The colonization of new adaptive zones is widely recognized as one of the hallmarks of adaptive radiation. However, the adoption of novel resources during this process is rarely distinguished from phenotypic change because morphology is a common proxy for ecology. How can we quantify ecological novelty independent of phenotype? Our study is split into two parts: we first document a remarkable example of ecological novelty, scale-eating (lepidophagy), within a rapidly-evolving adaptive radiation of Cyprinodon pupfishes on San Salvador Island, Bahamas. This specialized predatory niche is known in several other fish groups, but is not found elsewhere among the 1,500 species of atherinomorphs. Second, we quantify this ecological novelty by measuring the time-calibrated phylogenetic distance in years to the most closely-related species with convergent ecology. We find that scale-eating pupfish are separated by 168 million years of evolution from the nearest scale-eating fish. We apply this approach to a variety of examples and highlight the frequent decoupling of ecological novelty from phenotypic divergence. We observe that novel ecology is not always tightly correlated with rates of phenotypic or species diversification, particularly within recent adaptive radiations, necessitating the use of additional measures of ecological novelty independent of phenotype.
Citation: Martin CH, Wainwright PC (2013) On the Measurement of Ecological Novelty: Scale-Eating Pupfish Are Separated by 168 my from Other Scale-Eating Fishes. PLoS ONE 8(8): e71164. https://doi.org/10.1371/journal.pone.0071164
Editor: Nadir Alvarez, University of Lausanne, Switzerland
Received: February 12, 2013; Accepted: July 3, 2013; Published: August 19, 2013
Copyright: © 2013 Martin, Wainwright. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: Funding was provided by an NSF Graduate Research Fellowship, an NSF doctoral dissertation improvement grant, a UC Davis Dissertation Year fellowship, the Northern California ARCS Foundation, the Center for Population Biology, student grants from the Gerace Research Centre (2008, 2011), and NSF grant DEB-1061981 to PCW. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
Novel ecology, independent of any phenotypic novelty, is rarely adequately addressed in discussions of evolutionary novelty, except in the very general sense of an increase in niche diversity (–; but see –). Here we define novel ecology as the adoption of resources and a way of life (sensu Simpsons's adaptive zones ) not only unique within a given community (i.e. niche diversity), but often unique across a clade's global range and discontinuous with global niche diversity within that clade (Table 1). For example, only a single species of spider is known to be herbivorous, feeding on the Beltian bodies of an ant-plant . Likewise, blood-feeding, folivory, and tool-use are unique to Darwin's finches, despite the global abundance of these resources in all passerine communities , . Nonetheless, our perception of the novelty of any particular niche is still dependent on the size of the outgroup used for comparison and thus remains subjective (e.g. , , ).
There is currently no index for quantifying the rarity or novelty of an ecological niche within a clade, to our knowledge, despite the extensive literature on quantifying niche diversity (e.g. –). While approaches such as niche modeling have proven incredibly productive for measuring ecology , , it is easy to overlook discontinuous resource use (i.e. novelty) when examining only shared resource axes among taxa (e.g. ). Thus, at the macroevolutionary level, ecological novelty is only described qualitatively , , ,  or phenotypic diversification is used as an indicator of ecological diversification (e.g. , ), despite the fact that these axes of organismal diversification are sometimes decoupled (e.g. , –).
Foraging on the scales of other fishes is a specialized predatory niche surprisingly rare across the teleost tree of life despite the omnipresence of this resource in all fish communities. Specialized scale-eating has evolved at least 19 times independently (4–6 times in both African cichlids and South American characoids, plus examples in four additional freshwater families and seven marine families –) and is currently known in about 50 species of teleost  and the cookie-cutter shark (Isistius brasiliensis). Scale-eating is accomplished through a wide variety of morphological and behavioral solutions, including open-gaped ram feeding , one-sided lateral strikes with asymmetrical jaws or behavioral handedness , –, aggressive mimicry, rasping with external teeth, group hunting, cleaning and mucus-feeding (reviewed in ), and possibly deep-water pelagic ambush . Tooth shape is also exceptionally variable among scale-eaters, even within scale-eating clades , .
Despite the considerable diversity of morphological and behavioral strategies that underlie the scale-eating trophic specialization, there appear to be some universal features of this ecological strategy. First, all scale-eaters must be small relative to the size of their prey due to the high energy-to-resource ratio per strike . For example, juvenile facultative scale-eaters switch to piscivory after they grow larger than their prey . Conversely, the most specialized scale-eaters often switch completely to scales when they reach adult size and never grow larger than their prey , . Second, a corollary to this observation is that specialized scale-eaters never forage on both scales and whole fish (other than fish larvae, e.g. ) at the same time. Third, scale-eater populations always remain much smaller than their prey populations. Lastly, lateral jaw asymmetry has evolved in scale-eating specialists at least four times independently across a wide range of foraging strategies and habitats (from Amazonia to Lake Tanganyika to the mesopelagic ocean: , , , ; but also see ). This suggests that laterally asymmetric jaws may be a universally adaptive trait for scale-eaters by allowing lateral attacks while pursuing prey, whereas symmetrical jaws may require perpendicular alignment with the prey. The many scale-eating specialists with symmetric jaws (including the scale-eating pupfish) may be constrained by negligible genetic variation for jaw asymmetry.
Across the global distribution of approximately 1,500 species of atherinomorph fishes , no scale-eaters have been documented previously (see review in ). We investigated reports of a potential piscivore or scale-eater within a 10,000-year-old adaptive radiation of Cyprinodon pupfishes (Atherinomorpha: Cyprinodontidae) endemic to a single 11-mile long island in the Bahamas –. The spectacular natural history of the scale-eating pupfish inspired us to adopt a simple phylogenetic novelty index to quantify ecological novelty on a temporal scale: evolutionary distance to the most closely related species with convergent ecology. Our approach was inspired by several case studies of morphological novelties which extoll the novelty of a trait relative to the age of the clade from which it has emerged –.
Here we document the rapid evolution of lepidophagy within this radiation and test for three convergent features of the scale-eating niche: 1) reduced adult size relative to prey, 2) absence of piscivory, and 3) low frequency relative to prey population. We then apply the phylogenetic novelty index to a variety of examples of ecological novelty in order to place our discovery of a scale-eating pupfish in context. This is an example of a dramatic ecological transition that is largely decoupled from phenotypic divergence, and like other similar examples, has thus been previously overshadowed by a focus on phenotype as a proxy for ecology. We propose further that many recent adaptive radiations contain a previously overlooked dimension of diversification: ecological novelty (Table 1). We identify a 168-million-year ecological novelty within a recent adaptive radiation of Cyprinodon pupfishes, use a phylogenetic-distance metric to quantify this novelty, and find that the ecological novelty index within this radiation far exceeds that of a second Cyprinodon adaptive radiation with nearly 3-fold higher rates of morphological diversification. Our study addresses three major questions throughout: 1) Do clear examples of ecological novelty exist within extremely young species? 2) Can ecological novelty be measured independent of phenotype? 3) Is ecological diversification always strongly associated with morphological diversification?
