Proposal (#376) to South American Classification Committee

Eliminate Sporophila zelichi from the main list

Effect on South American CL: this proposal would eliminate Sporophila zelichi from the main SACC list.

Background: The current SACC note is as follows:

"32a. Narosky (1977) described Sporophila zelichi as a new species, but whether it is a valid species is controversial. Ridgely & Tudor (1989) and Sibley & Monroe (1990) maintained it as a species but noted that it was perhaps a localized color morph of S. cinnamomea or a hybrid population (S. cinnamomea X S. palustris), as suggested by Vuilleumier & Mayr (1987). Mazar Barnett & Pearman (2001) also continued to recognize it as a species. The observations of Azpiroz (2003) suggest that zelichi could be a valid species confined to marsh vegetation."

The taxonomy and systematics of the Capuchinos have been fertile grounds for the imagination of systematists, but there has been little effort in gathering comparative data to assess the species status of most forms. Thus, many different possibilities have been put forward, but none of them have been tested. In an attempt to bring some order into this complex situation, I have used an explicit hypothetic-deductive method to test several systematic scenarios regarding the validity of S. zelichi.

New information: In a forthcoming paper (Areta 2008) I propose that "Entre Ríos Seedeater" (Sporophila zelichi) (also called Zelich's Seedeater, White-collared Seedeater, and Narosky's Seedeater) is one of the rarest species in the Neotropics. However, doubts have been raised about the validity of this species. Therefore, I evaluated the systematic status of the Entre Ríos Seedeater based on analysis of previously unpublished vocal and habitat data. I tested four hypotheses regarding the systematic status of S. zelichi: Good Species Hypothesis (valid species), Hybridization Hypothesis (hybrid S. palustris X S. cinnamomea), Color Morph Hypothesis I (morph of S. cinnamomea), and Color Morph Hypothesis II (morph of S. palustris).

Briefly summarizing Areta (2008), the songs and preferred habitat of S. zelichi are indistinguishable from those of Marsh Seedeaters (S. palustris), and the songs of both forms have exhibited similar changes from the early 1990s to 2003-2007. In contrast, the songs and preferred habitat of Chestnut Seedeaters (S. cinnamomea) differ from those of S. zelichi. Therefore, the Good Species Hypothesis is rejected by vocalization and habitat data, the Hybrid Hypothesis is undermined by the absence of shared vocal characters and limited habitat overlap of the proposed parental forms S. cinnamomea/S. palustris, and Color Morph Hypothesis I is rejected by both song and habitat data. However, Color Morph Hypothesis II is supported by both song and habitat data. Thus, I propose that S. zelichi be considered a color morph of S. palustris.

Further discussion on the origin, predictions, and testing of each hypothesis can be found in the galley proof of the article, available for some time at:

Recommendation: I recommend a YES vote on this proposal. There is no evidence that would support treating zelichi as a full species under any species concept. Irrespective of whether the choice is to consider zelichi as a synonym of palustris (which seems the appropriate course) or as a possible hybrid between palustris and cinnamomea (or any other Capuchino, for that matter), zelichi is not a valid species. Voting for YES would eliminate Sporophila zelichi from the main list of the SACC.

Final comment: although some of you could be a little skeptical based on small sample sizes for voices of zelichi, it is worth mentioning that the results would be the same if I had taped only 3 palustris, 3 zelichi and 3 cinnamomea. The vocal differences are consistent regardless of sample size. It should also be considered that zelichi is a ¨Critically Endangered¨ species (50-250 individuals ¨estimated¨ by BirdLife Intl.), thus a genuine rarity. Further work is in course with the remaining ¨species¨ of the group (it is simply impossible to analyze the whole mess in a single publication), although some preliminary conclusions about other Capuchinos that might influence the status of zelichi are mentioned in the text.

Literature Cited

ARETA, J. I. 2008. The Entre Ríos Seedeater (Sporophila zelichi): a species that never was. Journal of Field Ornithology 79: 352-363.


