The Dutch taxonomic committee has ruled to hold on for the various redpolls for now.
I bet some of you won't be surprised, but to give you an idea of how the article that caused the lumps can also be interpreted, I've copied and pasted the reasoning:
----------
The AOS (Chesser et al 2024) has included Arctic Redpoll Acanthis hornemanni and Lesser Redpoll A cabaret in A flammea and recommended treatment of this group as a single species A flammea (Redpoll). The rationale for this change was that these taxa show ‘genomic homogeneity, continuous phenotypic variation, overlapping suitable habitat, and a lack of evidence supporting prolonged isolation among these three taxa (Mason & Taylor 2015, Funk et al 2021).’
We regard this lump as premature because 1 the two genomic studies (Mason & Taylor 2015, Funk et al 2021) lacked samples of A f rostrata; as a consequence, the genomic pattern of variation between A f rostrata and A h hornemanni has not been investigated; and 2 there is a lack of detailed quantitative studies of assortative mating in the areas where A f rostrata and A h hornemanni breed sympatrically, and where A f flammea and A h exilipes breed sympatrically. In addition, part of the case for conspecificity may be based on selective interpretation of the available genomic data – an interpretation that is inconsistent with other case studies of the genomics of speciation in birds.
During the past decade, several well-documented case studies have demonstrated that some taxa that are widely considered as valid species differ only in one or a few very small parts of the genome (‘islands of differentiation’). Examples include the parapatric Carrion Crow Corvus corone and Hooded Crow C cornix in Europe (Poelstra et al 2014), the parapatric Blue-winged Warbler Vermivora cyanoptera and Golden-winged Warbler V chrysoptera in North America (Toews et al 2016), the sympatric Iberá Seedeater Sporophila iberaensis and Tawny-bellied Seedeater S hypoxantha in South America (Turbek et al 2021), and the sympatric Plumbeous Seedeater S plumbea and Tropeiro Seedeater S beltoni in South America (Nguyen et al 2024).
In these pairs of species, very large parts of the genome of one species are almost entirely undifferentiated from the other species. Consequently, these pairs are characterised by very low genetic divergence (genome-wide Fst) values. Nevertheless, in each of these cases close examination of the genomic data showed that there is one relatively small part, or a few relatively small parts, in the genome that shows (nearly) complete genetic divergence between the species, and which contains genes coding for plumage differences, and sometimes other relevant differences (Poelstra et al 2014, Toews et al 2016, Turbek et al 2021, Nguyen et al 2024).
These case studies are highly relevant for species-level taxonomy, including that of the redpolls, because they demonstrate that: 1 valid species may show very low genome-wide divergence (ie, very low genome-wide divergence on its own is not evidence for conspecificity); 2 differences between species may be located on only one or a few islands of differentiation (ie, taxonomists should not conclude that two taxa ‘do not differ in their genomes’ if only (small) parts of the genomes have been sequenced, such as with ddRADseq or UCE sequencing); 3 sympatric populations with very low genomic divergence are not necessarily morphs of a single species; and 4 the case of Iberá Seedeater and Tawny-bellied Seedeater, which show strong assortative mating in sympatry and hence strongly reduced gene flow, further shows that very low genomic divergence does not necessarily mean extensive gene flow or lack of reproductive isolation, unless there is positive evidence for widespread hybridisation.
The pattern of genomic divergence among redpolls is similar to that of the four cases mentioned before. Funk et al (2021) demonstrated that differences between exilipes and flammea/cabaret are largely found on a relatively small part of the genome (a chromosomal inversion involving 55 million base pairs) and that this part includes candidate genes related to melanogenesis, carotenoid colouration, and bill shape. We see no reason to interpret the case of the redpolls as fundamentally different from the aforementioned cases.
The four aspects listed by Chesser et al (2024) in support of lumping the redpolls also apply to some or all well-studied cases cited above: genomic homogeneity in most of the genome (all four cases), continuous phenotypic variation (Corvus, Vermivora), overlapping suitable habitat (all except S plumbea/S beltoni), and a lack of evidence supporting prolonged isolation among taxa (all except S plumbea/S beltoni). Thus, the aspects listed in support of lumping are not convincing (see also Knox et al 2024).
We recommend that: 1 A hornemanni, A flammea and A cabaret are best retained as species pending the outcome of field studies; 2 field studies are conducted to assess levels of assortative mating between A h hornemanni and A f rostrata, between A h exilipes and A f flammea, and between A flammea and A cabaret; and 3 levels of song divergence between the three species (five taxa) are adequately studied (preferably where taxa are in contact).
----------