We broadly define lepidophage as any free-living, pursuit or ambush aquatic predator which non-lethally removes scales, skin, mucus, fins, eyes, or whole chunks of its prey (e.g. , ). We use the term ‘scale-eater’ for this niche, although no scale-eater derives nutriment exclusively from scales, but rather the protein-rich mucus and skin tissue surrounding scales (e.g. ). There are no obligate scale-eaters (but see one example of an obligate mucus-feeder: ); however, when a species spends the majority of its foraging time extracting scales from other fishes it generally shows specialized morphology or behaviors which are presumably adaptations for scale-eating. Thus, as with previous authors , we distinguish specialized scale-eaters from the many species which incidentally forage on scales removed during aggressive bouts, attempts at piscivory, or scavenged from dead fishes (e.g. , ).
We investigated reports of a “piscivorous form”  or a “scale-eater/piscivore”  within a recent adaptive radiation of Cyprinodon pupfishes endemic to San Salvador Island, Bahamas. This undescribed species (Cyprinodon sp. ‘bulldog’) is confined to several interior hypersaline lakes on the island in sympatry with two closely related Cyprinodon species (C. sp. ‘normal’ and C. sp. ‘durophage’ ) and only two other euryhaline fish species (Gambusia hubbsi and Atherinomorus stipes; , ). C. sp. ‘normal’ is nearly indistinguishable in morphology and habit to widespread Caribbean and Atlantic coast populations of C. variegatus. C. sp. ‘durophage’ possesses a novel nasal appendage formed from a skeletal extension of the head of the maxilla which may facilitate hard-shelled prey extraction (, CHM pers. obs.).
This radiation of three sympatric Cyprinodon species shows significant genetic differentiation within and among lakes (Fst = 0.1–0.31; ). Despite ongoing gene flow, as in many recent adaptive radiations (e.g. ), all three species are partially reproductively isolated due to strong ecological selection against hybrids with intermediate phenotypes . Initial field observations and laboratory trials also indicate that C. sp. ‘bulldog’ is further isolated by female mate choice for conspecific males [CHM pers. obs.]. Preliminary genomic-scale analyses of genetic structure among these species support a single origin of each specialist species followed by dispersal among lakes [CHM unpublished data]. Our initial results also indicate that C. sp. ‘bulldog’ individuals form a monophyletic clade across lakes, indicating strong reproductive isolation from C. variegatus and C. sp. ‘durophage’ [CHM unpublished data]. C. sp. ‘bulldog’ and C. sp. ‘durophage’ are currently being formally described as new species [Martin and Wainwright in revision].
In July 2008, all three Cyprinodon species were collected from two lake populations, Crescent Pond (CP) and Little Lake (LL), on San Salvador Island, Bahamas. Adults were sampled from 0.3–2 m depth using a hand net while snorkeling or by seine net. At the surface, individuals were immediately euthanized in an overdose of MS-222 (Finquel, Argent Laboratories Inc.). Animal care procedures followed the recommended guidelines for laboratory animal care and were approved by the University of California, Davis Animal Care and Use Committee (Protocols #15640, 15908). Further digestion was halted with an intra-peritoneal injection of 15% formalin, followed by fixing the whole specimen in 15% formalin. After fixing, specimens were moved to 70% ethanol for long-term storage. Research permits were approved by the Bahamas Environment, Science & Technology commission, Ministry of the Environment, and the Ministry of Agriculture (2011); and by the Department of Marine Resources, Ministry of Agriculture and Marine Resources (2008) with the support of the Gerace Research Centre.
In the laboratory, approximately 3 cm of the anterior gut was removed from each individual for stomach content analysis. Food items were spread on a Sedgwick-Rafter cell and sorted at 25× magnification under a stereoscope. Items were identified to class with additional categories for scales, whole fishes (comprising both Cyprinodon and Gambusia species), detritus, and crushed shells or silt. Major food items included various macroalgae (predominantly Batophora oerstedi and Cladophoropsis macromeres), widgeon grass (Ruppia maritima), ostracods, various gastropods, scales, whole fish, and polychaetes (Table 2). The total volume of each dietary component was estimated from the total number of 1 µl cells covered in the chamber. The proportion of each dietary component was calculated from its volume relative to the total volume of all components for each individual (as in ; Table 2). Individuals with empty stomachs (n = 22) were excluded.
Visual censuses of species abundance were also conducted in three lakes containing all three Cyprinodon species in July 2008. All adults and subadults were counted by a single observer within a 0.3 m×30-m transect while snorkeling. Censuses were repeated 4 times each in Crescent Pond, Little Lake, and Osprey Lake. Although population abundances vary seasonally and annually, these censuses estimate the relative abundance of scale-eaters and their prey.
In March and July 2011, adults (>2.0 cm) from all three species were collected from Crescent Pond and Little Lake for stable isotope analyses. After euthanasia, approximately 5 mg of muscle tissue was removed from the caudal peduncle of each specimen and dried at 60°C for 48 hours. Tissue samples were weighed and submitted to the UC Davis Stable Isotope Facility for measurement on a PDZ Europa ANCA-GSL elemental analyzer, interfaced to a PDZ Europa 20–20 isotope ratio mass spectrometer (Sercon Ltd., Cheshire, UK). For comparisons of standard length (SL), adults of each species from larger samples collected in 2008 and 2011 were measured using dial calipers.
We also performed a literature search to identify all Cyprinodontidae species with published data on stomach contents (reviewed in ). Additional species descriptions, field guides , , data from FishTraits , and community knowledge (e.g. American Killifish Association) confirmed that no other atherinomorphs are known scale-eaters.
Phylogenetic novelty index
Ideally, the rarity of any ecological niche within a clade could be estimated from the transition rate among niche states across a distribution of time-calibrated phylogenies with near-complete sampling at the species level (e.g. see  for transition rates among dietary categories in 1/3 of all mammal species). This approach would incorporate much of the uncertainty in phylogenetic estimation and time-calibration (assuming adequate prior estimates of node ages) by summing over a posterior distribution of models and phylogenies , . However, studying rare ecological transitions across the tree of life is still limited by the lack of large trees with near-complete sampling at the species level (or even large trees with 50% sampling, as investigated by ). Thus, it is rarely possible to find one chronogram containing multiple species with rare, convergent ecology.
Due to the current lack of large chronograms connecting rare convergent ecology, we used parsimony to reconstruct the origin of rare ecological niches. Parsimony ignores uncertainty in ancestral estimation of niche evolution (e.g. ); however, this uncertainty is relatively minor for rare, recent events across the tree of life (e.g. blood-feeding most likely evolved within the adaptive radiation of Darwin's finches). Furthermore, parsimony is not sensitive to incomplete lineage sampling (e.g. ).
