes of 20 µl including the following reagents: distilled water (10.2 µl), 2.5x cycle-sequencing buffer (6 µl), BigDye® Terminator v3.1 cycle-sequencing kit (Applied Biosystems, 2 µl), primer (0.8 µl), and purified PCR product (1 µl). Cycle-sequencing was performed using a Perkin Elmer GeneAmp PCR System 9600 (Applied Biosystems), and comprised 32 cycles of 96°C for 10 seconds, 50°C for 30 seconds, and 60°C for 4 minutes. Cycle-sequencing products were purified using either Ethanol/EDTA/NaAc precipitation or isopropanol precipitation. Nucleotide sequences were determined using an ABI PRISM 3100 Genetic 2

Correspondence: Tshifhiwa C. Nangammbi; e-mail: [email protected]

Analyser (Applied Biosystems), or sent to the MACROGEN Inc. sequencing facility in Seoul, South Korea. Sequence alignment For each taxon, sequences were edited and assembled using the Staden package 2002.0 (Staden et al., 2003) and MEGA version 6.0 (Tamura et al., 2013). COI sequences could be readily aligned by eye because no gaps or stop codons were present. The 16S rRNA sequences were aligned in MEGA using the ClustalW algorithm (Higgins et al., 1994). Ambiguously aligned sites were removed using default parameters at the GBLOCKS server v0.91b (Castresana, 2000), but for comparison, we also constructed a phylogeny that included all sites, with T. pullus as the sole outgroup taxon. Following sequence alignment and the insertion of gaps (where applicable), alignments were 612 bp for COI, and 564 and 429 bp for 16S rRNA before and after GBLOCKS reduction, respectively. Where possible, phylogenetic analysis included several individuals from different localities to represent each southern African Tricolia species, but for T. adusta, T. bicarinata and T. saxatilis, which are rarely found alive, we were only able to include single individuals. Phylogenetic analysis Maximum likelihood analyses were performed using the maximum likelihood algorithm implemented in MEGA with complete deletion of missing characters, nearest-neighbourinterchange and a neighbour-joining tree specified as the starting phylogeny. Clade support was based on 1000 bootstrap replications (BS) (Felsenstein, 1985). Nucleotide substitution models that best describe the substitution patterns of the data were selected in MEGA on the basis of the lowest Bayesian Information Criterion scores (Schwarz, 1978). Phylogenetic trees were reconstructed both for individual markers and for the combined data set. Congruence between data sets was tested using an incongruence length difference (ILD) test (Farris et al., 1995) in PAUP* v4b10 (Swofford, 2003). We included a single arbitrarily selected individual from each species and specified simple taxon addition, TRB searches, holding ten trees at each step, with maxtrees set to 100. The validity of applying tree-based analyses to the individual and combined data sets was assessed using SplitsTree v.4 (Huson & Bryant, 2006), which determines whether or not the data are compatible with a bifurcating tree. To this end, we generated 95% confidence phylogenetic networks based on 1000 bootstrap replications,

using default settings. The hypothesis that the data originated from a tree is supported if the 95% confidence network is a unique tree, and rejected if it contains a number of ambiguous phylogenetic signals, which are represented by parallel edges rather than single branches (Huson & Bryant, 2006). RESULTS Combined COI and 16S rRNA datasets The combined mtDNA (COI and 16S rRNA) data matrix contained 65 specimens and 1038 sites, of which 462 were variable. The ILD test indicated that the data sets were highly compatible (P = 1.0). The maximum likelihood phylogeny derived from this combined dataset recovered all 10 southern African representatives of the genus Tricolia for which data from both markers were available as a monophyletic lineage (Fig. 1). While bootstrap support for the monophyly of this group was weak (BS = 58%), this was a result of the outgroup species T. pullus not clustering in a basal position relative to the ingroup in the 16S rRNA phylogeny based on GBLOCKS-reduced data (Supplementary material, Fig. S2). In contrast, the phylogeny based on the complete 16S rRNA data recovered this species basal to the ingroup (Supplementary material, Fig. S3), and bootstrap support for the monophyly of the southern African species was very strong (BS = 95%) in the COI phylogeny (Supplementary material, Fig. S4). Four species in the combined phylogeny were recovered as being monophyletic lineages, namely T. elongata (BS = 100%), T. neritina (BS = 100%), T. kochii (BS = 99%), and T. formosa (BS = 99%). In contrast, T. bicarinata, T. insignis and T. kraussi were recovered as a mixed group (BS = 98%), as were T. africana and T. capensis (BS = 99%). Maximum likelihood phylogenies based on individual markers were congruent with the phylogeny based on the combined data matrix in failing to discriminate between T. africana and T. capensis, as well as among T. bicarinata, T. insignis and T. kraussi (trees provided in Supplementary material, Figs. S3 and S4). The 16S rRNA phylogenies included a specimen of T. adusta (no COI sequence was available for this species), which was recovered as sister to, but clearly separate from, T. formosa (Supplementary material, Figs. S2 and S3). While the 95% confidence network for the 16S rRNA data resulted in a single, unique tree (Supplementary material, Fig. S5b), several instances of parallel edges were found for both the COI data and the combined data (Supplementary material, Fig. S5a and c, respectively). In conjunction with low bootstrap values in Fig. 1, we conclude that our data do not contain sufficient signal to resolve deeper phylogenetic nodes.

