Abstract
Despite the amount of data on different aspects of biodiversity, such as species distributions, taxonomy, or phylogenetics, there are still significant gaps and biases in the available information. This is particularly true for life history traits, with fragmentary data for most taxa, especially those with complex life cycles. Anurans (frogs and toads) show larval (premetamorphic) stages that are in general radically decoupled from adult forms in most biological aspects. Our understanding of this group is highly uneven, as the main wide-scope investigations focus on adult specimens and larval stages remain unknown for a significant part of the anuran tree. The main purpose of this work was to estimate the extent of knowledge gaps regarding the diversity of tadpoles, interpret their biological and geographical patterns, and discuss possible explanations and implications for other large-scale analyses. Our findings show that more than half of the anuran species described to date still lack information on their embryonic/larval stages. Furthermore, knowledge varies among taxonomic groups, larval ecomorphological guilds, and world ecoregions. Description percentages generally decrease in lineages with a higher proportion of species known or suspected to have endotrophic development. Also, geographic areas with the highest levels of ignorance in larval biology (Tropical Andes and New Guinea) coincide with the highest diversity of endotrophic guilds. Among exotrophic larvae, generalized lentic-lotic tadpoles have the widest distribution and levels of knowledge, whereas specialized lotic, fossorial, and terrestrial forms are more taxonomically and geographically restricted. Further large-scale analyses on tadpole biology are crucial for their impact in varied scientific disciplines including anuran conservation. At a conceptual level, the discussion of the anuran biphasic life cycle is pertinent in the context of shortfalls of biodiversity knowledge and their interrelationships.
Similar content being viewed by others
Introduction
Achieving a comprehensive understanding of biodiversity is crucial for advancing applied science1. Today, in the BigData era, the availability of information about different aspects of biodiversity—such as species distributions, taxonomy, or phylogenetics—is astonishing. However, there are still significant gaps and biases in the available information often hindering accurate applications2. Furthermore, several essential aspects of biodiversity still need to be studied, with scarce or nonexistent information available3. This lack of understanding is particularly true for life history traits, with fragmentary data for most taxa but especially those with complex life cycles. The lack of information about the ontogeny of certain organisms, which often involves a complete knowledge gap regarding particular life stages, significantly impacts on our understanding of their general biology and interactions with the environment, hindering our ability to manage or protect these taxa. While this fact has been previously introduced4, it still needs to be quantified across groups5. A textbook example in this regard is anurans (frogs and toads), whose larval (premetamorphic) stages are in general radically decoupled from adult forms in most biological aspects6. Both life stages may face independent threats from human activities, and survival may depend on completing each stage successfully7. Regrettably, our understanding of this group is highly uneven, as primary wide-scope research focuses on the adult phase8,9, and larval stages still need to be discovered for a significant part of the anuran tree.
Tadpoles are a conspicuous part of the fauna that surrounds us. However, they were scarcely reported in traditional (i.e., morphology-based) taxonomy and virtually never included in the original description of anuran species (e.g., see the chronology of tadpole research in ref. 10). Identifications of subsequently described tadpoles were often hazardous, especially in species-rich regions, and often only based on the presence of adult specimens in the area. Innovating taxonomical practices such as molecular taxonomy and DNA barcoding in particular overcame the assignation impediment of tadpoles, boosted tadpole descriptions11,12, and enhanced the interest in this development stage. Far from a generalized idea (slippery things wriggling in muddy ponds), tadpoles have evolved an extraordinary diversity of shape (including unique adaptive characters), size, microhabitat use, and behavior, and this diversity may occur at higher taxonomic levels but also within smaller clades such as genera or species groups13. In this context, ecomorphological guilds were proposed to synthesize morphological diversity and emphasize the occurrence of similar trait sets in independent lineages. The most widely used scheme is that by Altig and Johnston14 (updated in ref. 10), which combines a heterogeneous set of characters (developmental modes, microhabitat, feeding behavior, and larval external morphology) to produce 21 ecomorphological categories (e.g., Fig. 1). Developmental mode is the first discriminator and distinguishes those tadpoles that feed actively on external sources (exotrophic) from those that develop exclusively from parental sources (endotrophic). Exotrophic tadpoles are divided following the type of water body they inhabit (lentic, lotic, or both) and then categorized according to microhabitat and behavior (e.g., benthic, arboreal, semiterrestrial) and trophic specializations (e.g., carnivorous, suspension feeder). Endotrophic species are classified into six groups that vary in the source of nutrition, the site of development, and the morphology of embryos and tadpoles (e.g., viviparous, direct developers). By definition, some of these latter species lack a tadpole, i.e., a “free-living larval stage” (although many typically larval characters and even a cryptic metamorphosis can be retained15), so premetamorphic stages are often referred to as embryos. The resulting 21 guilds provide a valuable tool to begin to explore larval ecomorphological diversity and its taxonomic and geographic distribution, and at the same time, may help to interpret patterns of knowledge gaps in large-scale analyses.
At first glance, the imbalanced knowledge of different life stages in anurans may have several underlying causes. Biological and ecological reasons would account for intrinsic features related to microhabitat use or development. For instance, many tadpoles inhabit ponds or pools and are relatively easy to collect, but several species develop in hidden, hard-to-access microhabitats. Some species have very short larval periods, and as mentioned, many frogs do not develop from free-living tadpoles. Geographical factors may involve remote areas with difficult access, and hotspots with an extraordinary, underestimated biodiversity, among others. Furthermore, historical and epistemological reasons may be related to the behavior of researchers, patterns of colonization and inventorying1, or access to regions with profound social conflict.
From all the above, the main purposes of this work are to estimate the extent of knowledge gaps regarding the diversity of tadpoles, interpret their biological and geographical patterns, and discuss possible causes and implications for other large-scale analyses. By framing our work in the theoretical context of shortfalls of biodiversity knowledge1,4, we expect to provide the first comprehensive assessment of our current level of ignorance on anuran tadpole biology.
Results
Described tadpoles per family and ecoregion
Our results show that 3088 out of 7507 species of frogs (41%) have larval stages described (Supplementary Table). Figure 2 illustrates percentages of descriptions in families and world ecoregions. At a taxonomic level (Fig. 2 top), 19 clades have more than 50% unknown larval diversity, and 68 out of 461 recognized genera (15%) have no tadpoles described to date. Although with a wide dispersion, most speciose families generally have lesser percentages described (see Supplementary Fig. 1), except for the highly diverse Hylidae (1036 species) with 65% of species with tadpoles characterized. Families with the highest (100%) and lowest (below 10%) percentages of tadpole descriptions are among the least speciose (e.g., Ascaphidae 2 spp., 100%; Allophrynidae 3 spp., 0%). Brachycephaloidea stands out as a highly diverse group (1228 spp.) with only 3% of premetamorphic stages described. In general, the tadpoles from non-neobatrachian families are fairly known. In Neobatrachia, the clade Indianura has mid-to-low percentages of description, except the small families Nasikabatrachidae, Odontobatrachidae, and Conrauidae with high values. Values are overall higher in Notogaeanura, with most species having tadpoles described among australobatrachians (Calyptocephalellidae, Limnodynastidae, and Myobatrachidae), Ceratophryidae, and Rhinodermatidae. Allophrynidae and brachycephaloid Ceuthomantidae are the sole families with premetamorphic stages undescribed to date.
