Hostname: page-component-cb9f654ff-p5m67 Total loading time: 0 Render date: 2025-09-06T19:42:30.104Z Has data issue: false hasContentIssue false
  • English
  • Français

Evaluating morphological species recognition in fossil and modern gastropods (Littorinidae, periwinkles)

Published online by Cambridge University Press:  27 August 2025

Caren P. Shin Open the ORCID record for Caren P. Shin [Opens in a new window]  and
Warren D. Allmon
Caren P. Shin*
Affiliation:
Department of Earth and Atmospheric Sciences, https://ror.org/05bnh6r87 Cornell University , Ithaca, New York 14853, U.S.A. Paleontological Research Institution, Ithaca, New York 14850, U.S.A.
Warren D. Allmon
Affiliation:
Paleontological Research Institution, Ithaca, New York 14850, U.S.A.
*
Corresponding author: Caren Shin; Email: cps257@cornell.edu
Rights & Permissions [Opens in a new window]

Abstract

Species recognition is an essential part of biological and paleontological study. In gastropods, although species are genetic entities, shell morphology continues to be used as the primary source of information to recognize most species. While there are few directly tested cases, variations in conchological characters for modern species are expected to reflect underlying genetic differences that define a biological species, an assumption that is also applied to identify species in the fossil record. Additionally, how consistently shell shape differentiates gastropod species remains poorly understood. In this study, shell shape of Recent and Pliocene–Pleistocene fossil specimens of well-known intertidal gastropods (Littorinidae, periwinkles: †Littorina petricola, Littorina keenae, and the sister-species pair Littorina plena and Littorina scutulata) from the east Pacific was analyzed using landmark-based morphometrics and compared with published molecular data. For the extant species, there is a general positive relationship between shell shape and genetic differences. Discriminant function analyses indicate distantly related species can be more reliably recognized from their shells, while closely related species have a higher error. Fossils and recent specimens were classified with similar consistency. More work is needed to illuminate whether this case applies more widely.

Information

Type
Article
Information
Paleobiology , First View , pp. 1 - 11
Check for updates
Creative Commons
Creative Common License - CCCreative Common License - BYCreative Common License - NCCreative Common License - SA
This is an Open Access article, distributed under the terms of the Creative Commons Attribution-NonCommercial-ShareAlike licence (http://creativecommons.org/licenses/by-nc-sa/4.0), which permits non-commercial re-use, distribution, and reproduction in any medium, provided the same Creative Commons licence is used to distribute the re-used or adapted article and the original article is properly cited. The written permission of Cambridge University Press must be obtained prior to any commercial use.
Copyright
© The Author(s), 2025. Published by Cambridge University Press on behalf of Paleontological Society

Non-technical Summary

Recognizing biological species is important—so much information is unlocked when you can find out the name of what you are observing. Part of identifying a fossil to species is using its shape. We assume visible shape differences indicate invisible genetic differences (not knowable for most fossils), which seems to be true for many living animals. Using marine snails (periwinkles), we find that species are indeed different in shell shape and genetics from each other. We also see that fossil and modern snails can be identified with similar consistency using their shells. These results give us more confidence when identifying species using their shapes.

Introduction

Species are fundamental biological units, whose accurate identification underlies much of ecological and evolutionary investigation. However, there is continued discussion of “the species problem,” as exemplified by a recent survey that suggested that “two randomly chosen respondents will most likely disagree on the nature of species” and use different concepts, largely depending on their research discipline and whether they focus on the micro- or macroevolutionary scale (Stankowski and Ravinet Reference Stankowski and Ravinet2021: R428). There are at least two distinct parts to the issue of “the nature of species”: conceptualizing what a species is in theory and the practical challenge of species recognition (Allmon Reference Allmon2013, Reference Allmon, Allmon and Yacobucci2016), both of which are often obscured by the widely applied but frequently undefined term “species” (e.g., Struck et al. Reference Struck, Feder, Bendiksby, Birkeland, Cerca, Gusarov and Kistenich2018; Shin and Allmon Reference Shin and Allmon2023). A frequently cited key element in species definition is reproductive isolation (after the biological species concept [Mayr Reference Mayr1942]), which might be defined as “a quantitative measure of the effect of genetic differences on gene flow” (Westram et al. Reference Westram, Stankowski, Surendranadh and Barton2022). However, reproductive isolation may be difficult to measure. As a result, generally only taxa suspected or known to have non-morphological distinguishing features (e.g., chemical or behavioral differences [Knowlton Reference Knowlton1993; Bickford et al. Reference Bickford, Lohman, Sodhi, Ng, Meier, Winker, Ingram and Das2007]) might be investigated in detail for an additional suite of data such as phenotypic, life-history, or ecological aspects that might also be considered to aid species discrimination, although this may often be outside the scope of non-taxonomic work. In this paper, we treat species as genetically and phenotypically distinct independent lineages (after the evolutionary species concept [Simpson Reference Simpson1951; Barraclough Reference Barraclough2019]) and concentrate on species recognition, which considers the types of information and methods used to delineate species.

In practice, many extant species have been, and continue to be, described based on morphological characters. For organisms only known from fossils or fossil specimens of extant taxa, species recognition is based largely on morphology (such species are sometimes called “morphological species” or “morphospecies”), in addition to considering biogeographic and geological age information. In the study of morphospecies, a certain degree of morphological distinction is assumed to mirror species-level genetic differences (“good species” [Allmon Reference Allmon, Allmon and Yacobucci2016]), and morphological variation and features of living species are assumed to similarly apply to fossils (following uniformitarianism).

The correspondence between reproductively isolated species identified in modern and fossil morphological species has only been explored in a few case studies, notably in the widely cited textbook example of bryozoans by Jackson and Cheetham (Reference Jackson and Cheetham1990, Reference Jackson and Cheetham1994; included in Foote and Miller [Reference Foote and Miller2006]; see discussion in Allmon [Reference Allmon, Allmon and Yacobucci2016] and Shin and Allmon [Reference Shin and Allmon2023]). In their study, differences in allozyme variation and skeletal features of extant colonies were found to have a statistically significant positive correlation value (~0.7, p < 0.05; see Jackson and Cheetham [Reference Jackson and Cheetham1994: fig. 4, table 3]). Additionally, from statistical analyses, the morphological classification of Recent and fossil colonies was almost the same as the genetic-based designations (for species with available data, 0.4% error [Jackson and Cheetham Reference Jackson and Cheetham1994: table 2]). Other studies have found that dependence on skeletal morphology generally leads to success in recognizing species in fossil and extant specimens of several taxa, although with variable uncertainty (e.g., gastropods [Michaux Reference Michaux1989]; corals [Budd and Pandolfi Reference Budd and Pandolfi2004]; crinoids [Purens Reference Purens2016]; brachiopods [López Carranza and Carlson Reference López Carranza and Carlson2019, Reference López Carranza and Carlson2021]; crocodilians [Brochu and Sumrall Reference Brochu and Sumrall2020]).

A major concern with relying on morphology to distinguish between species is that there are potential areas for inaccuracy, such as for genetically distinct species that do not have any diagnostic morphology (“cryptic species” or “cryptic species sensu stricto,” after Chenuil et al. [Reference Chenuil, Cahill, Délémontey, Du Salliant du Luc, Fanton, Casetta, da Silva and Vecchi2019]; definitions vary, see review by Shin and Allmon [Reference Shin and Allmon2023]) to be overlooked and not counted or for one phenotypically variable species to be identified as multiple species, leading to more species recognized than are present (taxonomic “oversplitting”). Morphologically identical or similar species are often associated with closely related taxa (e.g., “species complexes”) that may have recently diverged, but can also occur for species that are millions of years old due to evolutionary stasis (e.g., Cerca et al. Reference Cerca, Meyer, Stateczny, Siemon, Wegbrod, Purschke, Dimitrov and Struck2019), as well as in distantly related taxa from parallel or convergent evolution (Struck et al. Reference Struck, Feder, Bendiksby, Birkeland, Cerca, Gusarov and Kistenich2018). A lack of morphological differences in diverged species has sometimes been attributed to ecological or environmental influences (e.g., environmental preferences can differentiate sympatric taxa; the same morphology adaptive to a particular environment could occur in separate taxa), or if species may be differentiated based on systems (e.g., behavioral or chemical recognition among individuals, timing of life-history events) that are not reflected in their external morphology. Imprecise species recognition may obscure biodiversity metrics and understanding of evolutionary modes and rates through time (e.g., Bickford et al. Reference Bickford, Lohman, Sodhi, Ng, Meier, Winker, Ingram and Das2007; Allmon and Smith Reference Allmon and Smith2011; Fišer et al. Reference Fišer, Robinson and Malard2018; Struck et al. Reference Struck, Feder, Bendiksby, Birkeland, Cerca, Gusarov and Kistenich2018; Monro and Mayo Reference Monro and Mayo2022).

