Proposal (989) to South American Classification Committee

Treat Ocreatus underwoodii as multiple species

Background: In an effort to align world lists, WGAC has split the Ocreatus racket-tails into three species, a move followed by Clements’/eBird and IOC lists. The main reasons for this move are based on information provided by Schuchmann et al. (2016); plumages, morphometrics, display behavior, and biogeography. SACC has not yet addressed the issue and footnote 62a in Trochilidae says the following:

The two southern subspecies, annae and addae, were formerly (e.g., Cory 1918) each considered a separate species from Ocreatus underwoodii, but they were all treated as conspecific by Peters (1945). Ridgely & Greenfield (2001) suggested that addae might deserve recognition as a separate species. The subspecies peruana was also formerly (e.g., Cory 1918) considered a separate species from O. underwoodii, but they were treated as conspecific by Peters (1945); Cory (1918) used "cissiurus" for the species name, but not only does peruana have priority but Peters (1945) considered "cissiurus (= cissiura)" a synonym of peruana. Schuchmann et al. (2016) provided evidence from plumage and behavior that Ocreatus underwoodii should be treated as four species, with the subspecies addae, annae, and peruana elevated to species rank. SACC proposal needed.

According to Schuchmann et al. (2016), Ocreatus comprises between two and four Biological Species. The plumage characters that they deemed important were: 1) male ventral coloration, 2) female ventral coloration, 3) tibial tuft (“boot”) coloration, 4) undertail coloration, 5) male tail morphology (rectrix length and whether rectrices cross or not), and 6) male tail racket (“flag”) coloration. Morphological characters of interest were: 1) bill length of white-booted forms vs. buff-booted forms, 2) wing length of white-booted forms vs. buff-booted forms, 3) rectrix length (and shape) of white-booted vs. melanantherus/peruanus vs. annae vs. addae, and 4) flag size of white-booted forms vs. peruanus vs. addae/annae. Schuchmann et al (2016) provided observations on the courtship display from four taxa: incommodus (of the white-booted group, N=17), peruanus (of the buff-booted group, N=11), annae (of the buff-booted group, N=6), and addae (of the buff-booted group, N>15), in which incommodus had a three-stage display vs. a two-stage display exhibited by peruanus, and a one- or two-stage display exhibited by annae/addae. According to Schuchmann et al (2016, these displays “broadly correspond with the morphological groups, suggesting the existence of at least three different groups (underwoodii/melanantherus, peruanus, annae/addae).” Finally, Schuchmann et al. (2016) mentioned that a case could be made for sympatric occurrence in eastern (Pogio, near Loja) Ecuador for melanantherus and peruanus based on specimen material, noting that the specimens of melanantherus were of immatures, suggesting local breeding rather than vagrancy (which struck me as an odd statement, as most hummingbird vagrancy seems to be done by immatures!), but Zimmer (1951) dismissed this case suggesting instead that the two immature melanantherus specimens were mislabeled.

Using the Tobias scoring system, Schuchmann et al (2016) concluded that Ocreatus contained four biological species: white-booted O. underwoodii, and the buff-booted peruanus, annae, and addae. Schuchmann et al. (2016) did not delve into vocal characters given that few recordings were available on Xeno-canto or Macaulay at the time, and it wasn’t clear if the vocalizations present were comparable (i.e., homologous).