Distributional data (Fig. 2 bottom) show that areas with ca. 100% tadpoles described concentrate in the northern hemisphere and some areas of the southern hemisphere, such as southeast Australia, Tasmania and New Zealand, and the Uruguayan savannas of South America. Those areas with lower percentages are the Northwest and Central Andes, Greater Antilles, most of New Guinea and New Britain-New Ireland islands, some Indonesian islands (Lesser Sunda and Sumba deciduous forest), the northern part of Filipinas (Luzon rainforest), and in Europe, Cyprus. Africa generally has lower description percentages than other regions, with the smallest values in East Saharan and Somali xeric woodlands. Northern New Guinea montane rainforest and the Solomon Islands are the ecoregions with the lowest proportion of tadpoles described (below 15%).
Taxonomic and geographic distribution of larval exotrophic ecomorphological guilds
Among the 3088 species with premetamorphic stages described, 69% have generalized lentic-lotic tadpoles and 12% specialized lotic tadpoles, while terrestrial and fossorial guilds accumulate 9% and 3%, respectively. Taxonomic distribution of guilds (Fig. 3) shows that generalized lentic-lotic tadpoles are the single guild described in 12 families and predominate in other 22. Specialized lotic tadpoles occur exclusively in four families, including the earliest diverging anuran Ascaphidae, and neobatrachian Heleophrynidae, Nasikabatrachidae, and Conrauidae; these tadpoles also develop in other 11 families, but only predominate in Calyptocephalellidae. Some type of terrestrial development occurs in 17 families, being unique or majoritarian in five of them (e.g., Odontobatrachidae and Petropedetidae). In turn, fossorial tadpoles are described in eight distant families and are likely the single morph in Micrixalidae and Centrolenidae.
The geographic distribution of larval exotrophic guilds is illustrated in Fig. 4. Generalized lentic-lotic tadpoles (1917 species) are distributed worldwide, with the highest alpha diversity in the Brazilian Cerrado (201 species), Serra do Mar coastal forest (168 species), and Bahia interior forest (167 species). Specialized lotic tadpoles (336 species) concentrate in Madagascar lowland (humid) and subhumid forests (34 and 33 species), and with lower values in Borneo lowland rainforest (28 species), Qionglai-Minshan conifer forest in China (25 species), and Petén-Veracruz moist forest in America (22 species). Fossorial and terrestrial tadpoles are described from similar areas (e.g., mostly tropical, west to the Wallace line). Fossorial tadpoles (100 species) show highest alpha diversity in the South American Chocó-Darién (17 species) and nearby areas of Northwest Andean forests (15 species), Magdalena Valley montane forest (15 species), and Isthmian-Atlantic moist forests (14 species), and in Borneo rainforests (14 species). In turn, terrestrial tadpoles (249 species) have the highest alpha diversity values in the Northwest Andean and Ucayali forests (25 species each).
Ignorance about endotrophic ecomorphological guilds
Embryos and tadpoles with some type of endotrophic development are described for 226 (3%) of the 7507 anuran species. They are present in 23 taxonomic groups, being the single developmental type known in five of them (Fig. 3). However, at least 2303 species are known or suspected to have endotrophic nutrition (Supplementary Table), comprising about 30% of the anuran fauna. Among the 68 genera with unknown tadpoles, around 60% are likely endotrophic. The areas with the highest alpha diversity of species with known or suspected endotrophic development (>25 spp.) concentrate in South and Central America, Madagascar, Southwest India and Sri Lanka, and Borneo and New Guinea islands; however, these areas are those with the lowest percentages of embryos/tadpoles described (Fig. 5). Our estimation shows that in 398 ecoregions where endotrophic anurans are known or suspected to occur, less than a quarter of species have premetamorphic stages described. Among these, 132 (33%) show a total degree of ignorance (i.e., a presumed occurrence of that guild but no premetamorphic stage formally described).
Endotrophic guilds vary in their taxonomic and geographic distributions and the percentages of embryos/tadpoles described (Figs. 3 and 5). Direct development has been described in 112 species from nine clades but is also suspected in at least 1821 species plus two other clades (Petropedetidae and Sooglossidae). This accounts for approximately a quarter of all frogs, rendering direct development the most frequent endotrophic guild. Direct-developing species have wide distribution and are the single endotrophic guild in most of Central and North America, India, east of Wallace line, and northern regions of Australia. They explain exclusively the high diversity of endotrophic embryos in New Guinea, which remains mostly undescribed. In turn, nidicolous tadpoles are reported or presumed to be present in 18 families, in at least 156 species aside from the 87 where tadpoles are described. This accounts for 243 species, thus ca. 3% of frogs develop non-feeding tadpoles. The distribution of nidicolous tadpoles partly overlaps with that of direct-developing species, but it also reaches the southern Andes, Madagascar, Japan and New Zealand. The highest number of described tadpoles is in Malagasy’s humid and subhumid forests (ca. 12 species each), but more than half of the likely nidicolous larval diversity in these areas still needs to be discovered. Finally, some form of viviparity is reported or suspected in 7 clades (69 species, 1% of anurans), with the highest diversity in the northern regions of South America. Description percentages of these embryos are mid to low.
The threshold to dismiss commission errors of polygons ranges to species distributions did not reveal a significant effect at this level of analysis, and it also entails some artifacts in results (see Supplementary Figs. 2–4). For instance, it introduces certain noticeable omission errors, possibly due to the overlapping of small distributional ranges with ecoregions of varying extents, without accounting for transitional zones. As an example, it depicts no occurrences in Alaska when tadpoles of six species in that area are described. For these reasons, and while further exploratory analyses are pending, we opt for a parsimonious interpretation of distribution data as they were originally loaded and discuss main biogeographical patterns from maps accounting for species complete distributions.
Discussion
Our study provides a comprehensive overview of the current knowledge of tadpole diversity, revealing significant gaps in our understanding. Surprisingly, despite the extensive research on different aspects of anuran biology, our findings demonstrate that more than half of the described species still lack information on their larval stages. Furthermore, current knowledge is highly uneven among taxonomic groups, ecomorphological guilds, and ecoregions. In the paragraphs below, we (1) summarize possible explanations for patterns of knowledge about tadpole diversity and (2) pinpoint some effects on other large-scale analyses regarding anuran biology and conservation. Finally, we comment on some aspects of the biphasic ontogeny in frogs in the context of shortfalls of biodiversity knowledge.