Nacho Areta, November 2008



Comments from Stiles: "YES. The existence of an "odd" plumage in this group of Sporophila is not totally surprising, especially as such things do crop up elsewhere in the genus (v. gr., S. "insulata"); the field data definitely tip the balance towards eliminating zelichi as a species, irrespective of what its eventual status might be (and this might only be settled by genetic data). Although the cases are not wholly comparable, the fact that such data for S. insularis (370) indicate that the insulata plumage types show no indication of forming a monophyletic group within or beside telasco makes me leery of according zelichi species status at this time."


Comments from Nores:  “Yes, aunque solamente por el canto. Las características del hábitat, mencionadas en la propuesta, no son válidas para invalidar una especie. El hecho de que palustris y zelichi tengan hábitats similares no significa que sean la misma especie. Hay cantidades de ejemplos de especie congenéricas que tienen similar hábitat. En este caso, las características de hábitat servirían para separar zelichi de cinnamomea, pero no para invalidar a zelichi. Además, tampoco queda establecido que se trata de un morpho de palustris, ya que edad no fue evaluada. Como mencioné en la respuesta a la propuesta #245, dos aĖos atrás, un lugareĖo de Entre Ríos, que tenía un buen número de capuchinos en jaulas me dijo que en los ejemplares muy viejos de palustris el banco del pecho le avanzaba por el cuello formando un collar. O sea que tomaba el aspecto de zelichi. Por lo tanto, de ser así, se trataría solamente de ejemplares muy viejos de palustris, lo que explicaría también porque es tan escaso en la naturaleza.”

Comments from Santiago Claramunt: “Nacho's study is an important contribution towards resolving the species status of S. zelichi.  However, it is premature to draw a conclusion with any confidence.  Nacho's main argument is that the song and habitat of S. zelichi are very similar, if not identical, to those of S. palustris.  I consider the "habitat" evidence inconclusive.  Other good species of Sporophila share habitats, so I don't see why two species of seedeaters could not live in the same habitat. The vocal data are more suggestive of S. zelichi being a morph of S. palustris.  However, this is based on only three recordings of zelichi, and we do not know much about Sporophila vocalizations. Presumably, the primary songs are learned as in other oscines, and some species can mimic the song other birds (e.g. S. collaris, S. plumbea). Maybe zelichi males are singing palustris songs to maintain interspecific territoriality with a more abundant species that inhabit the same habitat. Maybe zelichi can sing other songs. A sample size of 3 is too small to draw any definitive conclusion.


“A piece of information that is missing is a detailed analysis of plumage. Plumage patterns in Sporophila are likely genetically determined and play a role in species recognition and mate choice. To me, the plumage of zelichi is inconsistent with Nacho's favored hypothesis of a morph of palustris. It seems that S. zelichi differ from palustris not only by the presence of a complete white collar but also by a darker rufous pigment, which is reddish chestnut like in cinnamomea, not light rufous like in palustris.  These differences may well function as species recognition cues for female zelichi. Alternatively, depending on the Mendelian nature of these two plumage characters, a hybrid palustris x cinnamomea may look just like zelichi. To me, the plumage is consistent with zelichi being either a valid species or a palustris x cinnamomea hybrid, but not a morph of palustris.


“Determining the species status of zelichi and other Sporophila is not going to be easy. I think more behavioral data is critical (vocalizations and mating behavior). Reproductive isolation and genealogical congruence may also be assessed by banding studies, or using highly variable molecular markers. Detailed analyses of plumage patterns, ontogeny, and heritability will also be useful. I think the evidence we have so far is not enough to reach a conclusion. The fact that zelichi is a critically endangered species is a good reason to be conservative and rigorous in this case.”

Comments from Cadena: “NO. The data presented by Nacho are suggestive, but given the small sample size (and that we are dealing with a critically endangered taxon), I think it is best to wait for more data. I also agree with comments by Manuel and Santiago that sharing habitats with other Sporophila provides little to no information about species status considering many Sporophila are syntopic. As a side note, I recently saw a BirdLife News item mentioning that a new population of S. zelichi has been found using habitat models (; perhaps additional data from this population will assist further taxonomic analyses.”


Additional comments from Nores: “Yo había aceptado eliminar a Sporophila zelichi sólo por el canto, pero después de leer los comentarios de Santiago Claramunt y de Cadena considero que eliminar la especie es prematuro. Por esa razón cambio mi voto a un NO.”