To address this issue, two main approaches have been taken: reviewing the number or information of nominal, described species compared with evidence for cryptic species at varying taxonomic scales or contrasting genetic with morphological classifications for incongruities, usually among related species or genera (e.g., for particular bryozoans [Jackson and Cheetham Reference Jackson and Cheetham1994]; gastropods [Puillandre et al. Reference Puillandre, Sysoev, Olivera, Couloux and Bouchet2010]; and corals [Sheets et al. Reference Sheets, Warner and Palumbi2018]; few studies have been at a global scale, e.g., planktonic foraminifera [Morard et al. Reference Morard, Darling, Weiner, Hassenrück, Vanni, Cordier and Henry2024]). Taxonomic reviews may include analyses at a class level (e.g., rotifers, where multiple mechanisms toward reproductive isolation are identified [Kordbacheh et al. Reference Kordbacheh, Rahimian and Fontaneto2023]), among orders (e.g., from data on multiple insect orders there may be ~3.1 cryptic species per insect morphospecies [Li and Wiens Reference Li and Wiens2023]), clades (e.g., marine gastropods, with ~2% to 30% of species being cryptic with variable confidence in cryptic species status [Shin and Allmon Reference Shin and Allmon2023]), and habitats (e.g., <1% nominal marine metazoan species have cryptic species [Cahill et al. Reference Cahill, Meglécz and Chenuil2024]). The degree of inaccuracy for species recognition does not seem to be consistent and may be group specific, based on a combination of methodological (e.g., sampling, taxonomic practice, scientific history) and biological factors (e.g., reproductive mode, type of fertilization [Pérez-Ponce de León and Poulin Reference Pérez-Ponce de León and Poulin2016; Shin and Allmon Reference Shin and Allmon2023; Cahill et al. Reference Cahill, Meglécz and Chenuil2024]).

For gastropods, conchological characters (e.g., shell shape, sculpture, and color pattern) remain the primary source of taxonomic information for species description (e.g., Bieler Reference Bieler1992). Indeed, most available material in collections is dry shell specimens, from which genetic measures may not be possible. For marine gastropods, there are 32,000 to 40,000 described species, a number estimated to only be 23% to 32% of their total diversity (Appeltans et al. Reference Appeltans, Ahyong, Anderson, Angel, Artois, Bailly and Bamber2012). Despite the large number of species, marine gastropods have had relatively few cryptic species reported (Knowlton Reference Knowlton1993; Pfenninger and Schwenk Reference Pfenninger and Schwenk2007; Pérez-Ponce de León and Poulin Reference Pérez-Ponce de León and Poulin2016; Chenuil et al. Reference Chenuil, Cahill, Délémontey, Du Salliant du Luc, Fanton, Casetta, da Silva and Vecchi2019; Shin and Allmon Reference Shin and Allmon2023; Cahill et al. Reference Cahill, Meglécz and Chenuil2024), perhaps even less than expected (as modeled by Cahill et al. Reference Cahill, Meglécz and Chenuil2024). A review of extant marine gastropods with adult shells found that while most studies used genetic analyses to separate species, they did not find very many cryptic species sensu stricto, perhaps suggesting a similarly high proportion of these species could be reliably identified from morphology, including fossils (Shin and Allmon Reference Shin and Allmon2023).

The present study evaluates the consistency of species identification using shell shape, and examines the relationship between shell shape and genetic differences for a set of closely related marine gastropods (Littorinidae, periwinkles: Littorina Férussac, Reference Férussac1822). As part of a well-studied extant genus (e.g., Reid Reference Reid1989, Reference Reid1996; McQuaid Reference McQuaid1996a, Reference McQuaidb; Rolán-Alvarez et al. Reference Rolán-Alvarez, Austin and Boulding2015; Johannesson et al. Reference Johannesson, Faria, Le Moan, Rafajlović, Westram, Butlin and Stankowski2024), east Pacific Littorina provide an opportunity to test gastropod species recognition on both modern and fossil specimens. While all living Littorina species are anatomically distinct, there are some species that can be difficult to differentiate from one another based on their shell morphology alone, leading to potential identification error. There may also not be consistent conchological features to diagnose species (e.g., high phenotypic variability due to ecological differentiation or spatial separation, as reviewed in Johannesson et al. [Reference Johannesson, Faria, Le Moan, Rafajlović, Westram, Butlin and Stankowski2024]), and shells may appear similar between species (e.g., between the North Atlantic sister species L. obtusata (Linnaeus, Reference Linnaeus1758) and L. fabalis (W. Turton, Reference Turton1825), which are most reliably differentiated by male reproductive organs), or are highly variable (e.g., between the North Atlantic direct developer L. saxatilis (Olivi, Reference Olivi1792) and its egg-laying relative L. arcana Hannaford-Ellis, Reference Hannaford-Ellis1978; Reid Reference Reid1996), contributing to misidentification.

From analyses of Littorina species with similar shell morphology, estimates of species misclassification vary widely depending on the species compared (e.g., up to ~15% for L. saxatilis and its close relatives [Caley et al. Reference Caley, Grahame and Mill1995; Conde-Padín et al. Reference Conde-Padín, Carvajal-Rodríguez, Carballo, Carballero and Rolán-Alvarez2007]; up to 53% between east Pacific L. plena A. Gould, Reference Gould1849 and L. scutulata A. Gould, Reference Gould1849 [Chow Reference Chow1987]). Nonetheless, there is concordance when anatomical and shell characters and molecular metrics are used in phylogenetic analyses of extant species (Reid Reference Reid, Johannesson, Raffaelli and Ellis1990, Reference Reid1996; Reid et al. Reference Reid, Rumbak and Thomas1996, Reference Reid, Dyal and Williams2012). We therefore expect that Littorina species will be largely recognizable using shell shape and that these distinguishable shell shapes correspond with genetic differences.

Materials and Methods

Case Study: East Pacific Intertidal Littorinidae

The fossil record of Littorina is relatively scarce, with the oldest certain Littorina fossil dating from the Oligocene (†L. sookensis B. L. Clark & Arnold, Reference Clark, Arnold, Lawson and Louderback1923, British Columbia, Canada). For extant species with a fossil record, most are known from the Pliocene and Pleistocene (Reid Reference Reid1996). In this study, four co-occurring east Pacific species were chosen: the extinct †Littorina petricola Arnold, Reference Arnold1908, and the extant L. keenae Rosewater, Reference Rosewater1978, L. plena, and L. scutulata (Figs. 1, 2). These species were chosen because they have relatively abundant and available fossil material and are sympatrically distributed, and the relationship between shell shape variation and species relatedness can be investigated.

Figure 1. Typical shell forms of well-preserved Littorina fossils (A, B, D, F) and modern specimens (C, E, G), all from California, USA. A, †Littorina petricola Arnold, Reference Arnold1908; PRI 76551, Kettleman Hills; Etchegoin Formation, Pliocene. B, Littorina keenae Rosewater, Reference Rosewater1978; LACMIP 5100.71, Point Vincente, Palos Verdes Estates; Quaternary terrace, late Pleistocene. C, Littorina keenae; LACM 66066, Shell Beach; collected in 1958. D, Littorina plena A. Gould, Reference Gould1849; LACMIP 7220.166, Point Loma, San Diego; late Pleistocene. E, Littorina plena; LACM 1948-39.5, Oyster Cove, Tomales Bay; collected from “sand, mud, and stones,” in 1948. F, Littorina scutulata A. Gould, Reference Gould1849; LACMIP 5100.80, Point Vincente, Palos Verdes Estates; Quaternary terrace, late Pleistocene. (G) Littorina scutulata; LACM 1948-37.11, Nick’s Cove, Tomales Bay; collected from intertidal “coarse to fine sand and rock,” in 1948. See Table 1 for institutional abbreviations.

Figure 2. A, Molecular phylogeny of extant Littorina (based on 28S rRNA, 12S rRNA, and cytochrome oxidase c subunit I [COI] genes with fossil calibrations; approximate divergence times as redrawn from Reid et al. [Reference Reid, Dyal and Williams2012: fig. 2]) and B, morphology-based phylogeny of all Littorina (from shell and anatomical characters for extant species and shell only for extinct species, with hypothesized branch times; redrawn from Reid [Reference Reid1996: fig. 115]). Species studied are highlighted. Dotted bars indicate possible ancestral species relationships to extant taxa, and their stratigraphic ranges. Ma, millions of years ago.