We calculated phylogenetic novelty from the amount of time separating two lineages with convergent ecology minus the inferred origination times of the convergent niche in each clade:where 1 and 2 correspond to the most closely related species or clades exhibiting convergent ecology (Fig. 1). Thus, novelty is measured as the time over which ancestors connecting two convergent ecological niches do not occupy the niche. This timespan should be inversely correlated to the probability of re-emergence of the niche which may fall off exponentially with increasing time, as expected under a diffusion process of continuous trait evolution (e.g. ). Unlike transition rates calculated from discrete character shifts across a tree , , , this temporal measure of novelty is unaffected by relative lineage diversification rates among the groups with convergent ecology.
The most closely related species with convergent ecology are a clade of scale-eating cichlids from Lake Tanganyika, Perissodus spp. A simplified cladogram connecting these groups is illustrated with numbers at tips corresponding to the number of scale-eating (black) and non-scale-eating (red) species within the Cyprinodon and Tanganyikan haplochromine clades (note that thousands of additional outgroup species have been pruned and these species numbers are not presented). Phylogenetic novelty index (indicated by the green line; 168 million years in Table 3) is calculated from twice the divergence time (t) minus the estimated origin of scale-eating in each clade (a and b). The stem age of the Perissodus clade is used as a conservative estimate of the origin of scale-eating (b). Note that the phylogenetic novelty index is not the same if applied to scale-eating cichlids (Table 3), which have repeatedly colonized this niche within each Great Lake radiation. Also note the aggressive mimicry in Perissodus straelini and the crypsis of female Cyprinodon sp. ‘bulldog’. Photo credits: Jennifer O. Reynolds, Tony Terceira.
Note that this index is not symmetric due to the hierarchical structure of phylogenies. Thus, scale-eating pupfish are equally distant from all cichlids, in which scale-eating has evolved multiple times, whereas scale-eating cichlids are more closely related to other lineages of scale-eating cichlids and exhibit a much lower novelty index (Table 3). This asymmetric property of the index usefully reflects the greater novelty of scale-eating within Cyprinodontiformes than within Cichlidae.
Complete knowledge of the most closely related species with convergent ecology (and therefore complete knowledge that all intervening species do not express the ecology) and a time-calibrated phylogeny connecting these species is necessary to calculate the novelty index. For common ecological niches this is a daunting task; however, the occurrence of highly unusual ecologies is usually much better known due to their noteworthy status. The identification of convergent ecology is still a subjective choice by the investigator which requires detailed knowledge of the natural history of the group and can only be decided on a case-by-case basis. Although identifying specialized scale-eating within pupfishes was straightforward, assignment of specialized ecology is more subjective when the use of a novel resource varies across populations (e.g. the vampire finch), similar niches can be grouped (e.g. is eye-biting a form of scale-eating or a different niche? ), or different species exploit the same resource in different ways (e.g. sawfish are piscivores ).
Due to the lack of a comprehensive chronogram containing multiple ecologically novel species, we pieced together node ages from different chronograms to apply this index to the examples in Table 3. This is not ideal and can introduce considerable bias when comparing node ages estimated from different datasets, particularly given the known discrepancies among fossil calibration points , , uncorrelated relaxed-clock models estimated independently among different taxa and genes , and often complex relationship among conflicting time-calibration priors, tree topology, and branch lengths (e.g. see supplemental methods to ). In cases where chronograms were lacking, we used fossil age estimates from the earliest known representative of a given clade or geological age estimates of the lake basin which contained the adaptive radiation (Table 3). Thus, our current estimates of the phylogenetic novelty index should be interpreted cautiously as they are dependent on the particular set of chronograms and fossils available. These estimates should serve mainly to illustrate application of our approach. Nonetheless, the exponential expansion of molecular sequencing (e.g. ) and trend toward increasing tree sizes with greater access to fossil calibrations , – should facilitate greater use of this index in the near future. Timetree also provides a useful database for searching existing chronograms .
To be conservative in calculating niche age, we took the stem age, rather than the crown age, of clades in which all species exhibit the focal ecology, such as the Tanganyikan Perissodus clade of scale-eaters , because the novel ecology likely evolved sometime along this stem lineage before the evolution of the crown group . For clades in which only one species colonized the novel niche during adaptive radiation, such as San Salvador pupfishes, we took the crown age of the radiation for the niche age because the novel niche was likely colonized sometime along this terminal branch within the radiation. However, formal ancestral reconstruction methods (e.g. ) for estimating a range of times for the origin of novel ecology would be more appropriate for more recently diverged taxa or when clade membership is uncertain. Formal ancestral state reconstruction methods allow inference of niche origination times across a distribution of trees incorporating phylogenetic uncertainty . Nonetheless, most of the uncertainty in estimation of the phylogenetic novelty index arises from the large variance in estimates of node age, even when comparing nodes ages estimated jointly within the same chronogram. Increasing sophistication of time-calibration methods (e.g. ) and increased incorporation of fossil data should greatly improve inference of clade ages.
We include common ecological niche transitions for comparison in Table 3; however, we stress that the use of parsimony for ancestral character reconstruction is never justified when considerable uncertainty is present and likelihood or Bayesian estimation of ancestral character states is needed. In practice, as a given niche becomes more commonly used across a clade (i.e. as niche transition rate increases), the phylogenetic novelty index rapidly goes to zero as uncertainty in reconstruction of ancestral states prevents robust inference of ancestral niche use. Common niche use also makes assignment of convergent ecology more subjective. Instead, we suggest this index is most useful for quantifying and comparing rare niche transitions which can be treated as discrete.
One weakness of this approach, common to all comparative phylogenetic analyses based only on extant taxa, is the lack of ecological knowledge of extinct lineages which may have also colonized the focal niche. This biases estimates of phylogenetic novelty upwards. In particular, very large estimates of phylogenetic novelty should be regarded with increased suspicion due to the greater probability of extinct lineages occupying the focal niche. However, if ecological inferences about extinct taxa are available, this information can easily be incorporated into the novelty index. For example, including fossil sawsharks and sawfishes substantially reduces the ecological novelty of this specialized mode of foraging relative to comparisons of only extant sawfish and sawsharks by increasing estimates of the origins (niche age) of this convergent foraging strategy within each lineage (Table 3).
40–51% of stomach contents of Cyprinodon sp. ‘bulldog’ were comprised of scales from other Cyprinodon and possibly Gambusia fishes in two lakes (Fig. 2, Table 2). Not a single bulldog individual had ingested whole fish, despite the occurrence of piscivory in Cyprinodon sp. ‘normal’ (Table 2). Stable isotope analyses confirmed that C. sp. ‘bulldog’ occupied the highest trophic position (δ15N) relative to sympatric Cyprinodon species in both winter and summer seasons in two lakes (Fig. 2).
a,b) Proportion of scales (mean ± SE) in stomach contents of each species in a) Crescent Pond and b) Little Lake populations. c,d) Relative trophic position (δ15N: mean ± SE) of each species from samples collected in March (first bar) and July (second bar) in a) Crescent Pond and b) Little Lake populations. Multiple samples of ‘bulldog’ and ‘durophage’ were not available in March in Little Lake.
