Proposal (1010) to South American Classification Committee

Treat Cranioleuca marcapatae weskei as a species

Note: This is a high-priority issue for WGAC.

Background: I (Remsen 1984) described weskei as an isolated subspecies of nearby ecological equivalent Cranioleuca marcapatae, based on their conspicuous differences in crown color. The rationale for subspecies rank was based in part by extrapolation to the situation in Cranioleuca albiceps, in which (1) the nominate subspecies shows individual variation in crown color from white to ochraceous, with white-crowned individuals looking superficially like weskei (to the point that Vaurie [1980] considered them to represent an isolated population of C. albiceps, although I regard that as a lapsus), and (2) the Cochabamba subspecies discolor has no white in the crown. The basic logic was that if crown color can vary dramatically within individuals of a single taxon as well as between taxa currently treated as subspecies, then weskei should also be treated at the subspecies rank.

Here is the plate from the OD (by Larry McQueen):

Here are some specimen from LSUMZ. I don’t think any of the specimens from dpto. La Paz are as ochraceous as those of C. a. discolor. Note the crown color of marcapatae differs from any of the others and matches the back color, unlike any of the others, including weskei. We don’t know (yet) whether the tawny crowns in nominate albiceps are a consequence of gene flow:

We do not have a SACC Note on weskei. What we do have is the following (which needs to be updated with respect to this proposal; although I have completed adding SACC notes on del Hoyo & Collar [2014; nonpasserines], I stalled out somewhere in the suboscines on the passerine volume [2016]:

49. The superspecies relationship proposed for Cranioleuca marcapatae and C. albiceps (e.g., Remsen 1984) is corroborated by genetic data (García-Moreno et al. 1999b, Derryberry et al. 2011); Fjeldså & Krabbe (1990) proposed that they might be best treated as conspecific.

Here’s how I set up the taxonomic problems posed by this complex:

“Should each of the four spinetail taxa in question be considered allospecies of a superspecies complex, subspecies of a single biological species, or some intermediate combination? In the absence of comparative data on vocalizations and other potential reproductive isolating mechanisms or on degree of genetic differentiation, any taxonomic decisions at this point for these primarily allopatric populations are tentative at best.”

Concerning the crown color differences, here is what I wrote:

“Current information indicates that crown color differences do not necessarily result in reproductive isolation. Specimens from an area (Choro, Prov. Ayopaya, Dpto. Cochabamba) geographically intermediate between populations of C. a. albiceps and C. a. discolor are intermediate in crown color between the white-crowned nominate subspecies and tawny-crowned discolor. Furthermore, the high frequency of buff-crowned individuals in the Dpto. La Paz population (see below) may indicate gene flow from the southern populations. Thus it seems best to disregard crown color, the character that varies most dramatically among populations, as a potential isolating mechanism (although this conclusion is based on the assumption that the populations are in secondary contact). Therefore, that marcapatae has a chestnut crown, in contrast to neighboring weskei and albiceps, in itself provides no evidence for reproductive isolation.”

I went on to conclude that changing currently recognized species limits was unwarranted given the lack of critical data.

New information: Del Hoyo & Collar (2014) treated marcapatae as a separate species based on the Tobias et al. point scheme as follows (provided by Pam Rasmussen):

“Hitherto treated as conspecific with C. marcapatae, but differs in its white crown (3); stronger black lateral crownstripe (1); and much more evenly delivered song, shorter in length (2), faster (at least 2), lower maximum frequency (ns[2]), and different note shape (ns[at least 1]) (1). Monotypic.”

This is a good example of the conceptual flaws of the Tobias et al. scheme, namely that the scores are derived independent of phylogeny. In this case, we already know in sister taxa that crown color is not associated with a barrier to gene flow, yet here it is accorded 3 points out of the 7 needed for species rank.

The vocal data are from Boesman (2016i), who analyzed 7 recordings of weskei and 9 of marcapatae:

“Song of weskei differs from marcapatae by much more uniform delivery, shorter length (score=2), lower max. frequency (score=2), faster pace (score=2 or 3), descending song strophe (vs. rising and falling, score 2) and different note shape (score 1). This would lead to a total vocal score of 4 or 5 when applying Tobias criteria.”

