Copyedited by: YS MANUSCRIPT CATEGORY: Systematic Biology Syst. Biol. 69(6):1106–1121, 2020 © The Author(s) 2020. Published by Oxford University Press, on behalf of the Society of Systematic Biologists. All rights reserved. For permissions, please email: [email protected] DOI:10.1093/sysbio/syaa018 Advance Access publication March 12, 2020 Lineage Identification Affects Estimates of Evolutionary Mode in Marine Snails , , , ,∗ FELIX VAUX 1 2 3 ,MICHAEL R. GEMMELL1,SIMON F.K. HILLS1,BRUCE A. MARSHALL4,ALAN G. BEU5, JAMES S. CRAMPTON6,STEVEN A. TREWICK1, AND MARY MORGAN-RICHARDS1 1Wildlife and Ecology Group, School of Agriculture and Environment, Massey University, Private Bag 11-222, Palmerston North 4410, New Zealand; 2Department of Fisheries and Wildlife, Coastal Oregon Marine Experiment Station, Hatfield Marine Science Center, Oregon State University, 2030 SE Marine Science Dr, Newport, OR 97365, USA; 3Department of Zoology, University of Otago, 340 Great King Street, Dunedin 9016, Otago, New Zealand; 4Museum of New Zealand Te Papa Tongarewa, Wellington, New Zealand; 5GNS Science, PO Box 30-368, Lower Hutt 5011, New Zealand; and 6School of Geography, Environment & Earth Sciences, Victoria University of Wellington, PO Box 600, Wellington 6012, New Zealand ∗ Correspondence to be sent to: Department of Zoology, University of Otago, 340 Great King Street, Dunedin 9016, Otago, New Zealand; E-mail: [email protected]. Downloaded from https://academic.oup.com/sysbio/article/69/6/1106/5803659 by Massey University user on 26 October 2020 Received 01 October 2019; reviews returned 13 February 2020; accepted 09 March 2020 Associate Editor: Ken Halanych Abstract.—In order to study evolutionary pattern and process, we need to be able to accurately identify species and the evolutionary lineages from which they are derived. Determining the concordance between genetic and morphological variation of living populations, and then directly comparing extant and fossil morphological data, provides a robust approach for improving our identification of lineages through time. We investigate genetic and shell morphological variation in extant species of Penion marine snails from New Zealand, and extend this analysis into deep time using fossils. We find that genetic and morphological variation identify similar patterns and support most currently recognized extant species. However, some taxonomic over-splitting is detected due to shell size being a poor trait for species delimitation, and we identify incorrect assignment of some fossil specimens. We infer that a single evolutionary lineage (Penion sulcatus) has existed for 22 myr, with most aspects of shell shape and shell size evolving under a random walk. However, by removing samples previously classified as the extinct species P. marwicki, we instead detect morphological stasis for one axis of shell shape variation. This result demonstrates how lineage identification can change our perception of evolutionary pattern and process. [Genotyping by sequencing; geometric morphometrics; morphological evolution; Neogastropoda; phenotype; speciation; stasis.] Our perception of trait evolution and change in biod- is the degree of concordance between morphological iversity is strongly influenced by our identification of species and separately evolving lineages as revealed species as delineated segments of an evolutionary lin- by genetic data. This relationship matters in particular eage in time (de Queiroz 1998; Sites and Marshall 2003). for extinct species (Stanley and Yang 1987; Gould 1991; Unsurprisingly, if living and fossil species derived from Eldredge 1993; Forey et al. 2004). Restricting analyses of these lineage segments are not identified consistently, the fossil record to only generic-level classification can analyses that rely upon taxonomic identification (i.e., help to avoid this issue, as morphology is more likely to most) are likely to be inaccurate. This issue is especially accurately reflect deeper evolutionary splits (Forey et al. important for the investigation of evolutionary tempo 2004; Foote et al. 2007; Foote 2011; Eronen et al. 2011). In and mode (Lande 1986; Agapow et al. 2004), where addition, generic-level analyses can help to circumvent change can be described differently (e.g., stasis, random statistical limitations due to the incompleteness of walk or punctuated) depending on whether one or many the fossil record, as it is easier to sample a greater species are recognized over a given length of a lineage. proportion of genera than species. However, estimates Morphology is the predominant evidence for species of evolutionary change using genera and species can be delimitation preserved in the fossil record, which means highly discrepant, and it can be theoretically difficult