New Information: A quick visit to our LSU collection confirms some of the issues outlined above, and throws a monkey wrench in the works for others. Overall, I see agreement between our specimens and the comments on ventral/boot coloration in both sexes across the various populations represented. I also see the shortening of the rectrices from northern to southern taxa as was pointed out in the paper. The sticking point is that there appears to be a population that was not assessed by Schuchmann et al (2016) from Cushi, in eastern Huánuco, and north into the Cordilleras Divisoria and Azul (extending from SW Loreto and adjacent San Martin south to eastern Huánuco and Ucayali depts). Interestingly, although the authors acknowledge the LSUMZ collection in their paper, and specifically mention the Cushi locality in their Appendix 1, they apparently did not review the specimens therein very closely, as they overlooked this population, all specimens of which should have been present during a revision of the LSUMZ collection in preparation for this paper. This population appears to display intermediate characters between peruanus and annae, including flags intermediate in size and shape between peruanus and annae and that are more iridescent blue than annae (more like peruanus), rectrix length intermediate between peruanus and annae, non-crossing racket-tipped rectrices (like peruanus; although the LSU specimens are prepared such that this character cannot be assessed, I am basing this character state on the following field images from the Cordillera Azul: https://macaulaylibrary.org/asset/205180711, https://macaulaylibrary.org/asset/205179041, https://macaulaylibrary.org/asset/611086271, https://macaulaylibrary.org/asset/603149731, https://macaulaylibrary.org/asset/171092391, but note that at least occasionally, peruanus CAN cross its rectrices as seen here: https://macaulaylibrary.org/asset/513125041), annae typically crosses its rectrices (e.g., near topotypical: https://macaulaylibrary.org/asset/242905071), and the outer web of the longest rectrices are not paler-edged as in peruanus, but dark, more like annae (a character mentioned by Zimmer 1951). It is unclear to me if these specimens from Cushi and the Divisoria/Azul represent a named taxon (possibly annae, type locality of which is Chanchamayo, Pasco), or not. Our holdings of male specimens from central and southern Peru are rather sparse and do not answer this question adequately. Interestingly, Schuchmann et al (2016) place the specimen from Cushi under the name “annae” and one from “Fundo Cinchana” (which appears to be near the “Abra Divisoria” locality of our LSU specimen) in “peruanus” in their Appendix 1, suggesting that they didn’t detect that these specimens were anomalous to their taxonomy, but rather considered them to represent two different taxa!

Unlike at the time of the writing of Schuchmann et al. (2016), the holdings of online sound archives are much better with regard to voices of these taxa, and whereas there is still little indication of what sounds can be best considered homologous, I can, by ear, detect patterns that strongly suggest that the white-booted forms (O. underwoodii by Schuchmann et al.’s designation) are vocally fairly distinctive compared to buff-booted forms. To my ear, the white-booted birds have a “springier” chase call (the rapid, descending trill) that often accelerates over the course of one trill and tends to be higher-pitched (e.g., https://xeno-canto.org/168323, https://xeno-canto.org/693722), whereas the buff-booted birds have a lower-pitched chase call (e.g., https://xeno-canto.org/226843, https://xeno-canto.org/257059, https://xeno-canto.org/251277) with a tighter, more even-pace (but can have variable speeds, probably depending on emotional state). Flat-trill calls (usually given when perched or foraging unprovoked, pers. obs.) are higher pitched in white-booted birds (e.g., https://xeno-canto.org/398384, https://xeno-canto.org/96236) and lower pitched in buff-booted birds (e.g., https://xeno-canto.org/22910, https://macaulaylibrary.org/asset/513125041). However, within the buff-booted forms, I cannot find any strong signal to suggest that there are distinctive vocal groups. Unfortunately, the Bolivian addae, which appears to have some of the strongest divergence in morphological characters within this group, is unrepresented by recordings at the time of this writing; however, I have heard it and was able to identify it as Ocreatus in the field in Bolivia, suggesting it is not strongly differentiated from the Peruvian buff-booted forms that I know well.

For your listening pleasure:

White-booted forms (O. underwoodii, sensu Schuchmann et al. 2016):

https://search.macaulaylibrary.org/catalog?taxonCode=boorat1&mediaType=audio&sort=rating_rank_desc&view=grid
https://xeno-canto.org/species/Ocreatus-underwoodii

“O. peruanus”*:

https://search.macaulaylibrary.org/catalog?taxonCode=boorat2&mediaType=audio&sort=rating_rank_desc&view=grid
https://xeno-canto.org/species/Ocreatus-peruanus

“O. addae”*:

https://search.macaulaylibrary.org/catalog?taxonCode=rubrat1&mediaType=audio&sort=rating_rank_desc&view=grid
https://xeno-canto.org/species/Ocreatus-addae

* The division of recordings among taxa by both archives appears to contain some error (for example, recordings from the Cordillera Azul are included under “peruanus” when they seem best considered “annae” or perhaps an undescribed form, based on the specimens I discussed above).

