Proposal (325) to South American Classification Committee

Recognize Formicivora grantsaui as a valid species

Effect on South American Check-List: This proposal would add a newly described species to our official list.

Background: Gonzaga, Carvalhaes and Buzzetti (2007) recently described a new species of antwren, Formicivora grantsaui (Sincorá Antwren), from Bahia, Brazil. The new species is known only from the campo rupestre vegetation of the Serra do Sincorá between 850-1100 m in the Chapada Diamantina, which is part of the Espinhaço Range, a noted area of endemism for birds, and particularly for plants. Morphology, vocalizations and ecology of the new species leave no doubt as to its generic assignment to Formicivora. Both the male and female plumage of grantsaui are similar in most respects to the respective sexes of F. rufa, and this similarity is likely the reason for grantsaui having been previously overlooked. In fact, subsequent to the recent fieldwork revealing the distinctiveness of the newly described taxon, an earlier (1965) specimen was located (Grantsau 1967). In places, the new species occurs sympatrically with Formicivora rufa, but the two occupy different habitats. Diagnosis of the new species is supported by 10 specimens (7 males, 3 females) and tape-recorded voices of at least 8 individuals.

Analysis: SACC currently recognizes 7 species (iheringi, erythronotos, grisea, serrana, littoralis, melanogaster, rufa) in the genus Formicivora, and an additional species, acutirostris, in the monotypic genus Stymphalornis. This taxonomy was followed by Zimmer & Isler (2003), although we cited a PhD dissertation by Luis Gonzaga (2001) that provided convincing evidence that the status quo taxonomy was incorrect. In that dissertation, Gonzaga presented an extensive phylogenetic analysis of the genus Formicivora and its possible relatives, based on a combination of external morphology, syringeal morphology, vocalizations and natural history characters. Results of that analysis showed that monophyly of Formicivora was supported only if Stymphalornis acutirostris was included, and F. iheringi was excluded from the assemblage. The species acutirostris, grisea and rufa were found to form a clade, sister to a clade consisting of melanogaster, serrana and littoralis. [As an aside, I believe that molecular work by John Bates and others has corroborated the rufa-grisea relationship proposed by Gonzaga, but I don't have the citation at my fingertips.]

Returning to the question of grantsaui, Gonzaga et al. (2007) do not really address the question of generic diagnosis, but I would agree unequivocally with their allocation of grantsaui to Formicivora. Gonzaga probably knows more about Formicivora than anyone, and the morphological characters (both biometric and plumage), vocal characters, and open-country habitat of grantsaui all mesh well with established patterns for the genus as a whole, and particularly for the grisea-rufa-acutirostris clade.

Diagnosis (Gonzaga et al. 2007) as a distinct species was based on a combination of morphological, vocal and ecological characters. The 10 specimens of grantsaui were compared to those of erythronotus (7), littoralis (22), melanogaster (38), serrana (26), acutirostris (7), grisea (62) and rufa (65). Comparative material included a representative geographic spread for virtually all species. The relatively small number of erythronotus and acutirostris specimens examined actually represents an impressive sample considering the rarity of these two species in collections (one species only recently described, the other only recently rediscovered). In summary: Females of grantsaui differ from those of littoralis, melanogaster, and serrana by lacking a black facial mask, and by having the sides of the head, throat and breast whitish with black streaks (versus no streaks); from acutirostris and erythronotus by possessing 12 rectrices (versus 10 in the former and 10 or 12 in the latter) with large (versus absent or narrower) white tips. Adult males of grantsaui differ from those of the aforementioned species in having a white eyestripe (versus no stripe, or, stripe present, but limited to postocular) continuing and broadening down the sides of the head, throat and breast, and separating the black sides of the head and underparts from the browner upperparts. Both sexes of the new species differ from those of F. rufa by having the flanks brown (versus yellowish-buff); the upperparts a darker, gray-brown (not rufescent); and, by having the underwing coverts dark gray and white (versus entirely white). The new species differs from all taxa in the F. grisea complex in being longer-tailed. F. grisea, as currently constituted, consists of eight recognized subspecies (Zimmer & Isler 2003), some of which, almost certainly warrant recognition as distinct species (that being a topic for future proposals). Of these eight subspecies, only orenocensis has a female plumage with extensive, bold, dark streaking on the underparts (females of most taxa in the grisea complex being either unmarked below, or very lightly spotted or streaked) like that of grantsaui, and it differs in having white flanks (brown in grantsaui) and extensively white outer rectrices (versus outer rectrices just edged white in grantsaui).

Vocal characters provide even stronger evidence of the distinctiveness of grantsaui. Gonzaga et al. (2007) analyzed vocalizations of all species in Formicivora (including acutirostris; excluding iheringi) and detail a two-part alarm call formed by more than two (rarely only two) notes and a short, distinctively modulated territorial (duet) call, both of which distinguish grantsaui from all other species in the genus. A similar call-type to the first call was used by Gonzaga (2001) as one of the characters defining the grisea-rufa-acutirostris clade, and a homologous call to the second call type has been recorded from both grisea and rufa. Gonzaga et al. (2007) also demonstrate that grantsaui differs strikingly from rufa in the pace of the loudsong, which is much slower in grantsaui (ca. 2 notes/sec versus ca. 14 notes/sec). All of the vocal characters used to diagnose grantsaui are illustrated spectrographically and compared with corresponding call types of rufa, grisea, and acutirostris. An example of the loudsong of grantsaui is available (and can be compared directly with the loudsongs of rufa and other members of the genus) on a commercial CD set (Isler & Whitney 2002), where it is listed as "Example 2 of Formicivora rufa".