As expected, ecomorphological, biogeographical, and historical-epistemological aspects may explain general patterns of knowledge about tadpole diversity. First, intrinsic aspects of tadpole biology constrain the extent of knowledge of tadpole diversity. Description percentages generally decrease in lineages with a higher proportion of species known or suspected to have endotrophic development. Three groups hypothesized to be entirely endotrophic (Brachycephaloidea, Ceratobatrachidae, Brevicipitidae) bear the lowest indices of embryo descriptions (below 10%). Although most tadpoles described in the speciose family Microhylidae are conspicuous suspension-feeders16 (e.g., Microhyla in Fig. 1), the low description percentage in the clade can be explained by the high proportion of species presumably endotrophic (about 65%). The distribution of cophyline microhylids restricted to Madagascar explains the high richness of known and suspected nidicolous tadpoles on the island. Likewise, although most tadpoles described in African Arthroleptidae and Afro-Asian Rhacophoridae are lentic-lotic, significant portions of the diversity in both families (ca. 30 and 45% respectively) likely develop nidicolous or direct-developing embryos. The relationship between endotrophic development and the lack of knowledge of tadpole diversity is clearly evident in geographic data, where zones of highest ignorance coincide with the highest diversity of endotrophic guilds (compare maps in Figs. 2 and 5). Both the Tropical Andes and New Guinea, with the lowest percentages of tadpole description, house a high diversity of direct-developing and viviparous species whose ontogeny is unknown (e.g., Strabomantidae and Hemiphractidae in the Andes and a wide variety of direct-developing Ceratobatrachidae and asterophryine Microhylidae in New Guinea).
Taxonomic and geographic distribution of exotrophic tadpoles also contributes to interpreting patterns of knowledge on tadpole biology. Most families with more than 50% of their tadpoles described develop exclusive or predominantly generalized lentic-lotic tadpoles (e.g., Alytidae, Myobatrachidae). What we coded as generalized lentic-lotic tadpoles in fact includes a wide diversity in microhabitat use and trophic ecology. Although a thorough analysis is pending, these differences apparently do not affect the detectability of tadpoles and the percentages of tadpole descriptions. For example, high percentages are described in Ceratophryidae (with mostly carnivorous tadpoles17; e.g., Lepidobatrachus in Fig. 1), Pipidae (with mostly suspension-feeding tadpoles18) or in some megophryine genera (e.g., Boulenophrys, Brachytarsophrys; with exclusively neustonic tadpoles19). Generalized lentic-lotic tadpoles are almost the sole known guild in the Holarctic region, and this, along with the broad species distributions and the low diversity in that area20,21, would explain in part the high percentages of descriptions.
Several families that develop specialized lotic tadpoles exclusively or in the majority are in general well-known, probably because they are scarcely diverse (2–8 species, e.g., Ascaphidae, Conrauidae) and tadpoles are large and conspicuous, with large, sucker-like mouthparts22. Also, a comparatively low diversity in the environments where these tadpoles occur could render identification easier than in, for example, ponds where the number of similar syntopic species is usually higher. Almost a quarter of Bufonidae likely develops specialized lotic tadpoles (e.g., suctorial Ansonia in Fig. 1, gastromyzophorous Atelopus23,24), and about half of these are already described. Similarly, ca. 40% of Ranidae likely develop lotic tadpoles (e.g., ref. 25, gastromyzophorous Huia in Fig. 1), of which 40% are known. The geographic distribution of specialized lotic tadpoles, concentrated so far in Madagascar (Mantellidae) and Borneo (e.g., Bufonidae and Ranidae), contributes to the general mid-values of description percentages in both regions.
Terrestrial guilds have a taxonomic pattern of knowledge similar to that of lotic specialized tadpoles. Groups with higher description percentages are small families where terrestrial tadpoles develop exclusively or predominantly (e.g., African Odontobatrachidae—all five species described—and Petropedetidae—8 out of 13 species described). In Neotropical Aromobatidae and Dendrobatidae, the high proportion of phoretic species could increase the detectability of tadpoles in nature and contribute to the mid-high percentages of description in both families and their distribution zones. Tadpoles of at least 13 families develop in micro-waterbodies contained in plants26 (e.g., Phyllodytes and Frostius in Fig. 1). Although this could appear a priori as a hard-to sample microhabitat, these phytotelmon-dwelling tadpoles represent the highest proportion of terrestrial tadpoles described (unpublished data). As with specialized lotic tadpoles, the low diversity in these environments may balance a potentially lower detectability with a straightforward identification of tadpoles. Lastly, semiterrestrial tadpoles likely predominate in Cycloramphidae27 (e.g., Thoropa in Fig. 1) and almost half of them are already described. The restricted distribution of this family in the Brazilian Atlantic rainforest may explain the comparatively high description percentages in that region.
Regarding fossorial tadpoles, their microhabitat preferences and behavior of hiding or burying in the substrate could explain the overall lack of knowledge about this guild. The Centrolenidae is the best-known group, with 30% of their apparently exclusively fossorial tadpoles described in Chocó-Darién and near montane forests. In other families, several genera suspected to have only this type of tadpoles (e.g., arthroleptid Cardioglossa in Central Africa and megophryid Leptobrachella in East Asia28,29) have also about a third of their larval diversity described. Conversely, the tadpoles of only two species are described in the monotypic family Micrixalidae, endemic to the Western Ghats in India. Furthermore, micrixalid tadpoles remained unknown for more than a century (the type species was described in 1882) until 2016, when the first larvae were unearthed30.
Hortal et al.1 highlight that bias in biodiversity data toward specific regions, habitats, and environments involves syndromes related to historical patterns of colonization and inventorying and also to the behavior of researchers. This crucial issue was addressed in several analyses at a global scale, dealing with different taxonomic groups in biological and paleontological research31,32. In our specific case, a thorough analysis is in order but some main patterns can be easily interpreted. For instance, although tadpoles were described along with the first formal descriptions of frog species (e.g., Pseudis in ref. 33), the first specific works of larval morphology occurred during the nineteen century in Europe and North America34,35. This pioneer research, which surveys tadpole morphology of several typically Nearctic and Palearctic frogs (e.g., Alytidae, Pelobatidae, Scaphiopodidae), may also account for the high percentage of descriptions in those areas. High values in regions such as some South American grasslands and Indian plains and hills may be related to early local inventories of tadpole diversity36,37,38.
Regarding researcher behavior, we identify at least three main aspects that could explain the imbalanced knowledge of larval anurans. First, well-established schools of herpetological research working historically on some areas or taxa may explain why some taxonomic groups are better known than others. For instance, the high proportion of tadpoles described in Hylidae, the most diverse family among anurans, may be explained by its cosmopolitan distribution and accessibility for different research groups worldwide39,40,41,42. Second, specialization within the discipline (there are tadpole experts, tadpole literature, even tadpole scientific meetings43) can explain variation in description percentages among endotrophic guilds, ca. six times lesser in direct-developing species than in viviparous and nidicolous tadpoles. Unlike these two latter, which in general can still be characterized following standard procedures as in exotrophic tadpoles (e.g., using the developmental table in ref. 44), collecting, raising, and describing direct-developing embryos with a drastically different bauplan implies at least an out-of-routine task for tadpole specialists. Third, some methodological trends in taxonomic research, such as species descriptions based on molecular data, may have contributed to amplifying the gap between species descriptions and tadpole descriptions. In fact, a quick plotting of the two latter along the years shows that curves highly diverge from yr. 2000 onward, after which species discrimination by molecules produced a noticeable speed-up in species descriptions (Supplementary Fig. 5). The recent inclusion of tadpole characterization in species descriptions (with unambiguous assignation by DNA barcoding45) and approaches through integrative taxonomy46 remain to be improved to compensate the boost of new species discovery.
