Morphological Stasis in an Ongoing Gastropod Radiation from Lake Malawi

Morphological stasis in an ongoing gastropod radiation from Lake Malawi

Bert Van Bocxlaera,b,c,1 and Gene Hunta

aDepartment of Paleobiology and bDepartment of Invertebrate Zoology, National Museum of Natural History, Smithsonian Institution, Washington, DC 20013-7012; and cResearch Unit Palaeontology, Department of Geology and Soil Science, Ghent University, 9000 Ghent, Belgium

Edited by David Jablonski, University of Chicago, Chicago, IL, and approved July 15, 2013 (received for review May 13, 2013) Evolutionary processes leading to adaptive radiation regularly (e.g., refs. 10 and 11), whereas the evaluation of evolutionary occur too fast to be accurately recorded in the fossil record but too patterns using time series remains underrepresented in the cur- slowly to be readily observed in living biota. The study of evo- rent literature (but see e.g., refs. 12 and 13). lutionary radiations is thereby confronted with an epistemological Here we report on a unique and unusual study system of a gap between the timescales and approaches used by neontolo- young, ongoing radiation in Bellamya gastropods in the African gists and paleontologists. Here we report on an ongoing radiation Rift Lake Malawi (14). In addition to its extant diversity, this of extant Bellamya species (n = 4) from the African Rift Lake Bellamya clade left abundant, well-preserved fossil shells in a Malawi that provides an unusual opportunity to bridge this gap. high-resolution, radiocarbon-dated sediment record that spans The substantial molecular differentiation in this monophyletic much of the Holocene (15). The study system presents a partic- Bellamya clade has arisen since Late Pleistocene megadroughts ularly promising opportunity to study radiations because some of in the Malawi Basin caused by climate change. Morphological time- the life-history traits that are thought to have led to the di- series analysis of a high-resolution, radiocarbon-dated sequence of vergence of the modern descendants affect shell morphology and, 22 faunas spanning the Holocene documents stasis up to the middle hence, can be traced in fossils. Holocene in all traits studied (shell height, number of whorls, and Below we analyze changes in morphological traits [shell size, two variables obtained from geometric morphometrics). Between the number of whorls (#whorls) and geometric shell shape] over deposition of the last fossil fauna (∼5 ka) and the present day, a dras- a sequence of 22 Bellamya assemblages spanning the Holocene tic increase in morphological disparity was observed (3.7–5.8 times) (i.e., 21 fossil faunas and the modern one). Upon fitting evolu- associated with an increase in species diversity. Comparison of the tionary models (stasis, random walk, and directional change), rates of morphological evolution obtained from the paleontological stasis received high support in all traits studied for the early and time-series with phylogenetic rates indicates that the divergence in middle Holocene fossil assemblages. These results suggest that two traits could be reconstructed with the slow rates documented in evolutionary trajectories within established lineages can be char- the fossils, that one trait required a rate reduction (stabilizing selec- acterized best as fluctuations around a steady long-term mean, tion), and the other faster rates (divergent selection). The combined even in recently and rapidly radiating clades. Morphological di- paleontological and comparative approach taken here allows rec- vergences map well to molecular differentiation observed in the ognition that morphological stasis can be the dominant evolu- extant Bellamya, enabling us to integrate paleontological, phylo- tionary pattern within species lineages, even in very young and genetic, and natural history information to interpret the ecological radiating clades. importance of the patterns observed in this ongoing divergence. Study System speciation | rates of evolution | punctuated equilibrium | Viviparidae Four taxonomically valid Bellamya species currently occur in Lake Bel- volutionary radiations are potentially responsible for much of Malawi (16, 17): three inhabit shallow water environments ( lamya capillata, Bellamya jeffreysi,andBellamya robertsoni), whereas Ethe ecological and phenotypic diversity on earth (1, 2). Be- Bellamya ecclesi yond rapid speciation through lineage splitting, radiations are the fourth ( ) is rare and occupies substrates below the euphotic zone (16, 18). Molecular analysis using three gene often characterized by exceptional diversification into a variety of fragments [mitochondrial cytochrome oxidase subunit I (COI), ecological niches, with divergence, at least in traits under selec- the mitochondrial large subunit of rRNA and the nuclear large tion, regularly occurring very rapidly (e.g., refs. 3 and 4). How- subunit of rRNA] indicates that all Bellamya from the Malawi ever, the controls on diversification may be complex and include Basin form a monophyletic clade endemic to the basin, and al- historical contingencies, as well as ecological, genetic, and de- though the species are young, great and highly significant genetic velopmental factors (5, 6). One major challenge is that processes differentiation (FST = 0.234) exists among the three shallow-water leading to evolutionary radiations occur regularly too fast to be species (14). Exact tests of differentiation demonstrate significant accurately recorded in the fossil record where temporal reso- B. capillata B. jeffreysi P = fi ∼ 4 differences between and ( 0.0486), be- lutions ner than 10 y are unusual, but perhaps still too crude tween B. capillata and B. robertsoni (P = 0.0003), but not between to see speciation and radiation unfold. On the other hand, spe- B. robertsoni and B. jeffreysi (P = 0.1105). (The restricted number ciation generally occurs too slowly to be readily observed in living of specimens from the deep-water species B. ecclesi prevented its “ ” biota, resulting in an epistemological gap between the time- inclusion in statistical tests.) Mismatch analyses using COI data scales and approaches used by neontologists and those that paleontologists adopt for the study of organismal evolution fi (7, 8). Moreover, it can be dif cult to assess the adaptive sig- Author contributions: B.V.B. designed research; B.V.B. performed research; B.V.B. and nificance of phenotypic variation among fossil taxa (e.g., ref. 9), G.H. analyzed data; and B.V.B. and G.H. wrote the paper. partly because ecological, behavioral, physiological, and other The authors declare no conflict of interest. data are lacking. Neontological approaches are more diverse, but This article is a PNAS Direct Submission. they are usually limited to studying the subset of radiating species Data deposition: Datasets and R code are deposited at Dryad, http://dx.doi.org/10.5061/ that is still alive, and with that offer only a partial view of the dryad.vm523. involved evolutionary processes, as adaptive radiations are un- 1To whom correspondence should be addressed. E-mail: [email protected]. likely to occur in the absence of extinction (4). Most recent This article contains supporting information online at www.pnas.org/lookup/suppl/doi:10. studies have adopted phylogenetic and comparative approaches 1073/pnas.1308588110/-/DCSupplemental.