†Littorina petricola is only found from the Pliocene of Oregon and California, USA. Although little studied since its description (see Keen and Bentson Reference Keen and Bentson1944; Groves and Squires Reference Groves and Squires2021), †L. petricola is thought to be more closely related to the north Pacific L. squalida Broderip & G. B. Sowerby I, Reference Broderip and Sowerby1829, which is recorded from the Miocene, than to other east Pacific species (Reid Reference Reid1996). While genetic data are not available for this species, †L. petricola is included to compare morphologies with extant species. For the three living species, morphology and molecular analyses have found that L. keenae is distantly related to L. scutulata and L. plena, which are sister species (Murray Reference Murray1979; Mastro et al. Reference Mastro, Chow and Hedgecock1982; Reid Reference Reid1996; Reid et al. Reference Reid, Rumbak and Thomas1996, Reference Reid, Dyal and Williams2012). Littorina keenae is the oldest extant species of the genus, with an estimated divergence from other congenerics in the Eocene, although its fossils are only found from the Pleistocene onward, while L. plena and L. scutulata are a sister-species pair that diverged in the Miocene (Reid Reference Reid1996; Reid et al. Reference Reid, Dyal and Williams2012). All three extant species have planktotrophic larvae and overlap in different intertidal zones, although each species may have its preferences (L. keenae on exposed high shores, L. plena in sheltered habitats, and L. scutulata on exposed coasts [Behrens Yamada Reference Behrens Yamada, Grahame, Mill and Reid1992; Reid Reference Reid1996; Rugh Reference Rugh1997; Hohenlohe Reference Hohenlohe2002, Reference Hohenlohe2003a, Reference Hohenloheb]). When conchologically compared with these extant species, †L. petricola can be distinguished by commonly exhibiting surface reticulation, defined whorls, a thicker shell, and no flattened parietal area, which is distinctive in L. keenae (Reid Reference Reid1996).

While the species status of these extant Littorina are clearly established (Fig. 2), there has been documented inaccuracy when sister species L. plena and L. scutulata are identified based only on their shells, as no single conchological feature, discrete or continuous, can reliably differentiate all individuals of the two species (Supplementary Table 1). Previous work has quantified the percentage of individuals likely misidentified based on shell shape alone using discriminant function analysis on different conchological aspects (e.g., presence or absence of basal band, shell height, whorl number) of L. plena and L. scutulata, with variable results (the lowest error of 4% misrecognized specimens [Murray Reference Murray1982]; the highest error of 53% [Chow Reference Chow1987]). This study uses similar statistical techniques to quantify the potential error of recognizing Littorina species based on shell morphology on a widely sampled fossil and Recent dataset and evaluates the correspondence between morphological and genetic differences for extant species.

Morphological data for three extant and one extinct Littorina species are based on specimens from museum collections (Fig. 3). In total, 1020 modern and 388 fossil specimens from 36 modern and 14 fossil samples were analyzed (Table 1, Supplementary Table 2). Pliocene †L. petricola were sampled from the Etchegoin Formation in California, and Pleistocene fossils of L. keenae, L. plena, and L. scutulata came from Quaternary terrace sites in California and Baja California, Mexico. Species identifications from collections’ labels were verified by examining shell morphology and recorded biogeographic ranges (after Reid Reference Reid1996; see Supplementary Table 1), with three samples redesignated from their museum label identifications (two L. scutulata samples reassigned as L. plena based on locality; one L. keenae sample as L. scutulata from morphology). As it was not possible to confirm the identity of all L. scutulata and L. plena specimens based on conchological features alone, the data for these two species may contain a percentage of mixed or misclassified specimens. The majority of fossil samples (from the Natural History Museum of Los Angeles County) had their identification confirmed by A. Hendy (current curator and expert on east Pacific Cenozoic mollusks [e.g., Hendy Reference Hendy2013]). Modern samples from the Natural History Museum, London, had their identification confirmed by D. G. Reid (formerly based at that institution and taxonomic authority on Littorina [e.g., Reid Reference Reid1996]).

Figure 3. Map of the west coast of North America showing samples, with dotted lines showing state/country boundaries. Species ranges are indicated by the colored bars, based on modern occurrences for extant Littorina keenae, L. petricola, and L. scutulata, and fossil †L. petricola records (after Reid Reference Reid1996). †Littorina petricola is only found in the Pliocene, while the extant species are sampled from the Pleistocene.

Table 1. Summary of studied samples and specimens, listed by species and age. Sample details are listed in Supplemental Table 2. Institutional abbreviations: AMNH, American Museum of Natural History, New York, NY, USA; LACM and LACMIP, Natural History Museum of Los Angeles County, Los Angeles, CA, USA; NHM, Natural History Museum, London, UK; PRI, Paleontological Research Institution, Ithaca, NY, USA; USNM, Smithsonian National Museum of Natural History, Washington, DC, USA

Samples were chosen to encompass the breadth of each species distribution (to account for the potential range in shell morphology and habitat occupied) and for larger sample sizes. Relatively large samples (more than 100 specimens from a locality) were available from the Pleistocene and modern collections for all the extant species. No preservational differences were observed between samples assigned to the middle or late Pleistocene, so further discussion treats all Pleistocene samples similarly. A sample size of approximately 30 specimens per sample was selected as sufficient for capturing shape variation after testing sample sizes 15 to 50 at three modern sites in California that included L. keenae, L. plena, and L. scutulata. From principal component analyses of these species at three sites using different sample sizes, average change in the variance explained by components 1 and 2 was <1% and ~2% for component 3, with a total variance of ~0.008. Covariance matrices between all pairwise comparisons of tested sample sizes were similar (matrix permutation test on landmarks, p < 0.01). To include samples from every part of the extant species’ ranges, smaller samples from underrepresented localities were included (British Columbia, Washington, and Oregon). Recent specimens did not have information on collection date or habitat, so potential sampling effects (e.g., collection from differently exposed environments) could not be controlled for. †Littorina petricola did not have a single sample with more than 30 shells, so specimens from multiple samples of the same locality (Kettleman Hills, California) were combined. Both fossil and modern samples are considered “time averaged” in this study, as only broad temporal bins are discussed (Pliocene, Pleistocene, modern), and fossils in each sample are expected to have accumulated over a period of time (most modern samples appear to be live collected). Incomplete specimens were not used. Each specimen was photographed by C.P.S. following a standard apertural view (after Callomon Reference Callomon2019), using a digital camera set in a copy stand for consistent angle and orientation. For digitization of shell shape from photographs, 12 landmarks were placed on each image (after Carvajal-Rodríguez et al. Reference Carvajal-Rodríguez, Conde-Padín and Rolán-Alvarez2005) in tpsDig2 (v. 2.31; Rohlf Reference Rohlf2015; Fig. 4). Landmarks have been shown to be adequate in capturing overall shell shape variation in Littorina compared with other digitization methods based on shell photographs (e.g., outlines [Doyle et al. Reference Doyle, Gammel and Nash2018]; shape parameters derived from a helicospiral model [Larsson et al. Reference Larsson, Westram, Bengmark, Lundh and Butlin2020]). Shape changes have been successfully investigated using landmarks, especially among the highly variable Littorina saxatilis (e.g., Conde-Padín et al. Reference Conde-Padín, Carvajal-Rodríguez, Carballo, Carballero and Rolán-Alvarez2007; Butlin et al. Reference Butlin, Saura, Charrier, Jackson, André, Caballero and Coyne2014) and when comparing multiple Littorina species (e.g., Maltseva et al. Reference Maltseva, Varfolomeeva, Ayanka, Gafarova, Repkin, Pavlova, Shavarda, Mikhailova and Granovitch2021). Digitization was conducted solely by C.P.S. Landmarking consistency was tested by digitizing the same sample (20 specimens of L. keenae) three times, with no statistically significant differences in means of attempts (T 2 test, all p > 0.05; average shape of each attempt shown in Fig. 4), and digitization attempt explained much less of the shape variation (from Procrustes ANOVA on shape, ~17% variance) than the variance contributed by individuals (~83%). The morphometric data were then Procrustes transformed and statistically analyzed in MorphoJ (Generalized Procrustes Analysis; v. 1.07a [Klingenberg Reference Klingenberg2011]) and R (R Core Team 2023; packages moments [Komsta and Novomestky Reference Komsta and Novomestky2022], ggpubr [Kassambara Reference Kassambara2023], and ggplot2 [Wickham Reference Wickham2016] were used). Centroid size, the square root of the sum of squared distances from the center to all landmarks, was used to approximate specimen sizes. Principal component analyses illustrate shape variation among specimens, and the wireframe diagrams (selected landmarks connected by straight lines for visualization) show the magnitude of shape change among landmarks for each principal component axes. Cross-validated discriminant function analysis between pairs of species was used to estimate the percentage of specimens that would be misclassified based on their shell shape. Mahalanobis distance, the distance between the means of two species shapes while considering their variance and covariance, was also calculated during discriminant function analysis.