The implications of missing information about the ontogeny of organisms with complex life cycles extend beyond what can be addressed in just a few paragraphs. Immediate consequences in several disciplines may be envisioned after comparing conclusions from large-scale analyses based on different life stages. This is crucial in the context of the global biodiversity crisis47 due to the assignation of taxa to threat categories and the definition of Priority Conservation Areas are strongly biased toward knowledge of the adult phase (e.g., see ref. 5). Given the vast extent of our knowledge gaps on tadpole biology and the current trends in habitat transformation, one of the most significant implications we can highlight is the potential extinction of complex organisms that have yet to be recognized by science. Several examples, many recently discovered, reveal an extraordinary ecomorphological diversity that cannot be adjusted in our present larval guild scheme (e.g., the unique gastromyzophorous, phytotelmon-dwelling larvae of Phyllodytes gyrinaethes48,49, or the exotrophic tadpoles of Limnonectes larvaepartus that develop initially within the mother oviducts50). The disappearance of further rare or even unknown ecomorphotypes is feasible. Although this would be a worldwide concern, effects could be more profound in many of the most important regions regarding anuran diversity and conservation.
In this regard, global hotspots for anuran diversity21 differ in the extent of knowledge about tadpole diversity. Among the richest areas, whereas the lowlands of Amazonia and the Brazilian Atlantic forest are reasonably well described, the Andean region of Amazonia stands out as markedly understudied. Land use intensification would have a decisive impact on local anurofauna51,52, specifically on the scarcely known terrestrial and endotrophic tadpoles concentrated in those areas. Habitat split, i.e., human-induced disconnection between habitats used by different life history stages of a species, and water pollution are identified as major threats for species with aquatic tadpoles in this ecoregion53. In the second place, recent studies revisit patterns of amphibian richness and the extraordinary amount of endemisms in Madagascar54,55. Larval diversity in this hotspot is relatively well-known mainly due to systematic prospecting in the last decades56. This diversity includes exceptional, highly divergent ecomorphotypes (e.g., sand-eating tadpoles57) and a high concentration of specialized lotic and nidicolous tadpoles. Given current habitat destruction and fragmentation threats, conservation efforts are crucial in Malagasy eastern rainforests that still harbor a significant portion of undescribed diversity55. Finally, although comparatively less studied, the African rainforest anuran fauna stands out with high richness and proportion of endemisms58. Tadpole diversity in Cameroonian-Congolian forests is described in about half, and among potential threats to conservation, specific effects on eggs and tadpoles are foreseen59.
On the other hand, assessing the vulnerability of anuran species while ignoring larval stages potentially susceptible to different threat drivers clearly represents a significant drawback and may lead to profound underestimation of extinction risks. It is concerning to note that tadpoles are entirely unknown for nearly half of the frogs categorized as Least Concern (unpublished data). This implies that the “Low” extinction risk has been estimated from a partial scenario, overlooking that threats faced by larval stages could indeed be a concrete cause of local or even total extinction. Paradoxically, anurans are the tetrapod group with the highest number of species assessed as Data Deficient60, which is one of the most significant challenges for accurate decision-making61. Considering the substantial number of species assessed without a tadpole description, we suggest that the number of DD species might be underestimated.
The need for morphological knowledge and characterization of premetamorphic stages of anurans significantly hampers progress in disciplines such as comparative anatomy, systematics, and evolution. Large-scale phylogenies of anurans incorporating morphological evidence have relied on larval characters as the primary source of phenotypic variation62,63. However, these studies have only included a relatively small number of representative species of the diversity of Anura. More recently, well-sampled phylogenies of the order were exclusively based on molecular information64,65, providing a relatively well-resolved framework of relationships among major lineages. Nevertheless, many nodes throughout the entire tree need to be better supported and lack phenotypic diagnosis (see ref. 66). Further efforts to fill the knowledge gaps regarding larval morphology in anurans will positively affect understanding the phylogenetic relationships and evolutionary patterns of many phenotypic traits.
To end with, biphasic ontogeny in anurans can be discussed in the context of shortfalls of biodiversity knowledge. These shortfalls are defined as the gaps between realized/extant knowledge within a biological domain at a given time and concern taxonomic (species identity), extrinsic (geographic distribution, population dynamics, evolutionary relationships) and intrinsic attributes of species (functional traits, abiotic niche, biotic interactions1). The recent definition of a new shortfall type, the Haeckelian shortfall by Faria et al.4, brought the unnoticed importance of ontogenetic development into discussion. This was imminent, especially because complex life cycles are prevalent among animal phyla67 and living organisms in general68. As stressed by Faria et al.4, relationships among the Haeckelian shortfall and the previously synthesized types exhibit two-way dynamics: knowledge on ontogenetic aspects contributes to filling gaps of knowledge in other dimensions (e.g., phenotypic characterization of semaphoronts helps to generate more comprehensive phylogenetic hypotheses) and at the same time, knowledge on different life stages must be nourished with specific information covered by the remaining shortfalls (e.g., intrinsic and extrinsic attributes may differ significantly along life stages). The latter is reflected in the hierarchical structure that Faria et al.4 give to the interrelation scheme among shortfall types.
In anurans, biological differences among larval and adult forms are astonishingly profound and concern not only morphology and physiology but also ecology and behavior, the very attributes that in adults merited the definition of specific shortfall types. To begin with, tadpoles in general have their description separated from that of the species (i.e., only a third of tadpole descriptions published in Zootaxa journal are part of studies describing new species69). Some anuran species have been described based on premetamorphic stages, always very conspicuous animals that are easily differentiated from other species (e.g., Pseudis paradoxa33; Clinotarsus penelope70; Calyptocephalella gayi, but see ref. 71 for a different interpretation), and numerous candidate species hypothesized from larval morphology await formal description (e.g., the strange tadpole of ref. 37 waited for almost 100 years to be assigned to the recently described adults of Nasikabatrachus sahyadrensis72). This makes ignorance about tadpoles as basic entities resemble a Linnean shortfall.
Excepting the Wallacean shortfall (we assume that knowledge about a species distribution based on adult forms does not differ from that inferred from its tadpoles), other shortfalls can be equally invoked. A Haeckelian-Prestonian shortfall relationship is interpreted since population dynamics in larval and adult phases may differ significantly. For instance, the spatial and temporal distribution of adults that breed during brief lapses of explosive reproduction while tadpoles remain in ponds73 may render very different scenarios in population ecology studies. Although evolutionary relationships are intrinsically the same for adults and tadpoles (disregarding that in practice hypotheses are contingent on datasets), the evolution of traits also covered by the Darwinian shortfall may reveal varied arrangements prior to and after metamorphosis. For instance, Sherratt et al.74 found rampant homoplasy linked to ecology in body shape evolution among tadpoles of Australian anurans, as opposed to the adult morphospace for which phylogenetic structure is stronger.
As to extrinsic attributes, given that in most anurans the metamorphosis implies a reset in the fundamental variables of the ecological niche (aquatic to terrestrial environment), aspects related to responses to abiotic conditions (Hutchinsonian shortfall), ecological functions (Raunkiæran shortfall) and biotic interactions (Eltonian shortfall) can be radically divergent in tadpoles and cannot be inferred from knowledge in the adult phase. For instance, Bolochio et al.51 addressed the first global-scale analysis for anuran conservation from a functional perspective. This study, framed in the Raunkiæran shortfall type, stands out as a clear example of how contrasting the categorization (i.e., adult ecomorphs differ considerably from larval ecomorphological guilds), interpretation, and eventual decision-making may be (e.g., the most vulnerable adult ecomorph does not include a high proportion of the most neglected larval guilds) when knowledge is grounded on tadpole vs. adult frog datasets.
