Habitat preference and diversification rates in a speciose lineage of diving beetles

Habitat preference and diversification rates in a speciose lineage of diving beetles Running title: Habitat type does not affect diversification Adrián Villastrigo1*, Pedro Abellán2 and Ignacio Ribera1† 1 Institute of Evolutionary Biology (CSIC-Universitat Pompeu Fabra), Barcelona, Spain 2 University of Seville, Seville, Spain † Deceased April 15, 2020 *Corresponding author: Adrián Villastrigo, [email protected] Keyword: Habitat preference, Habitat transitions, Diversification rates, Diving beetles, Hydroporini, Macroevolution Abstract The long-term geological stability of aquatic habitats has been demonstrated to be a determinant in the evolution of macroinvertebrate fauna, with species in running (lotic) waters having lower dispersal abilities, smaller ranges and higher gene flow between populations than species in standing (lentic) environments. Lotic species have been hypothesized to be more specialised, but the diversification dynamics of both habitat types have not been studied in detail. Using a speciose lineage of water beetles we test here whether diversification rates are related to the habitat preference of the species and its consequences on turnover, which we expect to be higher for lotic taxa. Moreover, we tested whether life in lotic environments is acting as an evolutionary dead-end as it is considered an ecological specialisation. We built a comprehensive molecular phylogeny with 473 terminals representing 421 of the 689 known species of the tribe Hydroporini (Coleoptera, Dytiscidae), using a combination of sequences from four mitochondrial and two nuclear genes plus 69 mitogenomes obtained with NGS. We found a general pattern of gradual acceleration of diversification rate with time, with 2–3 significant diversification shifts. However, habitat is not the main factor driving diversification in Hydroporini based on SecSSE analyses. The most recent common ancestor of Hydroporini was reconstructed as a lotic species, with multiple shifts to lentic environments. Most frequent transitions were estimated from lentic and lotic habitats to the category “both”, followed by transitions from lotic to lentic and lentic to lotic respectively, although with very similar rates. Contrary to expectations, we found little evidence for differences in diversification dynamics between habitats, with lotic environments clearly not acting as evolutionary dead-ends in Hydroporini. 1. Introduction The unevenness in species richness across the Tree of Life is one of the most striking features of biological diversity. Two major hypotheses have been proposed to explain differences in richness among clades (Wiens, 2017): time as the main diversification factor (clade age hypothesis, Li and Wiens, 2019), and/or differential diversification rates among clades (rates hypothesis, Hugall and Stuart-Fox, 2012, Baker et al., 2014, McGuire et al., 2014, Wiens, 2015a, Seeholzer et al., 2017). A possible cause for differential speciation or extinction rates between clades is the constraint posed by the habitats that the species live in (the “habitat templet” concept; Southwood, 1977). This idea has been tested multiple times in both terrestrial and aquatic environments, such as the shift from saline to freshwater in amphibians (Hou et al., 2011), the type of nesting of some birds (Irestedt et al., 2009), or the habitat preference in lizards (Collar et al., 2010). In aquatic environments, the main habitat constraint is the one set between running (lotic) and standing (lentic) waters, with species tending to be restricted to one of them (e.g. Ribera et al., 2001, Ribera et al., 2008, Kalkman et al., 2018, Ye et al., 2019). The instability at geological time-scales of small, isolated lentic water bodies forces a higher dispersal capability of the species living in them, and in consequence broader geographical ranges with higher gene flow between populations for lentic than for lotic species (Ribera and Vogler, 2000, Marten et al., 2006, Hof et al., 2006, Abellán et al., 2009, Sánchez-Fernández et al., 2012). The combination of larger geographic ranges and an increased gene flow between populations should lead to a lower probability of speciation and extinction in lentic species (Papadopoulou et al., 2008, Abellán et al., 2009; see Ribera, 2008 for a review). When habitat preferences are maintained