Figure 4. A, Landmarking scheme (after Carvajal-Rodríguez et al. [Reference Carvajal-Rodríguez, Conde-Padín and Rolán-Alvarez2005]; landmarks described below) illustrated on an example specimen, Littorina scutulata. LACMIP 6153.2 Rancho Palos Verdes, California, Quaternary terrace, middle Pleistocene. Landmark 1 = shell apex; 2 = right upper side of penultimate whorl’s suture; 3 = midpoint on curve, between landmarks 2 and 4; 4 = lower suture of penultimate whorl; 5 = end of suture; 6 = rightmost external point on shell lip edge; 7 = lowest point on shell base; 8 = internal border of aperture; 9 = external border of aperture, perpendicular to 8; 10 = midpoint on shell edge, between landmarks 7 and 11; 11 = leftmost external point on penultimate whorl; 12 = left side of penultimate whorl’s suture. B, Example of landmarking consistency with average configurations of three digitization attempts on the same L. keenae sample (black, purple, and pink outlines). LACMIP, Natural History Museum of Los Angeles County, Los Angeles, CA, USA.

For the genetic data, published sequences for the extant species were taken from GenBank (1 specimen per species, for 28S rRNA, 12S rRNA, and cytochrome oxidase c subunit I [COI] genes, respectively, as cited in Reid et al. [Reference Reid, Dyal and Williams2012]; Supplementary Table 3). Sequences were aligned with MUSCLE (Edgar Reference Edgar2004; with default parameters of −400 gap opening penalty and unweighted pair group method with arithmetic mean clustering), and pairwise p distances, the number of base pair differences per site between two sequences, were calculated in MEGA 11 (Tamura et al. Reference Tamura, Stecher and Kumar2021).

Results and Discussion

Fossil and Modern Littorina

Overall, Pleistocene fossils of the extant species were smaller and less variable in size than modern individuals, while Pliocene †L. petricola was larger (Fig. 5A). As previously documented, †L. petricola and L. keenae were commonly larger than L. plena or L. scutulata, and L. plena specimens tended to be smaller than L. scutulata (Supplementary Table 1). Centroid sizes were not normally distributed for L. keenae, L. plena, or L. scutulata (Shapiro-Wilk normality test, p < 0.05), but were for †L. petricola (p = 0.2). Among species, centroid sizes were significantly different (Kruskal-Wallis χ2= 45.1, df = 3, p < 0.05), and pairwise comparisons between all species were statistically significant (Wilcoxon rank-sum test with Bonferroni correction, p < 0.05), except between L. plena and L. scutulata (p = 0.08). Although Pleistocene fossil and Recent centroid sizes of all three extant species overlap, they were statistically different for each species (L. keenae, Kruskal-Wallis χ2 = 51.9, df = 1, p < 0.01; L. plena, Kruskal-Wallis χ2 = 33.1, df = 1, p < 0.01; L. scutulata, Kruskal-Wallis χ2 = 15.0, df = 1, p < 0.01), but it is unclear if these size differences reflect time-averaging taphonomic processes or indicate biological change over time. There was no statistically significant difference in sizes for L. keenae across sampled regions (Kruskal-Wallis χ2 = 1.2, df = 1, p = 0.3), but there was for L. plena (Kruskal-Wallis χ2 = 189.3, df = 5, p < 0.01) and L. scutulata (Kruskal-Wallis χ2 = 48.6, df = 5, p < 0.01; Fig. 5B). While there were no regions that had statistically significant differences with all others, L. plena from Alaska (the northern extent of the species) and L. scutulata from Baja California (the southern extent of the species) differed from the most regions when pairwise comparisons were conducted (Wilcoxon rank-sum test with Bonferroni correction, p < 0.05).

Figure 5. A, Centroid size distribution of studied Littorina; dotted plots are sizes of fossil specimens, solid-colored plots are for Recent specimens. B, Centroid size distribution of extant Littorina, grouped by region, listed from north to south. AK, Alaska; BC, British Columbia; WA, Washington; OR, Oregon; CA, California; MX, Baja California.

Principal component analyses of fossil and Recent Littorina specimens illustrate groups of shell shapes according to species, particularly separating L. keenae from closely overlapping L. plena and L. scutulata individuals on principal component 1 (PC 1; 52.7% variance), while there was slight differentiation of fossil from modern material on PC 2 (12.0%) and PC 3 (7.8%; Fig. 6). Pliocene †L. petricola did not form a discrete morphological cluster, with the landmarked shell shapes plotting among L. plena and L. scutulata specimens, and showed the least variability compared with the extant species, although this may also be due to the limited sampling. Adding L. squalida, the closest extant relative to †L. petricola in future analyses could provide another species pair comparison to evaluate this study’s approach.

Figure 6. Principal component analysis of landmark data for all specimens of the three extant and one extinct Littorina species. Polygons enclose the points for each species. Wireframe diagrams show shape changes for each principal component (PC 1, PC 2, PC 3), from minimum (purple) to maximum values (black) along that axis.

Centroid sizes of Littorina species were not correlated with the morphological data in PC 1, PC 2, or PC 3 (when a simple linear regression of principal component axes with each species’ sizes are conducted, all adjusted R 2 < 0.5; Supplementary Fig. 1). Principal component analyses that considered modern (55.4% variance on PC 1, 12.1% for PC 2, 7.5% for PC 3; Supplementary Fig. 2) or fossil (51.7% variance for PC 1, 10.4% on PC 2, 7.8% on PC 3; Supplementary Fig. 3) material separately exhibit results similar to the analysis that included all specimens. Shell shape variation along the principal component axes could not be attributed to any singular landmark, although for all analyses there is an overall rounder, turbinate shape, and apex height change on PC 1, which is reflective of L. keenae, compared with more conical forms of L. plena and L. scutulata for PC 2 and PC 3 (Fig. 6). It has been suggested that overall shell shape in marine gastropods may be adaptive for temperature regulation in different intertidal zones (e.g., taller spires at higher shore heights [Vermeij Reference Vermeij1973]), although this has not been shown for the extant species in this study (L. keenae [Lee and Boulding Reference Lee and Boulding2010]; no specific data on L. scutulata or L. plena). Spire height has been specifically associated with predation risk (e.g., crab predation selects for smaller spires [Seeley Reference Seeley1986]), varying growth rates (e.g., faster growth may result in taller spires [Kemp and Bertness Reference Kemp and Bertness1984; Boulding and Hay Reference Boulding and Hay1993]), and perhaps a larger foot area, which would be advantageous against wave action (for L. scutulata and L. plena [Hohenlohe Reference Hohenlohe2003a]). It is unclear whether species differences in spire height are adaptive for the studied taxa. Including semilandmarks on certain shell outline sections may complement our current shape-digitization approach.

Little morphological change was observed for extant Littorina species, as Pleistocene and Recent specimens had comparable ranges of Mahalanobis distances (Table 2). As expected, the more phylogenetically distant L. keenae had larger Mahalanobis distances to L. plena and L. scutulata (3.4 to 4.2) than between the sister species L. plena and L. scutulata (1.1 with modern specimens, 1.5 with fossils). Despite a smaller fossil (388 specimens) than Recent dataset (1020 specimens), Pleistocene and modern L. keenae display similar overall ranges in shell shape variation, while Pleistocene L. plena and L. scutulata appear to have less variable shell shapes compared with their Recent specimens (modern data in Supplementary Fig. 2, fossils shown in Supplementary Fig. 3). It is ambiguous whether there are definite biological or environmental factors reducing fossil morphological variation, other than potential sampling effects. However, L. plena and L. scutulata have analogous ranges of overall shell shape variation, which is slightly different from previous findings (compared with L. scutulata, L. plena had smaller shell shape variation [Murray Reference Murray1982]; or L. plena has larger shell variation [Reid Reference Reid1996]). The spread of morphological variation within L. plena and L. scutulata on PC 1 and to a lesser extent on PC 2 and PC 3 may be attributed to geographic region, although there do not seem to be any strong latitudinal trends or discrete groupings of shapes, as there is overlap in specimens from disjunct areas (e.g., individuals from British Columbia overlay with those from California). As noted by Reid (Reference Reid1996), higher-spired specimens were found at the northern range of L. plena, and individuals of L. plena from Alaska, British Columbia, and Washington did cluster more toward the maximal values of PC 2 and PC 3, although morphological clusters of both L. plena and L. scutulata overlapped with each other within every sampled region. Nevertheless, intraspecific variability in shell shape for L. plena and L. scutulata appears to be continuous, suggesting there may be little effect of time averaging on shape data (e.g., as in other studies, reviewed by Kidwell and Holland [Reference Kidwell and Holland2002]). No consistent shell forms associated with specific environments (ecotypes, defined as “a phenotypically and genotypically distinct form of a species that is adapted to a particular habitat” [Johannesson et al. Reference Johannesson, Faria, Le Moan, Rafajlović, Westram, Butlin and Stankowski2024]) have been described for any of the study species (Reid Reference Reid1996). Planktotrophic larval development with relatively high dispersal capabilities for all three extant species may explain the continuity in the extent of shell shapes, as populations in the species distribution are generally connected (no spatial population genetic structure was found for L. keenae, L. plena, or L. scutulata [Lee and Boulding Reference Lee and Boulding2007, Reference Lee and Boulding2009, Reference Lee and Boulding2010]).

