|
This Trinidad and Tobago frog, formerly regarded as
Dedropsophis minutus is now Dendopsophus
goughi (Boulenger). JCM
|
Cryptic genetic
diversity is now so commonly reported in molecular studies of amphibian species
that the existence of nominally widespread tropical species has been called
into question. However, supposedly widespread species occurring across multiple
biomes and countries are rarely comprehensively sampled across their complete
geographic range in screenings of genetic diversity or phylogeographic studies.
Sampling of species from across vast continental areas and across political
borders is often handicapped by financial, logistic and political factors.
In the Neotropics, nominal taxa such as the
toad Rhinella margaritifera (Bufonidae), the
thin-toed frog Leptodactylus fuscus (Leptodactylidae),
and the tree frog Scinax ruber (Hylidae)
are prominent examples of anuran species once considered to occur across nearly
the entire tropical lowlands of South America. Evidence has accumulated that
many such putatively widespread species could in fact be complexes of cryptic
taxa. However, given limited genetic sampling and the difficulty in reviewing
material from all countries hosting populations, their relationships and
systematics remain in many cases as unclear as they were decades ago.
Dendropsophus minutus (Peters, 1872) is a
small hylid frog, 21–28 mm snout-vent length, distributed in Cis-Andean South
America, including the Andean slopes, the Amazon Basin, the Guiana Shield, down
to the Atlantic Forests of southeastern Brazil, with an elevational record from
near sea level up to 2,000 m. Variation in coloration, osteology, advertisement
calls and larval morphology, along with molecular data from limited parts of
the species' distribution suggest that the nominal D. minutus might
represent a species complex. However, the sheer size of its supposed
geographical range along with nomenclatural and taxonomic complexity (six
junior synonyms) and unresolved relationships in the D. minutus species group have so far made these frogs
inaccessible to revision.
In a new study in PLoS ONE Gehara and colleagues (2014) use D. minutus to understand to what degree a tropical, small-sized
anuran has the potential to be continentally widespread with limited genetic
structure within its range, as expected for a single species. In addition to
conservation concerns, this question has important implications for South
American biogeography in general and amphibian systematics and evolution in
particular. Evidence is accumulating that body size in amphibians has a
positive correlation with range size, but contrary to this trend many Holarctic
amphibians occur with little genetic substructure across the vast ranges they
colonized after the last glaciation, despite sometimes moderate to small body
sizes. Whether such patterns also exist across vast ranges in tropical regions,
with their distinct historical climatic dynamics, is an open question.
Deciphering possible cryptic diversity within the nominal D. minutus would
also help inform conservation assessments which typically use species' geographic
distributions as criteria for conservation status.
The 16S tree containing all Dendropsophus for which sequences
were available recovered the monophyly of the D. minutus species
group. Within the group, the clade containing samples representing lineages
19–43 received a maximal posterior probability (1.0) and is defined here as the D.
minutus complex, given that lineage 25 contains samples from the type
locality of D. minutus.
Most of the mitochondrial lineages containing
more than one sample received strong nodal support. The lineages splitting off
from basal nodes of the tree (lineages 1–18) are distributed in the Guiana
Shield, and in the Andean region of Peru, Ecuador and Colombia, with an eastern
extralimital clade assembling disjunct localities in Mato Grosso and Pará.
The remaining lineages are in general more
widely distributed in central and eastern South America Lineages are largely
allopatric but several cases of sympatry were observed. The uncorrected
pairwise distances between lineages for the 16S gene ranged from 0.7 to 13%,
while within-lineage p-distances ranged from 0.0 to 1.8%.
Most of the lineages (45%) were found in only
one or two localities. Fifty per cent of the lineages were only found in areas
smaller than 10 km2, and more than 70% have known ranges smaller than 10,000 km2.
Eight out of the 43 lineages have a distribution larger than 100,000 km2.
Largest range sizes were found in northeastern Brazil (Caatinga domain; 997,262
km2, lineage 36), eastern Bolivia and western Brazil (Cerrado, Chaco and Dry
Forest domains; 293,321 km2, lineage 33) and the Guiana Shield (269,741 km2,
lineage 2).
Among the D. minutus species group members
external to the D. minutus complex,
lineages 1–6 are Guianan, while 7–18 are primarily distributed along the Andean
foothills, and all show well-pronounced molecular differentiation and
divergence. Among lineages 1–6, there is moderate genetic differentiation.
Considering mitochondrial reciprocal monophyly and GMYC results as criteria,
and being taxonomically conservative, one available name, Hyla goughi Boulenger, 1911 (type
locality: Trinidad), should likely be removed from the synonymy of D.
minutus and allocated to populations comprised by all or some of lineages
1–6. As a conservative estimate, the authors hypothesize that lineages 7–18
comprise seven distinct species, i.e., five named taxa and two undescribed
species (lineages 9+10 and 11+12).
Data
presented herein provide conclusive evidence for a strong genetic subdivision
of the nominal species Dendropsophus minutus as currently understood. Current
taxonomy conservatively assumes a putatively widespread species encompassing a
vast area of South America (from approximately latitude 11.0°N to 35.0°S),
distributed across several biomes. Our results, however, reveal high genetic
diversity within D. minutus that would suggest the existence of
numerous distinct species, leading to an important increase in number of
species. If this hypothesis is confirmed through further studies, the existence
of an increased number of species with decreased range sizes would have
important consequences for the definition of centers of endemism and for
assessing conservation status.
Despite revealing a
substantial amount of cryptic genetic diversity within D. minutus sensu lato, our results
also confirm the existence of widespread Neotropical species of anurans. While the
authors cannot yet confirm which of the mitochondrial lineages within the D. minutus complex will merit a
status as separate species, the authors suggest they can inversely conclude
that in most cases, all samples assigned to one mtDNA lineage should be
conspecific. Although cases of distinct amphibian species with low mtDNA
differentiation exist and phenomena of mtDNA introgression can potentially blur
species identities, such cases remain exceptional. Therefore, these factors are
unlikely to substantially affect the calculation of range sizes according to
which a total of eight mtDNA lineages have ranges >100,000 km2 (lineages 2,
19, 33, 34, 36, 39 41, 42). In the most inflationist taxonomic scenario, with
each of these lineages representing separate species, the dataset still
provides evidence for a species of lowland Neotropical amphibian (lineage 36)
occupying an area of almost one million km2, encompassing multiple biomes
across a distance of about 1,600 km between its two most distant populations.
The most widespread
lineages within D. minutus sensu
lato have distributions restricted to or centered in Brazil and occur within
rather open habitats, while lineage 2 of the D. minutus group (with a range of almost 270,000 km2) occurs in
rainforest. Several of the lineages known from only few or single sites (e.g.,
lineages 8, 11, 12, 13, 14, 15, 16) occur in the Andean foothills or on
mountain slopes. Nevertheless, in the Andean area, a higher sampling density is
needed before it can be concluded with certainty that those lineages are
restricted to small ranges. Hence, the distribution of mitochondrial lineages
in the D. minutus group indicate that in open lowland areas of South America, small-sized species of anurans can be widespread.
Citation
Gehara M, Crawford AJ, Orrico VGD, Rodríguez A,
Lötters S, et al. (2014) High Levels of Diversity Uncovered in a Widespread
Nominal Taxon: Continental Phylogeography of the Neotropical Tree Frog Dendropsophus
minutus. PLoS ONE 9(9): e103958.
doi:10.1371/