On the surface, these sound like impressive differences. I am assuming that the recordings represent different individuals, but I leave it to those more experienced with Cranioleuca voices to evaluate them.

As for the inevitable demand for genetic data, I continue to point out their limited utility for assigning taxon rank in cases like this, i.e. allopatric taxa in a monophyletic group. We already know what the results would be – the populations WILL differ, especially given the obvious phenotypic differences. So, how different do they have to be to be treated as species? Genetic data measure largely neutral loci not explicitly those genes associated with gene flow and reproductive isolation, and thus largely measure time-since-isolation and effective population size. The former is very roughly correlated with speciation, but we have concrete examples of dramatic differences in genetic distance among populations that are given different taxonomic ranks that no one would dispute (e.g. capuchino Sporophila seedeater species vs. Syndactylus bulbul populations that are phenotypically indistinguishable and not accorded any taxonomic rank) that provide evidence that characters associated with reproductive isolation do not evolve at a steady rate but instead idiosyncratically. Looking at branch lengths among close relatives eliminates some but not all of the problems. Anyway, see the SACC note on earlier genetic studies that only show that we’re dealing with a monophyletic group, although weskei to my knowledge has yet to be sampled.

Here is the tree from Harvey et al. (2020), which shows clearly that albiceps s.l. and marcapatae are sister taxa and that the node that unites them is as deep as those linking most Cranioleuca treated as species, thus in the crude comparative sense supporting species rank for them.

As for evidence for or against gene flow between marcapatae and weskei, these are strongly allopatric taxa. The aerial distance on the map between the Vilcabamba range and the main Andes might not be impressive to those not familiar with the area, but the ecological gap is huge for high-elevation cloud-forest birds whose short, rounded wings predict limited dispersal abilities. Although in my opinion the dispersal ability of nonmigratory birds is vastly underappreciated, nonetheless, I suspect that it would be so infrequent between these two as to be impractical to detect without massive sampling.

Recommendation: Given my involvement with these taxa, I think the right thing to do is recuse myself from voting, and so I have no recommendation, and Glenn Seeholzer, whose dissertation was on the Cranioleuca antisiensis group (e.g. Seeholzer & Brumfield 2018), has agreed to take my voting slot. I don’t have strong feelings either way at this point – I can see both sides.

English names: If the proposal passes, then I see no need for a separate proposal. This taxon has no phenotypic features that distinguish it uniquely from other Cranioleuca, and so Vilcabamba Spinetail seems not only the obvious choice but the only one. Of course, if you see it otherwise, please speak out and do a proposal.

References: (see SACC Bibliography for standard references)

BOESMAN, P. 2016i. Notes on the vocalizations of Marcapata Spinetail (Cranioleuca marcapatae). HBW Alive Ornithological Note 100. In: Handbook of the Birds of the World Alive. Lynx Edicions, Barcelona. https://doi.org/10.2173/bow-on.100100

REMSEN, J. V., JR. 1984. Geographic variation, zoogeography, and possible rapid evolution in some Cranioleuca spinetails (Furnariidae) of the Andes. Wilson Bulletin 96: 515-523.

SEEHOLZER, AND R. T. BRUMFIELD. 2018. Isolation by distance, not incipient ecological speciation, explains genetic differentiation in an Andean songbird (Aves: Furnariidae: Cranioleuca antisiensis, Line-cheeked Spinetail) despite near threefold body size change across an environmental gradient. Molecular Ecology 27: 279-296.

VAURIE, C. 1980. Taxonomy and geographical distribution of the Furnariidae (Aves, Passeriformes). Bulletin American Museum of Natural History 166: 1-357.

Van Remsen, June 2024

Comments from Robbins: “As far as new data, in my opinion, only the vocal data are potentially relevant. Listening to vocalizations I think Boesman accurately describes differences, but whether these mean anything who knows. I’m going to wait and see what Glenn has to say on this given his deeper perspective on Cranioleuca before I vote.”

Comments from Jaramillo: “NO for now, will wait until Glen Seeholzer gives and opinion, I am open to change.”