Scale-eating by C. sp. ‘bulldog’ was frequently observed in the wild and occurred approximately once per minute during daylight hours (CHM pers. obs.). Adult C. sp. ‘bulldog’ were slightly smaller (CP: 2.27±.05 (mean ± SE; n = 17), LL: 2.68±.04 cm SL (n = 21)) than their most common prey target, C. sp. ‘normal’ (CP: 3.08±.07 (n = 39), LL: 3.01±.06 cm SL (n = 62)). C. sp. ‘normal’ was, by far, the most abundant species in all three lakes surveyed (mean % ± SE: CP: 92.7±1.0; LL: 92.3%±1.1; OL: 93.8%±1.7). C. sp. ‘bulldog’ frequency ranged from 0.2–3.1% of all pupfishes across the three lakes surveyed (CP: 0.9±0.4; LL: 3.1±0.8; OL: 0.2±0.1).
Phylogenetic novelty index
The most recent maximum clade credibility ultrametric tree for Ovalentaria places cichlids + Pholidichthys + Polycentrids as the sister group to atherinomorphs , . This node is weakly supported and blennioids may also be the most closely related group containing a scale-eating species, the mimic fangblenny (Plagiotremus tapeinosoma); however, this alternative phylogenetic hypothesis has little impact on the estimated novelty index. Under the phylogenetic hypothesis of cichlids + Pholidichthys + Polycentrids as sister group to atherinomorphs, the most closely related specialized scale-eaters to Cyprinodon sp. ‘bulldog’ are found in several lineages of scale-eating cichlids within adaptive radiations in Lakes Victoria, Tanganyika, and Malawi , . Using the stem age of the oldest of these scale-eating cichlid clades, the Tanganyikan Perissodus clade , we estimate that approximately 168 million years separates the San Salvador scale-eating pupfish from the origins of scale-eating in the Perissodus stem lineage (Fig. 1, Table 3). Several classic examples of adaptive radiation display astonishing levels of ecological novelty by this index (e.g. the vampire finch, the false Chameleon Anolis: Table 3), far exceeding their phenotypic novelty (e.g. ). Conversely, classic radiations which only partition a subset of their ancestral resource base do not show unusual levels of ecological novelty due to frequent parallel evolution of the same niche across similar environments (e.g. threespine sticklebacks, Geospiza seed-eating groundfinches, Chichancanab pupfishes: Table 3).
We document the recent evolution of a scale-eating pupfish, a unique niche within over 1,500 atherinomorph species distributed globally (Fig. 2, Table 2). To place this novelty in perspective, we observe that the most closely related scale-eating specialist is separated by 168 million years of evolution from the scale-eating pupfish (Fig. 1). We propose this simple phylogenetic novelty index as a quantitative measure for comparing rare and novel ecological transitions across the tree of life and apply it to several classic and unusual niches (Table 3). This index generalizes estimates of novelty beyond the bias of restricting attention to a named clade or outgroup (e.g. Table 1; ) and complements qualitative descriptions of ecological novelty (e.g. , , ). Although identifying convergent ecology often remains subjective on a case-by-case basis, our approach provides a method for quantifying the rarity of unassailable examples of convergent ecology across the tree of life.
We document the evolution of a specialized scale-eating pupfish species, Cyprinodon sp. ‘bulldog’, within a sympatric adaptive radiation on San Salvador Island, Bahamas. 41–51% of its diet was composed of scales (Fig. 2, Table 2). The actual proportion was probably much higher as most of the remaining stomach contents in this species were detritus, which may consist mainly of digested skin and mucus tissue from scale-feeding (Table 2). Scale-eating was independently supported by the higher trophic position of C. sp. ‘bulldog’ across lakes and seasons as inferred from δ15N isotope ratios (Fig. 2). As predicted by the functional and ecological constraints on specialized scale-eating, C. sp. ‘bulldog’ was slightly smaller than its prey, never captured whole fish, and occurred at low frequency across all three populations surveyed.
Rates of morphological evolution in the San Salvador Cyprinodon radiation are exceptional outliers among Cyprinodon clades . For example, jaw length and adductor mandibulae muscle mass are diversifying 51 and 47 times faster than background rates in allopatric species, respectively . In a second, independent adaptive radiation of Cyprinodon pupfishes endemic to Laguna Chichancanab, rates of morphological diversification are two- to three-fold higher: tooth length and adductor muscle mass are diversifying 131 and 120 times faster than background rates, respectively . These exceptional rates of morphological diversification are not only due to the young age of these clades, but rather the colonization of novel ecological niches: scale-eating and durophagy on San Salvador Island, Bahamas and piscivory and zooplanktivory in Laguna Chichancanab, Mexico .
We show that the ecological novelty of the San Salvador adaptive radiation is even more exceptional than its morphological diversification rate. A single species within this clade has recently colonized an ecological niche, scale-eating, that is approximately 168 million years removed from the most closely related species with convergent ecology (Fig. 1, Table 3). The phylogenetic novelty index does not account for extinct scale-eating taxa; however, the rarity and complete absence of this feeding specialization outside lacustrine adaptive radiations in atherinomorphs suggests this may be an accurate estimate of the novelty of this niche. Although the Laguna Chichancanab Cyprinodon radiation is diverging nearly three-fold faster for certain morphological traits, these species have invaded ecological niches that are not particularly novel for Cyprinodontidae, piscivory and zooplanktivory. Piscivory is common within nearly all fish groups and the most closely related specialized piscivore is most likely Orestias cuvieri, the extinct top-predator from the Lake Titicaca adaptive radiation of Orestias pupfishes . Specialized zooplanktivores are also known from the Lake Titicaca radiation (e.g. O. ispi ) and Old World Aphanius pupfishes such as Aphanius anatoliae splendens . The phylogenetic novelty index highlights the exceptional ecological novelty of a scale-eating pupfish within the San Salvador radiation, despite the nearly 3-fold lower morphological diversification rates in this clade (Table 3).
On the evolution of ecological novelty
Application of the phylogenetic novelty index to several recent and classic examples of adaptive radiation (Table 3) illuminates dramatic examples of extremely rare, major ecological transitions which are easily overlooked when using morphological proxies for ecology. For example, the beaks of Darwin's finches are not particularly diverse relative to other domed-nest clade finches in the Caribbean . However, the vampire finch (Geospiza difficilis septentrionalis) has adopted a novel and extremely rare niche among birds: blood-feeding. This resource forms a major part of the diet in only one other specialized lineage of birds, the oxpeckers (Buphagus spp.), which are also known to open wounds and drink blood , . Bill shapes of the vampire finch and other ecologically novel Darwin's finches, such as the vegetarian finch and tool-using woodpecker finch, are not particularly unusual , . Conversely, the widely divergent beak shapes of Darwin's seed-eating ground finches (e.g. Geospiza magnirostra vs. Geospiza fuliginosa) correspond to specializations for eating seeds of various sizes , a very common niche across the domed-nest clade (; Table 2). Thus, Darwin's finches provide an excellent example of the decoupling of ecological novelty from dramatic morphological diversification.