13892–13897 | PNAS | August 20, 2013 | vol. 110 | no. 34 www.pnas.org/cgi/doi/10.1073/pnas.1308588110 Downloaded by guest on September 27, 2021 reveal a drastic spatial expansion of the entire Malawi group since 97 ka [95% confidence interval (CI): 0–215 ka], and taxa with stable mismatch distributions (B. capillata and B. robertsoni)also show remarkable demographic expansions since 200 ka (95% CI: 105–295 ka) and 102 ka (95% CI: 12–192 ka), respectively (14). These estimates correspond well to two major megadrought phases ∼135 and 75 ka ago, during which Lake Malawi’s water level (currently 706 m deep) dropped 600 m and 350 m below the current level, respectively (19). Over considerable periods during this 60,000-y interval Lake Malawi was a ∼125-m-deep saline, alkaline, well-mixed lake (19). Since 55 ka, the Malawi Basin consistently experienced much wetter conditions and lake levels were high and similar to those at present, except for a drop of ∼100 m below the current level at the last glacial maximum (19). These high lake levels have promoted organismal evolution in mollusks and cichlid fishes (20, 21), and the demographic history of Bellamya as inferred from the molecular analyses summarized above is consistent with this environmental history. Results and Discussion Principal component analysis (PCA) of morphological traits (shell height, #whorls, and geometric shell shape from semilandmark morphometrics) documents the morphospace occupation of the fossil and modern material (Fig. 1 and Fig. S1). We used model-based clustering to explore how many evolutionary lineages are present in our dataset (Fig. S2). Applying such clustering to the extant Bellamya directly in the 2D morphospace resulted in the highest support values for three clusters (=mor- photaxa; ΔBIC ≥ 6.1 over other models; BIC, Bayesian In- formation Criterion). The major separation in the modern taxa is observed between the deep-water species B. ecclesi and the shallow-water taxa, and a secondary separation within this latter group between B. robertsoni and the group of B. capillata/B. jeffreysi. If models are forced to assign four clusters, then all four taxonomically valid species are recognized (ΔBIC ≥ 17.4 over EVOLUTION other four cluster models, but ΔBIC = 11.7 below the best so- lution with three clusters) (Fig. 1B and Figs. S2 and S3). The best-supported model was the same in both cases and has trait variances and covariances that are shared across species, which is realistic for close relatives (SI Text). Thus, model-based clustering without a priori assumptions [i.e., independent of the earlier identification of the specimens using the criteria established by previous researchers (16, 17)] recovered meaningful, morphologically delimited groups that correspond to taxonomically valid, extant Bellamya species from Lake Malawi. The abovementioned molecular analyses (14) used the same morphological identification system and demonstrated highly significant genetic differentiation between the morphospecies, except for the pair of B. robertsoni and B. jeffreysi. However, Fig. 1. PCA showing the morphospace occupation of (A) fossil Bellamya B. robertsoni and B. jeffreysi represent the only species pair with derived from 21 early to middle Holocene assemblages and (B) the modern substantial overlap in morphospace occupation (Fig. 1B). This taxa, predominantly from the southeastern arm of Lake Malawi. The ma- overlap may indicate that the divergence within this species pair terial in A and B was analyzed together; vector loadings in red indicate the is young compared with that in others (and perhaps incomplete). contribution of shell height (Log10H), the number of whorls (Log10W) and Alternatively, the morphospace overlap of this species pair may the semilandmark variables (all others) to the morphospace occupation [we used the function envfit() to scale variable vectors]. (Insets) The results of explain the nonsignificant result of the molecular test of differ- fi model-based clustering, with higher BIC scores indicating better model sup- entiation, as misidenti cations would diminish apparent genetic port. The best-supported clustering is indicated on each plot. separation. The significant genetic differentiation in other species pairs may indicate their reproductive isolation. Kat (22) reported informally that B. capillata and B. jeffreysi hybridize Applying model-based clustering to each fossil assemblage freely in the south of the lake but that the hybrids are sterile. resulted in the highest support values for a single cluster in most However, haplotype networks including specimens of all Bellamya cases. For a few assemblages, multiple clusters were preferred, but morphospecies from Lake Malawi indicate that speciation pro- these results represented biologically unlikely scenarios (SI Text, cesses in this young radiation are still ongoing (14). More research and Figs. S1, S2,andS4). In no case were bimodal distributions is required, but the here-documented morphological differentia- observed (Fig. S1), nor was any reasonable and consistent separation and the concordant and highly significant molecular differ- tion detectable by eye. If all fossils are lumped in one sample and entiation between morphospecies suggest that multiple Bellamya model-based clustering is applied, a single component is highly species currently exist in Lake Malawi, and that these correspond supported (by 8 of the 10 models applied; ΔBIC ≥ 3.7 over sug- to the valid species in classic taxonomy. gestions with multiple components (Fig. 1A). In addition, the