Comments from Lane: “NO. This is a case that I think is a bit hobbled by limited information. The distribution of the taxon weskei is still largely thought to be solely in the Vilcabamba range of Cusco and Junin depts, but since the 00's, it's been known that birds with white crowns occur all the way to the north side of the Mantaro valley in Junin dept. This is the population that I believe the recordings used in Boesman's assessment are from <https://xeno-canto.org/species/Cranioleuca-marcapatae?query=ssp:%22weskei%22>. A confounding factor is that there seems to be a vocal shift *within* white-crowned birds across the Mantaro, as birds I recorded in 2010 in Huancavelica dept (south of the Mantaro <https://macaulaylibrary.org/asset/392841541>) and birds that Mark Robbins and Peter Hosner recorded in Ayacucho in 2012 <https://macaulaylibrary.org/asset/173820, https://macaulaylibrary.org/asset/173847> and in the southern Vilcabamba in Cusco in 2012 <https://macaulaylibrary.org/asset/174006,https://macaulaylibrary.org/asset/186972> sound considerably more like nominate marcapatae. We don't have recordings from the type locality of C. m. weskei, so we can only assume that these more southerly recordings are representative of the topotypical voice of the taxon. So what does this apparent vocal shift mean across the Mantaro mean? Well, we have no specimen material to review, but I suspect there may be an undescribed taxon involved with the birds north of the valley, so Boesman's assessment of voice doesn't actually apply to true C. m. weskei, but probably to another taxon altogether which looks very like it. In the end, I think that situation needs to be sorted out before we can make species-level changes in the taxonomy of the C. marcapatae group. So, NO to splitting for now.”

Comments from Glenn Seeholzer (voting for Remsen): “YES.

“At the time of its description Cranioleuca marcapatae weskei was only known from the isolated northern Cordillera Vilcabamba. C. m. weskei and C. m. marcapatae (hereafter weskei and marcapatae) display the same plumage polarity (white vs. chestnut crown) that was used to diagnose the subspecies of C. albiceps. The initial placement of weskei as a subspecies of marcapatae was therefore based on the comparative BSC approach - white vs. buffy crown coloration was not a pre-mating barrier in C. albiceps (widespread occurrence of intermediate crowns) therefore weskei and marcapatae would also not exhibit reproductive isolation should they come into contact. Specifically, the presence of extensive intermediate crown coloration between the geographic and phenotypic extremes of C. a. discolor and C. a. albiceps was considered a sign of gene flow. By this same logic, if populations of weskei and marcapatae come into contact and show no evidence of intermediate crown coloration (i.e. introgression) this would indicate that gene flow between the taxa is minimal and that weskei is a species.

“Variation in crown color in C. marcapatae

“Weskei was described based on crown color – white – relative to the chestnut-crowned nominate subspecies (Remsen 1984). Based on the specimens available at the time Remsen considered crown color to be a discrete trait in C. marcapatae in contrast to the continuous variation in crown color between the buffy discolor extreme and the white nominate subspecies of Cranioleuca albiceps. Examination of recent specimens and ML photos (N=191) reaffirms Van’s original assessment that there is no evidence of intermediate crown coloration in either subspecies of C. marcapatae, even where there come into closest contact in the southern Cordillera Vilcabamba.

“See the following ML image galleries

● weskei (north of the Río Montaro)

● weskei (south of the Río Montaro, west of the Río Apurímac)

● weskei near contact zone (east of the Río Apurímac)

● marcapatae near contact zone (west of -72°W)

● marcapatae away from contact zone (east of -72°W)

●

“The ML photos fill in a large gap in specimen sampling in the southern Cordillera Vilcambamba (Figure 1), and the two subspecies are now documented to occur within ~40 linear kilometers of each other (60-70 km following the 3000 m contour).

“Figure 1 - Map of vouchered records of Cranioleuca marcapatae. Color indicates subspecies, shape indicates documentation method. Dark lines are 3000 m elevational contours. Gray lines are departmental boundaries.

Below are the two closest photographs with clear views of the crown.