Recommendation: I think there is enough evidence put forth by Schuchmann et al. (2016) to divide Ocreatus into more than one species. However, I am not sure I agree that the complex warrants four species based on the publication and on the specimens and recordings I reference here. Between the plumage characters (boot tuft color first and foremost) and the display details, I believe that we can comfortably split the white-booted Ocreatus underwoodii (sensu Schuchmann et al. 2016) from the buff-booted forms (peruanus, annae, addae). The display behavior does not support division within the latter group, nor does there seem to be strong vocal signal to suggest finer divisions therein. So it falls on morphological characters to decide if the buff-booted forms require further species-level separation or not. Personally, I am not convinced that they do, particularly given the apparently intermediate population I mention above from eastern Huánuco and the Cordillera Divisoria/Azul area. Therefore, I recommend the following:

A. Split Ocreatus into white-booted (O. underwoodii, including polystictus, discifer, incommodus, and melanantherus) and buff-booted (O. addae, the oldest name, including peruanus, and annae) species. I recommend YES.

B. Split buff-booted forms into two species: northern O. peruanus and southern O. addae (including annae), as per the current treatment by Clements’/eBird, IOC, and WGAC. I recommend NO as per my reservations based on the Huánuco specimens discussed above.

C. Split buff-booted forms into three species: northern O. peruanus, central O. annae, and southern O. addae, as per the treatment recommended by Schuchmann et al. (2016). I recommend NO, for the same reasons as in “2” and the additional reason that addae is lacking archived voice information at present.

English names would have to be considered if any of these changes from a monospecific Ocreatus should be adopted. The names already used by Clements’/eBird and IOC are:

White-booted Racket-tail (O. underwoodii)

Peruvian Racket-tail (O. peruanus)

Rufous-booted Racket-tail (O. addae)

Schuchmann et al. (2016) suggested Anna’s R-t (O. annae) and Adda’s R-t (O. addae).

Should we decide to follow recommendation 1 above and not split the buff-booted forms any further, we could use “Rufous-booted R-t” as has been suggested by Ridgely and Greenfield (2001).

References:

García, N. C., K.L. Schuchmann, and P. F. D. Boesman (2022). White-booted Racket-tail (Ocreatus underwoodii), version 1.0. In Birds of the World (B. K. Keeney, Editor). Cornell Lab of Ornithology, Ithaca, NY, USA. https://doi.org/10.2173/bow.boorat1.01

García, N. C., K.L. Schuchmann, and P. F. D. Boesman (2022). Peruvian Racket-tail (Ocreatus peruanus), version 1.0. In Birds of the World (B. K. Keeney, Editor). Cornell Lab of Ornithology, Ithaca, NY, USA. https://doi.org/10.2173/bow.boorat2.01

García, N. C., K.L. Schuchmann, and P. F. D. Boesman (2022). Rufous-booted Racket-tail (Ocreatus addae), version 1.0. In Birds of the World (B. K. Keeney, Editor). Cornell Lab of Ornithology, Ithaca, NY, USA. https://doi.org/10.2173/bow.rubrat1.01

Gill, F., D. Donsker & P. Rasmussen (Eds). 2024. IOC World Bird List (v14.1). doi : 10.14344/IOC.ML.14.1. (accessed 5 January 2024)

Ridgely, R. S., and P. J. Greenfield (2001). The Birds of Ecuador: status, distribution, and taxonomy. Cornell University Press, Ithaca, NY, USA.

