Prager (2017) offered a series of criticisms of our recent publication providing a first phylogenetic hypothesis for the Ploceidae (De Silva et al., 2017). We were chagrined to note her correct indication that we misinterpreted the taxonomic identity of samples to which we referred as Northern Red Bishop (Euplectes franciscanus) versus Southern Red Bishop (E. orix), Montane Widowbird (E. psammocromius) versus Marsh Widowbird (E. hartlaubi), and Fan-tailed Widowbird (Euplectes axillaris) versus Long-tailed Widowbird (E. progne). We accept these three corrections fully: we overlooked differing taxonomic concepts between holdings of museum collections (the first two cases), and did not catch a misidentification of a specimen in one museum collection (the latter case). A fourth criticism was odd, involving a problem with the taxon labels of the ND2 sequences that we deposited in Genbank, which are scrambled— Prager (2017) admitted that our figures and conclusions are not affected; we have verified that the Genbank accession numbers reported in our paper are correct, although the names in the Genbank data records are not. We are in the process of correcting the problems with the Genbank data records.
We suggest that such issues of sequence metadata could be addressed much more flexibly if Genbank were to create a comments and feedback section, so that persons other than the creator of the record can provide input. Such comments can be offered, for example, for auditory submissions to a public database of avian vocalizations, Xeno-canto (
http://www.xeno-canto.org/). In our case, thanks to Prager’s comment, we are addressing making the appropriate changes in Genbank. However, such corrections are not always made, so Genbank metadata often remain uncorrected in spite of some researcher knowing that problems exist.
After the factual documentation of the problems with our publication, however, Prager (2017) proceeded to make remarks that are more ad hominem in nature. She asserted that we have been “careless” in our literature review, but no documentation is offered. She suggested that, because she found problems in Euplectes, problems are probably rife across our ploceid work—indeed, we have checked each taxonomic name used, and can confirm that no further problems of this sort exist. A final comment, directed at co-author Peterson (although co-author Bates is equally guilty!), was about our focus on the need for museum vouchers in phylogenetic studies ( Bates et al., 2004; Peterson et al., 2007). Prager (2017) says that she appreciates the importance of museum collections and voucher samples in such research, but that a voucher is no substitute for knowledge of the studied organisms. We agree, and for that reason have sustained long field careers, but we remind Prager that vouchered samples can be revisited and re-examined, as she was able to request in the case of our misidentification of E. axillaris). Yes, perhaps a mistake was made (not by us, but by the source museum); however, thanks only to the existence of voucher specimens, it can be revisited and verified; with non-vouchered material (see, for example, the sampling in Prager et al., 2008), such repeatability of scientific results is not possible, and corrections are possible only via inference and supposition.
On a more fundamental level, we consider it rather sad that Prager (2017) saw it necessary to publish an aggressive critique of our paper. We were in friendly communication with her group soon after publication of our paper, exploring possible coordination and/or cooperation, yet no mention of problems was made. The problems pointed out could easily have been corrected as a “friendly amendment,” rather than as an attack. Science can, and should, proceed in a more civilized fashion, but the present interchange is not an example of that possibility.