Additional comments from Areta: “I welcome Santiagoęs and Manuelęs criticisms on my proposal to invalidate S. zelichi, and I would like to further comment on their contributions. I will avoid providing much new evidence, since this would require a separate publication according to SACC standards. I will try to focus on analyzing what we already know to answer the comments on my proposal and paper, but some minor reference to yet unpublished work is impossible to avoid.


“1) Manuel and Santiago forgot to mention that a white collar is not the only feature separating palustris from zelichi, but also that zelichi has a rufous back. Young zelichi males (n=2) show the adult pattern, including a white collar, even before acquiring the darker rufous colors of the body (a typical maturation pattern in the Capuchinos, Areta unpubl.). Likewise, the bird breeder Ramón Buceta wrote in the work of Narosky (1977:346): "ninguno de los capuchinos cambia de color pasando por fases. Tardan 4 ó 5 aĖos en adquirir en cautividad el plumaje definitivo, pero desde el segundo aĖo comienzan a notarse las características que poseerá el macho adulto. Durante el invierno el colorido decae un tanto. Cuando usa un llamador en una 'trampera, es rarísimo que caiga un Sporophila de distinta especie." Thus, an age-related hypothesis is rejected in zelichi, and in all collared forms (and support is given to the use of voices as key to species taxonomy).


“2) Variation in color intensity is extensive in the Capuchinos (this was known since at least Hellmayręs 1904 work). Hellmayr (1938) and Meyer de Schauensee (1952) adequately grasped the relationship between color intensity and age, and palustris fits this pattern of color intensity variation as well. Neither the type of zelichi, nor the remainder of individuals that I have seen and examined (n=9) approach the intensity of cinnamomea in chestnut coloration; on the contrary, most (if not all) zelichi are well within the variation found in palustris. Likewise, the original description of zelichi reads: “Si comparamos S. zelichi con las especies más afines, observamos que ventralmente no muestra diferencias con palustris, pero dorsalmente posee un típico cuello blanco que no existe en éste. Además la espalda, castaĖo o castaĖo acanelado en zelichi, es netamente gris en la otra especie” (Narosky 1977: 347). The following pictures will illustrate the argument: ZELICHI:, PALUSTRIS;, CINNAMOMEA: Because plumage features were the almost exclusive aspect that brought about the dubious taxonomic status of zelichi, I briefly mentioned, in the beginning of my work, the clue plumage differences between cinnamomea, zelichi and palustris that were relevant to evaluate the systematics of zelichi (Areta 2008:353 “Based on plumage features (white collar, rufous body, and gray cap in S. zelichi; white throat, rufous belly, gray back, nape, and cap in S. palustris; and chestnut body and gray cap in S. cinnamomea), doubts were raised about the validity of this species even in its formal description, where it was considered to be “a hybrid of S. palustris and S. cinnamomea””). Santiago mentioned that “To me, the plumage of zelichi is inconsistent with Nacho's favored hypothesis of a morph of palustris”, and “To me, the plumage is consistent with zelichi being either a valid species or a palustris x cinnamomea hybrid”. It is worth noting that, initially, all the systematic hypotheses that I tested are consistent with the plumage of zelichi, and that is why they were proposed as different explanations by different authors (i.e., one plumage, four possible explanations= a good species, a morph of A, a morph of B, a hybrid [AxB or BxA]). The good species hypothesis is based on the diagnostic plumage pattern of zelichi, the morph hypotheses are based a) on the shared ventral pattern of zelichi with palustris (white throat, “reddish” belly), and b) on the shared pattern of “reddish” color in the back of zelichi and cinnamomea, whereas the hybrid hypotheses argue that the plumage features of zelichi are intermediate between cinnamomea and palustris (see pictures). It remains unclear to me how Santiago judges, a priori, what the meaning of plumages of zelichi is likely to be. To avoid more speculations on the meaning of the plumage of zelichi, I have referred to other sources of information that could help to overcome the limitations of the taxonomic interpretations of plumages in the group: voices and habitat use.