Final remarks
The results of our study underscore the pressing necessity for additional research on tadpole diversity. Whereas the data we provide contribute to filling knowledge gaps in aspects related to Haeckelian-Linnean (by quantifying tadpole descriptions over anuran species) and Haeckelian-Raunkiæran shortfalls (by analyzing taxonomic and geographic distribution of larval functional groups), many other aspects of larval biology remain underexplored, especially in a global context. Our initial dataset may serve as a basis for many future analyses that lay on more refined ecomorphological characterization75, ecomorphological evolution76, and quantification and comparison of patterns and determinants of knowledge shortfalls across vertebrate groups77,78,79. This kind of research is crucial not only for anuran conservation but also concerning its potential significant applications in various scientific disciplines.
Methods
Datasets
Species list and taxonomy
We worked with the species list, nomenclature, and taxonomic arrangement for Anura by Frost80 (as available in October 2022). This yields a dataset of 7507 species distributed into 51 families plus the superfamily Brachycephaloidea (with five families—Brachycephalidae, Ceuthomantidae, Craugastoridae, Eleutherodactylidae, Strabomantidae—and three genera not assigned to family).
Tadpole descriptions
For each 7507 species, we first consigned whether or not larval stages are known. We collected this information from varied sources, including the AmphibiaWeb and Amphibian Species of the World webpages80,81, manual Google searches, and our libraries and expertise. We considered published formal descriptions but also descriptions in comprehensive works such as phylogenetic analyses, keys, books, and available dissertations. We tried to check the most comprehensive and recent compilations (e.g., generic revisions, comparative studies) to avoid tadpole misidentification problems.
Larval ecomorphological guilds
We assigned each described tadpole to one of seven ecomorphological guilds summarized from the scheme by McDiarmid and Altig10 as follows:
(A) Exotrophic tadpoles:
-
(1)
generalized lentic-lotic: tadpoles that inhabit ponds or quiet zones of lotic systems. In this guild, we include a wide variety of ecomorphological guilds originally defined as benthic, nektonic, neustonic (with upturned mouthparts), macrophagous, carnivorous, suspension-feeding, and suspension-rasper tadpoles. For the sake of simplicity, we do not discriminate microhabitat or trophic ecology in this instance.
-
(2)
specialized lotic: stream tadpoles with morphological specializations for inhabiting water bodies with faster current. These include adherent, clasping, suctorial, and gastromyzophorous (with abdominal sucker) tadpoles as originally defined.
-
(3)
fossorial: stream tadpoles that live buried among substrate (sand, leaves, gravel), as originally defined. Psammonic tadpoles (bury in the sand) are a different ecomorphotype in the reference scheme, but here we include them among fossorial.
-
(4)
terrestrial: tadpoles that can be found outside main water bodies. These include phytotelmon-dwelling (inhabit water-filled tanks in plants) and semiterrestrial tadpoles (inhabit the spray zone in streams) as originally defined, plus tadpoles that develop in underground chambers or caves. We also include species whose adults transport tadpoles in their backs (phoretic) to highlight the out-of-water phase of their development.
(B) Endotrophic tadpoles/embryos:
-
(5)
nidicolous: non-feeding tadpoles that develop in various places, including nests, ground cavities, phytotelma, ponds, etc.
-
(6)
viviparous: embryos that develop within some part of the parent body. This includes viviparous, ovoviviparous, exoviviparous, and paraviviparous guilds as originally defined.
-
(7)
direct developers: embryos that develop within terrestrial eggs before hatching as small froglets.
The last two guilds lack a tadpole, i.e., a “free-living larval stage”; in these cases, we considered descriptions of embryonic morphology and refer to them as embryos when possible.
Phylogeny
For analysis and discussion, we followed the phylogenetic hypothesis of Jetz and Pyron64, which incorporated nuclear and mitochondrial sequences for 3449 anurans, including representatives from all extant families.
Distribution data
Species distributions were obtained from the IUCN database60. We downloaded digital range maps (extent of occurrence maps) for 6235 anuran species available at this source. Given the large number of species, the global extent of the analyses, and the bias associated with the source of the distributional data2, we decided to run the analyses using ecoregions as a biogeographical unit82. In our opinion, maps based on these biological units are more realistic in a biological sense and easier to interpret than the “traditional” maps based on a grid in which biogeographical units could be mixed. Furthermore, using range maps at finer resolutions would increase the biases related to overinterpreting the limited information contained in these maps83,84.
Analyses
Tadpole descriptions in a taxonomic context
We generated a first dataset containing 7507 anuran species, indicating for each its familial/suprafamilial level, whether or not tadpoles are described, and the ecomorphological guild these are assigned to (Supplementary Table). In a few species, both exotrophic and endotrophic tadpoles have been described (e.g., Incilius periglenes); to simplify, facultative conditions were coded prioritizing the exotrophic guild. We then calculated and plotted percentages of species with tadpoles described per family and guild diversity among described tadpoles.
Mapping general ignorance and guild diversity
We explored our general ignorance about tadpole description and distribution of larval ecomorphological guilds worldwide. First, using the “join by location function” of QGIS85 we performed a geographic match between the geographic distribution of each species and its recompiled developmental information. After the match, we recovered a database encompassing 6235 species (Supplementary Table), for which both kinds of information are available. Using QGIS [“join attributed by location (summary)”] we overlapped the terrestrial ecoregions of the world with the species attributes, and we counted the number of species per ecoregion to generate the following maps and estimations:
-
(1)
the percentages of species with tadpoles described per ecoregion (considering the total number of the species in our database), and
-
(2)
the species richness of each ecomorphological guild per ecoregion. In this case, we cannot estimate percentages because the total number of tadpoles of each guild per region is not available (it would require assigning guilds to unknown tadpoles; but see next section). Thus we filtered and split the former database generating a new sub-database per guild (a total of 2812 species) containing only those species with already described tadpoles. By repeating the above-described procedure we could estimate and map the alpha diversity of each guild in each terrestrial ecoregion of the world.
To explore the effect of commission errors arising from inaccuracies in polygon distributions, we employed a threshold approach. Specifically, for a given species we only included those ecoregions overlapping with a minimum of 5% of its total range. To achieve this, we first utilized QGIS to perform an intersection analysis, overlaying the species distribution data with the vector file representing ecoregions. This allowed us to calculate the percentage of each species distribution contained within each individual ecoregion. Subsequently, we excluded those portions of the species range overlapping with “marginal ecoregions” (i.e., below the 5% threshold of their entire distribution).