across speciation events, or the physical setting of an area favours the dominance of either lotic or lentic habitats for long periods of time, it is expected that the consequences of habitat constraints will extend to macroevolutionary patterns of whole lineages. Thus, it could be expected that the lower probability of speciation and extinction in lentic species would result in their lower evolutionary turnover. However, the total number of species within a clade at any given time will depend on the balance between speciation and extinction. As the same general factors promoting speciation also favour extinction (Stanley, 1979, Hubbell, 2001, Gould, 2002, Jablonski, 2007), there are no clear predictions as to which habitat type would promote higher species diversity at any given time. Limited efforts have been made to understand whether diversification rates are influenced by habitat preference in aquatic environments, perhaps due to the difficulties to estimate diversification rates (Rabosky, 2009) and the need for comprehensive phylogenies to account for precise rates (Barraclough and Nee, 2001). Ribera et al. (2001) compared species-level phylogenies of the lentic diving beetle Ilybius and the lotic Deronectes, but did not find any significant difference in the lineage through time plots of the linearized trees, intended to reflect the diversification pattern of a lineage (Barraclough and Nee, 2001). Deronectes displayed a higher frequency of recent splits than Ilybius, consistent with a higher evolutionary turnover and a higher frequency of recent species, but differences were not significant compared to expected patterns under a constant speciation rate null model. Letsch et al. (2016) demonstrated heterogeneous speciation rates across Odonata, with lentic environments linked to higher rates. However, an incomplete taxonomic sampling (c. 17% of recognized anisopteran species) and the heterogeneous complex biological traits across species may have affected their results. More recently, Désamoré et al. (2018) suggested no differential diversification rates between lotic and lentic environments in diving beetles, but again with a very incomplete sampling (ca. 4% of dytiscid species) that unavoidably underestimated the number of habitat shifts within genera (e.g. the 189 described species of Hydroporus have very heterogeneous habitats, but in Désamoré et al., 2018 only five of them were included and it was considered an homogeneous lentic genus). The higher persistence of local populations, together with smaller geographical ranges and reduced gene flow, led Ribera (2008) to hypothesize that lotic species may show the characteristics of an ecological specialisation. If this was the case, it seems likely that this could bias the probability of habitat shifts, as once a lotic specialisation is established the possibility of a transition to a lentic habitat may be impaired. In this work we tested whether diversification rates are related to the habitat preference of the species, and whether habitat transitions can be associated to changes in diversification rates. To test if life in lotic environments can lead to an ecological specialisation (Ribera, 2008), we estimated the probabilities of habitat transitions between habitats. To investigate these questions, we used the tribe Hydroporini of the diving beetle family Dytiscidae with a comprehensive dataset including ca. 65% of its 689 named species (Nilsson and Hájek, 2019). Species of Hydroporini are more homogeneous in morphology and general life style than other tribes within Dytiscidae, but they are highly diverse and with numerous species in either lotic or lentic environments. 2. Methods 2.1. Molecular data Molecular sequences were obtained from 473 Hydroporini specimens, with representatives of all genera except Amurodytes, Etruscodytes, Siamoporus and Tassilodytes, all of them monotypic (Nilsson and Hájek, 2019). We included in the analyses only species of Hydroporini as currently understood (Nilsson and Hájek, 2019), following the most recent phylogenies of Dytiscidae (Désamoré et al., 2018). We compiled available published sequences plus 500 newly obtained sequences (Table S1). DNA was extracted with a nondestructive protocol using commercial kits, mostly Qiagen DNeasy Tissue Kit (Hildesheim, Germany). Voucher specimens and DNA samples are kept in the collections of Institute of Evolutionary