Proposal (#363) to South American Classification Committee

Split Zimmerius chrysops into two or three species

Effect on South American checklist: This would leave Zimmerius chrysops and Z. viridiflavus as separate species, and split out two additional species from chrysops, Z. albigularis and Z. flavidifrons.

Background: The genus Zimmerius was erected by Traylor (1977) for a set of small tyrannulets placed in the genus Tyranniscus, which lacked true wingbars among other characters. There have been a number of taxonomic issues at the species-level within the genus. In this case, the taxon Zimmerius chrysops has been variously treated as a distinct species (Zimmer 1941, Ridgely and Tudor 1994, Ridgely and Greenfield 2001, Fitzpatrick et al 2004), or conspecific with Z. viridiflavus (Meyer de Schauensee 1966, 1970, Schulenberg et al 2007). SACC has previously considered the issue of chrysops and viridiflavus in Proposal 173, which proposed lumping chrysops and viridiflavus. It did not pass, but did engender extensive commentary relevant to the current proposal.

The voice of Ecuadorian chrysops is distinct from Peruvian viridiflavus (Ridgely and Tudor 1994). However, Schulenberg et al (2007) noted that birds morphologically like chrysops that are found in Peru south of the Rio Marañon sound like viridiflavus. The details of this mismatch between voice and plumage were described in detail in Proposal 173 to relump Z. chrysops and Z. viridiflavus. Further complicating the issue are flavidifrons from southwestern Ecuador and northwestern Peru and albigularis from western Ecuador and southwestern Colombia. These taxa look much like chrysops but also have distinct voices.

Analysis and new information: In Proposal 173, Dan Lane called for genetic work on this complex. That genetic work has now been done. Rheindt et al. (2008) examined pieces of mitochondrial (NADH) and nuclear DNA (Fibrinogen intron 5) for the genus Zimmerius.

His trees find that albigularis does not associate on the tree with chrysops or viridiflavus. Instead, it is sister to Zimmerius vilissimus. The rest of the chrysops/viridiflavus complex forms a clade that subdivides into two well-differentiated groups. One group, their Clade 1 contains populations of chrysops from north of the Rio Marañon. Their Clade 2 contains viridiflavus, chrysops from south of the Marañon, and flavidifrons. These two clades have 100% bootstrap, etc. support. Because they are sister to one another, this tree is consistent with either lumping or splitting viridiflavus and chrysops. This is also completely consistent with the vocal data described by Dan Lane in Proposal 173,.

Recommendation:

Proposal A. Treat Zimmerius albigularis as a separate species. I recommend a YES vote, Rheindt et al (2008) started with this change, because it is most strongly supported by the genetic work. Given that this taxon has a distinct voice from the members of the viridiflavus complex, and genetically is not part of this complex, but rather more closely related to Zimmerius vilissimus, this seems like an easy decision. Rheindt et al suggested an English name of Choco Tyrannulet for albigularis, which seems like a reasonable suggestion.

Proposal B. Treat Zimmerius chrysops as conspecific with Z. viridiflavus. I recommend a NO vote. SACC has already voted against this concept, and the genetic data supports the basic idea that the two vocally distinct groups, chrysops north of the Marañon, and viridiflavus plus chrysops from south of the Marañón, along with flavidifrons are monophyletic units. The definition of chrysops that we have been using based on morphology is not supported by the vocal or genetic data. However, the vocal and genetic data align very well, and the name chrysops is the correct one for the northern taxon.

Proposal C. Treat Zimmerius flavidifrons as a separate species. I recommend a YES vote, weakly. Zimmerius flavidifrons should only be treated as a distinct species, in my view, if chrysops and viridiflavus are treated as separate species. If they are lumped, then flavidifrons is genetically buried in the heart of that complex. Committee members should look at Mark Robbins' comments on Proposal 173 for information on the vocalizations of this taxon. Genetically and vocally, flavidifrons is associated with viridiflavus, but its plumage resembles chrysops. The voice, while sharing certain features with flavidifrons, is distinct, whereas genetically it is sister to viridiflavus + south of the Marañon chrysops. Treating it as a distinct species seems reasonable, but not a slam dunk. Rheindt et al. suggested treating flavidifrons as a subspecies of viridiflavus. They argued that although the voice is different, it is not very different from viridiflavus, and that genetically it is not very different from viridiflavus , <1% different, compared to ca. 7% difference between viridiflavus and northern chrysops.

