Mayer, M., L.F. Marin da Fonte & S. Lötters
In Issues 2019
Mind the gap! A review of Amazonian anurans in GenBank. pp. 89-96 plus Supplementary data:
Supplementary data 1. List of 25 WWF Terrestrial Ecoregions used to define ‘Amazonia’;
Supplementary data 2. List of 609 anuran species with distributions overlapping with Amazonia;
Supplementary data 3. List of recently described Amazonian anuran species in alphabetical order;
Supplementary data 4. Combined list of 512 Amazonian species names used for GenBank searches;
Supplementary data 5. Geo-referenced locality data for 308 Amazonian anuran species names.
Abstract. We studied the knowledge gap in GenBank with regard to the ca. 600 anuran species from Amazonia. The markers 12S, 16S, COI and cytb were examined, on which information was available for about half of all species. Both the number of sample sites and the number of samples per species varied greatly (best studied each in 16S: 4.85 ± 10.37; 11.19 ± 31.20), and merely one fifth of all species had at least 5 sample sites. This suggests that a considerable portion of species is underrepresented in GenBank. Representativeness is especially difficult to assess in widespread species that at the same time could well represent cryptic allopatric species (i.e., with smaller distributions). This is a well-known phenomenon in Amazonian anurans considering that truly widespread species do exist. Moreover, limited sampling may not necessarily be the result of limited representativeness, as numerous species are known to occupy relatively small localised to regional ranges only. Our study furthermore revealed that in a geographic context, major portions of Amazonia have as yet been undersampled. That is, the total of 453 sample sites (most with more than one species sampled) are spatially clustered, often in areas with increased anthropogenic activity. We conclude that there is a large knowledge gap in terms of spatial sampling, resulting in taxonomic deficiencies.
Key words. ‘Missing areas’, mitochondrial markers, sampling, spatial distribution.