“3) Extrapolating in the Capuchinos can be risky and misleading in several aspects, so I am trying to keep comparisons within the Capuchinos clade rather than within Sporophila. I briefly analyze why we should not extrapolate in three aspects that have to do with the systematics of the Capuchinos:


“HABITAT- Although many Sporophila share habitats (an issue that I acknowledge in my work), the case of the Capuchinos is very interesting, because the males of different species use (when evaluated at a wider scale) different habitats to establish and defend territories. This, of course, should not be taken as a rigid law, and a small portion of any species will be found outside of the species preferred habitat as is widely documented for birds worldwide (this small overlap is also reported in my study). My argument is that palustris and zelichi share a preferred habitat (essentially marshes), and this does not occur in any other valid species of Capuchino that I studied. Thus, the case of palustris and zelichi is not conventional, and requires an explanation.


“MIMICRY- Mimicking Sporophila (e.g., collaris, dark-billed and yellow-billed plumbea, intermedia; with which I have field experience and recordings) are not closely related to Capuchinos, so Santiagoęs argument seems to require, again, an unjustified extrapolation from some Sporophila to the Capuchinos (for example, not all species of Euphonia mimic, but some species are accomplished mimics). Moreover, no Capuchino is known to mimic (n=250 individuals from all known species in the southern Capuchinos, my recordings), and captive birds maintain the same (presumably crystallized) song for many years without any apparent change (n=8). If mimicry were found to be relevant, then I would of course have reported it in my work.


“GENETICS- Other Sporophila have deep genetic divergences, whereas the Capuchinos do not. Thinking along the extrapolation line from Sporophila to Capuchino, no Capuchino in the ruficollis clade would be considered a valid species (because there is no obvious genetic divergence), and so zelichi would not be a valid species, contradicting Santiagoęs ideas of zelichi as a valid species.


“CONCLUSION- Can we directly extrapolate to Capuchinos from what happens in other Sporophila in any other way?  To me, the answer is NO. And if we are to extrapolate, then it should be with extreme caution.


“4) Although I sympathize with Santiagoęs theoretical supposition that “Maybe zelichi males are singing palustris songs to maintain interspecific territoriality with a more abundant species that inhabit the same habitat”, it is my conviction that the systematics of the Capuchinos do not need more speculation, but some good data to avoid them. Solid taxonomic foundations should be based on data and not on speculations. So, although the idea put forward by Santiago could be true (this is the nature of any hypothesis together with its falsifiability), there is no available evidence to support that assertion. Based on Occam’s razor (which I nevertheless use with caution), I find the idea too complicated and unnecessary to explain the observed patterns of habitat use, plumage and voices that I report in my work. When compared to the data, Santiagoęs statement is an ad-hoc hypothesis to save S. zelichi as a species, and such theoretical constructions should be avoided.


“5) The assertion that “Maybe zelichi can sing other songs.” is essentially true but of little value. Despite song-templates, many oscines are able to sing other songs, if they are, let’s say for example, misimprinted. As far as we know with hard evidence, zelichi sings like palustris (3 recordings, 12 zelichi trapped with palustris males calling for which I do not have tape recordings). I wouldn’t be surprised if some males of zelichi appear singing a cinnamomea song. This will not make zelichi a valid species (which I understand is Santiagoęs main point) but might provide some evidence in favor of a) the hybridization hypothesis, or b) in favor of the cinnamomea-morph hypothesis, or even c) show that zelichi can misimprint on cinnamomea song. It is here interesting to note that I reported 3 zelichi singing like palustris, but no palustris (n=34) or cinnamomea (n=24) singing like any other species of Capuchino. Thus, it seems odd to find a presumed good species (zelichi) consistently singing the same song of another one (palustris), whereas cinnamomea (and the rest of the Capuchinos, of which I did not present data in my paper but mention having the data) sing their own songs without any case of misimprinting (which might nevertheless occur, albeit rarely).


“6) Although I am deeply concerned with the conservation of the Capuchinos, there is no evidence to maintain zelichi as a valid species. Conservation policies should not interfere with a taxonomic judgment; otherwise splitting would be the norm. Moreover, to conserve zelichi, the habitat of palustris should be protected, and trapping prohibitions enforced. Thus, it seems more relevant to dedicate attention to palustris, which will also grant zelichi protection, instead of dedicating attention to zelichi while palustris keeps declining. Despite this, I do not argue that zelichi must be eliminated TO improve the conservation of palustris. Because I recognize that this is not a taxonomic statement, I do not see why the inverse should be true (i.e., to keep zelichi for conservation reasons is not a taxonomically informative statement).