Biology (IBE, Barcelona, Spain) and Museo Nacional de Ciencias Naturales (MNCN, Madrid). DNA was amplified by PCR for six gene fragments in five reactions for a set of mitochondrial and nuclear genes: i) 5′ end of cytochrome c oxidase subunit I (COI-5, the barcode, Hebert et al., 2004), ii) 3′ end of cytochrome c oxidase subunit I (COI-3), iii) 5′ end of the large ribosomal RNA unit (16S) plus the complete leucine tRNA (tRNA-L1) plus the 5′ end of NADH dehydrogenase subunit I (NAD1), iv) the 5′ end of the small ribosomal unit (18S) and v) an internal fragment of the Histone 3 gene (H3). PCR primers and conditions can be found in Table S2. Additionally, we sequenced 69 mitochondrial genomes of selected species from an equimolar bulk of DNA following a Mitochondrial Metagenomic protocol adapted from Crampton-Platt et al. (2016). Our pipeline includes preparation of DNA libraries with TruSeq Nano DNA kit, sequencing in an Illumina MiSeq (insert size 300 bp) and de novo assemblies with different software to validate each contig. To assure correct identification of contigs, BLAST searches (Madden et al., 2009) were made against a custom database with Sanger sequences of the COI and 16S genes (see Villastrigo et al., 2020 for the detailed pipeline). All sequences were trimmed using Geneious v10.2.5 (Kearse et al., 2012) and uploaded to the ENA database (accession numbers LR998984-LR999418 and MW465237- MW465304, see Table S1 for a complete list of voucher specimens, mitochondrial genomes and accession numbers). 2.2. Phylogenetic reconstruction Sequences from both mitochondrial genomes and Sanger sequencing were aligned using MAFFT 7.409 (Katoh et al., 2009) with the G-INS-I algorithm. To avoid alignments with partitions with very unequal amounts of missing data, the partition scheme was based on the type and source of the sequence: (i) nuclear genes (18S plus H3), (ii) leucine tRNA, present in both Sanger and mitogenomes sequences, (iii) tRNAs present only in mitogenome sequences, (iv) coding sequences present in both Sanger and mitogenome sequences (COI plus NAD1), (v) coding sequences present only in mitogenome sequences, (vi) rRNA present in both Sanger and mitogenome sequences (16S) and (vii) rRNA present only in mitogenome sequences (12S). We opted for a simple model selection approach, considering two complex evolutionary models (HKY + G + I and GTR + G + I) in combination with two different clock models (strict and uncorrelated lognormal) following recent published recommendations (Nascimento et al., 2017, Abadi et al., 2019). We run BEAST v1.10.4 (Suchard et al., 2018) for 100 My generations using a Yule speciation process as prior. Phylogenies were calibrated with the estimated crown age of Hydroporini from the most recent phylogeny of Dytiscidae (103.6 Ma), obtained using a combination of fossils (Désamoré et al., 2018). To reduce computation time, the monophyly of the subtribes was constrained based on the results of a previous analysis without any topological constraint with IQ-TREE v2.0 (Minh et al., 2020) with the same partition scheme, allowing the software to estimate the best evolutionary model per partition (Kalyaanamoorthy et al., 2017) and assessing branch support using SH-aLRT and ultrafast bootstrap (Hoang et al., 2018) (Fig. S1). Convergence was assessed using Tracer v1.7 (Rambaut et al., 2018). 2.3. Evolution of habitat preference Habitat preferences of the species were compiled using published data and our own observations (Table S1). Ancestral habitat was reconstructed using Beast v1.10.4 under an asymmetrical model with three states: i) Lotic, ii) Lentic, and iii) species that are habitually found in both habitats (i.e. category ‘Both’). Phylogenetic signal of habitat preference was calculated using a likelihood ratio test that compared the maximized likelihood of a model with strong signal (λ = 1) and a model without signal (λ = 0). Transitions rates between habitat categories were obtained in Beast. 