Table 2. Mahalanobis distances between pairs of Littorina species (left quadrant, all pairwise comparisons p < 0.01, T 2 test) and percentage of misclassified specimens from cross-validated discriminant function analysis (right quadrant) based on all specimens, modern shells, or fossils only

Evaluating Morphological Species Recognition

Larger Mahalanobis distances between pairwise species comparisons corresponded with higher accuracy in classifying specimens from cross-validated discriminant function analysis, demonstrating that shell shape as captured through landmarking can sufficiently differentiate among species (Table 2). The largest differences between means involved †L. petricola and the three extant species (Mahalanobis distances from 6.4 to 7.0), and the smallest between L. plena and L. scutulata (0.7), with all pairwise comparisons being statistically significant (T 2 test, p < 0.01). Nearly all specimens of †L. petricola could be distinguished from L. keenae, L. plena, and L. scutulata (<1% misclassified), while the highest error was between L. plena and L. scutulata (37%). A few individuals of L. plena and L. scutulata were misclassified when compared with L. keenae (4%). A similar trend in Mahalanobis distances and misclassification percentages between species was also found when only fossils or modern specimens were considered. These comparisons of more distantly related species have larger shape differences, and lower identification error is expected, as demonstrated by previous phylogenetic work based on conchological and soft anatomical characters (Fig. 2).

The amount of error in conchologically recognizing L. plena and L. scutulata (32% to 37%) is generally higher but still in the range of those previously reported using solely modern material, suggesting there is more shape variability captured when sampling widely across the species distribution. However, some error may also be due to some misidentification of species from collections labels. With different data, several studies have used discriminant function analysis to quantitatively compare the accuracy of classifying confirmed L. plena and L. scutulata (collected live, anatomical features observed) with conchological identification. Murray (Reference Murray1982) reported a 16% error in distinguishing L. plena based on its generally smaller size and more frequent appearance of a basal band, although this misclassification percentage could be reduced to 4% if shell length, depth, whorl height, and whorl number are measured (samples from California and Oregon). Misclassification percentages in a similar range were found for the two species based on measured shell length, width, whorl number, and presence of an amber band and tessellations (5% for L. plena and 11% for L. scutulata, from Bodega Bay, California [Chow Reference Chow1987]). Much higher error percentages were reported for L. plena (34%) and L. scutulata (31%) from British Columbia and Washington based on measurements of shell height, spire angle, aperture angle, whorl ratio, and aperture shape (Hohenlohe and Boulding Reference Hohenlohe and Boulding2001). The highest misclassification documented was 53% when classifying L. plena between open and protected shores at the same site (Chow Reference Chow1987). Adding discrete conchological characters such as color (e.g., the presence of a pale basal band and checkered patterning) would increase the accuracy of modern species identification between L. plena and L. scutulata (Murray Reference Murray1982; Chow Reference Chow1987; Rugh Reference Rugh1997; Hohenlohe and Boulding Reference Hohenlohe and Boulding2001; Supplementary Table 1), although this would not be possible for most fossil specimens.

Species-Level Morphological and Genetic Correlation

Increasing morphological distances (Mahalanobis distance) correspond with genetic p distances (from 28S rRNA, 12S rRNA, COI), as seen from pairwise comparisons between L. keenae, L. plena, and L. scutulata, supporting the use of shell shape in recognizing species with modern or fossil material (Fig. 7). This reflects phylogenetic analyses of Littorina based on morphological and molecular metrics, which are in close agreement (Reid Reference Reid, Johannesson, Raffaelli and Ellis1990, Reference Reid1996; Reid et al. Reference Reid, Rumbak and Thomas1996, Reference Reid, Dyal and Williams2012; Fig. 2).

Figure 7. Morphological distance (Mahalanobis distance, the distance of a central point and distribution of landmark data from modern and fossil specimens for that species) and genetic distance (number of base differences per site from published 28S rRNA, 12S rRNA, and cytochrome oxidase c subunit 1 [COI] sequences for each species) from pairwise comparisons for Littorina keenae (k), L. plena (p), and L. scutulata (s). Note that morphological distance for all the plots is the same; only the x-axes have changed.

Our results are similar to the highly correlated morphological (Mahalanobis distance, from fossil and Recent bryozoans) and genetic distances (Nei’s unbiased D, from allozymes) in the frequently referenced study by Jackson and Cheetham (Reference Jackson and Cheetham1994). For the Littorina studied here, morphological differences are quantified by Mahalanobis distances, and uncertainty in species identification is gauged by number of misclassified specimens between species, with higher error found to be correlated with how closely related species are. When combined with other approaches to evaluating the potential inaccuracy of relying on morphospecies recognition, such as taxonomic reviews of species status and comparisons of morphospecies and genetic-based taxonomies, it seems that there is a range of potential error that will likely vary and depend on the biology of the taxonomic group and its history of study (e.g., Shin and Allmon Reference Shin and Allmon2023; Cahill et al. Reference Cahill, Meglécz and Chenuil2024), emphasizing the importance of evaluating the consistency of species identification for specific groups.

Conclusions

  1. 1. Shell shape, as quantified by landmarking, can be used to distinguish among east Pacific Littorina species. From cross-validated discriminant function analysis, the number of misclassified specimens between two species is lower when the species are more distantly related (from <1% to 5%) and highest between sister species (32% to 37%).

  2. 2. For extant Littorina, Pleistocene and Recent specimens are conchologically similar and can be recognized with comparable accuracy. Although Pleistocene Littorina specimens were less morphologically variable than modern individuals, the range of fossil variation was contained within the Recent data, and all species had continuous ranges of morphological variation. Although the extinct species from the Pliocene morphologically overlapped with the studied extant species, it could be distinguished with other shell characters.

  3. 3. Differences in Littorina shell shape are correlated with genetic distances, giving confidence to recognizing Littorina based solely on morphology, and this may also apply to other gastropod species. Quantifying the variability in the relationship between morphological and genetic differences for specific taxa may add to our understanding of the potential error when recognizing morphospecies, and elucidate prospective causes in cases where morphological and genetic differentiation are inconsistent.

Acknowledgments

For helpful discussion, we thank the Littorina research group, PRI Paleogeeks group, K. Johannesson, M. Johnson, and D. Reid. We appreciate the assistance from Cornell University Library staff. For their support with collections, we are grateful to G. Dietl, J. Hendricks, L. Skibinski, and V. Wang (Paleontological Research Institution); L. Groves, A. Hendy, and J. Hook (Natural History Museum of Los Angeles County); J. Ablett, K. Collins, and A. Salvador (Natural History Museum, London); J. Goodheart, B. M. Hussaini, H. Ketchum, and L. Rincón-Rodriguez (American Museum of Natural History); and N. Drew, S. Edie, K. Reed, A. Robinson, E. Strong, and Y. Villacampa (Smithsonian National Museum of Natural History). Travel to collections was supported by grants from Cornell University’s Einaudi Center and Graduate School and Conchologists of America. Thanks to W. Bemis, G. Dietl, J. Hendricks, and three reviewers for their comments, which improved this article.

Competing Interests

The authors declare no competing interests.