Ignorance about species with endotrophic development
We applied a different approach for three endotrophic guilds by estimating the total numbers of species per guild based on the literature. Thus, in addition to species with described endotrophic tadpoles, we included those species where the endotrophic guild is suspected based on reproductive biology aspects (e.g., oviposition site, size and number of eggs, etc.; Supplementary Table). Although endotrophic guild distribution appears to be highly conserved at several levels (e.g., all known ceratobatrachids are direct developers, and all known cophyline microhylids have nidicolous tadpoles86,87), variation is known even within genera (e.g., Gastrotheca includes tadpole-producing and viviparous species; Philautus includes species with nidicolous and direct-developing embryos88,89). We were conservative enough not to assign guilds in groups with significant variation, but we did generalize a single type in cases where a second guild is exceptional (e.g., only Eleutherodactylus jasperi is viviparous while remaining known brachycephaloids are known or suspected to be direct developers90). For those species where some form of endotrophic development is suspected but the specific type cannot be identified (e.g., some brevicipitid genera), we coded only endotrophic. Using this new dataset, we estimated the percentages of described tadpoles of each endotrophic guild per ecoregion.
Although with an unlikely profound impact, we acknowledge several potential sources of bias in our dataset, results, and interpretations. First, specific identification of tadpoles (i.e., assignation to a species and corresponding adult) is not minor and a prerequisite of any subsequent reliable description. Certainty in tadpole identification may vary among groups (e.g., it may be easier in areas with low diversity or species with parental care) and times (e.g., molecular identification is becoming more frequent11,91). Especially, some old literature deals with widely distributed species that have been subsequently split into several taxa, and it is not always possible to know accurately to which one the original mention refers; in those cases, we retained the tadpole description as originally assigned. Similarly, some tadpoles mentioned in the first editions of books that are no more reported in the following editions are subject to doubt. Although we checked for comments on misidentification and synonymies during our search, we may have overlooked some examples. Second, the definition and assignation to ecomorphological guilds may differ from other schemes in use, which may render different interpretations compared to other analyses. For instance, the definition of direct development we apply differs from that in Liedtke et al.76, and thus some calculations along the anuran tree may vary. Finally, even after a threshold approach, the geographic data we depict may overestimate the distribution of some species and guilds. For instance, northern records of viviparous Darwin’s frog Rhinoderma darwinii extend artificially the distribution of the species and its ecomorphological guild into the whole ecoregion of Chilean Matorral.
Data availability
All data generated or analyzed during this study are included in this published article and its supplementary information files.
References
Hortal, J. et al. Seven shortfalls that beset large‐scale knowledge of biodiversity. Annu. Rev. Ecol. Evol. Syst. 46, 523–549 (2015).
Nori, J., Villalobos, F., Osorio, L. & Loyola, R. Insufficient protection and intense human pressure threaten islands worldwide. Perspect. Evol. Conserv. 20, 223–230 (2022).
Nori, J., Osorio-Olvera, L., Cordier, J. M. & Hortal, J. Global gaps in the knowledge of the herptile response to land transformation. Front. Ecol. Environ. https://doi.org/10.1002/fee.2625 (2023).
Faria, L. R. R., Pie, M. R., Salles, F. F. & Soares, E. D. G. The Haeckelian shortfall or the tale of the missing semaphoronts. J. Zool. Syst. Evo. Res. 59, 359–369 (2021).
Taboada, A., Wehrden, H. V. & Assmann, T. Integrating life stages into ecological niche models: a case study on Tiger Beetles. PLoS ONE 8, e70038 (2013).
Wassersug, R. The adaptive significance of the tadpole stage with comments on the maintenance of complex life cycles in anurans. Am. Zool. 15, 405–417 (1975).
Ceron, K., Sales, L. P., Santana, D. J. & Pires, M. M. Decoupled responses of biodiversity facets driven from anuran vulnerability to climate and land-use changes. Ecol. Lett. 26, 869–882 (2023).
Nori, J. et al. Amphibian conservation, land‐use changes and protected areas: a global overview. Biol. Conserv. 191, 367–374 (2015).
Huang, N., Sun, X., Song, Y., Yuan, Z. & Zhou, W. Amphibian traits database: a global database on morphological traits of amphibians. Glob. Ecol. Biogeogr. 32, 633–641 (2023).
McDiarmid, R. W. & Altig, R. Tadpoles, the Biology of Anuran Larvae (Chicago Univ. Press, 1999).
Raharivololoniaina, L., Grosjean, S., Rasoamampionona Raminosoa, N., Glaw, F. & Vences, M. Molecular identification, description, and phylogenetic implications of the tadpoles of 11 species of Malagasy treefrogs, genus Boophis. J. Nat. Hist. 40, 1449–1480 (2006).
Hiobiarilanto, T. R., Randrianiaina, R. D., Glos, J., Strauß, A. & Vences, M. Larval morphology of 12 species and subspecies of genus Boophis from Madagascar. Xenopus 2694, 1–25 (2010).
Haas, A., Das, I., Hertwig, S. T., Bublies, P. & Schulz-Schaeffer, R. A Guide to the Tadpoles of Borneo (Tredition, 2022).
Altig, R. & Johnston, G. F. Guilds of anuran larvae: relationships among developmental modes, morphologies and habits. Herpetol. Monogr. 2, 81–109 (1989).
Elinson, R. P. Metamorphosis in a frog that does not have a tadpole. Curr. Top. Dev. Biol. 103, 259–276 (2013).
Donnelly, M. A., de Sá, R. O. & Guyer, C. Description of the tadpoles of Gastrophryne pictiventris and Nelsonophryne aterrima (Anura, Microhylidae), with a review of morphological variation in free-swimming microhylid larvae. Am. Mus. Nov. 2976, 1–19 (1990).
Ruibal, R. & Thomas, E. The obligate carnivorous larvae of the frog Lepidobatrachus laevis (Leptodactylidae). Copeia 1998, 591–604 (1988).
Vigny, C. Morphologie larvaire de 12 espèces et sous-espèces du genre Xenopus. Rev. Suisse Zool. 86, 877–891 (1979).
Tapley, B. et al. The tadpoles of five Megophrys horned frogs (Amphibia: Megophryidae) from the Hoang Lien Range, Vietnam. Zootaxa 4845, 35–52 (2020).
Jenkins, C. N., Pimm, S. L. & Joppa, L. N. Global patterns of terrestrial vertebrate diversity and conservation. Proc. Natl Acad. Sci. USA 110, 2602–2610 (2013).
BiodiversityMapping.org. https://biodiversitymapping.org (2022).
Gradwell, N. On the functional morphology of suction and gill irrigation on the tadpole of Ascaphus, and notes on hibernation. Herpetologica 29, 84–93 (1973).
Inger, R. F. Variation of apomorphic characters in stream-dwelling tadpoles of the bufonid genus Ansonia (Amphibia: Anura). Zool. J. Linn. Soc. 105, 225–237 (1992).
Marcillo-Lara, A., Coloma, I. A., Alvarez-Solas, S. & Terneus, E. The gastromyzophorous tadpoles of Atelopus elegans and A. palmatus (Anura: Bufonidae), with comments on oral and suction structures. Neotrop. Biodivers. 6, 1–13 (2020).
Arifin, U. et al. Revisiting the phylogenetic predicament of the genus Huia (Amphibia: Ranidae) using molecular data and tadpole morphology. Zool. J. Linn. Soc. 193, 673–699 (2021).
Lehtinen, R. M. Phytotelm-Breeding Frogs of the World, Version 1.11. https://sites.google.com/site/phytotelmbreedingfrogsworld/ (2022).