Once we are able to measure differences in ecological novelty among taxa, the highly uneven distribution of ecological novelty among clades and environments begs explanation. Ecological novelty does not appear to be distributed randomly among clades, but often reappears repeatedly within the same clades invading similar environments and may be particularly common within recent adaptive radiations (Table 1). Why are some populations able to exploit these novel resources but not others? Are transitions to novel niches necessary for population persistence or would partitioning more common resource axes be sufficient? What combination of environment and lineage-specific factors is necessary to drive the evolution of novel ecology? These questions are particularly clear when comparing similar lineages that have speciated and adapted to the same environment, but display exceptional differences in their propensity to evolve ecological novelty: for nearly every compelling example of island adaptive radiation, there are similar lineages that have colonized the same environment and speciated to some extent, but fail to evolve novel ecology, such as finches and mockingbirds in the Galapagos , , honeycreepers and thrushes in the Hawaiian Islands , , pupfishes and mosquitofish in Caribbean salt lakes , , and different lineages of cichlids within sympatric radiations , . This pattern also occurs when the same lineage is distributed across many similar environments: rapid ecological diversification is often restricted to a few places , .
The evolution of ecological novelty, a subset of many types of niche divergence between populations, remains unexplained by existing ecological divergence mechanisms (e.g. ); as with phenotypic diversification, we have a solid theoretical and empirical framework for microevolutionary change, but no clear link between these processes and larger patterns of macroevolutionary diversification and stasis (e.g. ). For example, why has such an exceptionally rare trophic niche evolved only on San Salvador Island in the Bahamas? One possibility is that adaptation to novel ecological niches is strongly influenced by the location of their corresponding fitness peaks on the adaptive landscape. Field experiments on San Salvador Island measuring the fitness landscape for thousands of F2 hybrids placed in field enclosures support this idea . Two fitness peaks for growth and survival within the range of hybrid phenotypes measured corresponded to the phenotypes of C. sp. ‘normal’ and C. sp. ‘durophage’ observed in the wild. In contrast, hybrids resembling the scale-eater had low growth and survival across two lakes at two different densities . If a fitness peak for scale-eating exists in this environment, it may require a highly specialized phenotype for successful performance which was not recovered within the F2 hybrids used in this experiment. Thus, rare novel niches may reflect distant, isolated, or narrow fitness peaks surrounded by a large fitness valley on the adaptive landscape.
More generally, how should the ecological novelty index be interpreted? This index is a quantitative measure of the rarity of a niche. Alternatively, with complete lineage sampling, we could estimate the transition rate into any niche from formal ancestral state reconstructions as a quantitative measure of rarity: lower transition rates correspond to rarer niches. However, with this approach, transition rates also depend on lineage diversification rates. Is a rare niche more novel within a clade of 100 species than within a clade of 10 species? We think a more relevant measure of novelty is the minimum distance to a species with convergent rare ecology, regardless of lineage diversification rates spanning this time period.
Second, the niche is a complex and dynamic mapping of population persistence onto a hyper-dimensional ecological space ,  and an emergent property of both the evolving organism and the shifting biotic and abiotic environments , , . The rarity of any ecological niche is a function of the global abundance of its ecological space and the abundance of taxa able to persist (‘fundamental niche’) and currently competing within that space (‘ecological opportunity’). Thus, the novelty index can also be interpreted as the ‘findability’ (see ) of a niche on the adaptive landscape across the global biosphere, given these constraints of abundance, persistence, and competition.
While evolutionary novelty is frequently addressed at other levels of biological organization, here we provide a framework for quantifying and comparing novelty at the ecological level. We define ecological novelty as a major ecological transition to a new adaptive zone, often unique across a clade's global range (Table 1). We use a phylogenetic novelty index for quantifying ecological novelty: the time separating the inferred origin of the novelty from the inferred origin of the most closely related species with convergent ecology (Fig. 1). A synthesis of previous scattered observations reveals that ecological novelty is particularly common within recent adaptive radiations in isolated environments (Table 1) and that specialized species colonizing novel ecological niches are not necessarily the most phenotypically divergent (Table 3). In particular, we document the rapid evolution of a scale-eating specialist within an incipient adaptive radiation of Cyprinodon pupfishes endemic to San Salvador Island (Fig. 2, Table 2). We estimate this species is separated by approximately 168 million years of evolution from the most closely related scale-eating specialist (Table 3). The phylogenetic novelty index should facilitate further comparative analyses of novelty across the tree of life and illustrate the previously overlooked dimension of exceptional ecological diversification.
Research permits were approved by the Bahamas Environment, Science & Technology commission, Ministry of the Environment, and the Ministry of Agriculture (2011); and by the Department of Marine Resources, Ministry of Agriculture and Marine Resources (2008) with the support of the Gerace Research Centre. We thank Samantha Price, Michael Turelli, and Thomas Schoener for helpful comments on the manuscript.
Conceived and designed the experiments: CHM. Performed the experiments: CHM. Analyzed the data: CHM. Contributed reagents/materials/analysis tools: CHM PCW. Wrote the paper: CHM. Commented on the manuscript: PCW.
- 1. Moczek AP (2008) On the origins of novelty in development and evolution. BioEssays 30: 432–47.
- 2. Lefebvre L, Reader SM, Sol D (2004) Brains, innovations and evolution in birds and primates. Brain Behav Evol 63: 233–46.
- 3. Muller GB, Wagner GP (1991) Novelty in evolution: restructuring the concept. Ann Rev Ecol Syst 22: 229–256.
- 4. Nitecki NH (1990) Evolutionary Innovations. University of Chicago Press, Chicago.
- 5. Givnish TJ, Sytsma KJ, Smith JF, Hahn WJ, Benzing DH, et al.. (1997) Molecular evolution and adaptive radiation in Brocchinia (Bromeliaceae: Pitcairnioideae) atop tepuis of the Guayana Shield. In: Molecular evolution and adaptive radiation (Givnish, TJ, Sytsma, KJ, eds), Cambridge University Press, Cambridge.
- 6. Blount ZD, Borland CZ, Lenski RE (2008) Historical contingency and the evolution of a key innovation in an experimental population of Escherichia coli. Proc Nat Acad Sci USA 105: 7899–906.
- 7. Martin CH, Wainwright PC (2011) Trophic novelty is linked to exceptional rates of morphological diversification in two adaptive radiations of Cyprinodon pupfishes. Evolution 65: 2197–212.
- 8. Price SA, Holzman R, Near TJ, Wainwright PC (2011) Coral reefs promote the evolution of morphological diversity and ecological novelty in labrid fishes. Ecol Lett 14: 462–469.
- 9. Losos JB (2009) Lizards in an evolutionary tree: ecology and adaptive radiation of Anoles. University of California Press, Berkeley.