Schuchmann, K. L., Weller, A. A., and D. Jürgens (2016). Biogeography and taxonomy of racket-tail hummingbirds (Aves: Trochilidae: Ocreatus): Evidence for species delimitation from morphology and display behavior. Zootaxa 4200:83–108. https://doi.org/10.11646/zootaxa.4200.1.3

Zimmer, J. T. (1951) Studies of Peruvian birds. No. 61. The genera Aglaeactis, Lafresnaya, Pterophanes, Boissonneaua, Heliangelus, Eriocnemis, Haplophaedia, Ocreatus, and Lesbia. American Museum Novitates 1540. http://digitallibrary.amnh.org/bitstream/handle/2246/3700//v2/dspace/ingest/pdfSource/nov/N1540.pdf?sequence=1

Dan Lane, January 2024

Fig 1. Plate 1 from Schuchmann et al (2016) showing illustration of the different taxa within Ocreatus, and demonstrating important morphological characters. Note particularly boot color, male tail length, flag shape and color, and whether rectrices cross or not.

Fig 2. Figure 2 from Schuchmann et al (2016) showing specimens of selected taxa, including both male and female plumages.

Fig 3: Males of buff-booted forms from Peru and Bolivia, LSUMZ collection. A: peruanus from San Martin dept (near type locality). B: specimen from Cordillera Divisoria, Huánuco dept. C: specimen from Cushi, Huánuco dept. D: annae from Cusco dept. E: addae from La Paz, Bolivia. Photo by D. F. Lane.

Comments from Zimmer:

“A. YES, for reasons nicely summarized by Dan. I think the distinctiveness of the boot-color differences provides an obvious signal, reinforced by noted display differences. After listening to archived vocal samples of white-booted birds versus buff-booted birds, I would agree that the vocal distinctions are also supportive of splitting these two groups.

“B. NO, based on the seemingly intermediate specimens cited by Dan in the Proposal, and the lack of obvious (at least to my ears) vocal differences or display differences.

“C. NO as per my reasons for voting NO on Part 2, and given that there are no archived vocal samples of Bolivian addae for comparison with the other buff-booted taxa.

Comments from Areta: “In principle, even with notable gaps, the two-species treatment makes sense, but I will vote NO given several problems, missing data, and uncertainties. 

“The morphological and plumage evidence suggests the existence of clinality in several features (discussed by Schuchmann et al. 2016). As per the behavioral data, I am not convinced that the lack of the "whip phase" (phase C) in the short-tailed forms has been sufficiently corroborated (see Schuchmann et al. 2016:101 and Schuchmann 1987). For example, intensity and completeness of displays might be contingent upon the stage of the breeding season, and studying complex phenotypes for reduced periods of time might not provide thorough descriptions of displays. Because the last phase apparently leads to copulation, it is expected that not all the displays will end in this behavior. We don´t have information on periods of time or places in which these displays were seen.

“The vocal data shows that some calls (whose role in mating is possible non-existent) seem to differ in pitch. This adds another line of support on the differentiation between buff-booted and white-booted forms, but it falls short to show vocal differentiation relevant to mating. Also, note that the name that takes precedence is addae, but there are no recordings for this buff-booted distinctive taxon.

“See the relevant part of the tree from McGuire et al. (2014), which is perhaps the best available genetic evidence, but only includes one white-booted and one buff-booted taxon, from distributional extremes. The JPLV sample, which should be JLPV after Juan Luis Parra Vergara [thanks to T. Schulenberg for working this out], is from Cundinamarca, Colombia, therefore incommodus? The LSUMZ sample is from La Paz, Bolivia therefore addae. The dotted line is at 10 my.”

“This is a case in which more fieldwork, formal bioacoustic analyses, and genetic analyses are needed to properly assess the taxonomy of these wonders.”

Comments from Bonaccorso: “NO to A, B, and C. I agree with Nacho on all points. Definitively, genetic data and formal vocal analyses are needed to decide. Also, a better taxonomic covering regarding displays and probably, more observations, would further support the case. This is a nice group of taxa for which a complete phylogeographic analysis is still pending.”

Comments from Stiles: “NO for reasons stated above, but additionally for a new wrinkle in the tapestry – a recently discovered population of buff-puffed birds in the Serranía de San Lucas, the northernmost spur of the Central Andes of Colombia. Andrés Cuervo et al. will hopefully obtain genetic data on these (n=3), but until these are analyzed, it´s impossible to know which (if any) applecarts might be upset.”