Treat Megascops atricapilla as conspecific with M. watsonii

Proposal (977) to South American Classification Committee

Treat Megascops atricapilla as conspecific with M. watsonii

Recent phylogenetic work showed that Megascops watsonii as currently defined is not monophyletic because it includes M. atricapilla (Dantas et al. 2016, 2021). A recent proposal to split M. atricapilla and M. watsonii each into multiple species, largely on the basis of fixed differences in mitochondrial DNA sequence data (Dantas et al. 2021), did not pass in this committee (SACC Proposal 937), and the two-species treatment has been maintained in most reference works. A possibility that has not been fully considered, and which I think may be the best option for now, is to treat the Megascops atricapilla-watsonii complex as a single species. The name atricapilla Temminck 1822 has priority over watsonii Cassin, 1849.

Harvey et al. (2017) showed, using genome-wide markers, that the Amazonian populations of the complex are largely undifferentiated in the nuclear genome. Dantas et al. (2021) sampled all populations and showed that the complex as a whole, including the Atlantic Forest populations, is undifferentiated in three different nuclear markers (MUSK, CHD, and BF5), despite sizeable divergence in mitochondrial markers (Fig. S2 of Dantas et al. 2021). Vocal variation seems very complex, and although differences in vocalizations between some populations (Dantas et al. 2021) suggest that multiple biological species may exist, it is not completely clear which populations should be separated and which ones should be grouped together. The results of Dantas et al. (2021) do support the recognition of multiple phylogenetic species or biological subspecies, but it is less clear if they support multiple biological species. I think that the best option for the time being is to treat the Megascops atricapilla-watsonii complex as a single biological species, until a more densely sampled study of geographic variation in both the phenotype and genotype is available. Given that genetic and acoustic information are particularly important in this morphologically cryptic group, utmost care should be taken to thoroughly sample the contact zones and in the classification of different vocalization types (which can be challenging in this group). Field playback experiments could be particularly informative regarding the taxonomic value of the vocal differences, although such experiments can be logistically difficult to carry out with owls.

Although paraphyly is not a problem per se in species delimited under the BSC, it must be noted that this treatment of atricapilla and watsonii as a single species eliminates paraphyly. As pointed out by Dantas et al. (2021), atricapilla and watsonii were historically treated as a single species, so this would not be the first time.

References:

Dantas, S.M., J.D. Weckstein, J. Bates, N.K. Krabbe, C.D. Cadena, M.B. Robbins, E. Valderrama, and A. Aleixo. 2016. Molecular systematics of the new world screech-owls (Megascops: Aves, Strigidae): biogeographic and taxonomic implications. Molecular Phylogenetics and Evolution, 94:626‒634. https://doi.org/10.1016/j.ympev.2015.09.025

Dantas, S.M., J.D. Weckstein, J. Bates, J. N. Oliveira, T.A. Catanach, and A. Aleixo. 2021. Multi-character taxonomic review, systematics, and biogeography of the Black-capped/Tawny-bellied Screech Owl (Megascops atricapilla-M. watsonii) complex (Aves: Strigidae). Zootaxa 4949:401-444. https://doi.org/10.11646/zootaxa.4949.3.1

Harvey, M.G., A. Aleixo, C.C. Ribas and R.T. Brumfield, 2017. Habitat association predicts genetic diversity and population divergence in Amazonian birds. American Naturalist, 190:631‒648. https://doi.org/10.1086/693856

Rafael D. Lima, July 2023

Comments from Areta: “NO. This complex is difficult to sort out. While I sympathize with the idea of merging watsonii and atricapillus, this seems to be hiding away variation that has been so far difficult to explain. To me, the vocal and genetic data at hand do not allow to either lump or split the complex convincingly. It may be the case that carefully designed playback studies can help elucidate species limits, but this might not be as straightforward as in other cases. Megascops (and other owls as well) often respond aggressively to heterospecific vocalizations. I remember well the night that I was almost hit on the face by a M. hoyi in Salta (Argentina) in response to the long song of M. atricapilla recorded in Misiones (Argentina). Other hoyi individuals were not that aggressive, but still came to playback of atricapilla.”

Comments from Robbins: “YES to treating Megascops atricapilla as conspecific with M. watsonii. For reasons outlined in the proposal (primarily nuclear genetic data) it does seem most prudent to apply this treatment. Because Megascops do respond to other species songs (pers. obs.), I have serious doubts on whether playback experiments will offer insight.”

Comments from Stiles: “YES. Although Nacho is probably right that much variation in this group is unexplained, without further data on genetics, vocalizations and hopefully, plumages, I agree with Mark that it is probably best to accept Lima´s proposal for now, rather than going ahead with splits based on insufficient data and getting it wrong, so YES on this one.”

Comments from Krabbe (voting for Remsen): "YES. I find that the natural consequence of the NO to Proposal 937 (treating watsonii as multiple species) is to treat watsonii and atricapilla as a single species. Genetically, atricapilla is embedded within watsonii, and vocally, it falls within the variation found in watsonii."