- 10. Simpson GG (1944) Tempo and mode in evolution. Columbia University Press, New York.
- 11. Meehan CJ, Olson EJ, Reudink MW, Kyser TK, Curry RL (2009) Herbivory in a spider through exploitation of an ant-plant mutualism. Curr Biol 19: R892–R893.
- 12. Grant PR, Grant BR (2008) How and why species multiply: the radiation of Darwin's finches. Princeton University Press, Princeton.
- 13. Burns KJ, Hackett SJ, Klein NK (2002) Phylogenetic relationships and morphological diversity in Darwin's finches and their relatives. Evolution 56: 1240–1252.
- 14. Carlquist SJ (1965) Island life: a natural history of the islands of the world. Natural History Press, Michigan.
- 15. Roughgarden J (1972) Evolution of niche width. Am Nat 106: 683–718.
- 16. Cooper VS (2002) Quantifying the fundamental and realized niche. 9: 1–9.
- 17. Webb CO, Ackerly DD, McPeek MA, Donoghue MJ (2002) Phylogenies and Community Ecology. Ann Rev Ecol Syst 33: 475–505.
- 18. Magurran A (2003) Measuring biological diversity. John Wiley & Sons, Ltd.
- 19. Warren DL, Glor RE, Turelli M (2008) Environmental niche equivalency versus conservatism: quantitative approaches to niche evolution. Evolution 62: 2868–83.
- 20. Holt R (2009) Bringing the Hutchinsonian niche into the 21st century: ecological and evolutionary perspectives. Proc Nat Acad Sci USA 106: 19659–19665.
- 21. Peterson AT, Soberon J, Pearson RG, Anderson RP, Martinez-Meyer E, et al.. (2011) Ecological niches and geographic distributions. Princeton University Press, Princeton.
- 22. Ackerly D (2009) Conservatism and diversification of plant functional traits: Evolutionary rates versus phylogenetic signal. Proc Nat Acad Sci USA 106: 19699–706.
- 23. Schoener T (1974) Resource partitioning in ecological communities. Science 185: 27–39.
- 24. Martin CH, Genner MJ (2009) High niche overlap between two successfully coexisting pairs of Lake Malawi cichlid fishes. Can J Fish Aquat Sci 66: 579–588.
- 25. Adams DC, Berns CM, Kozak KH, Wiens JJ (2009) Are rates of species diversification correlated with rates of morphological evolution? Proc Roy Soc Lond B 276: 2729–2738.
- 26. Webb JK, Shine R (1994) Feeding habits and reproductive biology of Australian Pygopodid lizards of the genus Aprasia. Copeia 1994: 390–398.
- 27. Alexandrou MA, Oliveira C, Maillard M, McGill RAR, Newton J, et al. (2011) Competition and phylogeny determine community structure in Müllerian co-mimics. Nature 469: 84–8.
- 28. Blankers T, Adams DC, Wiens JJ (2012) Ecological radiation with limited morphological diversification in salamanders. J Evol Biol 25: 634–646.
- 29. Fryer G, Iles TD (1972) The cichlid fishes of the Great Lakes of Africa: their biology and evolution. Oliver and Boyd, Edinburg.
- 30. Sazima I (1983) Scale-eating in characoids and other fishes. Env Bio Fish 9: 87–101.
- 31. Grubh A, Winemiller K (2004) Ontogeny of scale feeding in the Asian glassfish, Chanda nama (Ambassidae). Copeia 2004: 903–907.
- 32. Nakae M, Sasaki K (2002) A scale-eating triacanthodid, Macrorhamphosodes uradoi: prey fishes and mouth “handedness” (Tetraodontiformes, Triacanthoidei). Ichthy Res 49: 7–14.
- 33. Janovetz J (2005) Functional morphology of feeding in the scale-eating specialist Catoprion mento. J Exp Bio 208: 4757–68.
- 34. Hori M (1993) Frequency-dependent natural selection in the handedness of scale-eating cichlid fish. Science 260: 216–219.
- 35. Van Dooren TJM, van Goor HA, van Putten M (2010) Handedness and asymmetry in scale-eating cichlids: antisymmetries of different strength. Evolution 64: 2159–2165.
- 36. Hata H, Yasugi M, Hori M (2011) Jaw laterality and related handedness in the hunting behavior of a scale-eating characin, Exodon paradoxus. PLOS One 6: E29349.
- 37. Lee HJ, Kusche H, Meyer A (2012) Handed foraging behavior in scale-eating cichlid fish: its potential role in shaping morphological asymmetry. PLOS One 7: e44670.
- 38. Takahashi R, Moriwaki T, Hori M (2007) Foraging behaviour and functional morphology of two scale-eating cichlids from Lake Tanganyika. J Fish Bio 70: 1458–1469.
- 39. Nshombo M, Yanagisawa Y, Nagoshi M (1985) Scale-eating in Perissodus microlepis (Cichlidae) and change of its food habits with growth. Japan J Ichthy 32: 66–73.
- 40. Kusche H, Lee HJ, Meyer A (2012) Mouth asymmetry in the textbook example of scale-eating cichlid fish is not a discrete dimorphism after all. Proc R Soc Lond B 279: 4715–4723.
- 41. Setiamarga D, Miya M (2008) Interrelationships of Atherinomorpha (medakas, flyingfishes, killifishes, silversides, and their relatives): the first evidence based on whole mitogenome sequences. Mol Phylo Evol 49: 598–605.
- 42. Holtmeier CL (2000) Morphological and trophic diversification among pupfishes (Cyprinodontidae): dietary, genetic and ontogenetic effects. Doctoral dissertation, Cornell University.
- 43. Holtmeier CL (2001) Heterochrony, maternal effects, and phenotypic variation among sympatric pupfishes. Evolution 55: 330–338.
- 44. Turner BJ, Duvernell DD, Bunt TM, Barton MG (2008) Reproductive isolation among endemic pupfishes (Cyprinodon) on San Salvador Island, Bahamas: microsatellite evidence. Proc Roy Soc Lond B 95: 566–582.
- 45. Wiens JJ, Brandley MC, Reeder TW (2006) Why does a trait evolve multiple times within a clade? Repeated evolution of snakelike body form in squamate reptiles. Evolution 60: 123–41.
- 46. Wiens J (2011) Re-evolution of lost mandibular teeth in frogs after more than 200 million years, and re-evaluating Dollo's Law. Evolution 65: 1283–1296.
- 47. Brandley MC, Huelsenbeck JP, Wiens JJ (2008) Rates and patterns in the evolution of snake-like body form in squamate reptiles: evidence for repeated re-evolution of lost digits and long-term persistence of intermediate body forms. Evolution 62: 2042–64.
- 48. Britz R, Conway K (2009) Spectacular morphological novelty in a miniature cyprinid fish, Danionella dracula n. sp. Proc Roy Soc Lond B 1665: 2179–2186.
- 49. Rubinoff D, Schmitz P (2010) Multiple aquatic invasions by an endemic, terrestrial Hawaiian moth radiation. Proc Nat Acad Sci USA 107: 5903–5906.