Van Bocxlaer and Hunt PNAS | August 20, 2013 | vol. 110 | no. 34 | 13893 Downloaded by guest on September 27, 2021 magnitude of morphological variation within the fossil samples is data suggest a rapid demographic expansion in the Bellamya from comparable to that seen in the living species (Fig. S5). These Lake Malawi (14), given that net diversification is expected to be findings confirm the earlier qualitative result that all Holocene highest early in the speciation process, followed by a slowdown Bellamya fossils from Lake Malawi belong to a single morpho- (25, 26, but see also ref. 10), and given that divergence was species (23). expected to have occurred very rapidly (3, 4) with initially large The existence of a single fossil Bellamya lineage in the Chipa- trait changes that progressively become smaller (27). It is gener- lamawamba Beds is somewhat surprising because these as- ally accepted that rates of evolutionary change can be exception- semblages record three dispersal events (15) from a source region ally high over the course of a few generations (e.g., refs. 28 and that is currently inhabited by three shallow-water Bellamya species, 29); however, these rates normally drop off steeply over the course which regularly occur in sympatry (16, 22) (Table S1). No tapho- of multiple generations (5 ka = ∼5,000 Bellamya generations) (24, nomic biases exist that could explain the absence of multiple shal- 30, 31), because selection pressures rarely maintain the same di- low-water Bellamya species in the fossil assemblages if they were rection over the course of hundreds of generations or more (e.g., existing locally at the time of deposition, and except for Bellamya, refs. 32–34). The large increase in disparity that we document here all other currently diversified mollusk genera show diversities in the does not imply that selection pressures were always present, that Chipalamawamba Beds similar to modern levels (15) (Table S2). they would have acted consistently in the same direction (the These observations are based on the identification of more than morphospace expansion is not unidirectional), or with equal 40,000 fossil specimens from diverse shallow-water habitats and strength for all traits studied. Nevertheless, considerable and rapid exclude inadequate sampling as explanation (SI Text and Tables morphological differentiation occurred between the modern spe- S1–S3). Given the data at hand, the most parsimonious option is cies, even though our fossil lineage documents ∼5,000 y of mor- that at least two speciation events have occurred among the shal- phological stasis for all traits studied. low-water species since the deposition of the fossil material. Nev- Stasis does not imply that traits are invariant; some fluctua- ertheless, we cannot exclude that multiple lineages existed in the tions around the long-term mean are expected (11, 35, 36), and early Holocene but were not captured by the Chipalamawamba indeed are observed in the traits studied. Under stasis, these Beds, and we accounted for a range of divergence scenarios in our fluctuations do not accumulate, and hence the bounds of stasis analyses below (Materials and Methods and Fig. S6). are independent of the time elapsed (95% probability interval in Fossils of the deep-water taxon cannot reasonably be expected in Fig. 2). Can the fossil stasis be reconciled with the divergence the Chipalamawamba Beds, which accumulated in shallow waters. observed among the extant species? This question can be ad- B. ecclesi currently occurs in our sampling area but is adapted to dressed by comparing the trait means of the extant species to the a habitat that is markedly dissimilar from those occupied by shal- bounds of stasis, or by computing model support values for low-water species (18). Because this habitat did not exist in the scenarios in which the extant species are added to the fossil se- Malawi Basin during the Late Pleistocene megadroughts (19), nor ries as end members (Fig. 2). Among the extant species, the trait perhaps at the last glacial maximum, it is unlikely that lineage values of only B. capillata fall close enough to the fluctuations splitting and the divergence toward B. ecclesi started consider- observed in the fossils for the stasis model to remain best sup- ably earlier than that in the shallow-water community. ported. All other species have one or more traits that lie well The fossil specimens occupy a rather more restricted region of outside the confidence limits of the stasis observed in the fossils morphospace than the modern ones (Fig. 1 and Fig. S1). Although (Fig. 2) and are accordingly better accounted for by models, such the fossil material is intermediate to the three extant shallow-water as the random walk, that accumulate evolutionary change over taxa in some morphological features, it also differs substantially in time. Hence, the stasis documented for the fossil Bellamya se- others: part of the morphospace occupied by the fossils is devoid of quence is not adequate to account for the larger divergences modern taxa, and an even more substantial part of the morpho- between the extant species, implying that morphological change space occupied by the extant fauna is not represented by fossil departed from the normal dynamics of stasis to produce intervals Bellamya. Variable loadings show that PC1 is largely correlated with of elevated change. size and several shape variables, whereas PC2 represents #whorls With data on the living species and a high-resolution fossil and other shape variables. Comparing the morphological disparity sequence, it becomes possible to compare the pace of trait of the fossil material with that of our extant taxa using two variance- evolution within and between evolutionary lineages. The exact based disparity indices, we observed a 3.66 ± 0.30-times (or 2.27 ± phylogenetic relationship between the extant species in the Bellamya 0.05-times if B. ecclesi is excluded) increase in morphological dis- radiation from Lake Malawi is unresolved, so we used a large parity since deposition of the last fossil bed. However, the mor- number of randomly resolved, time-scaled trees of the four phological disparity in the fossil material is likely inflated by time modern species to calculate rates of morphological trait evolu- averaging over the >5,000-y period (10,630 ± 66 until 5,348 ± 57 cal tion according to the Brownian motion model (Materials and yr BP) that these fossils span. Comparing individual fossil Methods). These phylogenetic rates were compared with pale- assemblages to the modern fauna suggests that the increase in ontological rates obtained by fitting the equivalent model to the disparity since the middle Holocene could be up to 5.84 ± 0.30- fossil time series. We report results across a wide range of di- times (or 3.62 ± 0.05-times if B. ecclesi is excluded). vergence times (Fig. S6) but focus on a relatively literal reading Upon fitting evolutionary models to the sequence of 21 fossil of the fossil record (i.e., the scenario in which diversification of populations for all univariate traits studied (shell height, #whorls, the lineages occurs after the last fossil sample at ∼5.5 ka) (Fig. morphoPC1, and morphoPC2), the model of stasis (i.e., fluctua- 3). For two traits (#whorls and morphoPC1) the paleontological tions around an unchanging, long-term mean) consistently gained rates from the fossil time series correspond well to the phylo- a much higher support than other models. The dominance of genetic rates obtained when the depth of the root node was set to stasis was overwhelming for shell height, #whorls, and mor- coincide with the last fossil sample (Fig. 3). Hence, the paleon- phoPC2 [Akaike Information Criterion (AIC) weights >0.984; tological rates suffice to explain the divergence in these traits unity represents absolute support for one fitted model over the toward the modern taxa over 5,500 y, assuming that Brownian others], and somewhat weaker for morphoPC1 (AIC weight = motion applied throughout. However, the two other characters, 0.640) (Fig. 2). Not only do these trait trajectories match the shell height and morphoPC2, behave differently. For morphoPC2, fluctuating pattern of stasis, the magnitudes of the fluctuations the paleontological rates were faster than phylogenetic rates over around the mean are low as well (i.e., generally less than one unit ∼5.5 ka (negative log-ratio in Fig. 3), which remains true even if of within-sample SD) (Fig. 2) (see also refs. 11 and 24). The the depth of the root node is only 1.5 ka (Fig. S6). The difference finding of predominant stasis is surprising given that molecular in rates suggests stronger stabilizing selection for this trait when

13894 | www.pnas.org/cgi/doi/10.1073/pnas.1308588110 Van Bocxlaer and Hunt Downloaded by guest on September 27, 2021 Fig. 2. Time-series plots of fossil Bellamya for the four univariate traits examined (shell height, number of whorls, morphoPC1, and morphoPC2) and the trait EVOLUTION values of the extant species (symbols correspond to those in Fig. 1). Variation in the traits was standardized to within-sample SDs, three evolutionary models were ﬁtted to these data [stasis (St), unbiased random walk (URW), and directional change (DIR)] and the model with the best ﬁt is reported; AIC weights in subscript indicate model support. Values for the extant taxa are obtained by considering each of these taxa independently as a direct end member of the time series. Black horizontal lines indicate the trait means of the fossil specimens and shaded areas represent 95% probability intervals. Error bars represent 1 SE (x ± 1seÞ.