Data Availability Statement

Supplementary Tables 1–3 and Supplementary Figs. 1–3 are available at: https://zenodo.org/records/16884479

Footnotes

Handling Editor: John Huntley

References

Literature Cited

Allmon, W. D. 2013. Species, speciation, and paleontology up to the Modern Synthesis: persistent themes and unanswered questions. Palaeontology 56:1199–1223.10.1111/pala.12054CrossRefGoogle Scholar
Allmon, W. D. 2016. Studying species in the fossil record: a review and recommendations for a more unified approach. Pp. 59–120 in Allmon, W. D. and Yacobucci, M. M., eds., Species and speciation in the fossil record. University of Chicago Press, Chicago.10.7208/chicago/9780226377582.003.0003CrossRefGoogle Scholar
Allmon, W. D., and Smith, U. E.. 2011. What, if anything, can we learn from the fossil record about speciation in marine gastropods? Biological and geological considerations. American Malacological Bulletin 29:247–276.10.4003/006.029.0214CrossRefGoogle Scholar
Appeltans, W., Ahyong, S. T., Anderson, G., Angel, M. V., Artois, T., Bailly, N., Bamber, R., et al. 2012. The magnitude of global marine species diversity. Current Biology 22:2189–2202.CrossRefGoogle ScholarPubMed
Arnold, R. 1908. Descriptions of new Cretaceous and Tertiary fossils from the Santa Cruz Mountains, California. Proceedings of the United States National Museum 34:345–389.CrossRefGoogle Scholar
Barraclough, T. G. 2019. The evolutionary biology of species, Oxford University Press, Oxford.10.1093/oso/9780198749745.001.0001CrossRefGoogle Scholar
Behrens Yamada, S. 1992. Niche relationships in northeastern Pacific Littorines. Pp. 281–291 in Grahame, J., Mill, P. J., and Reid, D. G., eds., Proceedings of the Third International Symposium on Littorinid Biology. Malacological Society of London, London.Google Scholar
Bickford, D., Lohman, D. J., Sodhi, N. S., Ng, P. K. L., Meier, R., Winker, K., Ingram, K. K., and Das, I.. 2007. Cryptic species as a window on diversity and conservation. Trends in Ecology and Evolution 22:148–155.10.1016/j.tree.2006.11.004CrossRefGoogle ScholarPubMed
Bieler, R. 1992. Gastropod phylogeny and systematics. Annual Review of Ecology, Evolution, and Systematics 23:311–338.CrossRefGoogle Scholar
Boulding, E. G., and Hay, T. K.. 1993. Quantitative genetics of shell form of an intertidal snail: constraints on short-term response to selection. Evolution 47:576–592.CrossRefGoogle ScholarPubMed
Brochu, C. A., and Sumrall, C. D.. 2020. Modern cryptic species and crocodilian diversity in the fossil record. Zoological Journal of the Linnean Society 189:700–711.10.1093/zoolinnean/zlaa039CrossRefGoogle Scholar
Broderip, W. J., and Sowerby, G. B. I.. 1829. Observations on new or interesting Mollusca contained, for the most part, in the Museum of the Zoological Society. Zoological Journal 4:359–379.Google Scholar
Budd, A. F., and Pandolfi, J. M.. 2004. Overlapping species boundaries and hybridization within the Montastraea “annularis” reef coral complex in the Pleistocene of the Bahama Islands. Paleobiology 30:396–425.10.1666/0094-8373(2004)030<0396:OSBAHW>2.0.CO;22.0.CO;2>CrossRefGoogle Scholar
Butlin, R. K., Saura, M., Charrier, G., Jackson, B., André, C., Caballero, A., Coyne, J. A., et al. 2014. Parallel evolution of local adaptation and reproductive isolation in the face of gene flow. Evolution 68:935–949.10.1111/evo.12329CrossRefGoogle ScholarPubMed
Cahill, A. E., Meglécz, E., and Chenuil, A.. 2024. Scientific history, biogeography, and biological traits predict presence of cryptic or overlooked species. Biological Reviews 99:546–561.10.1111/brv.13034CrossRefGoogle ScholarPubMed
Caley, K. J., Grahame, J., and Mill, P. J.. 1995. A geographically-based study of shell shape in small rough periwinkles. Hydrobiologia 309:181–193.10.1007/BF00014486CrossRefGoogle Scholar
Callomon, P. 2019. Standard views for imaging mollusk shells. American Malacological Society. https://ams.wildapricot.org/resources/Documents/Standard%20views%20for%20imaging%20mollusk%20shells%20for%20web%202019.pdf.Google Scholar
Carvajal-Rodríguez, A., Conde-Padín, P., and Rolán-Alvarez, E.. 2005. Decomposing shell form into size and shape by geometric morphometric methods in two sympatric ecotypes of Littorina saxatilis. Journal of Molluscan Studies 71:313–318.10.1093/mollus/eyi037CrossRefGoogle Scholar
Cerca, J., Meyer, C., Stateczny, D., Siemon, D., Wegbrod, J., Purschke, G., Dimitrov, D., and Struck, T. H.. 2019. Deceleration of morphological evolution in a cryptic species complex and its link to paleontological stasis. Evolution 74:116–131.CrossRefGoogle Scholar
Chenuil, A., Cahill, A. E., Délémontey, N., Du Salliant du Luc, E., and Fanton, H.. 2019. Problems and questions posed by cryptic species. A framework to guide future studies. Pp. 77–106 in Casetta, E., da Silva, J. M. and Vecchi, D., eds., From assessing to conserving biodiversity: conceptual and practical challenges. History, Philosophy and Theory of the Life Sciences 24. SpringerOpen.CrossRefGoogle Scholar
Chow, V. 1987. Morphological classification of sibling species of Littorina (Gastropoda: Prosobranchia): discretionary use of discriminant analysis. The Veliger 29:359–366.Google Scholar
Clark, B. L., and Arnold, R.. 1923. Fauna of the Sooke Formation, Vancouver Island with description of a new coral by T. Wyland Vaughan. Pp. 123–234 in Lawson, A. C. and Louderback, G. D., eds., University of California Publications in Geological Sciences, Vol. 14. University of California Press, Berkeley.Google Scholar
Conde-Padín, P., Carvajal-Rodríguez, A., Carballo, M., Carballero, A., and Rolán-Alvarez, E.. 2007. Genetic variation for shell traits in a direct-developing marine snail involved in a putative sympatric ecological process. Evolutionary Ecology 21:635–650.CrossRefGoogle Scholar
Doyle, D., Gammel, M. P., and Nash, R.. 2018. Morphometric methods for the analysis and classification of gastropods: a comparison using Littorina littorea. Journal of Molluscan Studies 84:190–197.10.1093/mollus/eyy010CrossRefGoogle Scholar
Edgar, R. C. 2004. MUSCLE: multiple sequence alignment with high accuracy and high throughput. Nucleic Acids Research 32:1792–1797.10.1093/nar/gkh340CrossRefGoogle ScholarPubMed
Férussac, A. E. 1822. Tableaux systématiques des animaux mollusques classés en familles naturelles, dans lesquels on a établi la concordance de tous les systèmes; suivis d’un prodrome général pour tous les mollusques terrestres ou fluviatiles, vivants ou fossils. A. Bertrand. https://www.biodiversitylibrary.org/item/41533.Google Scholar
Fišer, C., Robinson, C. T., and Malard, F.. 2018. Cryptic species as a window into the paradigm shift of the species concept. Molecular Ecology 27:613–635.CrossRefGoogle ScholarPubMed
Foote, M., and Miller, A. I.. 2006. Principles of paleontology, 3rd ed. Freeman, New York.Google Scholar
Gould, A. 1849. Proceedings of the Boston Society of Natural History, Vol. 3. Boston Society of Natural History. https://www.biodiversitylibrary.org/item/35741.Google Scholar
Groves, L. T., and Squires, R. L.. 2021. Checklist of California Paleogene–Neogene marine Mollusca since Keen and Bentson (1944). PaleoBios 38:1–360.CrossRefGoogle Scholar
Hannaford-Ellis, C. J. 1978. Littorina arcana sp. nov.: a new species of winkle (Gastropoda: Prosobranchia: Littorinidae. Journal of Conchology 29: 304.CrossRefGoogle Scholar
Hendy, A. J. W. 2013. Spatial and stratigraphic variation of marine paleoenvironments in the middle-upper Miocene Gatun formation, isthmus of Panama. Palaios 28:210–227.CrossRefGoogle Scholar
Hohenlohe, P. A. 2002. Life history of Littorina scutulata and L. plena, sibling gastropod species with planktotrophic larvae. Invertebrate Biology 121:25–37.10.1111/j.1744-7410.2002.tb00126.xCrossRefGoogle Scholar
Hohenlohe, P. A. 2003a. Distribution of sister Littorina species, I: tenacity and the wave-exposure gradient. The Veliger 46:162–168.Google Scholar
Hohenlohe, P. A. 2003b. Distribution of sister Littorina species, II: geographic and tidal-height pattern do not support sympatric speciation. The Veliger 46:211–219.Google Scholar
Hohenlohe, P. A., and Boulding, E. G.. 2001. A molecular assay identifies morphological characters useful for distinguishing the sibling species Littorina scutulata and L. plena. Journal of Shellfish Research 20:453–457.Google Scholar
Jackson, J. B. C., and Cheetham, A. H.. 1990. Evolutionary significance of morphospecies: a test with cheilostome Bryozoa. Science 248:579–583.10.1126/science.248.4955.579CrossRefGoogle ScholarPubMed
Jackson, J. B. C., and Cheetham, A. H.. 1994. Phylogeny reconstruction and the tempo of speciation in cheilostome Bryozoa. Paleobiology 20:407–423.10.1017/S0094837300012902CrossRefGoogle Scholar
Johannesson, K., Faria, R., Le Moan, A., Rafajlović, M., Westram, A. M., Butlin, R. K., and Stankowski, S.. 2024. Diverse pathways to speciation revealed by marine snails. Trends in Genetics 40:337–351.10.1016/j.tig.2024.01.002CrossRefGoogle ScholarPubMed
Kassambara, A. 2023. ggpubr: ‘ggplot2’ based publication ready plots. R package version 0.6.0. https://rpkgs.datanovia.com/ggpubr.Google Scholar
Keen, A. M., and Bentson, H.. 1944. Checklist of California Tertiary marine Mollusca. Geological Society of America Special Paper 56:1–280.CrossRefGoogle Scholar
Kemp, P., and Bertness, M. D.. 1984. Snail shape and growth rates: evidence for plastic shell allometry in Littorina littorea. Proceedings of the National Academy of Sciences USA 81:811–813.CrossRefGoogle ScholarPubMed
Kidwell, S. M., and Holland, S. M.. 2002. The quality of the fossil record: implications for evolutionary analyses. Annual Review of Ecology and Systematics 33:561–588.10.1146/annurev.ecolsys.33.030602.152151CrossRefGoogle Scholar
Klingenberg, C. P. 2011. MorphoJ: an integrated software package for geometric morphometrics. Molecular Ecology Resources 11:353–357.10.1111/j.1755-0998.2010.02924.xCrossRefGoogle ScholarPubMed