- 50. Gorlick D (1980) Ingestion of host fish surface mucus by the Hawaiian cleaning wrasse, Labroides phthirophagus (Labridae), and its effect on host species preference. Copeia 1980: 863–868.
- 51. Winemiller KO, Yan HY (1989) Obligate mucus-feeding in a South American Trichomycterid catfish (Pisces: Ostariophysi). Copeia 1989: 511–514.
- 52. Winemiller KO, Kelso-Winemiller LC, Brenkert AL (1995) Ecomorphological diversification and convergence in fluvial cichlid fishes. Env Biol Fish 44: 235–261.
- 53. Winemiller KO (1991) Ecomorphological diversification in lowland freshwater fish assemblages from five biotic regions. Ecol Monograph 61: 343–365.
- 54. Grant PR, Grant BR (2009) The secondary contact phase of allopatric speciation in Darwin's finches. Proc Nat Acad Sci USA 106: 20141–20148.
- 55. Martin CH, Wainwright PC (2013) Multiple fitness peaks on the adaptive landscape drive adaptive radiation in the wild. Science 339: 208–211.
- 56. Miller RR, Minckley WL, Norris SM (2005) Freshwater Fishes of Mexico. University of Chicago Press, Chicago.
- 57. Seegers L (2000) Killifishes of the World: New World Killies. Verlag A. C. S. GmbH, Germany.
- 58. Frimpong EA, Angermeier P (2009) Fish traits: A database of ecological and life-history traits of freshwater fishes of the United States. Fisheries 34: 487–495.
- 59. Price SA, Hopkins SSB, Smith KK, Roth VL (2012) Tempo of trophic evolution and its impact on mammalian diversification. Proc Nat Acad Sci USA 109: 7008–7012.
- 60. Huelsenbeck J, Rannala B, Masly J (2000) Accommodating phylogenetic uncertainty in evolutionary studies. Science 288: 2349–2350.
- 61. Burnham KP, Anderson A (2002) Model selection and multi-modal inference. 2nd Ed. Springer, New York.
- 62. FitzJohn RG, Maddison WP, Otto SP (2009) Estimating trait-dependent speciation and extinction rates from incompletely resolved phylogenies. Syst Bio 58: 595–611.
- 63. Huelsenbeck JP, Nielsen R, Bollback JP (2003) Stochastic mapping of morphological characters. Syst Bio 52: 131–158.
- 64. Li G, Steel M, Zhang L (2008) More taxa are not necessarily better for the reconstruction of ancestral character states. Syst Bio 57: 647–653.
- 65. O'Meara BC, Ane C, Sanderson MJ, Wainwright PC (2006) Testing for different rates of continuous trait evolution using likelihood. Evolution 60: 922–933.
- 66. FitzJohn RG, Maddison WP, Otto SP (2009) Estimating trait-dependent speciation and extinction rates from incompletely resolved phylogenies. Syst Bio 58: 595–611.
- 67. Maddison WP, Midford PE, Otto SP (2007) Estimating a binary character's effect on speciation and extinction. Syst Bio 56: 701–710.
- 68. Wueringer BE, Squire L, Collin SP (2009) The biology of extinct and extant sawfish (Batoidea: Sclerorhynchidae and Pristidae). Rev Fish Bio Fisheries 19: 445–464.
- 69. Near TJ, Sanderson MJ (2004) Assessing the quality of molecular divergence time estimates by fossil calibrations and fossil-based model selection. Phil Trans Roy Soc 75.
- 70. Ho SY, Phillips MJ (2009) Accounting for calibration uncertainty in phylogenetic estimation of evolutionary divergence times. Syst Bio 58: 367–380.
- 71. Drummond AJ, Ho SYW, Phillips MJ, Rambaut A (2006) Relaxed phylogenetics and dating with confidence. PLOS Biol 4: e88.
- 72. Lemmon AR, Emme SA, Lemmon EM (2012) Anchored hybrid enrichment for massively high-throughput phylogenomics. Syst Bio 61: 727–744.
- 73. Near TJ, Eytan RI, Dornburg A, Kuhn KL, Moore JA, et al. (2012) Resolution of ray-finned fish phylogeny and timing of diversification. Proc Nat Acad Sci USA 109: 13698–13703.
- 74. Wainwright PC, Smith WL, Price SA, Tang KL, Sparks JS, et al. (2012) The evolution of pharyngognathy: a phylogenetic and functional appraisal of the pharyngeal jaw key innovation in labroid fishes and beyond. Systematic Biology 61: 1001–1027.
- 75. Alfaro ME, Santini F, Brock C, Alamillo H, Dornburg A, et al. (2009) Nine exceptional radiations plus high turnover explain species diversity in jawed vertebrates. Proc Nat Acad Sci USA 106: 13410–13414.
- 76. Hedges SB, Dudley J, Kumar S (2006) TimeTree: a public knowledge-base of divergence times among organisms. Bioinformatics 22: 2971–2972.
- 77. Koblmuller S, Egger B, Sturmbauer C, Sefc KM (2007) Evolutionary history of Lake Tanganyika's scale-eating cichlid fishes. Mol Phylo Evol 44: 1295–1305.
- 78. Magallon S, Sanderson MJ (2001) Absolute diversification rates in angiosperm clades. Evolution 55: 1762–1780.
- 79. Heath TA (2012) A hierarchical Bayesian model for calibrating estimates of species divergence times. Syst Bio 61: 793–809.
- 80. Parenti LR (1984) A taxonomic revision of the Andean killifish genus Orestias (Cyprinodontiformes, Cyprinodontidae). Bull Am Mus Nat Hist 178: 107–214.
- 81. Vaux P, Wurtsbaugh W, Trevino H, Marino L, Bustamante E, et al. (1988) Ecology of the pelagic fishes of Lake Titicaca, Peru-Bolivia. Biotropic 20: 220–229.
- 82. Wildekamp RH, Kucuk F, Unlusayin M, Neer WV (1999) Species and subspecies of the genus Aphanius Nardo 1897 (Pisces: Cyprinodontidae) in Turkey. Turkish J Zoo 23: 23–44.
- 83. McElligott AG, Maggini I, Hunziker L, Konig B (2004) Oxpeckers and black rhinos in captivity. Zoo Biol 23: 347–354.
- 84. Weeks P (2000) Red-billed oxpeckers: vampires or tickbirds? Behav Ecol 11: 154–160.
- 85. Arbogast BS, Drovetski SV, Curry RL, Boag PT, Seutin G, et al. (2006) The origin and diversification of Galapagos mockingbirds. Evolution 60: 370–382.
- 86. Lovette IJ, Bermingham E, Ricklefs RE (2002) Morphological diversification and adaptive radiation in Hawaiian songbirds. Proc Roy Soc B 269: 37–42.
- 87. Amadon D (1950) The Hawaiian honeycreepers (Aves, Drepaniidae). Bull Am Mus Nat Hist 95: 151–262.