“marcapatae

● ML440773301 - Inca Trail above Macchu Picchu (-13.226, -72.509)

● ML217863001 - Camino Salcantay above Colcapampa (-13.318, -72.609) - even closer than above but can’t confirm that the crown is not intermediate

“weskei

● ML492931841 - headwaters of the Rio Vilcabamba near Lucma (-13.046, -72.934)

“These localities occur at the headwaters of the Ríos Vilcabamba and Urubamba which converge downstream of Machu Picchu. Examination of the satellite imagery of the massif between these two rivers shows no obvious break in this species’ habitat (2500-3500 m montane forest). Unvouchered eBird observational records (S59715216, S33942238, S127451093) from near Mutuypampa (-13.181, -72.718) occur midway along the 3000 m contour between the closest vouchered localities. Given that weskei occurs across significant habitat breaks elsewhere in its range (e.g. Río Apurimac, Río Montaro), it’s a safe bet that some flavor of C. marcapatae continuously occupies the 2500-3500 m habitat band between these rivers making this region a contact zone.

“For comparison, specimens of Cranioleuca albiceps from southern Puno to Cochabamba show the full spectrum of intermediate crown coloration from white to buffy (see Van’s specimen series) while spanning almost 500 linear km and numerous breaks in the 2500-3500 m habitat band. Lest we doubt the ability of photographs to document intermediate crown coloration, see this photo of two C. albiceps individuals next to each other showing two different points on the white to buffy spectrum or this intermediate individual.

“If there were intermediates between weskei and marcapatae outside of the unsampled contact zone we would probably know it by now.

“Contact zone scenarios

“As explained above, it is likely that weskei and marcapatae come into contact somewhere along the 2500-3500 m habitat band between the Ríos Vilcabamba and Urubamba. There is a spectrum of possible patterns that we could expect in this contact zone. At one extreme is assortative mating indicating reproductive isolation, a slam dunk for treating weskei as a species.

“At the other extreme is a hybrid zone with panmixia creating a highly variable population of individuals with intermediate crown coloration as seen in C. albiceps. If there is a hybrid zone, it is a narrow one. From a strict BSC interpretation a hybrid zone with panmixia demonstrates a lack of reproductive isolation and the interacting taxa should be considered the same species. However, in practice we are highly tolerant of this kind of hybridization between ‘good’ species.

“In Cranioleuca, Seeholzer and Brumfield (2017) documented a ~1000 km gradient in body size, plumage, and genetics in C. antisiensis which resulted in the southern baroni and northern antisiensis groups being lumped. This gradient was highly correlated with elevation and temperature and is best considered a cline, not a hybrid zone. The only well-documented hybrid zone in Cranioleuca is between C. pyrrhophia and obsoleta. A thorough phenotypic study showed it to be 100s of kilometers wide with extensive intermixing of pyrrhophia and obsoleta phenotypic traits. Claramunt (2002) concluded that either the hybrid zone was young or that some form of reproductive isolation exists maintaining the phenotypes of the taxa in the face of gene flow. In either case, Claramunt recommended the taxa continue to be treated as separate species.

“More broadly we treat the members of many other super-species complexes with well-documented hybrid zones as separate species. Manacus candei, vitellinus, aurantiacus, and manacus all hybridize where their ranges abu and despite diverse treatments we currently consider these all separate species. Other examples include Rhegmatorhina berlepschi/hoffmansi, Myioborus melanocephalus/ornatus, Ramphocelus carbo/melanogaster, Thraupis episcopus/sayaca, and of course most of the taxa that hybridize across the Great Plains suture zone.

“I’m too checked out of the speciation and taxonomy literature to parse the meandering rationales for maintaining hybridizing taxa as species but if we’re going to be consistent, then a narrow hybrid zone between weskei and marcapatae would not be incompatible with treating them as species.

“Vocalizations

“After listening to recordings, I agree with Dan that southern weskei and nominate marcapatae sound very similar. Here are links to four galleries of ML recordings of the short song (the most commonly recorded vocalization) so others can assess for themselves.

● weskei (north of the Río Montaro)

● weskei (south of the Río Montaro, west of the Río Apurímac)

● weskei near contact zone (east of the Río Apurímac)

● marcapatae near contact zone (west of -72°W)

● marcapatae away from contact zone (east of -72°W)

“I don’t have time to do a quantitative analysis, but I also don’t think it’s necessary in this case. Absent an obvious diagnostic plumage feature like crown color, the minimal vocal differentiation between southern weskei and nominate marcapatae would indeed point us to one species. BUT we do have an obvious diagnostic plumage feature that shows no evidence of introgression despite close contact. A diagnostic vocal difference would help the case for species status, but the absence of vocal differences doesn’t detract from it.