There are actually 3 potential treatments of this taxon. One is to raise it to a separate species. One is to continue to treat it as a subspecies of chrysops. This would match morphology, but be in conflict with genetics and voice. It is implicitly our current treatment. The third treatment would be to lump it into viridiflavus. This last treatment would make more sense if flavidifrons is not considered a distinct species. It is the treatment recommended by Rheindt et al. For flavidifrons, if split, Ridgely and Greenfield suggest the name Loja Tyrannulet, which seems appropriate.

Rheindt et al also suggested that if northern chrysops is split, and viridiflavus, southern chrysops, and flavidifrons are treated as a single species, then the name Peruvian Tyrannulet is not appropriate for viridiflavus. They suggest Tschudi's Tyrannulet, which has a history of being applied to viridiflavus. I am not sure I agree with them. However, if, when the dust settles, we have created that taxon, then I will do a proposal for such a name change.

Further recommendation: Because the SACC list is strictly at the species level and we provide no geographic range for the species, the assignment of intraspecific populations to particular species is not always clear. Such is the case in this instance. If we recognize (as we currently do) chrysops and viridiflavus as distinct, the treatment of the chrysops from south of the Marañón and of flavidifrons is ambiguous. I think, implicitly, we currently follow the treatment of taxa and populations suggested by Ridgely and Tudor (and followed by Howard and Moore), in which flavidifrons and all of chrysops are included in chrysops.

A more reasonable treatment would be to continue to recognize two species, but include the southern populations of "chrysops" in viridiflavus along with flavidifrons (the suggested treatment of Rheindt et al). However, vote of No on both Proposals B and C do not distinguish between the two alternatives. So I would like to ask that committee members that vote No on Proposal B specify whether they accept the Rheindt tree which places southern chrysops with viridiflavus and flavidifrons or are maintaining the status quo of all of chrysops together. Similarly, if committee members vote No on Proposal C, they should indicate whether they want to associate flavidifrons with chrysops or with viridiflavus. I would recommend associating it with viridiflavus. I would hope that we can then include the information on the assignment of these populations by the committee in the notes online, assuming that one or the other of these votes require this specification.

One additional issue with respect to the placement of populations within the named species is that Rheindt et al in their suggested treatment (table 5 in the paper) treat the southern populations of chrysops as part of their subspecies flavidifrons. This doesn't seem to match the genetics, voice or plumage.

One final note. The population of "chrysops" from south of the Marañón appears to have no name. If this group is split off from the northern chrysops and not considered part of flavidifrons, it will require a subspecific name.

References:

FITZPATRICK, J. W. 2004. Tyrannidae. Pp. 170-462 in "Handbook of the Birds of the World, Vol. 9. Cotingas to pipits and wagtails." (J. del Hoyo et al., eds.). Lynx Edicions, Barcelona.

MEYER DE SCHAUENSEE, R. 1966. The species of birds of South America and their distribution. Livingston Publishing Co., Narberth, Pennsylvania.

MEYER DE SCHAUENSEE, R. 1970. A guide to the birds of South America. Livingston Publishing Co., Wynnewood, Pennsylvania.

RHEINDT, F. E., J. A. NORMAN, AND L. CHRISTIDIS. 2008. DNA evidence shows vocalizations to be better indicator of taxonomic limits than plumage patterns in Zimmerius tyrant-flycatchers. Molecular Evolution and Phylogenetics 48:150-156.

RIDGELY , R. S., AND P. J. GREENFIELD. 2001. The birds of Ecuador. Vol. I. Status, distribution, and taxonomy. Cornell University Press, Ithaca, New York.