2.4. Patterns of diversification and their relationship with habitat Diversification rates were calculated for all genera with more than two sampled terminals in BAMM v2.5.0 (Rabosky, 2014). In addition to the consensus tree obtained in Beast, we randomly selected a set of 100 trees from the post-burnin posterior distribution to account for phylogenetic uncertainty, considering in both cases the sampled fraction of species following BAMM specifications. We also estimated diversification rates of the same clades using the method of moments estimator (MM-estimator, Magallón and Sanderson, 2001) using the geiger R-package (Harmon et al., 2008) for both crown and stem ages under different extinction scenarios (no extinction, ε = 0; intermediate, ε = 0.5; and high extinction, ε = 0.9). To test whether habitat preference types had different diversification rates we used a statedependent speciation and extinction approach that account for alternative models and concealed states with the R package SecSSE (Herrera-Alsina et al., 2019). We tested three alternative models: i) a constant rate model (CR) where all species share the same speciation and extinction rates, ii) the examined-trait dependent diversification model (ETD), where the speciation and extinction rates may vary only between the examined states and iii) a concealed-trait-dependent diversification model (CTD) where speciation an extinction rates may vary only between concealed states. The three models were compared using Akaike Information Criterion weights (Wagenmakers and Farrell, 2004). We also estimated the contemporary diversification pattern in each habitat type. Mean diversification rates of extant species (tip rates) plus their variance were obtained from the set of 100 post-burnin trees, and the slouch R-package (Hansen et al., 2008) was used to obtain pairwise comparison of each habitat type trends. We used the estimated tip rates and their variance, and habitat preference as reconstructed in Beast was included as a fixed factor. The slouch package can integrate a measure of variance within the response, minimizing errors attributed to an uneven number of measures across taxa. 3. Results 3.1. Molecular phylogeny and taxonomic status The phylogenetic reconstruction included 473 terminals representing 423 species and 21 subspecies, plus 29 unnamed terminals (Table S3), covering ca. 62% of the recognized taxa of Hydroporini (688 species plus 27 subspecies, Nilsson and Hájek, 2019). Analyses under uncorrelated lognormal clock did not reach an adequate effective sample size, whilst analyses with a strict clock did with essentially the same topology (results not shown). The HKY + G + I model obtained a larger effective sample size than the GTR + G + I and therefore was selected for subsequent analyses. The consensus phylogeny (Fig. S2) fully resolved the relationships between major clades (posterior probability [pp] > 1 for subtribe relationships) and recovered most genera as monophyletic, except for Tiporus (which included the monotypic genus Sekaliporus), a clade including the American genera Neoporus plus Heterosternuta plus Haideoporus, and a clade with Hydrocolus plus Hydroporus. The only unexpected result was the placement of Lioporeus within Siettitiina (Fig. S1, S2), while it is currently considered within Hydroporina (Nilsson and Hájek, 2019). A collapsed phylogeny showing major clades is provided in Fig. 1. 3.2. Evolution of habitat preference Beast reconstructed the most recent common ancestor of Hydroporini as a lotic species, with multiple shifts to lentic environments (Fig. 2). In addition, some genera reconstructed as lotic such as Graptodytes, Megaporus or Strenopriscus contained nested subclades that represented reversals to other habitat state. Transitions from Lentic to Lotic occurred multiple times mostly on terminal branches (except in a clade within Hydroporus), with a similar pattern for species of category Both. 3.2. Evolution of habitat preference Beast reconstructed the most recent common ancestor of Hydroporini as a lotic species, with multiple shifts to lentic environments (Fig. 2). In addition, some genera reconstructed as lotic such as Graptodytes, Megaporus or Strenopriscus contained nested subclades that represented reversals to other habitat state. Transitions from Lentic to Lotic occurred multiple times mostly on terminal branches (except in a clade within Hydroporus), with a similar pattern for species of category Both. The impact of habitat preference on diversification rates was tested on SecSSE. The examined-trait dependent model estimated a higher diversification rate for species living in both habitats, followed by lentic species, which have a slightly higher diversification rate than lotic species (Table 1). It was also estimated a contrasting turnover between lotic and lentic species, being more