Proposal (717) to South American Classification Committee


Recognize the new genus Mazaria for “Synallaxis” propinqua


“The best definition of a genus seems to be one based on the honest admission of the subjective nature of this unit…” (Mayr 1999:283)


Background: Synallaxis propinqua has been always included in the genus Synallaxis, classification that nobody has questioned since the morphology and habits of this species seem typical of members of this genus. However, in the comprehensive molecular phylogeny of Derryberry et al. (2011), propinqua appears as sister to Schoeniophylax phryganophilus, and together they form the sister group of a clade including Certhiaxis and other Synallaxis. Lack of statistical support for the relevant nodes prevented making taxonomic rearrangements before the surprising relationship could be confirmed.


New Information: I recently revisited the problem with and enhanced molecular dataset including: 1) the mitochondrial ND2 gene (already analyzed by Derryberry et al. 2011), 2) two introns from the Z chromosome (ACO1 and MUSK, analyzed together); 3) introns 5-11 of the G3PDH gene (and intervening exonic sequences); and 4) introns 5 and 7 of the beta fibrinogen gene FGB (Claramunt 2014).


When analyzed separately, the four genomic regions independently showed a sister relationship between propinqua and Schoeniophylax (bootstrap support: G3PDH: 51, ND2: 86, Z-linked: 96, FGB: 100). There was less agreement regarding the position of this pair of species within the larger clade, but G3PDh and Z-linked genes showed propinqua and phryganophilus sister to Certhiaxis with strong bootstrap support (94 and 99 respectively), whereas ND2 showed propinqua and phryganophilus sister to Synallaxis but with low support (73), and FGB did not resolve basal relationships.  A sister relationship between the pair propinqua-phryganophilus and Certhiaxis was also found when genes were analyzed jointly either by concatenation (ML bootstrap: 92) and using a species-tree (STAR) method that accounts for potential incomplete lineage sorting.


Given these results, the classification needs to be changed. Among several options, I opted for describing a new genus for propinqua, which I named Mazaria, honoring our dear Juan Mazar-Barnett.





Analysis and Recommendation: Although a sister relationship between propinqua and phryganophilus was unexpected given the phenotypic differences between these two birds and the similarities between the former and other Synallaxis, the molecular-phylogenetic evidence showing this relationship is overwhelming: all four genomic regions independently corroborated the sister relationship between propinqua and phryganophilus. Plumage similarities between propinqua and other Synallaxis species must be ancestral characteristics compared to the more-derived morphology of phryganophilus. However, propinqua has more attenuated (pointed) rectrices compared to other Synallaxis and, regarding vocalizations, shares with phryganophilus the inclusion of low-pitched guttural rattles as part of some of their vocalizations (something that was pointed to me long ago by Brian O’Shea and Luciano Naka). The divergence between propinqua and phryganophilus is a relatively old event: between 7 and 10 million years ago, using biogeographic calibrations (Derryberry et al. 2011), and between 4 and 12 million years ago using a mitochondrial clock (Claramunt 2014). This explains the accumulation of phenotypic differences between these sister species.


There are three ways in which these relationships can be represented in a revised classification:


A: Merge Certhiaxis and Schoeniophylax into an expanded Synallaxis.

B: Place propinqua in the genus Schoeniophylax.

C: Place propinqua in its own genus Mazaria, the option advocated in this proposal.


Option A is the least convenient, in my opinion. Synallaxis is already very diverse and heterogeneous (it already includes the former genera Siptornopsis and Gyalophylax, for example); inclusion of Certhiaxis and Schoeniophylax within Synallaxis will increase this heterogeneity even more. Schoeniophylax was previously merged within Synallaxis by Vaurie (1980), but this treatment did not gain general acceptance. Moreover, merging Certhiaxis into Synallaxis creates further nomenclatorial problems as the Yellow-chinned Spinetail, S. cinnamomeus (Gmelin 1788), would become homonym with the Stripe-breasted Spinetail S. cinnamomea Lafresnaye 1843.