RIDGELY, R. S., AND G. TUDOR. 1994. The birds of South America, vol. 2. Univ. Texas Press, Austin.

TRAYLOR, M. A., JR. 1977. A classification of the tyrant-flycatchers (Tyrannidae). Bulletin, Museum of Comparative Zoology 148:128-184.

ZIMMER, J. 1941. Studies of Peruvian birds, No. 37. The genera Sublegatus, Phaeomyias, Camptostoma, Xanthomyias, Phyllomyias and Tyrannus. American Museum Novitates 1109: 1-25.

Doug Stotz, July 2008

============================================================

Comments from Stiles:

"363A. YES. Genetic and vocal data (plus my limited experience with
albigularis in SW Colombia) support this change.
363B. NO. Again, two species-level taxa are supported by both genetic and vocal data.
363C. NO, tentatively. This one is harder, since flavidifrons is much ore closely associated with viridiflavus both genetically and vocally. On present evidence, it might be best to retain flavidifrons as a distinctive subspecies of viridiflavus (with consequent naming of the S-of-Marañón chrysops as a third subspecies - actually, this must be done in any case. I am reluctant to split flavidifrons as a species at the present time - there are several other such cases of vocally (but not, or barely, in plumage) distinctive subspecies in tyrannid species like "Myiozetetes similis" and "Tolmomyias sulphurescens" that need genetic work. When this is done, we will have a better "yardstick" to judge such cases. For now, I favor a conservative approach."

Comments from Robbins: "Doug has done a good job of summarizing this complicated group.

"363A. I vote "YES" for recognizing albigularis as a species, as both genetic and vocal data support this.

"363B. I vote "NO" for considering chrysops and viridiflavus as conspecific, as genetic and vocal data indicate that they should be considered species.

"363C. I vote "YES" in recognizing these as species, because the voice of flavidifrons appears quite distinct from viridiflavus (see my comments in proposal 173) and they differ in plumage morphology."

Comments from Schulenberg:

 "363A (treat Zimmerius albigularis [no subtle guiding of our votes there] as a separate species): YES.

"363B NO. I'm thinking here only of "true" (northern) chrysops; the southern chrysops I would place with viridiflavus.

"363C: YES. This is complicated, but I think it's the right approach. I would leave no-name chrysops ("chrysops" in northern Peru south of the Maranon) in with viridiflavus, which of course is what it sounds like."

Comments from Nores: "

"A. YES. Como señala Stotz resulta una fácil decisión, ya que tanto por la voz como genéticamente se separa totalmente de chrysops.

"B. NO. Genéticamente hay dos grupos de chrysops bien separados: el de Ecuador y norte de Perú (chrysops) y el del sur de Peru (viridiflavus).

"C. NO. Genéticamente y por vocalizaciones está relacionado con viridiflavus y no debería ser separado como especie. Aunque por coloración se parece a chrysops esto no parece tener mayor importancia, ya que como señalan Rheindt et al. las vocalizaciones resultan mejor indicador de especie que la coloración en este grupo de tiránidos. Aunque esta opinión va en contra de mi modo de pensar como taxónomo, los análisis moleculares indican que los criterios usados en la taxonomía clásica no siempre son correctos."

Additional comments from Frank Rheindt:

"In his well-summarized proposal to carry out taxonomic splits within the Zimmerius chrysops complex, Dr Stotz referred to our recent paper (Rheindt et al. 2008. MPE 48: 150), in which we advocated the elevation to species rank of albigularis and called for a transfer of flavidifrons and "southern chrysops" to Z. viridiflavus. All SACC committee members who have so far cast their verdict have agreed with the albigularis split and the transfer of "southern chrysops" to Z. viridiflavus, but some have argued that the Tumbesian taxon flavidifrons should also be elevated to species rank rather than be lumped with Z. viridiflavus.