Knowlton, N. 1993. Sibling species in the sea. Annual Review of Ecology and Systematics 24:189–216.10.1146/annurev.es.24.110193.001201CrossRefGoogle Scholar
Komsta, L., and Novomestky, F.. 2022. moments: moments, cumulants, skewness, kurtosis and related tests. R package version 0.14.1. http://doi.org/10.32614/CRAN.package.moments.CrossRefGoogle Scholar
Kordbacheh, A., Rahimian, H., and Fontaneto, D.. 2023. Mechanisms of reproductive isolation among cryptic species in monogonont rotifers. Hydrobiologia 850:4705–4718.10.1007/s10750-022-05131-0CrossRefGoogle Scholar
Larsson, J., Westram, A. M., Bengmark, S., Lundh, T., and Butlin, R. K.. 2020. A developmentally descriptive method for quantifying shape in gastropod shells. Journal of the Royal Society Interface 17:20190721.10.1098/rsif.2019.0721CrossRefGoogle Scholar
Lee, H. J., and Boulding, E. G.. 2007. Mitochondrial DNA variation in space and time in the northeastern Pacific gastropod, Littorina keenae. Molecular Ecology 16:3084–3103.10.1111/j.1365-294X.2007.03364.xCrossRefGoogle ScholarPubMed
Lee, H. J., and Boulding, E. G.. 2009. Spatial and temporal population genetic structure of four northeastern Pacific littorinid gastropods: the effect of mode of larval development on variation at one mitochondrial and two nuclear DNA markers. Molecular Ecology 18:2165–2184.10.1111/j.1365-294X.2009.04169.xCrossRefGoogle ScholarPubMed
Lee, H. J., and Boulding, E. G.. 2010. Latitudinal clines in body size, but not in thermal tolerance or heat-shock cognate 70 (HSC70), in the highly dispersing intertidal gastropod Littorina keenae (Gastropoda: Littorinidae). Biological Journal of the Linnean Society 100:494–505.10.1111/j.1095-8312.2010.01450.xCrossRefGoogle Scholar
Li, X., and Wiens, J. J.. 2023. Estimating global biodiversity: the role of cryptic insect species. Systematic Biology 72:391–403.10.1093/sysbio/syac069CrossRefGoogle ScholarPubMed
Linnaeus, C. 1758. Systema Naturae per regna tria naturae, secundum classes, ordines, genera, species, cum characteribus, differentiis, synonymis, locis, Vol. 1. Impensis Direct. Laurentii Salvii. https://www.biodiversitylibrary.org/item/10277.Google Scholar
López Carranza, N., and Carlson, S. J.. 2019. Testing species assignments in extant terebratulide brachiopods: a three-dimensional geometric morphometric analysis of long-looped brachidia. PLoS ONE 14:e0225528.CrossRefGoogle ScholarPubMed
López Carranza, N., and Carlson, S. J.. 2021. Quantifying shell outline variability in extant and fossil Laqueus (Brachiopoda: Terebratulida): are outlines good proxies for long-looped brachidial morphology and can they help us characterize species? Paleobiology 47:149–170.10.1017/pab.2020.56CrossRefGoogle Scholar
Maltseva, A. L., Varfolomeeva, M. A., Ayanka, R. V., Gafarova, E. R., Repkin, E. A., Pavlova, P. A., Shavarda, A., Mikhailova, N. A., and Granovitch, A. I.. 2021. Linking ecology, morphology, and metabolism: niche differentiation in sympatric populations of closely related species of the genus Littorina (Neritrema). Ecology and Evolution 11:11134–11154.10.1002/ece3.7901CrossRefGoogle ScholarPubMed
Mastro, E., Chow, V., and Hedgecock, D.. 1982. Littorina scutulata and Littorina plena: sibling species status of two prosobranch gastropod species confirmed by electrophoresis. The Veliger 24:239–246.Google Scholar
Mayr, E. 1942. Systematics and the origin of species: from a viewpoint of a zoologist. Columbia University Press, New York.Google Scholar
McQuaid, C. D. 1996a. Biology of the gastropod family Littorinidae. I. Evolutionary aspects. Oceanography and Marine Biology: An Annual Review 34:233–262.Google Scholar
McQuaid, C. D. 1996b. Biology of the gastropod family Littorinidae. II. Role in the ecology of intertidal and shallow marine ecosystems. Oceanography and Marine Biology: An Annual Review 34:263–302.Google Scholar
Michaux, B. 1989. Morphological variation of species through time. Biological Journal of the Linnean Society 38:239–255.10.1111/j.1095-8312.1989.tb01577.xCrossRefGoogle Scholar
Monro, A. K., and Mayo, S. J., eds. 2022. Cryptic species: morphological stasis, circumscription, and hidden diversity. Cambridge University Press, Cambridge.10.1017/9781009070553CrossRefGoogle Scholar
Morard, R., Darling, K. F., Weiner, A. K. M., Hassenrück, C., Vanni, C., Cordier, T., Henry, N., et al. 2024. The global genetic diversity of planktonic foraminifera reveals the structure of cryptic speciation in plankton. Biological Reviews 99:1218–1241.10.1111/brv.13065CrossRefGoogle ScholarPubMed
Murray, T. 1979. Evidence for an additional Littorina species and a summary of the reproductive biology of Littorina from California. The Veliger 21:469–474.Google Scholar
Murray, T. 1982. Morphological characterization of the Littorina scutulata species complex. The Veliger 24:233–238.Google Scholar
Olivi, G. 1792. Zoologia Adriatica, ossia catalogo ragionato degli animali del golfo e della lagune di Venezia. https://www.biodiversitylibrary.org/item/123881.Google Scholar
Pérez-Ponce de León, G., and Poulin, R.. 2016. Taxonomic distribution of cryptic diversity among metazoans: not so homogenous after all. Biology Letters 12:20160371.10.1098/rsbl.2016.0371CrossRefGoogle ScholarPubMed
Pfenninger, M., and Schwenk, K.. 2007. Cryptic animal species are homogenously distributed among taxa and biogeographical regions. BMC Evolutionary Biology 7:121.CrossRefGoogle ScholarPubMed
Puillandre, N., Sysoev, A. V., Olivera, B. M., Couloux, A., and Bouchet, P.. 2010. Loss of planktotrophy and speciation: geographic fragmentation in the deep-water gastropod genus Bathytoma (Gastropoda, Conoidea) in the western Pacific. Systematics and Biodiversity 8:371–394.10.1080/14772001003748709CrossRefGoogle Scholar
Purens, K. J. S. 2016. Detecting comatulid crinoid cryptic species in the fossil record. Palaeogeography, Palaeoclimatology, Palaeoecology 446:195–204.10.1016/j.palaeo.2015.12.027CrossRefGoogle Scholar
R Core Team. 2023. R: a language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. https://www.R-project.org.Google Scholar
Reid, D. G. 1989. The comparative morphology, phylogeny and evolution of the gastropod family Littorinidae. Philosophical Transactions of the Royal Society B 324:1–110.Google Scholar
Reid, D. G. 1990. A cladistic phylogeny of the genus Littorina (Gastropoda): implications for evolution of reproductive strategies and for classification. Pp. 1–19 in Johannesson, K., Raffaelli, D. G., and Ellis, C. J. Hannaford, eds., Progress in littorinid and muricid biology. Kluwer Academic, Dordrecht, Netherlands.Google Scholar
Reid, D. G. 1996. Systematics and evolution of Littorina. Ray Society, London.Google ScholarPubMed
Reid, D. G., Rumbak, E., and Thomas, R. H.. 1996. DNA, morphology and fossils: phylogeny and evolutionary rates of the gastropod genus Littorina. Philosophical Transactions of the Royal Society B 351:877–895.Google ScholarPubMed
Reid, D. G., Dyal, P., and Williams, S. T.. 2012. A global molecular phylogeny of 147 periwinkle species (Gastropoda: Littorininae). Zoologica Scripta 41:125–136.10.1111/j.1463-6409.2011.00505.xCrossRefGoogle Scholar
Rohlf, F. J. 2015. The tps series of software. Hystrix 26: 9–12. [Software is distributed by the author at: https://sbmorphometrics.org/index.html.]Google Scholar
Rolán-Alvarez, E., Austin, C. J., and Boulding, E. G.. 2015. The contribution of the genus Littorina to the field of evolutionary biology. Oceanography and Marine Biology: An Annual Review 53:157–214.Google Scholar
Rosewater, J. 1978. A case of double primary homonymy in eastern Pacific Littorinidae. The Nautilus 92:123–125.Google Scholar
Rugh, N. S. 1997. Differences in shell morphology between sibling species Littorina scutulata and Littorina plena (Gastropoda: Prosobranchia). The Veliger 40:350–357.Google Scholar
Seeley, R. H. 1986. Intense natural selection caused a rapid morphological transition in a living marine snail. Proceedings of the Natural Academy of Sciences USA 83:6897–6901.10.1073/pnas.83.18.6897CrossRefGoogle Scholar
Sheets, E. A., Warner, P. A., and Palumbi, S. R.. 2018. Accurate population genetic measurements require cryptic species identification in corals. Coral Reefs 37:549–563.10.1007/s00338-018-1679-9CrossRefGoogle Scholar
Shin, C. P., and Allmon, W. D.. 2023. How we study cryptic species and their biological implications: a case study from marine shelled gastropods. Ecology and Evolution 13:e10360.10.1002/ece3.10360CrossRefGoogle ScholarPubMed
Simpson, G. G. 1951. The species concept. Evolution 5:285–298.10.2307/2405675CrossRefGoogle Scholar
Stankowski, S., and Ravinet, M.. 2021. Quantifying the use of species concepts. Current Biology 31:R415–R429.CrossRefGoogle ScholarPubMed
Struck, T. H., Feder, J. L., Bendiksby, M., Birkeland, S., Cerca, J., Gusarov, V. I., Kistenich, S., et al. 2018. Finding evolutionary processes hidden in cryptic species. Trends in Ecology and Evolution 33:153–163.CrossRefGoogle ScholarPubMed
Tamura, K., Stecher, G., and Kumar, S.. 2021. MEGA 11: Molecular Evolutionary Genetics Analysis version 11. Molecular Biology and Evolution 38:3022–3027.CrossRefGoogle Scholar
Turton, W. 1825. Description of some new British shells. Zoological Journal 2:361–367.Google Scholar
Vermeij, G. J. 1973. Morphological patterns in high-intertidal gastropods: adaptive strategies and their limitations. Marine Biology 20:319–346.10.1007/BF00354275CrossRefGoogle Scholar
Westram, A. M., Stankowski, S., Surendranadh, P., and Barton, N.. 2022. What is reproductive isolation? Journal of Evolutionary Biology 35:1143–1164.10.1111/jeb.14005CrossRefGoogle ScholarPubMed
Wickham, H. 2016. ggplot2: elegant graphics for data analysis. Springer-Verlag, New York.10.1007/978-3-319-24277-4CrossRefGoogle Scholar
Figure 0