Option B. Transferring propinqua from Synallaxis to Schoeniophylax is less disrupting than A regarding nomenclatorial changes, although the White-bellied Spinetail would become Schoeniophylax propinquus, to match the masculine genus (David & Gosselin 2002). The drawback of this option is that the resultant genus would combine two species that are phenotypically very distinct, with no other transitional species that fills the phenotypic gap between them. The resultant genus would be difficult to characterize other than listing the sum of characteristics of both species. Also note that these sister species are not particularly closely related according to divergence time estimates. On the other hand, note that the argument regarding phenotypic differences could be turned around in support of Option B by arguing that the new taxonomy would help showing a relationship that is otherwise difficult to recognize.


Option C, is minimally disrupting other than introducing a new generic name. The disadvantage of Option C is the creation of a monotypic genus that is not informative regarding relationships and redundant in the sense that the genus Mazaria and the species Mazaria propinqua would contain exactly the same taxon. On the other hand, monotypic taxa cannot be completely avoided in general, and in this case, the monotypic genus would help represent the phenotypic distinction of propinqua in relation to its sister species and highlight that theses two species are not particularly closely and have long history of independent evolution.


At the end, I think that Option C would result in a classification that is more in consonance with traditional conceptualizations of what an avian genus is (Mayr 1999) and therefore I recommend recognizing the new genus Mazaria for propinqua (a YES on this proposal).



Literature Cited:


Claramunt, S. 2014. Phylogenetic relationships among Synallaxini spinetails (Aves: Furnariidae) reveal a new biogeographic pattern across the Amazon and Parana river basins. Molecular Phylogenetics and Evolution 78:223–231.

David, N. & M. Gosselin 2002. The grammatical gender of avian genera. Bull. Brit. Orn. Cl. 122: 257-282.

Derryberry, E. P., S. Claramunt, R. T. Chesser, J. V. Remsen Jr., J. Cracraft, A. Aleixo, & R. T. Brumfield. 2011. Lineage diversification and morphological evolution in a large-scale continental radiation: the Neotropical ovenbirds and woodcreepers (Aves: Furnariidae). Evolution 65(10):2973-2986.

Mayr, E. 1999. Systematics and the origin of species from the viewpoint of a zoologist. Harvard University Press, Cambridge, MA.

Vaurie, C., 1980. Taxonomy and geographical distribution of the Furnariidae (Aves, Passeriformes). Bull. Am. Mus. Nat. Hist. 166, 1–357.






Comments from Remsen: “YES.  I have been aware of this one for several years and strongly concur with Santiago on all of this.  The classification must be changed, and Option C is clearly the best, in my view.”


Comments from Areta: “YES. In the most Mayrian sense of a genus, I endorse the placement of propinqua in Mazaria. Vocally and phenotypically it does sound-look like a Schoeniophylax-Certhiaxis. However, my subjective heart is pleased to see this new genus dedicated to the memory of Juancito.”


Comments from Bret Whitney: “I much prefer Option B, in this case, for the primary reason Santiago recognized: one genus for these two clarifies and highlights their sister relationship within this huge family.  That information would be essentially obscured by assigning propinqua a new genus, especially because I don’t fully agree that phryganophilus and propinqua are “phenotypically very distinct”.  Their plumages and tail morphologies are fairly different, and the head/neck plumage pattern of phryganophilus is practically unique in the Furnariidae (approached, however, by Poecilurus, which got lumped into Synallaxis).  Similarly, Siptornopsis is now considered a Synallaxis, together with sister stictothorax (= Option A) — but I’d certainly prefer to see the two distinctive taxa in that clade in a genus apart (Siptornopsis), in the same manner as phryganophilus and propinqua (propinquus) would be best, I think, comprising Schoeniophylax.


“Phenotypic distinctiveness of phryganophilus and propinqua comes down a couple of notches when vocalizations are considered.  They uniquely share a harsh, grating quality in songs and calls that is practically unique in the part of the phylogeny presented in this proposal, and they both occasionally perform fairly complex duets.  Some of the others in this part of the tree might be considered to give duets, in that the members of the pair sometimes vocalize in tandem (e.g., Certhiaxis, also somewhat “harsh and grating"), but they are not two-parted, synchronized vocalizations like those occasionally delivered by pairs of Schoeniophylax.  (I’ve never seen “Synallaxis" hypochondriaca in life, and it’s been so long since I’ve seen stictothorax that I can’t remember much about its vocalizations — so I don’t know about possible dueting in that clade.)”