lineages diversified than in the fossil lineage over 5,000 y, where one [B. ecclesi; adult specimens with a height of ∼52 mm weigh on fluctuations around the long-term mean where already limited average 3.4 g; shells of ∼42 mm are about 1.5 g, which is about 0.5 g (Fig. 2). Hence, this result testifies to the commonness of stabi- less than equally sized B. phthinotropis, the thin-walled deep-water lizing selection in nature (24). For shell height, phylogenetic rates species from Lake Victoria; (16)]. Similarly, extant B. capillata from are faster than paleontological ones if the root node dates to Lake Malawi are among the smallest African Bellamya together 5.5 ka (Fig. 3). Upon increasing the depth of the root node, the with B. costulata and B. constricta from Lake Victoria; they be- rates become more similar, then equal each other for root node come mature at ∼17-mm and adults average ∼20-mm shell depths of ∼50 ka, after which they diverge again (Fig. S6). Two height (16, 17). B. ecclesi, with an average height of 54 mm, is by explanations are conceivable. The firstisthatdivergenceinshell far the largest one (17). Selection on shell height in the Bellamya height started considerably earlier than the deposition of the fossil from Lake Malawi is particularly likely because females breed material, but that this divergence remains unsampled in the fossil their young in a brood pouch and the size and volume of the sequence. The second explanation is that divergence in shell brood pouch is directly and positively correlated with the size of height started more recently and that the evolutionary rates as- the shell and the inflation of the body whorl. Differences in adult sociated with it are, consistent with divergent selection, faster than size are moreover correlated with marked differences in life the paleontological ones we documented. Regardless of the actual history traits [e.g., the size at which juveniles are released in the timing of divergence between these species, our results indicate environment (<6mminB. capillata, >8mminB. robertsoni and that different traits have evolved differently during this young B. jeffreysi, and ∼14 mm in B. ecclesi) (16)]. radiation (see also ref. 13). The Bellamya from Lake Malawi represent an ongoing radia- External evidence for divergent selection is provided by the tion (14), and may be undergoing “explosive” diversification, fact that the young Bellamya species from Lake Malawi occupy which has been observed regularly early in the process of radi- extreme positions for several traits among African Bellamya (18 ation (37). The term hints to several lineage splits that are closely species) (17). The southeastern arm of Lake Malawi is home to spaced in time—thousands to a few 10 thousands of generations. the most strongly calcified Bellamya species alive today [B. jeffreysi; Yet, despite this active evolutionary milieu, all traits examined adult specimens with a shell height of ∼38 mm weigh on average experienced morphological stasis within a densely sampled fossil 5.8 g; that is, twice as much as any other African Bellamya of equal lineage. This pattern is consistent with the long questioned (38, 39) size (except for B. robertsoni)], and to the most weakly calcified suggestion of punctuated equilibrium proponents that morphological

Van Bocxlaer and Hunt PNAS | August 20, 2013 | vol. 110 | no. 34 | 13895 Downloaded by guest on September 27, 2021 method allows identifying morphological clusters without a priori species designations and was applied to each of the 22 assemblages, and to the ag- gregated dataset of all fossils. This approach requires assumptions about the degree to which covariance patterns are shared across groups. We used the 10 default parameterizations, which differ in the degree of commonality of the shape, volume, and orientation of trait variances and covariances. Support for different models was compared with a BIC. Statistically valid cluster sol- utions may not be biologically realistic; a few such instances with high BIC support are discarded here but are discussed in SI Text. Morphological disparity was calculated for each fossil population and for the modern species using the sum and the product of the variances on PC1 and PC2 (48, 49). We performed time-series analyses of morphological evolution on the 21 fossil samples, and subsequently with each of the modern taxa as end member of the time sequence (paleoTS 0.4-4) (50, 51). These methods are designed for univariate variables and we performed a separate PCA on the multivariate morphometric dataset (landmarks, semilandmarks) from which the two first PCs were extracted and analyzed similar to the univariate characters in our dataset (shell height, #whorls). We standardized the variation in traits to within-sample SD units to facilitate comparisons among traits. Subsequently we fit three evolutionary models (stasis, an unbiased random walk, and directional change) to each univariate variable (shell height, #whorls, mor- Fig. 3. Comparison of paleontological rates with simulated phylogenetic phoPC1, morphoPC2). Stasis was modeled as uncorrelated, Gaussian variation rates of character change for 1,000 randomly resolved phylogenetic trees of around a stable mean trait value (50, 52). A random walk is a simple model in Bellamya since deposition of the fossil beds (root node = ∼5.5 ka). Log-ratios which evolutionary increments are independent of each other and equally demonstrate that phylogenetic rates were faster (for shell height), equal (#whorls, morphoPC1) and slower (morphoPC2) than the paleontological likely to be increases and decreases. The directional evolution model is similar rates. Whiskers represent the 95% central simulated rates, boxes the 50% to the random walk, except that increases are more probable than decreases, central rates with indication of the median. Modeling results for a wide or vice versa (50). Model support was determined via an AICc with AIC range of root node depths are shown in Fig. S6. weights of 0 and 1 representing no and absolute support among the candi- date models, respectively. In-depth explanations of the models, the AICc calculation, and their implementation are provided in ref. 50. change occurs predominantly in association with events of branching The phylogeny of the extant species remains unresolved (one polytomy speciation (35, 40), even though we cannot currently disentangle that may be hard), but only (2n-3)!!