to that the majority of all organisms that have existed integrate data from well-defined living species and fossil are likely to be defined exclusively using phenotype genera (Roy et al. 1996; Wilkinson 2011; Hendricks et al. (Allmon and Yacobucci 2016; Marshall 2017). Fossils 2014; Weise et al. 2016). can preserve morphology for many different internal It is probably appropriate to recognize distinct spe- and external anatomical components, and “morphos- cies where fossils are separated by geologically sig- pecies” can be distinguished based on variation for a nificant periods of time, or if they originate from wide variety of continuous and discrete morphological geographically distant provenances. However, many characters. Shape and size are usually key for species extinct morphospecies are classified from adjacent or delimitation though because these morphological fea- overlapping geological time intervals, and it is com- tures are typically preserved best in the fossil record. mon for fossil sites to contain multiple species from However, diversity among living organisms assessed the same genus. Previous studies have highlighted using molecular markers frequently demonstrates the sampling biases (e.g., Guillerme and Cooper 2018; existence of crypsis, polymorphism, and morphological Hopkins et al. 2018), and arbitrary limits set for the convergence (e.g., Trewick 1998; Bickford et al. 2007; temporal duration of a species (e.g., Aze et al. 2013), as Coppard and Lessios 2017). Consequently, a perennial potential sources of error when attempting to analyze interest for evolutionary biologists and paleontologists evolutionary change and fossil diversity. Where there 1106 [12:42 8/10/2020 Sysbio-OP-SYSB200018.tex] Page: 1106 1106–1121 Copyedited by: YS MANUSCRIPT CATEGORY: Systematic Biology 2020 VAUX ET AL. — LINEAGE IDENTIFICATION IN MARINE SNAILS 1107 is no clear temporal or spatial disjunction, or substan- In this study, we use New Zealand Penion as an tial morphological differences between fossil taxa, a exemplar to integrate extinct and extant material in genetic context for morphological variation is advant- order to identify evolutionary lineages and estimate ageous. Studies can gather this context by integrating evolutionary mode. Multilocus nuclear genetic variation genetic data from living relatives and the morpho- among six extant Penion taxa is investigated, which is logical analysis of both extant and fossil specimens used to determine species delimitation between P. chath- (e.g., Stanley and Yang 1987; Jackson and Cheetham amensis and P. fairfieldae (Powell, 1947), and between 1994; Hills et al. 2012; Smith and Hendricks 2013; P. jeakingsi (Powell, 1947) and P. ormesi (Powell, 1927). Hopkins et al. 2018). Using a geometric morphometric method, we assess Siphon whelks from New Zealand provide an oppor- the extent of shell shape and size variation within tunity to identify evolutionary lineages using genetic extant species and apply this knowledge to identify and morphological data, and this knowledge can be fossil material. Two subsampled data sets are used to applied to fossil material. Siphon whelks of the genus analyze morphological variation among extant: 1) P. Penion P. Fischer, 1884 are large marine snails (Fig. 1a) chathamensis and P. fairfieldae and 2) P. jeakingsi and P. Downloaded from https://academic.oup.com/sysbio/article/69/6/1106/5803659 by Massey University user on 26 October 2020 with a rich fossil record in New Zealand, including 19 ormesi. These morphometric results are compared with extinct species, found in high abundance from many variation in nuclear DNA sequence data and the signal paleontological sites spanning the last 66 myr (Beu from previous molecular phylogenetics results (Vaux and Maxwell 1990). Although many fossils are well- et al. 2018; Fig. 1b). The taxonomic identification of preserved (Fig. 1c), evolutionary relationships among fossil specimens is investigated at two paleontological extant and fossil taxa are unclear and the taxonomic sites that putatively exhibit multiple species of Penion identity of specimens at many fossil sites is uncertain (Beu and Maxwell 1990): 1) Wanganui Beach (0.97–0.38 Ma) with putative fossils of P. sulcatus, P. ormesi, and P. (Beu and Maxwell 1990). Shells exhibit considerable jeakingsi and 2) Te Piki, Cape Runaway (2.40–1.63 Ma) morphological variation in shape, size, and sculp- with specimens currently classified as P. sulcatus and P. ture within and among the six extant species (and cuvierianus (Powell 1927) (Fig. 1a). Fossil specimens from one subspecies) currently recognized in New Zealand these sites are analyzed along with extant populations waters (Powell 1979; Supplementary Fig. S1 available on from the surrounding geographic