"In response, I would like to draw attention to the disappointingly small sample size of our cited study. The Zimmerius paper was the fortuitous side product of a PhD thesis that primarily focused on another tyrannid genus, so a more thorough sampling regime was not feasible. The geographical sampling did not do full justice to the interesting phylogenetic patterns within the Zimmerius chrysops complex, and I hope the publication of these preliminary results will not dishearten other researchers from revisiting this genus with a more thorough sampling regime.

"While increased sampling is unlikely to change conclusions with regard to the solid albigularis split and the transfer of "southern chrysops" to Z. viridiflavus, much more geographical sampling is required to evaluate the final status of flavidifrons under the biological species concept. We advocated a subsumption of flavidifrons under Z. viridiflavus mainly based on the small mtDNA divergence (1%) that falls much below the typical level of divergences between tyrannid sister species. Since many SACC committee members may be reluctant to accept genetic divergences as guidance in taxonomic decisions where there is little comparative material from closely related taxa, it is worth noting that there are now a few studies that place this divergence in a comparative framework and that have consistently documented mtDNA divergences in excess of 4% in a wide range of tyrannid sister species pairs, including closely related elaeniine genera (e.g. Johnson & Cicero, 2002, Mol Ecol 11: 2065; Joseph et al, 2003, MPE 31: 139; Chesser, 2000, MPE 15: 369; Rheindt et al, 2008, BMC Evol Biol 8: 193; Rheindt et al, 2008, Emu 108: 261). The divergence between Z. chrysops (sensu our study) and expanded Z. viridiflavus is 7%!

"Those who prefer to discount the low mtDNA divergence as a yardstick may want to consider that the morphological and vocal cases for species status are not that strong either:
(1) The most pronounced break in plumage pattern does not fall between flavidifrons and the east-slope taxa, but within the east-slope taxa (i.e. between unnamed "southern chrysops" and viridiflavus); a flavidifrons split on morphological grounds would therefore have to include a transfer of "southern chrysops" to flavidifrons, which would counter-argue vocal data.
(2) There is much disagreement over just how different flavidifrons vocalizations are, and whether these differences merit biological species status; see the excellent discussion in SACC Proposal 173 between people with some of the most field experience in this genus. Some argue that the break of the single flavidifrons call into 2-3 vocal elements within the east-slope taxa is substantial enough to prevent them from interbreeding in hypothetical non-allopatric settings. However, the ranges of these taxa may not be allopatric at all, as they closely approach each other in the Maranon Gap. Vocal and genetic sampling in this region may show intermediacy in call types/repertoire and mtDNA, which would argue for a cline rather than distinct species. Even in the absence of a continuous range, vocal and genetic characters may approach each other where flavidifrons comes close to the Maranon. I guess a particularly interesting area to sample would be the Zumba/Palanda region of Zamora-Chinchipe Province in Ecuador and neighboring parts of Peruvian Cajamarca (San Ignacio down to the Maranon Gap). Maybe some SACC members or other field workers already have access to recordings or even genetic data from this area? In any case, I think more thorough vocal and/or genetic sampling from this region should be carried out before separating flavidifrons at the biological species level."

Comments from Jaramillo:

"A - YES
B - NO
C - YES. This is the most problematic of the choices. Voice and genetic work show it is a distinct, albeit relatively young lineage. It seems to me that the cleanest thing to do is separate it as a species for the reasons Stotz outlines."

Comments from Cadena:

"A: YES
B: NO
C: NO. I think more work is necessary to define species limits in the viridiflavus group before species status can be assigned to flavidifrons. In relation to F. Rheindt's comments, he mentions that committee members who had voted thus far had accepted "the transfer of southern chrysops to viridiflavus". Because southern chrysops does not (yet?) have species rank, this is something that would not be reflected in our classification, but I just thought I'd should note we are actually not voting for this in the current proposal, or am I missing something? Also, just for the record: although Rheindt et al. (2008) did look at variation in a nuclear gene, this was not informative about relationships in the clade, so their molecular phylogeny is based entirely on mtDNA."

Comments from Pacheco:

"A YES
B  NO
C NO. Sobretudo por conta da opinião emanada pelo Daniel, prefiro ser conservador neste item e aguardar por novos estudos."