Dias, P. H. S. et al. Life on the edge: tadpoles of Cycloramphidae (Amphibia; Anura), anatomy, systematics, functional morphology, and comments on the evolution of semiterrestrial tadpoles. J. Zool. Syst. Evol. Res. 59, 1297–1321 (2021).
Hirschfeld, M., Barej, M. F., Legrand Gonwouo, N. & Rödel, M. O. Tadpole descriptions of three Cardioglossa species from southwestern Cameroon (Amphibia: Anura: Arthroleptidae). Salamandra 48, 147–156 (2012).
Vassilieva, A. B. Larval external morphology and chondrocranium of Leptobrachella bidoupensis (Anura: Megophryidae), an endemic litter toad from southern Vietnam. Zootaxa 5068, 587–594 (2021).
Senevirathne, G., Garg, S., Kerney, R., Meegaskumbura, M. & Biju, S. D. Unearthing the fossorial tadpoles of the Indian dancing frog family Micrixalidae. PLoS ONE 11, e0151781 (2016).
Moura, M. R. et al. Geographical and socioeconomic determinants of species discovery trends in a biodiversity hotspot. Biol. Conserv. 220, 237–244 (2018).
Raja, N. B. et al. Colonial history and global economics distort our understanding of deep-time biodiversity. Nat. Ecol. Evol. 6, 1445���154 (2022).
Linnaeus, C. Systema Naturae per Regna Tria Naturae, Secundum Classes, Ordines, Genera, Species, cum Characteribus, Differentiis, Synonymis, Locis (L. Salvii, 1758).
Hinckley, M. H. On some differences in the mouth structure of tadpoles of the anourous batrachians found in Milton, Massachusetts. Proc. Boston Soc. Nat. Hist. 21, 307–315 (1881).
van Bambeke, C. Research on the oral structure in tadpoles of batrachian anurans. Bull. Acad. R. 16, 339–354 (1863).
Annandale, N. Some undescribed tadpoles from the hills of southern India. Rec. Indian Museum 15, 17–24 (1918).
Annandale, N. & Rao, C. R. N. Indian tadpoles. Proc. Asiat. Soc. Bengal 13, 185–186 (1917).
Fernández, K. & Fernández, M. On the biology and reproduction of some argentinean batrachians I. Cystignathidae. An. Soc. Cient. Argent. 91, 97–140 (1921).
Altig, R. A key to the tadpoles of the continental United States and Canada. Herpetologica 26, 180–207 (1970).
Anstis, M. Tadpoles and Frogs of Australia (New Holland Publishers, 2013).
Duellman, W. E. The Hylid Frogs of Middle America (SSAR, 2001).
Fei, L., Changyuan, Y. & Jianping, J. Colored Atlas of Chinese Amphibians and Their Distributions (Sichuan Publ. Group, 2012).
Rossa-Feres, D. C. et al. Taking tadpole biology into the 21st century: a consensus paper from the First Tadpoles International Workshop. Herpetol. Bras. 4, 48–59 (2015).
Gosner, K. L. A simplified table for staging anuran embryos and larvae with notes on identification. Herpetologica 16, 183–190 (1960).
Poyarkov, N. A. Jr, Nguyen, T. V., Yang, J. H. & Gorin, V. A. A new species of Micryletta (Amphibia: Anura: Microhylidae) from the Langbian Plateau in southern Vietnam. Zool. Res. 42, 726–733 (2021).
Vacher, J. P. et al. Cryptic diversity in Amazonian frogs: integrative taxonomy of the genus Anomaloglossus (Amphibia: Anura: Aromobatidae) reveals a unique case of diversification within the Guiana Shield. Mol. Phyl. Evol. 112, 158–173 (2017).
Bongaarts, J. I. P. B. E. S. Summary for policymakers of the global assessment report on biodiversity and ecosystem services of the Intergovernmental Science‐Policy Platform on Biodiversity and Ecosystem Services. Popul. Dev. Rev. 45, 680–681 (2019).
Peixoto, O. L., Caramaschi, U. & Freire, E. M. X. Two new species of Phyllodytes (Anura: Hylidae) from the state of Alagoas, northeastern Brazil. Herpetologica 59, 235–246 (2003).
Vera Candioti, F., Haas, A., Altig, R. & Peixoto, O. Cranial anatomy of the amazing bromeliad tadpoles of Phyllodytes gyrinaethes (Hylidae: Lophyohylini), with comments about other gastromyzophorous larvae. Zoomorphology 136, 61–73 (2017).
Kusrini, M. D., Rowley, J. J. L., Khairunnisa, L. R., Shea, G. M. & Altig, R. The reproductive biology and larvae of the first tadpole-bearing frog, Limnonectes larvaepartus. PLoS ONE 10, e116154 (2015).
Bolochio, B. E., Lescano, J. N., Cordier, J. M., Loyola, R. & Nori, J. A functional perspective for global amphibian conservation. Biol. Conserv. 245, 108572 (2020).
Carvalho-Rocha, V., Peres, C. A. & Neckel-Oliveira, S. Habitat amount and ambient temperature dictate patterns of anuran diversity along a subtropical elevational gradient. Divers. Distrib. 27, 344–359 (2020).
Becker, C. G., Fonseca, C. R., Hadda, C. F. B., Batista, R. F. & Prado, P. I. Habitat split and the global decline of amphibians. Science 318, 1775–1777 (2007).
Gehring, P. S. et al. Biodiversity Hotspots. Distribution and Protection of Conservation Priority Areas (eds Zachos, F. E. & Habel, J. C.) 235–254 (Springer, 2011).
Vieites, D. R. et al. Vast underestimation of Madagascar’s biodiversity evidenced by an integrative amphibian inventory. Proc. Natl Acad. Sci. USA 106, 8267–8272 (2009).
Vences, M. et al. The New Natural History of Madagascar (ed. Goodman, S. M.) 1323–1325 (Princeton Univ. Press, 2022).
Grosjean, S., Randrianiaina, R. D., Strauß, A. & Vences, M. Sand-eating tadpoles in Madagascar: morphology and ecology of the unique larvae of the treefrog Boophis picturatus. Salamandra 47, 75–88 (2011).
Poynton, J. C. Patterns of Distribution of Amphibians—A Global Perspective (ed. Duellman, W. E.) 483–539 (John Hopkins Univ. Press, 1999).
Herrmann, H. W. et al. African biodiversity hotspots: the amphibians of Mt. Nlonako, Cameroon. Salamandra 41, 61–81 (2005).
IUCN. The IUCN Red List of Threatened Species. Version 2021-3 (No. 2020–1). https://www.iucnredlist.org (2021).
Nori, J. & Loyola, R. On the worrying fate of data deficient amphibians. PLoS ONE 10, e0125055 (2015).
Frost, D. R. et al. The amphibian tree of life. Bull. Am. Mus. Nat. Hist. 297, 1–291 (2006).
Haas, A. Phylogeny of frogs as inferred from primarily larval characters (Amphibia: Anura). Cladistics 19, 23–89 (2003).
Jetz, W. & Pyron, R. A. The interplay of past diversification and evolutionary isolation with present imperilment across the amphibian tree of life. Nat. Ecol. Evol. 2, 850–858 (2018).