Comments from Zimmer: “NO. I agree wholeheartedly with not adopting the additional splits within watsonii that were recommended by Dantas et al (2016), purely on the basis of genetic results, for reasons already thoroughly discussed by others. However, despite the apparent paraphyly of atricapillus being embedded within watsonii (something I would like to see stronger corroboration of), I regard lumping all taxa into atricapillus (which has priority) as the proverbial “bridge too far”. There is simply too much vocal, morphological and ecological variation within this wide-ranging complex (which spans the entirety of the Amazonian and Atlantic Forest biomes) for me to accept that there is only 1 biological species involved, no matter what the genetic data are telling us. Just because we haven’t been able to explain or properly delimit the variation in a way that allows us to move forward, doesn’t justify, in my opinion, sweeping it all under the rug and taking a destabilizing step backwards. According to comments provided by Mike Harvey (presented in Proposal #937 on species-limits in this group), his analysis was congruent with that of Dantas et al (2021) in recovering each of the three oldest clades, but differed in that the internal branches of his tree were much shorter, suggesting that the divergence times in Dantas’s tree were highly overestimated, likely due to deep mitochondrial coalescence. So, although both analyses were concordant in identifying geographic structure to genetic differences, it appears as though the time of divergence of these clades is much more recent than suggested by Dantas et al (2021). Keep in mind, that Mike did not have access to samples from the Atlantic Forest (Dantas’s clades D, E and F, which pertain to currently recognized atricapillus), which led him to acknowledge that “Without samples from Dantas’s clades D, E, and F, I can’t evaluate those more recent divergences.” He went on to say that “If the clades were supported by clear phenotypic/vocal differences, then I would be okay with considering them as separate taxa, but it looks like this is not really the case in their analysis.” I would agree that Dantas’s analysis failed to demonstrate clear phenotypic/vocal differences with respect to justifying any of the newly proposed splits (although I suspect this is largely because we still haven’t seen a densely sampled vocal, morphological and genetic analysis to make sense of what is obviously a complex situation), but neither do I think there is convincing published evidence of an absence of phenotypic/vocal differences between Atlantic Forest atricapillus and Amazonian watsonii to justify destabilizing the current taxonomy by lumping all of the observed variation under a single species. It is widely conceded that vocal differences are likely to provide more meaningful taxonomic characters in nocturnal species such as owls and nightjars, and even more so in groups such as Megascops, in which the majority of species are known to include multiple discrete color-morphs and all manner of intermediate plumages. I haven’t attempted any quantitative vocal analyses in the watsonii-atricapillus complex, but qualitatively, to my ears, the Atlantic Forest birds are vocally more similar to geographically far-removed Megascops vermiculatus/guatemalae than they are to geographically more proximate and intervening S Bank Amazonian populations of M. watsonii usta, in pitch, pace and pattern (particularly at the ending of the song). It doesn’t make sense to me, to treat these Atlantic Forest populations as conspecific with vocally dissimilar widespread Amazonian forms, just because the genetic data indicate that time of divergence is recent, and that some populations of watsonii are closer to atricapillus than they are to other populations of watsonii. No matter how recent the divergence, the Atlantic Forest populations of atricapillus, given habitat fragmentation and widespread deforestation along the southern arc of Amazonia, would seem to be on an independent evolutionary trajectory from watsonii, so even if one doesn’t consider them different enough now, they are only going to continue to diverge. As for the argument that atricapillus and watsonii have been treated as conspecific in the past, I would note that sanctaecatarinae was much more recently treated as conspecific with atricapillus, and we now know that they are not only genetically, vocally and morphologically distinct, but they also occur syntopically with atricapillus. In sum, I think it is better in this case to ‘First, do no harm’ and wait to make further changes to the status quo until we have geographically broader sampling and more robust analyses. With respect to the question of using reciprocal playback trials to test the importance of vocal differences, I would join with others in expressing my doubts about how helpful that would be in this case, exactly for the reasons already mentioned (e.g. the prevalence of heterospecific aggressive responses to playback, as well as the logistical difficulties of playback trials under nocturnal conditions).”

Comments from Lane: “NO. I agree with Kevin on this, and refer back to some proposals I made years ago (173, 174) that were shot down because they would have to be reversed when the respective taxonomies were finally ironed out. I now see the wisdom in those decisions and believe it applies here. Whereas I think we don't yet have enough to act upon to divide up M. watsonii into additional biological species in a satisfactory way, it will be inevitable, so taking a step back here is not going to improve the situation. Leave it as is until watsonii can be resolved.”

Comments from Del-Rio: “NO. I vote for keeping the status quo until we have a clearer picture of what is going on in this complicated genus. It might be one of those cases where whole genomes might shed some light on species limits and levels of gene flow.”

Comment from Josh Beck: “I am personally fascinated by the vocal variation in M watsonii. I have a few times played the "wrong" version of a standard song to a bird that was already vocalizing. Several times I have heard the bird seem to alter its song - either to conform more to the new playback stimulus, or just generally becoming more aggressive - with the "regional incorrect" stimulus. I've never done this in a structured way and unfortunately don't have demonstrative before/after recordings, so this is purely anecdotal. However, this is in line with others' observations here and I agree that playback experiments in this group would be very difficult to interpret.”

Comments from Bonaccorso: “NO. For all the reasons exposed by Kevin´s detailed account. More vocal and morphological data, along with better genetic data are needed to take such an important step. The tree in Dantas et al. 2021 is probably driven by mitochondrial signal, which may or may not be consistent with the evolutionary history of these lineages.

“In the end, if more nuclear data corroborate the mitochondrial topology and some of those clades are effectively diagnosable, several splits will be guaranteed, including recognizing M. watsonii as a species. If so, we would return to point zero after disturbing the taxonomy of Megascops for no good reason. For now, it seems more reasonable to maintain Megascops atricapilla separate from M. watsonii.”