“The erection of a monotypic genus should almost always be considered quite disruptive, just as is the lumping of phenotypically distinctive clades/species traditionally considered separate genera into related genera, always in the pursuit of avoiding paraphyly."


Comments from Stiles: “YES. I definitely prefer diagnosable genera, even if monotypic; the only real alternative, including both in Certhiaxis, produces an undiagnosable soup.”


Comments from Pacheco: “YES.  I prefer C, the option that keeps a distinctive genus for this taxon.”


Comments from Robbins: “NO.  I support option B, agreeing with comments made by Bret.  Moreover, the continued movement for making anything that looks different and/or has a relatively long branch in the tree as a monotypic genus is undermining the purpose of nomenclature, i.e., effective communication and conveying relationships.”


Comments from Cadena: “NO. I largely agree with Bret, and embracing the subjectivity noted by Mayr I subjectively prefer classifications that are informative about relationships over the recognition of monotypic genera except when dealing with real oddballs. Subjectively, admittedly, I think this is not the case here.”


Comments from Jaramillo: “YES – Inclusion into Schoeniophylax is not palatable to me, that is such a distinctive bird. Given that this pair is not all that closely related to each other, I am ok with the creation of a monotypic genus and happy that it is called Mazaria.”


Additional comments from Remsen: “I am a little surprised that this proposal met any opposition, so I am adding some extra comments.  Including propinqua in the same genus with Schoeniophylax phryganophilus creates an indefensible morphological grouping in addition to the two lineages being separated as long ago as many current genera.  In my opinion, these two species basically bear no resemblance to one another in plumage or morphology (in contrast to the Siptornopsis group) other than sharing generalized synallaxines features.


“As for voice, here’s a Dan Lane recording of Schoeniophylax phryganophilus:  Here is a Lane recording of propinqua:, and one by Mitch Lysinger:  Although I would agree with Bret that they share some similarities with respect to other Synallaxis that in retrospect are consistent with a sister relationship (length and pace), I also would not describe phryganophilus song as harsh and grating, but rather, in my subjective opinion, amazingly rich and melodious, at least compared to any other spinetail.”


Comments from Zimmer: “YES” on C.  I’ve waffled back and forth on this one, sharing some of the concerns raised by Bret, Mark and Daniel regarding erecting a monotypic genus for a bird that is not that distinctive phenotypically or vocally from other members of the larger clade.  But Schoeniophylax really is, to my thinking, a unique bird, and lumping propinqua into that genus results in a dilution of clarity regarding the uniqueness of Schoeniophylax that, in my opinion, outweighs any informative gains of recognizing the sister status of the two species by placing them in the same genus.  I think Van is spot-on in his comments regarding the alleged vocal similarities of Schoeniophylax and propinqua.  I too, find the song of Schoeniophylax to be “rich and melodious”, having a liquid, gurgling tonal quality that is difficult to describe, and, which is unique within the Synallaxines.  Some of the abbreviated contact-type or agonistic vocalizations of Schoeniophylax might be considered “harsh or grating” but no more so than are the contact vocalizations and agonistic vocalizations of several species of Synallaxis.  To my ears, the harsh, grating vocalizations of propinqua are much closer in quality to those of Certhiaxis cinnamomea and C. species novum (from the Araguaia Basin) than they are to Schoeniophylax.  And, to follow up on Bret’s musings regarding Siptornopsis vocalizations, my experience with the former Siptornopsis species (stictothorax and hypochondriaca) is that both species routinely duet, as does Certhiaxis.  So neither the harsh, grating vocalizations of propinqua, nor the fact that they are given in duet, is enough in my mind to justify placing them with Schoeniophylax.  That leaves morphological characters, and, as Van points out, any morphological similarities between propinqua and Schoeniophylax are really just generalized synallaxine features.  Transferring propinqua to Schoeniophylax would result in a tiny, amorphous genus that would defy any attempts to produce a coherent diagnosis, solely to make clear that the two species are sister taxa.”