- 88. Langerhans RB, DeWitt TJ (2004) Shared and unique features of evolutionary diversification. Am Nat 164: 335–349.
- 89. Seehausen O (2006) African cichlid fish: a model system in adaptive radiation research. Proc Roy Soc Lond B 273: 1987–1998.
- 90. Martin C (2012) Weak disruptive selection and incomplete pheonytpic divergence in two classic examples of sympatric speciation: Cameroon crater lake cichlids. American Naturalist 180: E90–109.
- 91. Schluter D (2000) The ecology of adaptive radiation, Oxford University Press, Oxford.
- 92. Uyeda JC, Hansen TF, Arnold SJ, Pienaar J (2011) The million-year wait for macroevolutionary bursts. Proc Nat Acad Sci USA 108: 15908–15913.
- 93. Hutchinson GE (1957) Concluding remarks. Cold Spring Harbor Symposium Quantitative Biology XXII: 415–427.
- 94. Leibold MA, McPeek MA (2006) Coexistence of the niche and neutral perspectives in community ecology. Ecology 87: 1399–1410.
- 95. Schoener TW (1989) The ecological niche. In: Ecological concepts: the contribution of ecology to an understanding of the natural world (ed. Cherrett, J. M.). pp. 79–113. Blackwell Scientific Publications, Oxford
- 96. McCandlish DM (2013) On the findability of genotypes. Evolution In press.
- 97. Freed LA, Conant S, Fleischer RC (1987) Evolutionary ecology and radiation of Hawaiian passerine birds. TREE 2: 196–203.
- 98. Herrel A, Holanova V (2008) Cranial morphology and bite force in Chamaeleolis lizards–adaptations to molluscivory? Zoology 111: 467–475.
- 99. Joyce DA, Lunt DH, Bills R, Turner GF, Katongo C, et al. (2005) An extant cichlid fish radiation emerged in an extinct Pleistocene lake. Nature 435: 90–95.
- 100. Kontula T, Kirilchik SV, Vainola R (2003) Endemic diversification of the monophyletic cottoid fish species flock in Lake Baikal explored with mtDNA sequencing. Mol Phylo Evol 27: 143–155.
- 101. Sideleva VG (2003) The endemic fishes of Lake Baikal. Backhuys Publishers, Netherlands.
- 102. Macdonald KS, Yampolsky L, Duffy JE (2005) Molecular and morphological evolution of the amphipod radiation of Lake Baikal. Mol Phylo Evol 35: 323–343.
- 103. Vainola R, Witt JDS, Grabowski M, Bradbury JH, Jazdzewski KM, et al. (2008) Global diversity of amphipods (Amphipoda; Crustacea) in freshwater. Hydrobio 595: 241–255.
- 104. Kambysellis MP, Craddock EM (1997) Ecological and reproductive shifts in the diversification of the endemic Hawaiian Drosophila. In: Molecular Evolution and Adaptive Radiation (Givnish TJ and Sytsma KJ eds.), Cambridge University Press, Cambridge.
- 105. Heed WB (1968) Ecology of the Hawaiian Drosophilidae, University of Texas Publishers. 6818: 387–419.108.
- 106. Hardy DE (1965) Insects of Hawaii, Vol. 12. Diptera: Drosophilidae. University of Hawaii Press, Honolulu.
- 107. Gillespie RG (2004) Community assembly through adaptive radiation in Hawaiian spiders. Science 303: 356–359.
- 108. Roderick GK, Gillespie RG (1998) Speciation and phylogeography of Hawaiian terrestrial arthropods. Mol Ecol 7: 519–531.
- 109. Gillespie RG, Croom HB, Palumbi SR (1994) Multiple origins of a spider radiation in Hawaii. Proc Nat Acad Sci USA 91: 2290–2294.
- 110. Baldwin BG (1997) Adaptive radiation of the Hawaiian silversword alliance: congruence and conflict of phylogenetic evidence from molecular and non-molecular investigations. In: Molecular evolution and adaptive radiation (Givnish TJ and Sytsma KJ, eds), pp. 103–128. Cambridge University Press, Cambridge.
- 111. Hrbek T, Meyer A (2003) Closing of the Tethys Sea and the phylogeny of Eurasian killifishes (Cyprinodontiformes: Cyprinodontidae). J Evol Bio 16: 17–36.
- 112. McCune A (1997) How fast is speciation? Molecular, geological, and phylogenetic evidence from adaptive radiations of fishes. In: Molecular Evolution and Adaptive Radiation (eds. Givnish TJ and Sytsma KJ), pp. 585–610. Cambridge University Press, Cambridge.
- 113. Reichenbacher B, Kowalke T (2009) Neogene and present-day zoogeography of killifishes Aphanius and Aphanolebias in the Mediterranean and Paratethys areas. Paleogeo, Paleoclim, Paleoecol 281: 43–56.
- 114. Zuccon D, Cibois A, Pasquet E, Ericson PGP (2006) Nuclear and mitochondrial sequence data reveal the major lineages of starlings, mynas and related taxa. Mol Phylo Evol 41: 333–344.
- 115. Castillo P, Batiza R, Vanko D, Malavassi E, Barquero J, et al. (1988) Anomalously young volcanoes on old hot-spot traces: I. Geology and petrology of Cocos Island. Bull Geo Soc Am 9: 1400–1414.
- 116. Noonan BP, Sites JW (2010) Tracing the origins of iguanid lizards and boine snakes of the Pacific. Am Nat 175: 61–72.
- 117. Estes R, Williams EE (1984) Ontogenetic variation in the molariform teeth of lizards. J Vert Paleo 4: 96–107.
- 118. Givnish TJ, Barfuss MHJ, Van Ee B, Riina R, Schulte K, et al. (2011) Phylogeny, adaptive radiation, and historical biogeography in Bromeliaceae: insights from an eight-locus plastid phylogeny. Am J Bot 98: 872–895.
- 119. Poschmann M, Anderson LI, Dunlop JA (2005) Chelicerate arthropods, including the oldest phalangiotarbid arachnid, from the early Devonian (Siegenian) of the Rhenish Massif, Germany. J Paleo 79: 110–124.
- 120. Wueringer BE, Squire L, Collin SP (2009) The biology of extinct and extant sawfish (Batoidea: Sclerorhynchidae and Pristidae). Rev Fish Bio Fisheries 19: 445–464.
- 121. Gottfried MD, Rabarison JA (1997) First Mesozoic Gondwanan record of a sawshark (Chondrichthyes, Pristiophoriformes), from the Late Cretaceous of Madagascar. J Vert Paleo 17: 750–751.
- 122. Hallstrom BM, Janke A (2009) Gnathostome phylogenomics utilizing lungfish EST sequences. Mol Bio Evol 26: 463–471.
- 123. Wagner CE, Harmon LJ, Seehausen O (2012) Ecological opportunity and sexual selection together predict adaptive radiation. Nature 487: 366–369.