Figure 1. Typical shell forms of well-preserved Littorina fossils (A, B, D, F) and modern specimens (C, E, G), all from California, USA. A, †Littorina petricola Arnold, 1908; PRI 76551, Kettleman Hills; Etchegoin Formation, Pliocene. B,Littorina keenae Rosewater, 1978; LACMIP 5100.71, Point Vincente, Palos Verdes Estates; Quaternary terrace, late Pleistocene. C,Littorina keenae; LACM 66066, Shell Beach; collected in 1958. D,Littorina plena A. Gould, 1849; LACMIP 7220.166, Point Loma, San Diego; late Pleistocene. E,Littorina plena; LACM 1948-39.5, Oyster Cove, Tomales Bay; collected from “sand, mud, and stones,” in 1948. F,Littorina scutulata A. Gould, 1849; LACMIP 5100.80, Point Vincente, Palos Verdes Estates; Quaternary terrace, late Pleistocene. (G) Littorina scutulata; LACM 1948-37.11, Nick’s Cove, Tomales Bay; collected from intertidal “coarse to fine sand and rock,” in 1948. See Table 1 for institutional abbreviations.

Figure 1

Figure 2. A, Molecular phylogeny of extant Littorina (based on 28S rRNA, 12S rRNA, and cytochrome oxidase c subunit I [COI] genes with fossil calibrations; approximate divergence times as redrawn from Reid et al. [2012: fig. 2]) and B, morphology-based phylogeny of all Littorina (from shell and anatomical characters for extant species and shell only for extinct species, with hypothesized branch times; redrawn from Reid [1996: fig. 115]). Species studied are highlighted. Dotted bars indicate possible ancestral species relationships to extant taxa, and their stratigraphic ranges. Ma, millions of years ago.

Figure 2

Figure 3. Map of the west coast of North America showing samples, with dotted lines showing state/country boundaries. Species ranges are indicated by the colored bars, based on modern occurrences for extant Littorina keenae, L. petricola, and L. scutulata, and fossil †L. petricola records (after Reid 1996). †Littorina petricola is only found in the Pliocene, while the extant species are sampled from the Pleistocene.

Figure 3

Table 1. Summary of studied samples and specimens, listed by species and age. Sample details are listed in Supplemental Table 2. Institutional abbreviations: AMNH, American Museum of Natural History, New York, NY, USA; LACM and LACMIP, Natural History Museum of Los Angeles County, Los Angeles, CA, USA; NHM, Natural History Museum, London, UK; PRI, Paleontological Research Institution, Ithaca, NY, USA; USNM, Smithsonian National Museum of Natural History, Washington, DC, USA

Figure 4

Figure 4. A, Landmarking scheme (after Carvajal-Rodríguez et al. [2005]; landmarks described below) illustrated on an example specimen, Littorina scutulata. LACMIP 6153.2 Rancho Palos Verdes, California, Quaternary terrace, middle Pleistocene. Landmark 1 = shell apex; 2 = right upper side of penultimate whorl’s suture; 3 = midpoint on curve, between landmarks 2 and 4; 4 = lower suture of penultimate whorl; 5 = end of suture; 6 = rightmost external point on shell lip edge; 7 = lowest point on shell base; 8 = internal border of aperture; 9 = external border of aperture, perpendicular to 8; 10 = midpoint on shell edge, between landmarks 7 and 11; 11 = leftmost external point on penultimate whorl; 12 = left side of penultimate whorl’s suture. B, Example of landmarking consistency with average configurations of three digitization attempts on the same L. keenae sample (black, purple, and pink outlines). LACMIP, Natural History Museum of Los Angeles County, Los Angeles, CA, USA.

Figure 5

Figure 5. A, Centroid size distribution of studied Littorina; dotted plots are sizes of fossil specimens, solid-colored plots are for Recent specimens. B, Centroid size distribution of extant Littorina, grouped by region, listed from north to south. AK, Alaska; BC, British Columbia; WA, Washington; OR, Oregon; CA, California; MX, Baja California.

Figure 6

Figure 6. Principal component analysis of landmark data for all specimens of the three extant and one extinct Littorina species. Polygons enclose the points for each species. Wireframe diagrams show shape changes for each principal component (PC 1, PC 2, PC 3), from minimum (purple) to maximum values (black) along that axis.

Figure 7

Table 2. Mahalanobis distances between pairs of Littorina species (left quadrant, all pairwise comparisons p < 0.01, T2 test) and percentage of misclassified specimens from cross-validated discriminant function analysis (right quadrant) based on all specimens, modern shells, or fossils only

Figure 8

Figure 7. Morphological distance (Mahalanobis distance, the distance of a central point and distribution of landmark data from modern and fossil specimens for that species) and genetic distance (number of base differences per site from published 28S rRNA, 12S rRNA, and cytochrome oxidase c subunit 1 [COI] sequences for each species) from pairwise comparisons for Littorina keenae (k), L. plena (p), and L. scutulata (s). Note that morphological distance for all the plots is the same; only the x-axes have changed.