“Dan’s intriguing observation that weskei north of the Montaro in Junín sound different is not relevant to the question of reproductive isolation or diagnostic characters between weskei and marcapatae where their ranges abut. I will add that in addition to that population having a higher, faster short song, the image galleries for this population also seem to show a more extensive buffy wash to their throats compared to individuals east of the Río Apurímac. Certainly seems like it could be an undescribed taxon.

“Summary

“Weskei and marcapatae show a discrete difference in crown coloration (white vs. chestnut) with closely abutting distributions and a high likelihood of a contact zone. This contact zone does not occur at an ecotone or biogeographic barrier. Song divergence is minimal but not relevant to assessing gene flow when crown color can be used as a proxy. There is no evidence of introgression in crown coloration suggesting that in the contact zone there is 1) assortative mating or 2) a narrow hybrid zone without gene flow into either taxon. Clearly additional specimens and photographs from the presumed contact zone are desirable but this would not change the patterns outlined above which are enough for me to consider weskei a distinct species from marcapatae.

“Thoughts

“It was fun to think about Cranioleuca again after so many years. Specimens are always the most desirable form of documentation. but I was impressed with how useful ML photographs were for this system. For assessing the distribution of a simple binary trait like crown color, they certainly came in handy!

“A modern genomics treatment C. marcapatae/weskei and C. albiceps would make a great dissertation project. Replicate systems with what appear to be two different evolutionary outcomes. For the future intrepid graduate student inspired by this system, here are some juicy looking sampling localities in the contact zone.

● Above Huayrani (-13.070, -72.829)

● Achirayoc (-13.062, -72.765) and trails above (-13.119, -72.771)

● Mutuypampa (-13.181, -72.718)

● Colcapampa (-13.319, -72.669)

“Additionally, C. marcapatae and C. albiceps presumably come into contact somewhere between Marcapata, Cusco, and the Sandia Valley, Puno along with potential contact zones for many other Bolivian Yungas endemics and their northern counterparts. High priority region for some adventure ornithology.”

Comments from Areta: “" I vote NO to the split. After seeing the case of C. baroni-antisiensis, taking a look at the intergradation in crown color in C. albicapilla (albicapilla grades into albigula, according to the HBW), knowing about the drastic geographic variation in plumage and the dramatic within-locality vocal variation in C. pyrrhophia, thinking on the geographic variation in crown color as exemplified by C. albiceps by Remsen, and being absolutely convinced that the Boesman analysis is woefully incomplete for a Cranioleuca I don´t think there is a strong case here.

“Note: there are sheer amounts on variation of speed of delivery, frequency patterns, note shape, duration and pretty much any aspect one may want to consider even in successive "songs" of the same individuals of Cranioleuca; therefore, any analysis will need to a) have large sample sizes and b) take extra care to compare the same vocalization types under the same motivational context [even then, it might be mare]). So, I think that the vocal analysis needs to be seriously downplayed here, given the high chances that it mischaracterizes the differences and similarities in the voices of weskei and marcapatae. Based on a cursory investigation of sounds, I don´t find any striking difference in the "double-calls" or in the "short-song" of these taxa. Of course, this is just a quick-and-dirty analysis, but I am underwhelmed by the differences (unsurprisingly, given how difficult to separate on vocalizations are several species of Cranioleuca...).