Finally, grantsaui and rufa appear to be locally sympatric (occurring in some cases within potential hearing distance of one another) but separated by habitat differences, with grantsaui inhabiting campo rupestre vegetation on the rocky outcrops above 800 m elevation on the slopes of stream valleys or high plateaus and at exposed ridges, and rufa occupying savanna or cerrado habitats in flatter valleys with lateritic soils below (Gonzaga et al. 2007). Similar ecological segregation exists between F. rufa and F. grisea when they occur in sympatry (Sick 1955, Silva et al. 1997, Gonzaga 2001)

Recommendation: I recommend a "YES" vote on accepting this distinctive antwren as a new species to our list. Although both the male and female plumages are similar to the corresponding plumages of F. rufa, the degree of difference is at least equal to, or exceeds that, found between other pairs of taxa in the same genus that are currently accepted as representing separate species (e.g. serrana interposita/littoralis, or melanogaster/serrana). More importantly, vocalizations of grantsaui are unlike anything given by rufa throughout its rather wide range. My personal field experience with the taxa involved extends to all 7 species of Formicivora currently recognized by the SACC, plus acutirostris and grantsaui, and I have no doubts concerning the distinctiveness of the latter. In fact, on 29 January 1997, just 27 days after Buzzetti and Carvalhaes made their initial field discovery of grantsaui near Lençois, Andy Whittaker and I independently discovered a family group (2 adults, 2 fledged juveniles) of Formicivora below Morro do Pai Inácio (same region) that we immediately recognized as different from typical rufa on the basis of vocalizations, dorsal color, and flank color. I sent my tape recordings of those birds to Mort and Phyllis Isler, who compared them with their more extensive archive of Formicivora recordings, and they could find no match. It was not until 1-2 years later that we became aware that Gonzaga et al. were already working on the complex. Since then, I have tape-recorded and videotaped multiple grantsaui in the Lençois region, and have found the diagnostic vocal and morphological characters described by Gonzaga et al. (2007) to be consistent.

Note: My recommendation would be to follow the suggestion of the authors, and use "Sincora Antwren" as the English name for Formicivora grantsaui. The name is informative in describing the localized range of the species, and the morphological characters that diagnose it are too subtle (relative to congeners) to make for a good descriptive name.

Literature cited:

GONZAGA, L. P. 2001. Análise Filogenética do Gênero Formicivora Swainson 1825 (Aves: Passeriformes: Thamnophilidae) Baseada em Caracteres Morfológicos e Vocais. PhD dissertation. Universidade Federal do Rio de Janeiro, Rio de Janeiro, xv + 215 pp.

GONZAGA, P. D., A. M. P. CARVALHAES & D. R. C. BUZZETTI. 2007. A new species of Formicivora antwren from the Chapada Diamantina, eastern Brazil (Aves: Passeriformes: Thamnophilidae). Zootaxa 1473:25-44.

GRANTSAU, R. 1967. Sôbre o gênero Augastes, com a descrição de uma subespécie nova (Aves, Trochilidae). Papéis Avulsos de Zoologia de São Paulo, 21(3): 21-31.

ISLER, P. R., & B. M. WHITNEY. 2002. Songs of the Antbirds. Thamnophilidae, Formicariidae and Conopophagidae. Macaulay Library of Natural Sounds, Ithaca, New York, audio CD.

SICK, H. 1955. O aspecto fitofisionômico da paisagem do médio rio das Mortes, Mato Grosso, e a avifauna da região. Arquivos do Museu Nacional, 42 (part 2), 541-576.

SILVA, J. M. C., D. C. OREN, J. C. ROMA & L. M. P. HENRIQUES. 1997. Composition and distribution patterns of the avifauna of an Amazonian upland savanna, Amapá, Brazil. Ornithological Monographs 48:743-762.

ZIMMER, K. J., & M. L. ISLER. 2003. Family Thamnophilidae (Typical Antbirds). In: del Hoyo, J., A. Elliott & D. A. Christie (Eds.), Handbook of the Birds of the World, Vol. 8: Broadbills to Tapaculos. Lynx Edicions, Barcelona, pp. 448-681.

Kevin J. Zimmer, December 2007

______________________________________________________________________________________________________

Comments from Stiles: "YES. The evidence for recognition of this species seems solid, especially considering that grantsaui and rufa have been found to be sympatric but occupying different habitats."

Comments from Stotz: "YES. Appears to be a straightforward split."

Comments from Pacheco: "YES. Um enfático sim! A descrição foi magistralmente realizada. Aproveito para registrar que o naturalista alemão Rolf Grantsau, radicado há muito no Brasil, completou recentemente 80 anos."

Comments from Nores: "YES aunque no muy convencido; pienso más en una subespecie que en una especie diferente. Sin embargo, la diferencia en el canto, mencionada por los autores y la simpatría (aunque en distintos hábitats) parece indicar de qué se trata de especies diferentes."