Portik, D. M. et al. Redefining possible: combining phylogenomic and supersparse data in frogs. Mol. Biol. Evol. 40, msad109 (2023).
Blotto, B. L., Pereyra, M. O., Grant, T. & Faivovich, J. Hand and foot musculature of Anura: structure, homology, terminology, and synapomorphies for major clades. Bull. Am. Mus. Nat. Hist. 443, 1–156 (2020).
Raff, R. A. Origins of the other metazoan body plans: the evolution of larval forms. Philos. Trans. R. Soc. 363, 1473–1479 (2008).
Albecker, M. A. et al. Does a complex life cycle affect adaptation to environmental change? Genome-informed insights for characterizing selection across complex life cycle. Proc. R. Soc. B 288, 20212122 (2021).
Rivera-Correa, M. et al. Amphibians in Zootaxa: 20 years documenting the global diversity of frogs, salamanders, and caecilians. Zootaxa 4979, 57–69 (2021).
Grosjean, S., Bordoloi, S., Chuaynkern, Y., Chakravarty, P. & Ohler, A. When young are more conspicuous than adults: a new ranid species (Anura: Ranidae) revealed by its tadpole. Zootaxa 4058, 471–498 (2015).
Muzzopappa, P. Stories about lizards and frogs: the Salamandra aquatique et noire de Feuillé (1714). Hist. Nat. 10, 107–115 (2020).
Raj, P. et al. Larval morphology and ontogeny of Nasikabatrachus sahydrensis Biju & Bossuyt, 2003 (Anura, Nasikabatrachidae) from Western Ghats, India. Zootaxa 3510, 65–76 (2012).
Marangoni, F. & Baldo, D. Life-history traits of three syntopic species of the South American redbelly toads (Anura: Bufonidae: Melanophryniscus) from the Atlantic Forest of Argentina. Herpetol. Conserv. Biol. 18, 213–228 (2023).
Sherratt, E., Vidal-García, M., Anstis, M. & Keogh, S. Adult frogs and tadpoles have different macroevolutionary patterns across the Australian continent. Nat. Ecol. Evol. 1, 1385–1391 (2017).
Schiesari, L., Rossa-Feres, D. C., Menin, M. & Hödl, W. Tadpoles of Central Amazonia (Amphibia: Anura). Zootaxa 5223, 1–149 (2022).
Liedtke, H. C., Wiens, J. J. & Gomez-Mestre, I. The evolution of reproductive modes and life cycles in amphibians. Nat. Commun. 13, 7039 (2022).
Etard, A., Morrill, S. & Newbold, T. Global gaps in trait data for terrestrial vertebrates. Glob. Ecol. Biogeogr. 29, 2143–2158 (2020).
Moura, M. R. & Jetz, W. Shortfalls and opportunities in terrestrial vertebrate species discovery. Nat. Ecol. Evol. 5, 631–639 (2021).
Guedes, J. J. M., Moura, M. R. & Diniz-Filho, J. A. F. Species out of sight: elucidating the determinants of research effort in global reptiles. Ecography 2023, e06491 (2023).
Frost, D. R. Amphibian Species of the World: An Online Reference. Version 6.1. https://amphibiansoftheworld.amnh.org/index.php (American Museum of Natural History, 2022).
AmphibiaWeb. https://amphibiaweb.org (University of California, 2022).
Olson, D. M. et al. Terrestrial ecoregions of the world: a new map of life on Earth. BioScience 51, 933–938 (2001).
Hurlbert, A. H. & Jetz, W. Species richness, hotspots, and the scale dependence of range maps in ecology and conservation. Proc. Natl Acad. Sci. USA 104, 13384–13389 (2007).
Peterson, A. T. Problems with reductive, polygon‐based methods for estimating species’ ranges: reply to Pimm et al. 2017. Conserv. Biol. 31, 948–951 (2017).
QGIS Development Team. QGIS Geographic Information System. Open Source Geospatial Foundation Project. http://qgis.osgeo.org (2022).
Blommers-Schlösser, R. M. A. Observations on the larval development of some Malagasy frogs, with notes on their ecology and biology (Anura: Dyscophinae, Scaphiophryninae and Cophylinae). Beaufortia 24, 7–26 (1975).
Jiang, K., Wang, K., Wang, Y. F., Li, C. & Che, J. A new species of the endemic Himalayan genus Liurana (Anura, Ceratobatrachidae) from southeastern Tibet, China, with comments on the distribution, reproductive biology, and conservation of the genus. Zool. Res. 40, 175–184 (2019).
Carvajal-Endara, S. et al. Phylogenetic systematics, ecology, and conservation of marsupial frogs (Anura: Hemiphractidae) from the Andes of southern Ecuador, with description of four new biphasic species. Zootaxa 4562, 1–102 (2019).
Hertwig, S., Das, I., Schweizer, M., Brown, R. & Haas, A. Phylogenetic relationships of the Rhacophorus everetti-group and implications for the evolution of reproductive modes in Philautus (Amphibia: Anura: Rhacophoridae). Zool. Scr. 41, 29–47 (2011).
Wake, M. H. The reproductive biology of Eleutherodactylus jasperi (Amphibia, Anura, Leptodactylidae), with comments on the evolution of live-bearing systems. J. Herpetol. 12, 121–133 (1978).
Schulze, A., Jansen, M. & Köhler, G. Tadpole diversity of Bolivia’s lowland anuran communities: molecular identification, morphological characterization, and ecological assignment. Zootaxa 4016, 1–111 (2015).
Acknowledgements
We deeply thank R. Altig, R. McDiarmid and P. Dias, whose selfless contribution (Bibliography for the identification, morphology and development of amphibian gametes and larvae. Distributed electronically) made our initial search much easier. We also thank M.L. Ponssa and S. Barrionuevo for discussions about developmental biology and C. Borteiro for language editing. F.V.C., D.B. and M.O.P. acknowledge the support from Agencia Nacional de Promoción Científica y Tecnológica (ANPCyT PICTs 2018-3349, 2019-346, 2019-2519, 2019-2557 and 2019-3895) and Consejo Nacional de Investigaciones Científicas y Técnicas (CONICET PIP 11220200102800CO). J.N. thanks CONATURAR, Redes Federales de Alto Impacto, Ministerio de Ciencia y Tecnología de Argentina (2023-102072649-APN-MCT), and CONACyT-CF 2023-I-11226 and PA-PIIT-IA203922 for financial support.
Author information
Authors and Affiliations
Contributions
F.V.C., J.N. and D.B. designed the study. F.V.C., S.G., D.B., M.O.P. and J.N. gathered data. F.V.C. and J.N. carried out the analyses and wrote the first version of the manuscript. All authors completed, edited and approved the final version.
Corresponding authors
Ethics declarations
Competing interests
The authors declare no competing interests.
Additional information
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Supplementary information
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/.
About this article
Cite this article
Vera Candioti, F., Baldo, D., Grosjean, S. et al. Global shortfalls of knowledge on anuran tadpoles. npj biodivers 2, 22 (2023). https://doi.org/10.1038/s44185-023-00027-1
Received:
Accepted:
Published:
DOI: https://doi.org/10.1038/s44185-023-00027-1