3 Correspondence: Tshifhiwa C. Nangammbi; e-mail: [email protected]

DISCUSSION The mtDNA phylogeny reconstructed in this study supports the delineation of only some of the regionally endemic Tricolia species that have been identified on conchological grounds. In the case of T. elongata, T. formosa, T. kochii, and T. neritina, which were represented by multiple individuals, the taxa grouped as separate, reciprocally monophyletic clades. For T. adusta and T. saxatilis, which were represented by single individuals, the specimens emerged as separate and clearly distinct genetic entities, particularly given the evidence for very low genetic variation within lineages. We are confident, therefore, that in these instances the existing morphology-based species hypotheses are sound and based on genuinely diagnostic characters. Our mtDNA sequence data, however, failed to discriminate between T. bicarinata, T. insignis and T. kraussi, as well as between T. africana and T. capensis. Although each of these multi-species groups appeared as a wellsupported lineage, they exhibited no internal resolution below this level. To explain this conflict between our genetic and morphological data for the nominal species concerned we have to consider that the two mixed-species groups might each represent a single, morphologically variable species. As in many parts of the world, intraspecific variability is common in South African marine gastropods (numerous examples given in Kilburn & Rippey, 1982 and Branch et al., 2010). It may sometimes appear random (e.g. Trochita cingulata, Vaughtia scrobiculata and Bursa granularis), but more commonly it is correlated with ecological or physical parameters such as depth (e.g. Gibbula cicer and Nucella wahlbergi), substratum type (e.g. Semicassis labiata zeylanica), wave action (e.g. Nucella dubia) or water temperature (e.g. Argobuccinum pustulosum, Conus algoensis, Conus mozambicus, Dendrofissurella scutellum, Nassarius kraussianus, Nucella squamosa and Turbo cidaris) (Kilburn & Rippey, 1982; Branch et al., 2010). When correlated with such environmental variables, this plasticity creates what are in fact phenotypically diagnosable ecomorphs, rather than evolutionary distinct species (though in some cases they may have been afforded the status of subspecies). In the case of the multi-species cluster comprising Tricolia bicarinata, T. insignis and T. kraussi, the three nominal species are conchologically similar to each other, with somewhat overlapping or intergrading shell characters. For example, typical specimens of T.

bicarinata and T. insignis are biangulate and possess close-set spiral ridges on the body whorl, though the shell coloration of each is distinctive. In contrast, T. kraussi is at most slightly biangulate and has finer spiral sculpture; in coloration it is closer to T. insignis. The three taxa have parapatric distributions (Fig. 2): T. bicarinata is a coldtemperate west coast species ranging from the Atlantic coast of the Cape Peninsula to Namibia, T. insignis is a warm-temperate south coast species ranging from the KwaZulu-Natal/Eastern Cape border to Hawston (just east of False Bay), and T. kraussi is known only from False Bay. The distributions of T. bicarinata and T. insignis thus coincide with two distinct marine biogeographical provinces, where differing water temperature is known to have a strong influence on the biota (Stephenson & Stephenson, 1972; Brown & Jarman, 1978). T. kraussi occupies an intermediate position in the more sheltered environment of False Bay, at the western end of the Cape Point–Cape Agulhas transitional zone between the warm- and cold-temperate provinces (Brown & Jarman, 1978; Bolton & Anderson, 1997; Teske et al., 2011). For this multi-species group therefore, we clearly have a situation where the regional variation manifest in shell morphology is spatially concordant with regional variation in environmental conditions and might in fact lack a genetic signature. Three similarly distributed morphs are recognised in Conus algoensis (Kilburn & Rippey, 1982).

