Proposal (1013) to South American Classification Committee

Treat Stilpnia whitelyi (Black-hooded Tanager) as a separate species from S. cyanoptera

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

Background: Our SACC note on this is as follows:

36a. Ridgely & Tudor (1989), Hilty (2003, 2011), and Restall et al. (2006) noted that the subspecies whitelyi of the Tepui region might deserve treatment as a separate species from Stilpnia cyanoptera. Del Hoyo & Collar (2016) treated these two as separate species.

The two taxa involved are:

S. c. cyanoptera of the mountains of ne. Colombia and n. Venezuela

S. c. whitelyi of the Tepui region of s. Venezuela and w. Guyana, as well as a limited area in adjacent extreme n. Brazil

Ridgely & Tudor (1989) simply stated that “Whitelyi may well represent a separate species”; they described whitelyi as:

“ … notably duller with less opalescence generally and a rather strong mottled effect below, resulting in a much less neat appearance; it also lacks the blue wing-edging. Female whitelyi very dingy, with pronounced flammulated effect below.”

Hilty (2003) simply stated that the two might be separate species but did not provide details and did not illustrate whitelyi separately in the plate. Restall et al. (2006) cited Hilty for possibility that the two are separate species and described whitelyi as:

“male noticeably duller with less opalescence, strong mottled effect below, lacks blue wing-fringes, female very dingy with pronounced flammulation on underparts.”

Here is the HBW plate by Hilary Burn from Hilty (2011):

New information: Del Hoyo & Collar (2014) treated whitelyi and cyanoptera as a separate species based on the Tobias et al. point scheme as follows (provided by Paul Donald):

"See T. argentea [aka cyanoptera*], with which hitherto treated as conspecific; differs in its more extensive black hood, extending onto nape and upper breast (2); lack of a deep cobalt wingpanel (3); paler, much less yellow underparts, mantle and back, with more grey smudging below (2); larger bill (effect size 2.5, score 2); and shorter wing (effect size –2.03, score 2). Monotypic."

* Del Hoyo & Collar (2014) maintained a broad Tangara, and thus had to use argentea in place of cyanoptera. We avoided this problem by placing cyanoptera in Stilpnia; see SACC Note 34aa under Thraupidae.

Thus, the threshold 7 points was achieved (and then some -- 11 points) by awarding 2 points to “larger bill”, 2 points to “shorter wing”, and 2 points to “more extensive black hood”. In my opinion, those differences in bill and wing are barely perceptible, weakly supported, and nearly irrelevant at that scale to species limits in birds. Whether the differences in hood warrant two points is debatable – the differences are barely apparent in the HBW plate. So, even within the Tobias et al. point scheme, which I have twice criticized in print as fundamentally flawed, this one is really stretching the limits of the point system. One fundamental flaw is evident here. Tanagers, especially this set of genera, have a famously rich palette of color and pattern differences and thus have many more “opportunities” for racking up points than do most groups. Applying a pan-phylogenetic scoring system doesn’t make sense unless one believes that plumage phenotype is the one and only reproductive isolating mechanism in birds.

In response to my first draft of this proposal, Peter Boesman graciously provided the following information to include with the proposal:

A comparison of voice

There are currently 38 recordings in ML and 15 in XC (of which 8 are shared, and ML has 5 additional edited versions). A quick glance reveals that voice of both taxa is clearly different, and within each taxon there is consistency (which for a tanager, is worth highlighting).

An overview per taxon (starting with the commonest vocalization):

A. cyanoptera

1. tseek

A slightly squeaky (seemingly polyphonic) note

This vocalization is documented in ML620105829, ML608506168, ML608316543, ML575551461, ML431738041, ML301940221, ML300375481, ML287229501, ML228231241, ML167288, ML67241, XC164093, XC12661)

In a couple of cases the shape of the note is different, downslurred:

(ML247845, ML300375481 mixed in with ‘normal’ notes)

2. squeal

A very distinctive vocalization, sounding like ‘a nail scrubbing over a blackboard’.

(This vocalization is documented in ML67241, ML215019801, ML246270151, ML279203111, ML28517861, ML287229501, ML300375481,ML301940221, ML315744251, ML390578971, ML431738041, ML431741961, ML519155821, ML552841281, ML554409511,ML554409571, ML608316543, ML608506168)

The above two vocalizations are often used intermittently by the same bird or group in a recording.

3. Song

There is only a single recording of song, apparently a slowly descending series of high-pitched notes

(ML77541481)

B. whitelyi

1. Spik!

A very short rising crisp note spik!

(This vocalization is documented in ML30410, ML182439, ML77541321, ML290815, ML306736361, ML371633841, ML372185661, ML386453901, XC247120)

2. Pseeuw

Sometimes the note is slightly longer in duration and more dagger-shaped, uttered usually in combination with the previous:

(This vocalization is documented in ML30410, ML182439, ML77541321, ML306736361, ML371633841, ML372185661, ML386453901, XC247120, XC214716)

3. Tsee

There is a single recording of a similar high-pitched tsee

(ML306736371)

4. Song

There is a single recording of song, a fast series of notes at stable high pitch

(ML372185561)

(I have discarded 3 recordings because of poor quality or uncertain identification)

From this quick and brief overview, it is clear that vocalizations of both taxa are entirely different:

· None of the call notes resembles one another, and none of them are shared

· Song is rarely heard in both taxa, but is also seemingly different. More recordings are desirable, however.

While it is always possible that more recordings reveal a wider vocabulary, it is reasonably certain that a distinctive vocalization such as the squeal, heard in about half of the available recordings of cyanoptera, and not documented in whitelyi, is unique to the former taxon.