than four times higher for species inhabiting lotic habitats as a result of fivefold extinction rate than lentic habitats. Nonetheless, when alternative state-dependent models were compared, SecSSE estimated that the concealed-trait dependent model (CTD) was the best scoring model based on AIC weights (Table 2). The contemporary diversification trends estimated in slouch for the terminals considered, as in the ETD model by SecSSE (Table 1), that lentic species tended to increase their diversification rates in comparison with lotic species (contrast 0.0004 ± 0.0002) or species in category ‘both’ (contrast 0.0007 ± 0.0006); raw contemporary rates for extant taxa are available at Table S5. Differences between states Lotic and Both were not significant due to large standard errors. The impact of habitat preference on diversification rates was tested on SecSSE. The examined-trait dependent model estimated a higher diversification rate for species living in both habitats, followed by lentic species, which have a slightly higher diversification rate than lotic species (Table 1). It was also estimated a contrasting turnover between lotic and lentic species, being more than four times higher for species inhabiting lotic habitats as a result of fivefold extinction rate than lentic habitats. Nonetheless, when alternative state-dependent models were compared, SecSSE estimated that the concealed-trait dependent model (CTD) was the best scoring model based on AIC weights (Table 2). The contemporary diversification trends estimated in slouch for the terminals considered, as in the ETD model by SecSSE (Table 1), that lentic species tended to increase their diversification rates in comparison with lotic species (contrast 0.0004 ± 0.0002) or species in category ‘both’ (contrast 0.0007 ± 0.0006); raw contemporary rates for extant taxa are available at Table S5. Differences between states Lotic and Both were not significant due to large standard errors. 4. Discussion 4.1. Molecular phylogeny We found some unexpected phylogenetic relationships, in disagreement with the current classification of Hydroporini. Previous authors noticed the morphological resemblance of Neoporus and Heterosternuta (e.g. Larson et al., 2000, Miller and Bergsten, 2016), considering them as closely related genera with overlapping distributions (Larson et al., 2000) and similar habitat preferences (Table S2). The main morphological character separating both genera is the apex of the aedeagus (bifid in Heterosternuta), and our samples were identified based on this character. Our results point to a more complex scenario, with Heterosternuta as a mononophyletic lineage derived within a larger clade including both Neoporus and Haideoporus, suggesting the need for a taxonomic revision of the group. Miller et al. (2013) reported a similar result, with a clade Bloom, D.D., Weir, J.T., Piller, K.R. Lovejoy, N.R., 2013. Do freshwater fishes diversify faster than marine fishes? A test using state-dependent diversification analyses and molecular phylogenetics of new world Silversides (Atherinopsidae), Evolution, 67(7), 2040-2057. Cieslak, A., Fresneda, J., Ribera, I., 2014. Life-history specialization was not an evolutionary dead-end in Pyrenean cave beetles. P. R. Soc. B.-Biol. Sci., 281(1781), 20132978. Collar, D.C., Schulte, J.A., O’meara, B.C., Losos, J.B., 2010. Habitat use affects morphological diversification in dragon lizards. J. Evolution. Biol., 23, 1033–1049. Crampton-Platt, A., Yu, D.W., Zhou, X., Vogler, A.P., 2016. Mitochondrial metagenomics: letting the genes out of the bottle. GigaScience, 5, 15. Cusimano, N., Renner, S.S., 2010. Slowdowns in diversification rates from real phylogenies may not be real. Syst. Biol., 59, 458–464. Darwin, C. (1859). On the origin of species by means of natural selection, or the preservation of favoured races in the struggle for life. John Murray, London. Day, E.H., Hua, X., Bromham, L., 2016. Is specialization an evolutionary dead end? Testing for differences in speciation, extinction and trait transition rates across diverse phylogenies of specialists and generalists. J. Evolution. Biol., 29, 1257–1267. Désamoré, A., Laenen, B., Miller, K.B., Bergsten, J., 2018. Early burst in body size evolution is uncoupled from species diversification in diving beetles (Dytiscidae). Mol. Ecol., 27, 979– 993. Everson, K.M., Soarimalala, V., Goodman, S.M., Olson, L.E., 2016. Multiple