In the case of the other multi-species group, where there is again very limited genetic differentiation, it is noteworthy that when describing Tricolia africana, Bartsch (1915) commented on similarity with T. capensis, but observed that it differed significantly in size. Certainly T. capensis is considerably larger than T. africana, but compared to T. capensis, typical examples T. africana are also more elongate with a proportionately higher spire and smaller, more circular aperture (e.g. they exhibit a lower rate of aperture expansion), and have more strongly convex whorls (Supplementary material, Fig. S6). They are also commonly more brightly coloured and frequently possess light blue spots below the suture. In addition, the two species differ in terms of their habitat preferences: T. africana lives on and under rocks in mid-shore pools, whereas T. capensis occurs amongst seaweed in low-tide pools. The two taxa are also allopatric; T. africana is a warm-temperate south coast species ranging from the Eastern Cape (Qora River mouth) to the southern Western Cape (Struis Bay, east of Cape Agulhas), whereas T. capensis is a cold-temperate west coast taxon, ranging from the south-western Cape (Hermanus) to northern Namibia. The ranges of the two species are thus separated by a distance of approximately 85 km, 4

Correspondence: Tshifhiwa C. Nangammbi; e-mail: [email protected]

between Struis Bay and Hermanus, an interval that straddles Cape Agulhas, the southernmost point of Africa. As in the preceding case, we thus have a situation where the boundary between the ranges of the two nominal species coincides with a biogeographical break. In this instance, however, the break lies at the eastern end of the transitional zone between the warm- and coldtemperate provinces, rather than in False Bay. In addition, there are behavioural/ecological as well as morphological differences on either side of the boundary. A similarly located break is evident between populations of Conus mozambicus mozambicus and C. m. lautus (Kilburn & Rippey, 1982). To date, the region intermediate between the ranges of T. africana and T. capensis has not been surveyed intensively, and more sampling is needed in order to confirm whether the hiatus between their ranges is real or an artefact resulting from insufficient sampling. In this regard it is interesting to note that although shell coloration differs widely between Eastern Cape (T. africana) and Atlantic Cape (T. capensis) populations, some of the colour patterns seen in shells from either side of Cape Agulhas show distinct similarities (Supplementary material, Fig. S6 H, J), although the size difference remains. The latter, however, may be related to differences in temperature (Atkinson, 1994; Trussell, 2000). Teske et al. (2007) found that specimens of Nassarius kraussianus from the Atlantic Cape coast had significantly larger shells than those from the south and east coasts, and considered this to be temperature-related. Similarly, the colour differences between west and south coast specimens may be influenced by diet (Creese & Underwood, 1976; Underwood & Creese, 1976; Robertson, 1985), and the dull coloration of west coast specimens in both of our multi-species clades (e.g. T. bicarinata and T. capensis) is a feature seen in populations of other vetigastropod taxa from the Atlantic Cape when compared to conspecific populations or sister taxa occurring on the south coast (e.g. Dendrofissurella scutellum scutellum, Gibbula beckeri, Gibbula capensis and Gibbula cicer – Herbert, pers. obs.). Variation in rate of aperture expansion has also been noted in populations of the South African dogwhelk Nucella dubia, where it is thought to be environment-related, the high-spired form acutispira occurring in sheltered habitats (Kilburn & Rippey, 1982; Houart et al., 2010). Clearly there remain questions to address regarding the status of the nominal species comprising these multi-species clades. We believe, however, that the very limited genetic variability evident in these clades (comparable