Discussion and Recommendation: Undecided. The information on vocal differences provided by Peter Boesman are informative, but how comfortable are we in using unpublished data? In a sense, they are “published” here?. Genetic data would only tell us that they differ to some degree at neutral loci and might provide a rough idea of divergence time and comparative divergence at neutral loci for other Stilpnia ranked as species or subspecies and thus would be of limited use, in my opinion, in assigning taxon rank. The differences in plumage between the two are eclipsed within a number of species in Tangara s.l. treated as conspecific (for better or worse), and are of unknown significance as barriers to gene flow.

English names: BLI retained “Black-headed” for cyanoptera/argentea and used “Black-hooded” for whitelyi. Retain the parental name Black-headed for the daughter cyanoptera s.s. fits with our SACC guidelines on English names, so I suggest we adopt those if the proposal passes; but if someone wants to write a proposal on this, feel free to do so.

References: (see SACC Bibliography for standard references)

Van Remsen, June 2024

Comments from Peter Boesman: “I had a quick look at vocalizations, and my conclusion is that there is a clear and consistent difference in the entire vocabulary of both taxa -- quick note is attached, please bear in mind I made that in a short time. I think this changes the case. We have now clear vocal differences, morphological differences of a magnitude similar to a few other Stilpnia pairs, and no genetic data (but since long allopatric taxa). I hope my small effort contributes to a better judgment of this case.”

Comments from Jaramillo: “YES – The point system is flawed, but it does give some “heads up” I think, similarly to DNA fingerprinting. It alone does not do much, but it highlights taxa that need a more serious look. Here is my thought, the reality is that these types of situations are subjective. How much weight do you give to certain aspects of plumage, voice, or genetics. If we had genetic information that showed a strong divergence, we would buy it and most would think two species were involved. But the crux of the matter, would these two hybridize in secondary contact? Would they share genes if parapatric? That we cannot know, and two species may reach that threshold before much genetic separation is evident. Scenario two, that same genetic work comes back and shows they are close molecularly, we would keep as subspecies. Yet we still don’t know, at least not as clearly as in parapatric populations. No matter what information we have on these types of scenarios, it is subjective. With that in mind, I do think the plumage differences are substantial in males. More importantly the plumage differences in the females attract my attention. Divergence in the “dull” plumages of a species is of interest to me, and it has not really been looked at all that closely. My assumption (again subjective) is that these differences take more time to develop than flashy male plumage changes in tanagers. Sexual selection works quickly at times and leads to divergence in the showy sex. But in the non-showy sex, perhaps less so. Therefore, noticeable plumage differences in male and females I find interesting. Also, the Tepuis are singular, an old set of mountains which are isolated and have a history of having taxa that are related to but separate from Andean taxa. It happens over and over again, creating the endemic rich zone that are the Tepuis. So in my subjective logic, having the different looking disjunct population in the Tepuis is important, if they were in another section of the Andes that would not be as important to me. That would imply a closer history perhaps, and more chance for gene flow historically or even in the future. Roundabout reason why I am voting YES on this one.”

Comments from Robbins: “YES. After listening to the audio recordings that Peter has highlighted in this proposal, I agree that the vocalizations are consistently different between these two taxa. I was particularly impressed with how unique the squeal (if truly absent from whitelyi) vocalization is of cyanoptera. I don’t recall hearing an analogous vocalization in any of the other Stilpnia or related genera. Thus, the vocalizations coupled with the differences in plumages in both male and female support recognition as species.  This would be consistent with how other related taxa are treated. Finally, I suspect if the male plumage was examined in the UV spectrum the differences would be even more pronounced.”

Comments from Areta: “YES. I remember well the drab taxon whitelyi from my visits to the Gran Sabana. This has been a long-time suspect. The plumage data is incontrovertible, as these two are quite different. The minor morphological differences add some nuance. But the vocal data seems to me the clincher. I have before voted against the split given the paucity of formally analysed data. Although this is pretty much the case at present, the available collection of information tips the scale towards the split. Of course, the Tepuis are a zone full of endemic species, and this one should now enlarge that list.”

Comments from Stiles: “YES. Plumage and vocal data coincide (not the case in several other proposals) making a stronger case for splitting whitelyi from cyanoptera.”

Comments from Lane: “YES. The group of Stilpnia this complex belongs to has a rather constrained vocabulary in my experience, so to have sisters that differ as much as these do is noteworthy.”

Comments from Claramunt: “YES. Phenotypic differences in multiple traits (including songs) and in both males and females suggest a long history of independent evolution.”

Comments from Steve Hilty: “YES. I have looked over the comments of those voting on this Stilpnia issue. Jaramillo's comments express pretty well my feelings on this issue. There is a lot of subjectivity in almost everything as regards these two entities. 

“Voice is the most persuasive evidence in this example and, although most of these little former Tangara or ex-Tangara in whatever genus they now reside, do respond, surprisingly well to playback of their voices (if you have a decent speaker)—they generally don't seem to be particularly vocal, and the songs, to human ears (or at least to mine) are so high-pitched and weak that they are easily and probably often overlooked. But I suppose they hear them just fine!

“Add plumage differences in both sexes to this mix of Andean and Tepui forms and the case gets a little stronger, as does the likely case for long isolation of so many groups in the tepuis.”

“That said, I don't feel this is the strongest data set (by a long shot) I've seen for recognizing two species, but it seems the best we have at present. The future might still prove all of us wrong.”