loci and complete taxonomic sampling resolve the phylogeny and biogeographic history of tenrecs (Mammalia: Tenrecidae) and reveal higher speciation rates in Madagascar’s humid forests. Syst. Biol., 65, 890–909. Fall, H.C., 1923. A revision of North American species of Hydroporus and Agaporus. J.D. Sherman, Mt. Vernon, New York. García‐Vázquez, D., Bilton, D. T., Alonso, R., Benetti, C. J., Garrido, J., Valladares, L. F., Ribera, I., 2016. Reconstructing ancient Mediterranean crossroads in Deronectes diving beetles. J. Biogeogr., 43(8), 1533-1545. García-Vázquez, D., Bilton, D.T., Foster, G.N., Ribera, I., 2017. Pleistocene range shifts, refugia and the origin of widespread species in western Palaearctic water beetles. Mol. Phylogenet. Evol., 114, 122–136. Gould, S. J., 2002. The structure of evolutionary theory. Harvard University Press. Hansen, T.F., Pienaar, J., Orzack, S.H., 2008. A comparative method for studying adaptation to a randomly evolving environment. Evolution, 62, 1965–1977. Harmon, L.J., Weir, J.T., Brock, C.D., Glor, R.E., Challenger, W., 2008. GEIGER: Investigating evolutionary radiations. Bioinformatics, 24, 129–131. Hebert, P. D., Penton, E. H., Burns, J. M., Janzen, D. H., Hallwachs, W., 2004. Ten species in one: DNA barcoding reveals cryptic species in the neotropical skipper butterfly Astraptes fulgerator. P. Natl. Acad. Sci. USA, 101(41), 14812-14817. Henao Diaz, L.F., Harmon, L.J., Sugawara, M.T.C., Miller, E.T., Pennell, M.W., 2019. Macroevolutionary diversification rates show time dependency. P. Natl. Acad. Sci. USA, 1– 6. Herrera-Alsina, L., van Els, P., Etienne, R. S., 2019. Detecting the dependence of diversification on multiple traits from Phylogenetic Trees and Trait Data. Systematic Biology, 68, 317–328. Hof, C., Brändle, M., Brandl, R., 2006. Lentic odonates have larger and more northern ranges than lotic species. J. Biogeogr., 33, 63–70. Hoang, D. T., Chernomor, O., Von Haeseler, A., Minh, B. Q., Vinh, L. S. (2018). UFBoot2: improving the ultrafast bootstrap approximation. Mol. Biol. Evol., 35(2), 518-522. Hou, Z., Sket, B., Fiser, C., Li, S., 2011. Eocene habitat shift from saline to freshwater promoted Tethyan amphipod diversification. P. Natl. Acad. Sci. USA, 108, 14533–14538. Hubbell, S. P., 2001. The unified neutral theory of biodiversity and biogeography (MPB-32). Princeton University Press. Hugall, A.F., Stuart-Fox, D., 2012. Accelerated speciation in colour-polymorphic birds. Nature, 485(7400), 631-634. Irestedt, M., Fjeldså, J., Dalén, L., Ericson, P.G.P., 2009. Convergent evolution, habitat shifts and variable diversification rates in the ovenbird-woodcreeper family (Furnariidae). BMC Evol. Biol., 9, 1–13. Jablonski, D., 2007. Scale and hierarchy in macroevolution. Palaeontology, 50(1), 87-109. Janz, N., Nylin, S., Wahlberg, N., 2006. Diversity begets diversity: host expansions and the diversification of plant-feeding insects. BMC Evol. Biol., 6(1), 4. Kalkman, V.J., Rafał, J.B., Geert, B., Suhling, F., Termaat, T., 2018. Diversity and conservation of European dragonflies and damselflies (Odonata). Hydrobiologia, 811, 269–282. Kalyaanamoorthy, S., Minh, B. Q., Wong, T. K., von Haeseler, A., Jermiin, L. S., 2017. ModelFinder: fast model selection for accurate phylogenetic estimates. Nat. Methods, 14(6), 587. Katoh, K., Asimenos, G., Toh, H., 2009. Multiple alignment of DNA sequences with MAFFT, in: Bioinformatics for DNA sequence analysis. pp. 39–64. Kearse, M., Moir, R., Wilson, A., Stones-Havas, S., Cheung, M., Sturrock, S., Buxton, S., Cooper, A., Markowitz, S., Duran, C., Thierer, T., Ashton, B., Meintjes, P., Drummond, A., 2012. Geneious Basic: An integrated and extendable desktop software platform for the organization and analysis of sequence data. Bioinformatics, 28, 1647–1649. Kelley, S.T., Farrell, B.D., 1998. Is specialization a dead end? The phylogeny of host use in Dendroctonus bark beetles. Evolution, 52, 1731–1743. Larson, D.J., Alarie, Y., Roughley, R.E., 2000. Predaceous diving beetles (Coleoptera: Dytiscidae) of the Nearctic Region with emphasis on the fauna of Canada and Alaska. NRC Research Press, Ottawa. Letsch, H., Gottsberger, B., Ware, J.L., 2016. Not going with the flow: A comprehensive timecalibrated phylogeny of dragonflies (Anisoptera: Odonata: Insecta) provides evidence for the role of lentic habitats on diversification. Mol. Ecol., 25, 1340–1353. Li, H., Wiens, J.J., 2019. Time