=15 possible bifurcating trees with n = 4 whether the elevated morphological changes occurred anageneti- terminal taxa are conceivable. To allow calculating rates of evolution along cally shortly before or after lineage splitting, or whether they co- each randomly resolved tree, we set the depth of the root node to 50 incided directly with cladogenesis. Either way, stasis appears to equally spaced times between 1.5 ka and 200 ka, each separated by 3,970 y. be a common pattern of trait evolution within species lineages, even For each, we randomly resolved the phylogeny 1,000 times into bifurcating trees using the packages ape 3.0-4, phytools 0.2-1 and paleotree 1.4 (53–57). during an ongoing and rapid radiation. Branch lengths were randomly assigned with a 500-y minimum for the pe- Materials and Methods riod between two subsequent nodes and, hence, speciation events. A Brownian motion model (package geiger 1.3-1) (58) was fit to calculate The paleontological material comes from abundant shell beds that were deposited in a wide variety of shallow-water habitats at the southern end of phylogenetic rates of morphological evolution for each of the four variables Lake Malawi during three high lake-level phases in the early to middle along the randomly resolved phylogenies. These rates were then compared fi Holocene (15) (SI Text). Radiocarbon dating indicates minimal time averag- with the paleontological rates obtained by tting an unbiased random walk ing; a single bed spans a few decades maximum (15). The modern material to the same variables over the fossil time series. We made comparisons for was predominantly collected from a shoreline stretch of 50 km along the all root-node depths (Fig. S6) but highlight a literal reading of the fossil southwestern shores of the southeastern arm of Lake Malawi to sample the record in which the root node was coeval with the youngest fossil assem- same diversity of shallow-water habitats in close proximity to the geo- blage sampled (∼5.5 ka) (Fig. 3). Such comparisons are possible because graphically confined fossil sites (SI Text). A total of 912 fossil Bellamya Brownian motion and random walks are closely related models, with specimens from a well-dated, high-resolution sequence of 21 fossil assem- equivalent parameters (31), that are commonly used as a rate metric in blages and 383 modern shells of the four extant species were digitized in comparative methods (10, 59) and paleontology (31). The advantage of us- apertural view conforming to the methods in ref. 41 for morphometric ing this metric is that it can provide a rate estimate that incorporates in- analyses. We used a semilandmark approach with 12 landmarks and four formation on all populations of a time-series or phylogeny, and that it is not open curves of equally spaced semilandmarks (60, 20, 20, and 15, re- constrained to document trait change from the pairwise comparison of spectively) that were anchored by starting and ending landmarks (41). populations, as traditional rate metrics do (see e.g., ref. 60). When traits are Landmarks and semilandmarks were digitized in TpsDig 2.16 (42), then scaled by within-sample SDs and time is measured in generations (i.e., ∼1y subjected to Procrustes superimpositioning in CoordGen6h (43), after which for Bellamya), the Brownian motion/random walk rate metric is equal to the semilandmarks were slid along their respective curve in SemiLand6, us- twice the Δ rate metric proposed by Lynch (29). ing the minimum Procrustes distance criterion (43). Additionally, we in- cluded data on shell height and the #whorls, because these characters are ACKNOWLEDGMENTS. We thank the chiefs of Chipalamawamba, Kwitambo, considered taxonomically important (16). These measurements were log10- and Kazembe for allowing us to work in outcrops near their villages; the tranformed, combined with the Procrustes superimposition coordinates and Cultural and Museum Centre Karonga for structural support; Friedemann imported in R 2.15.0 (44) for further analyses. Schrenk, Harisson Simfukwe, Joseph Tembo, Jacques Verniers, Nore Praet, PCA was applied to the final morphological dataset, after which we cal- and Wout Salenbien for providing assistance in the field; Jon Ablett, Didier culated variable loadings (shell height, #whorls, landmarks, and semiland- Van den Spiegel, Oliver Sandrock, Thomas Kristensen, Daniel Graf, and marks) and plotted their contribution to the morphospace occupation (stats Ellen Strong for providing access to museum specimens; and Roland 2.15.0; vegan 2.0–4) (44, 45). A scree plot with the broken-stick expectation Schultheiß, Achilles Gautier, Pieter Kat, Ellen Strong, Thomas Ezard, two anonymous referees, and the editor for providing constructive discussion. suggested that the first three PC axes contained significant information, but This work was supported in part by the Flanders Research Foundation, the no clear separation in morphospace was observed on PC3, and only PCs 1 Ghent University Research Council, a Jessup Award from the Academy of and 2 were used for subsequent analyses. Natural Sciences in Philadelphia, the Leopold III-fund, and Fellowships of Model-based clustering was performed on the PC1 vs. PC2 coordinates using the Belgian American Educational Foundation and the Smithsonian Insti- normal mixture models [mclust 3.5, (46); see ref. 47 for model descriptions]. The tution (to B.V.B.).