“7) The question of whether zelichi is a hybrid or a color morph remains unsettled (although I advocated for the morph hypothesis based on several congruent lines of evidence). I fail to see any valid reason to keep zelichi as a good species. Also, in Areta (2008:359) we read “These findings are paralleled by the existence of an undescribed black-collared and rusty-backed seedeater that is indistinguishable from S. ruficollis in vocalizations and habitat (Areta, unpubl data).” I would like to pose the question of whether modern systematists and taxonomists would accept a description of a bird with the following features as a valid species: 1) plumage like that of zelichi but with black-collar (i.e., the black-collared form is to ruficollis as zelichi is to palustris), 2) the bird sings exactly like ruficollis, 3) the bird shares a preferred habitat with ruficollis, 4) the bird has an extremely low density, while ruficollis is common in the same habitat. I doubt that such a species would gain acceptance by the standards of modern taxonomy and systematics. However, this case is the exact systematic and taxonomic parallel of the case of zelichi. With at least 250 Capuchinos tape-recorded, I have been unable to find any species-specific voice of either zelichi or this dark-collared form.


“8) Finally, the finding of a single individual of zelichi that Daniel mentions does not constitute the finding of a new population of these birds. It is interesting that, to date, there is no single area where a “population” (understood as several breeding territories in a circumscribed area) of zelichi can be found, whereas even the rarest and certainly valid Sporophila seedeater (or even Capuchino, to avoid risky extrapolations) is common when the habitat is the adequate (e.g., Sporophila nigrorufa). It would be interesting to know if the model mentioned by Daniel is able to predict the presence of zelichi without predicting at the same time the presence of palustris. In Areta (2008: 359) I wrote “Both S. zelichi and S. palustris were always found in marshes or wet grasslands and the presence of S. palustris predicts the presence of S. zelichi better than any other species”.


“RECOMMENDATION: In my view, the number of exceptions and ad hoc hypotheses that need to be accepted to keep zelichi as a good species is overwhelming. Other than Narosky’s original description, I do not see any argument (based on hard data) that could argue for zelichi as a valid species. I generally support cautionary approaches, but I do not support unfounded decisions. Keeping S. zelichi as a valid species seems an unfounded decision. Thus, I keep proposing a YES vote to this proposal.”


Comments from Remsen: “YES.  I find Areta’s arguments convincing that the burden-of-proof now rests with those who would maintain zelichi as a valid species.”

Comments from Stotz: “Given that the original describer of zelichi expressed his own doubts about this "species" and that Areta has shown through a variety of means that zelichi falls within palustris in a suite of characteristics including habitat and song, I think it is impossible to maintain as a distinct species.  Areta's suggestion of it as a color morph of palustris seems like the best fit for the data.”

Comments from Jaramillo: “YES – I have looked for this creature and have never seen it, although I am familiar with most of the other Capuchinos. They are a bewildering group, and certainly one that needs much more careful work like that of Nacho. Song does seem to be a key feature for this very confusing group, and I think that it makes sense to follow Nacho’s lead on this. No one has studied this taxon in such a detailed manner, in evaluating various possible alternatives for what zelichi is in a rigorous fashion. The burden of proof is on those who want to retain zelichi as a species, but Nacho’s paper certainly clarifies many questions I have had about this group and it makes sense on various levels. This is indeed why zelichi, since its description, had an element of doubt associated with it. It just does not “act” like a good species, and the available data suggests strongly it is a morph, although there is room for it to be of hybrid origin – either way it does not seem to be a valid species. [Entirely off the discussion, I met Manuel Nores looking at my first Sporophila cinnamomea at El Palmar National Park!].

Comments from Pacheco: "YES.  Sou de opiničo que o estudo do Nacho é convincente e satisfatório. Os argumentos extras expostos aqui pelo autor sedimentaram ainda mais essa opiničo.”