explains regional richness patterns within clades more often than diversification rates or area. Am. Nat., 193(4), 514-529. Liebherr, J. K., Hajek, A. E., 1990. A cladistic test of the taxon cycle and taxon pulse hypotheses. Cladistics, 6(1), 39-59. Madden, T.L., Camacho, C., Ma, N., Coulouris, G., Avagyan, V., Bealer, K., Papadopoulos, J., 2009. BLAST+: architecture and applications. BMC Bioinformatics, 10(1), 421. Magallón, S., Sanderson, M.J., 2001. Absolute diversification rates in Angiosperm clades. Evolution, 55, 1762–1780. Marten, A., Brändle, M., Brandl, R., 2006. Habitat type predicts genetic population differentiation in freshwater invertebrates. Mol. Ecol., 15, 2643–2651. McGuire, J.A., Witt, C.C., J. V. Remsen, J., Corl, A., Rabosky, D.L., Altshuler, D.L., Dudley, R., 2014. Molecular phylogenetics and the diversification of hummingbirds. Curr. Biol., 24, 910–916. McPeek, M.A., 2008. The ecological dynamics of clade diversification and community assembly. Am. Nat., 172, E270–E284. Meyer, A.L.S., Román-Palacios, C., Wiens, J.J., 2018. BAMM gives misleading rate estimates in simulated and empirical datasets. Evolution, 72(10), 2257–2266. Meyer, A.L.S., Wiens, J.J., 2018. Estimating diversification rates for higher taxa: BAMM can give problematic estimates of rates and rate shifts. Evolution, 72(1), 39–53. Miller, K.B., Bergsten, J., 2016. Diving beetles of the world: Systematics and biology of the Dytiscidae. JHU Press, Baltimore. Miller, K.B., Bergsten, J., 2014. The phylogeny and classification of predaceous diving beetles, in: Yee, D.A. (Eds.), Ecology, Systematics and the Natural History of Predaceous Diving Beetles (Coleoptera: Dytiscidae). Springer, pp. 49–172. Miller, K.B., Jean, A., Alarie, Y., Hardy, N., Gibson, R., 2013. Phylogenetic placement of north American subterranean diving beetles (Insecta: Coleoptera: Dytiscidae). Arthropod Syst. Phylo., 71, 75–90. Minh, B.Q., Schmidt, H.A., Chernomor, O., Schrempf, D., Woodhams, M.D., von Haeseler, A., Lanfear, R., 2020. IQ-TREE 2: New models and efficient methods for phylogenetic inference in the genomic era. Mol. Biol. Evol. Moen, D., Morlon, H., 2014. Why does diversification slow down? Trends Ecol. Evol., 29, 190– 197. Monaghan, M. T., Gattolliat, J. L., Sartori, M., Elouard, J. M., James, H., Derleth, P., Glaizot, O., de Moor, F., Vogler, A. P. (2005). Trans-oceanic and endemic origins of the small minnow mayflies (Ephemeroptera, Baetidae) of Madagascar. P. Roy. Soc. B-Biol. Sci., 272(1574), 1829-1836. Moore, B.R., Huelsenbeck, J.P., Rannala, B., May, M.R., Höhna, S., 2016. Critically evaluating the theory and performance of Bayesian analysis of macroevolutionary mixtures. P. Natl. Acad. Sci. USA, 113, 9569–9574. Morlon, H., Potts, M.D., Plotkin, J.B., 2010. Inferring the dynamics of diversification: A coalescent approach. PLoS Biol., 8(9). Nascimento, F. F., dos Reis, M., Yang, Z., 2017. A biologist’s guide to Bayesian phylogenetic analysis. Nat. Ecol. Evol., 1(10), 1446-1454. Nürk, N.M., Uribe-Convers, S., Gehrke, B., Tank, D.C., Blattner, F.R., 2015. Oligocene niche shift, Miocene diversification - Cold tolerance and accelerated speciation rates in the St. John’s Worts (Hypericum, Hypericaceae). BMC Evol. Biol., 15, 1–13. Papadopoulou, A., Bergsten, J., Fujisawa, T., Monaghan, M. T., Barraclough, T. G., Vogler, A. P., 2008. Speciation and DNA barcodes: testing the effects of dispersal on the formation of discrete sequence clusters. Philos. T. R. Soc. B, 363(1506), 2987-2996. Rabosky, D.L., 2019. Phylogenies and diversification rates: Variance cannot be ignored. Syst. Biol., 68(3), 538-550. Rabosky, D.L., 2018. BAMM at the court of false equivalency: A response to Meyer and Wiens. Evolution, 72(10), 2246–2256. Rabosky, D.L., 2014. Automatic detection of key innovations, rate shifts, and diversitydependence on phylogenetic trees. PLoS ONE, 9(2). Rabosky, D.L., 2009. Ecological limits and diversification rate: Alternative paradigms to explain the variation in species richness among clades and regions. Ecol. Lett., 12, 735–743. Rabosky, D.L., Mitchell, J.S., Chang, J., 2017. Is BAMM flawed? Theoretical and practical concerns in the analysis of multi-rate diversification models. Syst. Biol., 66, 477–498. Rambaut, A., Drummond, A.J., Xie, D., Baele, G., Suchard, M.A., 2018. Posterior summarization in Bayesian phylogenetics using Tracer 1.7. Syst. Biol., 67, 901–904. Ribera, I., 2008. Habitat