with that in the single species clades) and the intermixing of the nominal species within them is sufficient justification to consider these clades to be single species entities. Consequently, given that the distributions of the morphologically-delimited nominal taxa are congruent with marine biogeographic provinces, we propose that they be considered phenotypically diagnosable ecomorphs of single species, the oldest available names for which are T. bicarinata and T. capensis. The resultant taxonomic synonymy that this involves will be formalised in a full taxonomic revision of southern African phasianellids that is currently in preparation. Instances such as this, where the analysis of molecular data results in the synonymisation of species, are common in molluscs due to their high levels of phenotypic plasticity (Knowlton, 2000). If our interpretation is correct, then both of these examples run counter to other phylogeographic data from southern Africa, which indicate that most species whose distributions span one or more marine biogeographic disjunctions do in reality comprise multiple genetic lineages, the ranges of which are delimited by the disjunctions (Teske et al., 2011). Thus even under the synonymy proposed above, one would expect some phylogeographic structure congruent with the morphological variation to be evident. The above notwithstanding, there are other southern African examples, including intertidal gastropods, where DNA sequence data reveal no phylogeographic structure across the Atlantic/Indian Ocean biogeographical disjunction (Teske et al., 2007; Matumba, 2013; Mmonwa et al., 2015). However, these studies, like ours, involve only mtDNA and there are examples where mtDNA and nuDNA (and, by extension, the morphological characters coded for by the latter) may produce conflicting phylogenies (Larmuseau et al., 2010; Teske et al., 2014). Such situations may result either from incomplete lineage sorting of mtDNA when sister lineages/species have diverged very recently, or because of mitochondrial capture, e.g. the complete replacement of the mitochondrial genome of one species with that of a closely related species with which it has hybridised (Funk & Omland, 2003). The lack of genetic resolution that we found in our two multi-species clades, where single mtDNA lineages span a pronounced biogeographic disjunction, may thus be a result of mtDNAspecific artefacts (Teske et al., 2014) that do not reflect the environment-related differences that are manifest in morphology, and which might in fact be evident in nuDNA. Investigation of this question in more detail will require a phylogeographic, multi-locus approach including nuDNA data and sequences from more individuals.

5 Correspondence: Tshifhiwa C. Nangammbi; e-mail: [email protected]

Although peripheral to the question of species delineation, the clustering of the conchologically unusual T. neritina amongst the more conventionally shaped Tricolia species is noteworthy. On account of its neritiform shell shape (Fig. 1), this species was afforded its own genus, Chromotis Adams & Adams, 1863, but recent researchers have treated this as a synonym of Tricolia (Keen & Robertson, 1960; Robertson, 1985). Although the relationships of T. neritina are not strongly supported in our phylogeny, the mtDNA data provide no evidence to suggest that the taxon is distinctive at anything above species-level. Our findings are thus consistent with the synonymy of Chromotis with Tricolia. From a biodiversity perspective, the results of this study and the proposed synonymy indicate that the endemic southern Africa Tricolia radiation comprises a total of 10 species: T. adusta, T. bicarinata, T. capensis, T. elongata, T. formosa, T. kochii. T. neritina and T. saxatilis, plus T. retrolineata and T. striolata, for which no sequence data are yet available. If one adds to this the three tropical species extending into KwaZulu-Natal (e.g. T. ios, T. (H.) variabilis and Phasianella solida), then the southern African phasianellid fauna is the richest in the world. ACKNOWLEDGEMENTS We would like to thank the following individuals and their respective institutions for assisting with diving and intertidal collecting: L. Cilliers and colleagues (Oceanographic Research Institute); M. Wallace and V. Fraser; M. Els and M. Jearey; K. Sink; C. Lawrence (Ezemvelo KZN Wildlife); A. Plos and C. Griffiths (University of Cape Town); M. Cole, V. Ndibo, D. Hodgkinson and I. Hartwell (East London Museum); F. Fouché (Ndlabe Environmental Conservation); K. and R. Cook (Knysna). We would also like to thank R. Bowie (Museum of Vertebrate Zoology, Berkeley), A. Moussalli (Victoria Museum) and A. Mitchell (Australian Museum) for assistance in generating molecular sequence data during the initial stages of this project. We thank Suzanne Williams (Natural History Museum of London) for providing additional COI sequence data for Tricolia pullus, Phasianella solida and Hiloa variabilis, and also for specimens of Eulithidium perforatum. Daniel Geiger (Santa Barbara Museum of Natural History, California) also kindly provided specimens of Tricolia variabilis from French Polynesia. We thank the Department of Environmental Affairs and Tourism, marine and coastal management for issuing permits. This research was funded by the South African National Research Foundation, via a South African Biosystematics Initiative grant to DGH (GUN: 2069213). REFERENCES ATKINSON, D. 1994. Temperature and organism size – a biological law for ectotherms? Advances in Ecological Research, 25: 1–58.

