Proposal (932) to South American Classification Committee

Merge Sporophila murallae into S. americana

Effect on SACC classification: This merger would remove a species from the roster and expand the distribution of the resulting redefined Sporophila americana west across Amazonia to the Andean foothills.

Background: Proposal 287, which split Sporophila americana into three species and additionally suggested that S. intermedia was inside this complex, was based on the work published in Stiles (1996). That publication was largely based on fieldwork and museum work with a distinctly Colombian focus, understandably, as that country houses most of the potential nearest points of contact among the various members of the complex. Stiles (1996) used male plumage characters and morphometrics as the main basis of comparison among the various taxa involved, demonstrating that all members of the complex showed a certain amount of variation in the former. In the proposal, the decision to split the cis-Andean populations into two species—Sporophila americana of the Atlantic coast and eastern Amazon and S. murallae of western Amazonia—hinged on some minor plumage and morphometric differences, and (more importantly) on the apparent 500 km gap in the distributions of the two forms in Amazonia. Furthermore, Stiles (1996) demonstrated that previous claims that S. murallae was generally darker than S. americana because of some geneflow (current or historic) across the Andes with Pacific populations of S. corvina were flawed and this also supported the 3-way split of the S. americana complex. This decision was accepted by the SACC members and has become the present taxonomic status of these populations.

In the past decade or so, sightings of male Sporophila belonging to either S. americana or murallae have been increasing in central and southwestern Amazonia, and online repositories such as eBird and WikiAves have gained extensive photo galleries of individuals of this complex from areas that are outside of their “traditional” areas of distribution (as defined by e.g., Ridgely and Tudor 1989, 2009; Jaramillo 2011a,b). The identification of these individuals, I would guess, has been based on locality (e.g., which core population is closer) and plumage characters such as number of wing bars and completeness of collar. However, even within “traditional” areas of distribution, it is clear that both forms, but perhaps especially murallae, show a fair amount of variation—enough to show broad overlap in these characters between the two taxa.

Stiles (1996) was hampered by relatively small specimen samples of both Sporophila americana (sensu stricto, including subspecies dispar, with type locality Santarem, Amazonas, Brazil) and S. murallae, which weakens the punch of the morphometric distinctions that were among his reasons to suggest the two be separated as species. But more importantly, the supposed “500 km gap” between the two taxa seems to have been closed either by increasing fieldwork in the intervening areas, or perhaps a recent expansion of one or both of the original populations in response to anthropogenic land clearance in Amazonia. Regardless, the result has been that there appears not to be any perceivable gap between the western Amazonian S. murallae and the eastern S. americana:

WikiAves map showing combined distributions within Brazil of birds identified as S. americana and S. murallae:

Thus, one of the main arguments for the maintenance of the two forms as species, a 500km gap in their distributions, is no longer relevant.

Following up on the point of variation in plumages, Figure 6 of Stiles (1996) shows the following images of male plumages of the two forms:

… Suggesting that S. americana rarely, if ever, shows a complete collar, and S. murallae rarely doesn’t and rarely has two well-defined wingbars. However, the following photographs taken from the “traditional” areas of distribution of the two forms should show that there is indeed wider variation in the male plumages of both forms such that there is extensive overlap in these characters. Indeed, many of the reports of members of this complex from outside their respective traditional distributions have been plagued by uncertainty in how to distinguish the two species satisfactorily (this is not helped by the illustrations that appear in the Handbook of Birds of the World showing a single example of each taxon from the extremes of the examples in the above figure).

The following links are to Macaulay images of birds from Amazonian Colombia and Ecuador, which I think should safely represent “true” S. murallae:
https://macaulaylibrary.org/asset/204996841 <note two bold wingbars>

https://macaulaylibrary.org/asset/104502771 <note incomplete collar>

https://macaulaylibrary.org/asset/163570831 <note incomplete collar>

https://macaulaylibrary.org/asset/163987151 <note incomplete collar>

https://macaulaylibrary.org/asset/373313041 <note incomplete collar and double wingbars>

And the following links are of “true” S. americana from the Guianas:

https://macaulaylibrary.org/asset/155169561 <note complete collar>

https://macaulaylibrary.org/asset/275841171 <note complete collar>

https://macaulaylibrary.org/asset/254234081 <note complete collar>

No doubt additional delving into the photo galleries of eBird and WikiAves would produce more individuals showing characters not typical of their respective taxa as illustrated in Stiles’ Figure 6 above, but I think the idea should be clear from the above selection.

Analysis and Recommendation:

It now seems to us that we are dealing with a single population that tends to be darker at its western end (in northwestern Amazonia) and paler at its eastern end (on the Atlantic coast of northeastern South America) in a manner that follows Gloger’s Rule to some extent, but demonstrating extensive within-population variation that shows extensive overlap in characters. As such, we think it is perhaps best that we consider murallae to be a subspecies under S. americana at best, and perhaps it may not warrant taxonomic recognition at all if these morphologic characters show such broad overlap among populations across the range of the cis-Andean portion of the Sporophila americana complex. From what we can tell, there do not appear to be any great voice distinctions between S. murallae and S. americana (taking into account dialect forming), but there are so few examples of the latter available that the sample sizes are simply too small to do any useful analyses..