constraints and the generation of diversity in freshwater macroinvertebrates, in: Lancaster, J., Briers, R.A. (Eds.), Aquatic Insects: Challenges to populations: Proceedings of the Royal Entomological Society’s 24th Symposium. pp. 289– 311. Ribera, I., Barraclough, T.G., Vogler, A.P., 2001. The effect of habitat type on speciation rates and range movements in aquatic beetles: Inferences from species-level phylogenies. Mol. Ecol., 10, 721–735. Ribera, I., Castro, A., Díaz, J.A., Garrido, J., Izquierdo, A., Jäch, M.A., Valladares, L.F., 2011. The geography of speciation in narrow-range endemics of the “Haenydra” lineage (Coleoptera, Hydraenidae, Hydraena). J. Biogeogr., 38, 502–516. Ribera, I., Vogler, A.P., 2000. Habitat type as a determinant of species range sizes: the example of lotic – lentic differences in aquatic Coleoptera. Biol. J. Linn. Soc., 71, 33–52. Ribera, I., Vogler, A.P. (2004). Speciation of Iberian diving beetles in Pleistocene refugia (Coleoptera, Dytiscidae). Mol. Ecol., 13(1), 179-193. Ribera, I., Vogler, A.P., Balke, M., 2008. Phylogeny and diversification of diving beetles (Coleoptera : Dytiscidae). Cladistics, 24, 563–590. Sánchez-Fernández, D., Lobo, J.M., Millán, A., Ribera, I., 2012. Habitat type mediates equilibrium with climatic conditions in the distribution of Iberian diving beetles. Global Ecol. Biogeogr., 21, 988–997. Seeholzer, G.F., Claramunt, S., Brumfield, R.T., 2017. Niche evolution and diversification in a Neotropical radiation of birds (Aves: Furnariidae). Evolution, 71, 702–715. Sorenson, L., Santini, F., Alfaro, M.E., 2014. The effect of habitat on modern shark diversification. J. Evolution. Biol., 27, 1536–1548. Southwood, T.R.E., 1977. Habitat, the templet for ecological strategies? J. Anim. Ecol., 46, 337–365. Stanley, S. M., 1979. Macroevolution, pattern and process. WH Freeman. 332pp. Stern, D.B., Breinholt, J., Pedraza-Lara, C., López-Mejía, M., Owen, C.L., Bracken-Grissom, H., Fetzner, J.W., Crandall, K.A., 2017. Phylogenetic evidence from freshwater crayfishes that cave adaptation is not an evolutionary dead-end. Evolution, 71, 2522–2532. Suchard, M.A., Lemey, P., Baele, G., Ayres, D.L., Drummond, A.J., Rambaut, A., 2018. Bayesian phylogenetic and phylodynamic data integration using BEAST 1.10. Virus Evolution, 4, 1–5. Toussaint, E.F.A., Condamine, F.L., Hawlitschek, O., Watts, C.H., Porch, N., Hendrich, L., Balke, M., 2015. Unveiling the diversification dynamics of australasian predaceous diving beetles in the cenozoic. Syst. Biol., 64, 3–24. Villastrigo, A., Arribas, P. Ribera, I., 2020. Irreversible habitat specialization does not constraint diversification in hypersaline water beetles. Mol. Ecol., 29, 3637-3648. Wagenmakers, E., Farrell, S. (2004). AIC model selection using Akaike weights. Psychonomic Bulletin & Review, 11, 192–196. Wiens, J.J., 2017. What explains patterns of biodiversity across the Tree of Life?: New research is revealing the causes of the dramatic variation in species numbers across branches of the Tree of Life. BioEssays, 39, 1–10. Wiens, J.J., 2015a. Explaining large-scale patterns of vertebrate diversity. Biol. Lett., 11, 10–13. Wiens, J.J., 2015b. Faster diversification on land than sea helps explain global biodiversity patterns among habitats and animal phyla. Ecol. Lett., 18, 1234–1241. Ye, Z., Damgaard, J., Yang, H., Hebsgaard, M.B., Weir, T., Bu, W., 2019. Cladistics Phylogeny and diversification of the true water bugs (Insecta: Hemiptera: Heteroptera: Nepomorpha). Cladist, 1–16. Tables Table 1. Estimated speciation and extinction rates for habitat states using SecSSE sensu ETD model. Model Speciation Extinction Diversification Turnover Lotic 0.0519 0.0101 0.0418 0.1935 Both 0.0806 8.40 x 10-14 0.0806 1.04 x 10-12 Lentic 0.0489 0.0021 0.0468 0.0433 Table 2. Maximum Likelihood estimations of SecSSE for different models of trait-dependent diversification Model k ML AICw CR 8 -2258.928 6.035 x 10-10 ETD 12 -2245.494 7.549 x 10-6 CTD 12 -2233.699 0.99999 Figures Figure 1. Collapsed molecular phylogeny of the genera of Hydroporini, obtained with BEAST. Circles in nodes represent posterior probabilities = 1. The monophyly of the subtribes was constrained according to the results of a previous IQTREE search, which obtained maximum support for all of them (Fig. S1). Figure 2. Reconstruction of the habitat preference in Hydroporini using BEAST with an asymmetric transition probability model.