BARNARD, K.H. 1963. Contributions to the knowledge of South African marine Mollusca. Part IV. Gastropoda: Prosobranchiata: Rhipidoglossa, Docoglossa. Tectibranchiata. Polyplacophora. Solenogastres. Scaphopoda. Annals of the South African Museum, 47: 201–360. BARTSCH, P. 1915. Report on the Turton collection of South African marine mollusks with additional notes on other South African shells contained in the United States National Museum. Bulletin of the United States National Museum, 91: i–xii, 1– 305, pls 1–54. BOGI, C. & CAMPANI, E. 2007. Tricolia landinii, una nuova specie per le coste orientali della Sicilia. Iberus, 25(1): 27–31. BOLTON, J.J. & ANDERSON, R.J. 1997. Marine vegetation. In: Vegetation of southern Africa (R.M. Cowling, D.M. Richardson & S.M. Pierce, eds), pp. 348-370. Cambridge University Press, Cambridge. BRANCH, G.M., GRIFFITHS, C.L., BRANCH, M.L. & BECKLEY, L.E. 2010. Two Oceans: A guide to the marine life of southern Africa. Revised edition. David Philip, Cape Town. BROWN, A.C. & JARMAN, N. 1978. Coastal marine habitats. In: Biogeography and ecology of southern Africa (M.J.A. Werger, ed.), pp. 1239– 1277. W. Junk, The Hague. CASTRESANA, J. 2000. Selection of conserved blocks from multiple alignments for their use in phylogenetic analysis. Molecular Biology and Evolution, 17: 540–552. CREESE, R.G. & UNDERWOOD, A.J. 1976. Observations on the biology of the trochid gastropod Austrocochlea constricta (Lamarck) (Prosobranchia). I. Factors affecting shellbanding pattern. Journal of Experimental Marine Biology and Ecology, 23: 211–228. FARRIS, J.S. KӒLLERSJӦ, M., KLUGE, A.G. & BULT, C. 1995. Testing significance of incongruence, Cladistics 10: 315–319. FELSENSTEIN, J. 1985. Confidence limits on phylogenies: An approach using the bootstrap. Evolution, 39: 783–791. 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. Molecular Marine Biology and Biotechnology, 3: 294–299. FUNK, D.J. & OMLAND, K.E. 2003. Species-level paraphyly and polyphyly: frequency, causes and consequences, with insights from animal mitochondrial DNA. Annual Review of Ecology, Evolution and Systematics, 34: 397–423. GOFAS, S. 1982. The genus Tricolia in the Eastern Atlantic and the Mediterranean. Journal of Molluscan Studies, 48: 182–213. GOFAS, S. 1986. Taxonomie des Tricolia méditerranéennes. Atti del I congresso della Societá Italiana di Malacologia, Palermo 13-15 settembro 1984. Lavori della Societá Italiana di Malacologia, 22: 179–184. GOFAS, S. 1993. Notes on some Ibero-Moroccan and Mediterranean Tricolia (Gastropoda, Tricoliidae), with descriptions of new species. Journal of Molluscan Studies, 59: 351–361.

6 Correspondence: Tshifhiwa C. Nangammbi; e-mail: [email protected]

HASEGAWA, M., KISHINO, H. & KANO, T. 1985. Dating the human-ape split by a molecular clock of mitochondrial DNA. Journal of Molecular Evolution, 22: 160–174. HICKMAN, C.S. & MCLEAN, J.H. 1990. Systematic revision and suprageneric classification of trochacean gastropods. Science Series, Natural History Museum of Los Angeles County, 35: 1–169. HIGGINS, D., THOMPSON, J., GIBSON, T., THOMPSON, J.D., HIGGINS, D.G. & GIBSON, T.J. 1994. CLUSTAL W: improving the sensitivity of progressive multiple sequence alignment through sequence weighting, position-specific gap penalties and weight matrix choice. Nucleic Acids Research, 22: 4673–4680. HOUART, R., KILBURN, R.N. & MARAIS, J.P. 2010. Muricidae. In: Identification Guide to the Seashells of South Africa (A.P. Marais & A.D. Seccombe, eds), vol. 1, pp. 176–270. Centre for Molluscan Studies: Groenkloof. HUSON, D.H. & BRYANT, D. 2006. Application of phylogenetic networks in evolutionary studies. Molecular Biology and Evolution, 23: 254– 267. KEEN, A.M. & ROBERTSON, R. 1960. Family Phasianellidae Swainson, 1840. In: Moore, R.C. ed. Treatise on invertebrate paleontology. Part I (Mollusca 1), pp. 274– 275. Geological Society of America, Lawrence, Kansas. KILBURN, R.N. & RIPPEY, E. 1982. Sea shells of southern Africa. Macmillan, Johannesburg. KNOWLTON, N. 2000. Molecular genetic analyses of species boundaries in the sea. Hydrobiologia, 420: 73–90. LARMUSEAU, M.H., RAEYMAEKERS, J.A., HELLEMANS, B., VAN HOUDT, J.K. & VOLCKAERT, F.A. 2010. Mito-nuclear discordance in the degree of population differentiation in a marine goby. Heredity, 106: 532–542. MATUMBA, T.G. 2013. Genetics and thermal biology of littorinid snails of the genera Afrolittorina, Echinolittorina and Littoraria (Gastropoda: Littorinidae) from temperate, subtropical and tropical regions. Ph.D. thesis, Rhodes University, Grahamstown, South Africa. MOUSSALLI, A., HUGALL, A.F. & MORITZ, C. 2005. A mitochondrial phylogeny of the rainforest skink genus Saproscincus Wells & Wellington (1984). Molecular Phylogenetics and Evolution, 34: 190–202. MMONWA, K.L., TESKE, P.R., MCQUAID, C.D. & BARKER, N.P. 2015. Historical demography of southern African patellid limpets: congruence of population expansions, but not phylogeography. African Journal of Marine Science, DOI:10.2989/1814232X.2015.1009165. NANGAMMBI, T.C. 2010. Systematics of the Phasianelloidea in southern Africa (Mollusca: Gastropoda: Vetigastropoda). Ph.D. thesis, University of KwaZulu-Natal, Pietermaritzburg, South Africa. NICHOLLS, J.A., DOUBLE, M.C., ROWELL, D.M. & MAGRATH, R.D. 2000. The evolution of