13896 | www.pnas.org/cgi/doi/10.1073/pnas.1308588110 Van Bocxlaer and Hunt Downloaded by guest on September 27, 2021 1. Simpson GG (1953) The Major Features of Evolution (Columbia Univ Press, New York). 33. Grant PR, Grant BR (2002) Unpredictable evolution in a 30-year study of Darwin’s 2. Schluter D (2000) The Ecology of Adaptive Radiation (Oxford Univ Press, Oxford). finches. Science 296(5568):707–711. 3. Losos JB, Miles DB (2002) Testing the hypothesis that a clade has adaptively radiated: 34. Grant PR, Grant BR (2006) Evolution of character displacement in Darwin’s finches. Iguanid lizard clades as a case study. Am Nat 160(2):147–157. Science 313(5784):224–226. 4. Grant PR, Grant BR (2008) How and Why Species Multiply: The Radiation of Darwin’s 35. Gould SJ (2002) The Structure of Evolutionary Theory (Belknap, Cambridge, MA). Finches (Princeton Univ Press, Princeton, NJ). 36. Eldredge N, et al. (2005) The dynamics of evolutionary stasis. Paleobiology 31(2): 5. Gavrilets S, Losos JB (2009) Adaptive radiation: Contrasting theory with data. Science 133–145. 323(5915):732–737. 37. Kocher TD (2004) Adaptive evolution and explosive speciation: The cichlid fish model. 6. Wagner CE, Harmon LJ, Seehausen O (2012) Ecological opportunity and sexual se- Nat Rev Genet 5(4):288–298. lection together predict adaptive radiation. Nature 487(7407):366–369. 38. Levinton JS (2001) Genetics, Paleontology, and Macroevolution (Cambridge Univ 7. Kemp TS (1999) Fossils and Evolution (Oxford Univ Press, Oxford). Press, New York). 8. Reznick DN, Ricklefs RE (2009) Darwin’s bridge between microevolution and macro- 39. Levinton JS, Simon CM (1980) A critique of the punctuated equilibria model and evolution. Nature 457(7231):837–842. implications for the detection of speciation in the fossil record. Syst Zool 29(2): 9. Lieberman BS (2012) Adaptive radiations in the context of macroevolutionary theory: 130–142. A paleontological perspective. Evol Biol 39(2):181–191. 40. Gould SJ, Eldredge N (1977) Punctuated equilibria: The tempo and mode of evolution – 10. Harmon LJ, et al. (2010) Early bursts of body size and shape evolution are rare in reconsidered. Paleobiology 3(2):115 151. comparative data. Evolution 64(8):2385–2396. 41. Van Bocxlaer B, Schultheiß R (2010) Comparison of morphometric techniques for 11. Uyeda JC, Hansen TF, Arnold SJ, Pienaar J (2011) The million-year wait for macro- shapes with few homologous landmarks based on machine-learning approaches to – evolutionary bursts. Proc Natl Acad Sci USA 108(38):15908–15913. biological discrimination. Paleobiology 36(3):497 515. 12. Hunt G (2007) The relative importance of directional change, random walks, and 42. Rohlf FJ (2010) TpsDig 2 (Department of Ecology and Evolution, State University of stasis in the evolution of fossil lineages. Proc Natl Acad Sci USA 104(47):18404–18408. New York at Stony Brook, NY), v2.16. 13. Hopkins MJ, Lidgard S (2012) Evolutionary mode routinely varies among morpho- 43. Sheets HD (2008) IMP: Integrated Morphometrics Package (Department of Physics, logical traits within fossil species lineages. Proc Natl Acad Sci USA 109(50):20520–20525. Canisius College, Buffalo, NY). 14. Schultheiß R, Wilke T, Jørgensen A, Albrecht C (2011) The birth of an endemic species 44. R-Development-Core-Team (2012) R: A language and environment for statistical flock: Demographic history of the Bellamya group (Gastropoda, Viviparidae) in Lake computing (R Foundation for Statistical Computing). Available at http://www.R-project.org, v2.15.0. Accessed June 18, 2012. Malawi. Biol J Linn Soc Lond 102(1):130–143. 45. Oksanen J, et al. (2012) Vegan: Community Ecology Package: Ordination, Diversity 15. Van Bocxlaer B, Salenbien W, Praet N, Verniers J (2012) Stratigraphy and paleo- and Dissimilarities, version 2.0-4. Available at http://cran.r-project.org/web/packages/ environments of the early to middle Holocene Chipalamawamba Beds (Malawi Basin, vegan/index.html. Accessed June 18, 2012. Africa). Biogeosciences 9(11):4497–4512. 46. Fraley C, Raftery AE (2012) MCLUST: Normal Mixture Modeling and Model-Based 16. Mandahl-Barth G (1972) The freshwater Mollusca of Lake Malawi. Rev Zool Bot Afr Clustering, version 3.5. Available at http://cran.r-project.org/web/packages/mclust/ 86(3-4):257–289. index.html. Accessed July 6, 2012. 17. Brown DS (1994) Freshwater Snails of Africa and Their Medical Importance (Taylor 47. Fraley C, Raftery AE (2007) Bayesian regularization for normal mixture estimation and and Francis, London), 2nd Ed. model-based clustering. J Classif 24(2):155–181. 