“Also, there is this report by Hosner et al. (2015 WJO 127:563–581 , p. 575):

‘White-crowned' Marcapata Spinetail (Cranioleuca marcapatae weskei).—We found this spinetail uncommon to fairly common at all forested sites at 2,800–3,600 m (KU 112900, 112843, 122581–2 and CORBIDI LCA 2012-24, MCCF 476, MBR 8315, MBR 8339; MJA 322, ML 173820, 173847, 174006–7, 174037, 186915, 186972). In June at Ccano, a few non-vocalizing individuals were encountered. However, in September and October at Chupón and above Paccaypata, presumed pairs and family groups (up to 4 individuals) were seen and heard daily as they foraged in Chusquea bamboo understory at 3,000–3,600 m. The Ccano and Chupón records are the first for Ayacucho. Cranioleuca m. weskei has also been documented recently in Junín (e.g., XC 74478, 155740). Specimens from Paccaypata differed from typical C. m. weskei in that the face and lores were plain gray, rather than buffy, and a few white crown feathers had faint rufous edging. These characters suggest that birds in Paccaypata may be intergrades between C. m. weskei and nominate C. m. marcapatae from further east in Cuzco. Apparent intergrades with a mix of rufous and white crown feathers have been observed around Vitcos (13.111º S, 72.938º W, 3,000 m; Jul 2014; B. Walker, pers. comm.)’. (Thanks to Tom Schulenberg for pointing this out in our WGAC discussion).

“Thus, I see no reason to split weskei at present.”

Comments from Stiles: “NO on current information. Given the variation in crown color within the close relative albiceps, and with its white crown the primary basis for recognizing weskei as a separate species, at least more genetic data and information on exactly where recordings of several (and which?) taxa were made will be needed to sort this out.”

Comments from Robbins: “After reading all the comments, including a number since I first reviewed this proposal has me still on the fence. Nonetheless, if we must make a decision based on current information I think it is premature to treat weskei as a species, thus I vote NO for now.”

Comments from Claramunt: “YES. A difficult case as there are many in the genus Cranioleuca. But I would ask those that voted “NO” to reconsider. Glen explained several key aspects very well. My impression is that in the context of Cranioleuca taxonomy, weskei is clearly a different species. I totally understand why Van described it as a subspecies: an isolated peripheral population with a single (albeit conspicuous) plumage difference that happened to be a variable trait in the sister species. Now we know that the distribution of weskei is much larger, even larger than that of marcapatae (still a minor point). But consider the following:

1) It is not uncommon for good species in Cranioleuca to differ in a single trait affecting the color of the head: C. obsoleta – C. pallida, C. erythrops – C. curtata. C. marcapatae and C. weskei would be just another case of that.

2) Crown color has great potential for affecting mate choice and determining reproductive isolation in Cranioleuca and spinetails in general. As pointed out before, different species of Cranioleuca typically differ more in plumage than in songs.

3) While albiceps shows a gradation in the color of the crown, the crown varies from white to light cinnamon only, not to rufous like the crown of marcapatae. The magnitude of the difference is much greater in the case of marcapatae and weskei.

4) Also in contrast to the albiceps case, the difference between marcapatae and weskei is discrete; intermediate phenotypes have not been described so far, with the exception of the mention of specimens of weskei from one locality in which “… a few white crown feathers had faint rufous edging,” and a report of another population with birds with crowns with mixed crowns (Hosner et al. 2015). I would take those observations with a grain of salt until more details are published.

5) As pointed out by Glenn, if there is indeed some hybridization, it seems very limited geographically.

“Future work on these species will tell what is exactly going on, but in the context of our current taxonomy of Cranioleuca, I think that weskei should be treated as a species.”

Additional comments from Stiles: “I find Santiago's comments very interesting, especially his overview of the main part of the distribution of Cranioleuca - hence, I am willing to change my vote to YES for proposal 1010.

Comments from Ryan Terrill (voting for Del Rio): “YES -- I think this is a very borderline case that warrants more data and could go either way, but Glenn's comments help a lot here and are appreciated. It does seem like there is likely to be a contact zone somewhere in the headwaters of the Vilcabamba/Urubamba rivers, which would be great to visit and sample. But, as noted, if they do come into contact, they either don't interbreed, or if they do, form a very narrow hybrid zone. Other examples of variation within Cranioleuca species show a lot of variation and trait introgression – notably the C. antisiensis and C. albiceps complexes. In reference to Santiago's comments about crown color as a trait affecting mate choice, I have seen pairs of C. albiceps segregate by crown color within their contact zone in Cochabamba, see my notes in this checklist -- https://ebird.org/checklist/S42501143  "A pair of birds near the top of the road had pale yellow crowns, noticeably and distinctively paler than the orange crowned birds just down the road". Of course, this is only one anecdote.”