cooperative and pair breeding in thornbills Acanthiza (Pardalotidae). Journal of Avian Biology, 31: 165–176. PALUMBI, S.R., MARTIN, A., ROMANO, S., MCMILLAN, W.O., STICE, L. & GRABOWSKI, G. 1991. The simple fool’s Guide to PCR, Version 2.0. University of Hawaii, Privately published. ROBERTSON, R. 1985. Archaeogastropod biology and the systematics of the genus Tricolia (Trochacea: Tricoliidae) in the Indo-West-Pacific. Monographs of marine Mollusca, 3: 1–103. SCHWARZ, G. 1978. Estimating the dimension of a model. Annals of Statistics, 6: 461–464. STADEN, R., JUDGE, D.P. & BONFIELD, J.K. 2003. Analysing Sequences Using the Staden Package and EMBOSS. Introduction to Bioinformatics. A Theoretical and Practical Approach (A.K. Stephen & D.D. Womble, eds), pp. 393-410. Humana Press Inc., Totowa. STEPHENSON, T.A. & STEPHENSON, A. 1972. Life between Tidemarks on Rocky Shores. W.H. Freeman, San Francisco. SWOFFORD, D.L. 2003. PAUP*: phylogenetic analysis using parsimony, version 4.0 b10. TAMURA, K. 1992. Estimation of the number of nucleotide substitutions when there are strong transition-transversion and G+C content biases. Molecular Biology and Evolution, 9: 678–687. TAMURA, K. & NEI, M. 1993. Estimation of the number of nucleotide substitutions in the control region of mitochondrial DNA in humans and chimpanzees. Molecular Biology and Evolution, 10: 512–526. TAMURA, K., STECHER, G., PETERSON, D., FILIPSKI, A. & KUMAR, S. 2013. MEGA6: Molecular Evolutionary Genetics Analysis version 6.0. Molecular Biology and Evolution, 30: 2725– 2729. TESKE, P.R., PAPADOPOULOS, I., BARKER, N.P., MCQUAID, C.D. & BEHEREGARAY, L.B. 2014. Mitonuclear discordance in genetic structure across the Atlantic/Indian Ocean biogeographic transition zone. Journal of Biogeography, 41: 392–401. TESKE, P.R., PAPADOPOULOS, I., MCQUAID, C.D., NEWMAN, B.K. & BARKER, N.P. 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, 7: e614. TESKE, P.R., VON DER HEYDEN, S., MCQUAID, C.D. & BARKER, N.P. 2011. A review of marine phylogeography in southern Africa. South African Journal of Science, 107: 43–53. TRUSSELL, G.C. 2000. Phenotypic clines, plasticity, and morphological trade-offs in an intertidal snail. Evolution, 54: 151–166. TURTON, W.H. 1932. The marine shells of Port Alfred, South Africa. Oxford University Press, London. UNDERWOOD, A.J. & CREESE, R.G. 1976. Observations on the biology of the trochid gastropod Austrocochlea constricta (Lamarck) (Prosobranchia). II. The effects of available food on shell-banding pattern. Journal of Experimental Marine Biology and Ecology, 23: 229–240.

7 Correspondence: Tshifhiwa C. Nangammbi; e-mail: [email protected]

8 Correspondence: Tshifhiwa C. Nangammbi; e-mail: [email protected]