18. Van Bocxlaer B, Schultheiß R, Plisnier P-D, Albrecht C (2012) Does the decline of 48. Foote M (1991) Morphological and taxonomic diversity in a clade’s history: The gastropods in deep water herald ecosystem change in Lakes Malawi and Tanganyika? blastoid record and stochastic simulations. Contributions from the Museum of Pale- Freshw Biol 57(8):1733–1744. ontology, the University of Michigan 28(6):101–140. 19. Cohen AS, et al. (2007) Ecological consequences of early Late Pleistocene mega- 49. Van Valen LM (1974) Multivariate structural statistics in natural history. J Theor Biol droughts in tropical Africa. Proc Natl Acad Sci USA 104(42):16422–16427. 45(1):235–247. 20. Schultheiß R, Van Bocxlaer B, Wilke T, Albrecht C (2009) Old fossils-young species: 50. Hunt G (2006) Fitting and comparing models of phyletic evolution: Random walks and Evolutionary history of an endemic gastropod assemblage in Lake Malawi. Proc Biol beyond. Paleobiology 32(4):578–601. – Sci 276(1668):2837 2846. 51. Hunt G (2012) paleoTS: Modeling evolution in paleontological time-series, version 21. Genner MJ, Knight ME, Haesler MP, Turner GF (2010) Establishment and expansion of 0.4-4. Available at http://cran.r-project.org/web/packages/paleoTS/index.html. Ac- Lake Malawi rock fish populations after a dramatic Late Pleistocene lake level rise. cessed June 18, 2012. EVOLUTION – Mol Ecol 19(1):170 182. 52. Sheets HD, Mitchell CE (2001) Why the null matters: Statistical tests, random walks 22. Kat PW (1986) Origins of the molluscan faunas of the African Great Lakes: New evi- and evolution. Genetica 112–113(1):105–125. dence. Am Malacol Bull 4(1):107. 53. Paradis E, et al. (2012) APE: Analyses of phylogenetics and evolution, version 3.0-4. 23. Van Bocxlaer B (2005) Changes in the malacofauna of Lake Malawi since mid-Holo- Available at http://cran.r-project.org/web/packages/ape/index.html. Accessed June 18, cene times. Geol Belg 8(3):124. 2012. 24. Estes S, Arnold SJ (2007) Resolving the paradox of stasis: Models with stabilizing se- 54. Paradis E (2012) Analysis of Phylogenetics and Evolution with R (Springer, New York), – lection explain evolutionary divergence on all timescales. Am Nat 169(2):227 244. 2nd Ed. fi 25. Rabosky DL, Lovette IJ (2008) Density dependent diversi cation in North American 55. Revell LJ (2012) PhyTools: Phylogenetic tools for comparative biology (and other wood-warblers. Proc Biol Sci 275(1649):2363–2371. things), version 0.2-1. Available at http://cran.r-project.org/web/packages/phytools/ 26. Seehausen O (2006) African cichlid fish: A model system in adaptive radiation re- index.html. Accessed November 12, 2012. search. Proc Biol Sci 273(1597):1–12. 56. Bapst DW (2012) Paleotree: Paleontological and phylogenetic analyses of evolu- 27. Hunt G, Bell MA, Travis MP (2008) Evolution toward a new adaptive optimum: Phe- tion, version 1.4. Available at http://cran.r-project.org/web/packages/paleotree/ notypic evolution in a fossil stickleback lineage. Evolution 62(3):700–710. index.html. Accessed June 18, 2012. 28. Hendry AP, Kinnison MT (1999) The pace of modern life: Measuring rates of con- 57. Bapst DW (2012) Paleotree: An R package for paleontological and phylogenetic temporary microevolution. Evolution 53(6):1637–1653. analyses of evolution. Methods Ecol Evol 3:803–807. 29. Lynch M (1990) The rate of morphological evolution in mammals from the standpoint 58. Harmon L, et al. (2009) Geiger: Analysis of evolutionary diversification, version 1.3-1. of the neutral expectation. Am Nat 136(6):727–741. Available at http://cran.r-project.org/web/packages/geiger/index.html. Accessed June 30. Gingerich PD (1983) Rates of evolution: Effects of time and temporal scaling. Science 18, 2012. 222(4620):159–161. 59. Ackerly D (2009) Conservatism and diversification of plant functional traits: Evolu- 31. Hunt G (2012) Measuring rates of phenotypic evolution and the inseparability of tionary rates versus phylogenetic signal. Proc Natl Acad Sci USA 106(Suppl 2): tempo and mode. Paleobiology 38(3):351–373. 19699–19706. 32. Gibbs HL, Grant PR (1987) Oscillating selection on Darwin’s finches. Nature 327(6122): 60. Gingerich PD (1993) Quantification and comparison of evolutionary rates. Am J Sci 511–513. 293A(Suppl):453–478.

Van Bocxlaer and Hunt PNAS | August 20, 2013 | vol. 110 | no. 34 | 13897 Downloaded by guest on September 27, 2021