So in conclusion, we believe that the recommendation in Proposal 287 to recognize Sporophila murallae as a separate species from S. americana was premature and increased information available to us today would suggest that the two are not strongly differentiated in plumage patterns and there is no gap in their distributions across Amazonia. Thus we recommend a YES to lumping murallae back into Sporophila americana.

Literature cited:

JARAMILLO, A. 2011a. Caqueta Seedeater. P. 650 in “Handbook of the Birds of the World, Vol. 15. Tanagers to New World Blackbirds." (J. del Hoyo, A. Elliot, and D. A. Christie, eds.). Lynx Edicions, Barcelona.

JARAMILLO, A. 2011b. Wing-barred Seedeater. P. 650 in “Handbook of the Birds of the World, Vol. 15. Tanagers to New World Blackbirds." (J. del Hoyo, A. Elliot, and D. A. Christie, eds.). Lynx Edicions, Barcelona.

RIDGELY, R. S., AND G. TUDOR. 1989. The birds of South America, Vol. 1. University Texas Press, Austin.

RIDGELY, R. S., AND G. TUDOR. 2009. Field guide to the songbirds of South America. University Texas Press, Austin.

STILES, F. G. 1996. When black plus white equals gray: the nature of variation in the Variable Seedeater complex (Emberizinae: Sporophila). Ornitologia Neotropical 7: 75-108.

Dan Lane and Gary Stiles, February 2022

Comments from Remsen: “YES. New data require a change in taxonomy. Burden of proof is clearly on a two species treatment, for which the original evidence was somewhat weak anyway.”

Comments from Robbins: “YES for subsuming Sporophila murallae into S. americana based on information provided by Dan Lane and Gary Stiles (who originally proposed recognizing murallae as a species).

Comments from Areta: “YES to treating murallae as conspecific with americana. I´ve always found this split to be based on very slim evidence.

“It is difficult to understand where the main differences based on bill shape (longer and shallower bills in murallae than in americana) come from in Stiles (1996), because there is not a list of measured specimens and no measurements for americana are listed in Table 5 (if I understand correctly, 7 americana were studied by Gary in the Naturhistorisches Museum in Vienna and 5 additional specimens were measured by Manuel Marín at LSU; I wonder whether some of the effect on bill shape has to do with different people measuring them?. americana has the lowest sample size in this study). It might be that overall, the NW murallae have slimmer bills, but this is not strikingly evident when looking at pictures across the range of murallae (some look distinctly thin-billed [for a Sporophila], others look thick-billed). Even so, there may also be clinal variation in bill shape (or not, we don´t know), and a modest difference in bill shape does not seem enough for a split.

“There are apparently no diagnostic plumage features of murallae in comparison to americana, even if they differ "on average" in some features. Because plumage variation is high and we do not know enough on the sequence of plumages in these forms, some (or a lot) of the variation might be age-related, and care must be exercised when interpreting plumage variation.

“The vocalizations are complex, and no one has performed a rigorous analysis, but it is clear that both share the same singing style.

“The gap between murallae and americana has been closed (or never existed). I wonder whether the continuous distribution has something to do with seasonal movements or wandering (which would affect the interpretations on the geographic distribution of the phenotypes), in addition to the species breeding across this range.

“Although we are not dealing with subspecies at present, it might be good to think on whether murallae deserves treatment at the subspecific level or not. It is part of the lumping deal. My preference would be to keep it murallae a subspecies, for the time being. Although there are many unanswered questions, I find the current evidence to be more consistent with the existence of a single species.”

Comments from Pacheco: “YES. Faced with such variation and overlapping of plumage characters, I agree with the treatment of Sporophila murallae as a subspecies of S. americana. Perhaps, as warned by Nacho, murallae will prove not to be even a valid subspecies.”

Comments from Bonaccorso: “NO. I think the information provided is not sufficient. The Wiki Aves map is not informative. Even if there is no gap, it would be important to relate the points to the different plumage colors. Yes, differences could be of the Gloger´s rule type, but I think the proposal is based on anecdotical information (sorry Dan and Gary). I would like to see extensive hard data on morphology, song, and genetics to lump this two (which I agree, are probably the same species, but I think we need to make very informed decisions when it comes to lumping taxa).”

Comments from Claramunt: “YES. I think it was premature to make this split. Evidence for species status is not clear.”

Comments from Zimmer: “YES. I voted to split these on the original Proposal 287, but we now have a lot more data to fill in the previous gaps (we didn’t know what we didn’t know). This does look like a case of significant intraspecific plumage variation, possibly clinal in nature, to me, and does call into question whether murallae should be formally recognized at any taxonomic level. However, until we get a better grasp on the nature of the apparent cline, the conservative thing to do would be to maintain murallae as a upper Amazonian subspecies of americana.”

Additional comments from Stiles: “At best, murallae could be considered as a subspecies of americana (but even this might be unnecessary).”

Comments from Jaramillo: “YES – I am a bit hesitant though. Perhaps it is because given that we have a continuous range, it would be great to have more information on the change from one morphology to the other. Is it a step cline? Is it a continuous gradation? That would be nice. But if that information comes along in the future and we need to reassess, that is ok too.”