Spix's Macaw Cyanopsitta spixii Scientific name definitions

Christopher J. Sharpe, Guy M. Kirwan, Nigel Collar, and Peter F. D. Boesman
Version: 2.1 — Published June 27, 2022



The species is now restricted to a small area of the lower Rio São Francisco basin, close to the town of Curaçá in northern Bahia, east-central Brazil (13, 14). Historically, its range might have been rather wider, with reports embracing an area as large as 300,000 km² according to Collar et al. (13), although not all of these are believable and even some of the more confident of these sightings cannot be considered confirmed (see Historical Changes to the Distribution). Wild populations in all of these areas, including northwest Bahia, are now considered extinct. The last known wild individual near Curaçá disappeared in 2000 (15), but for another 20 years occasional local reports and unexplored potential habitat provided some hope for its continued survival (16). The first releases in a long-planned reintroduction project, based on the now-substantial captive population (14), occurred in June 2022.

Historical Changes to the Distribution

As noted by Collar et al. (13) many other localities have been claimed for the species, by both local people and ornithologists, also encompassing parts of southern Pernambuco (see 17, 18, F. B. Pontual in 13), southern Piauí (7, 17, 19, 20), southern Maranhão (21, 22), and (what is now) eastern Tocantins (23, 24, 22), but from all of these claims few stand up to definite scrutiny, with many of them being from the Cerrado biome, where there is no real evidence that Spix's Macaw ever occurred.

Occurrence in southern Pernambuco is based exclusively on local reports (17, 18), but cover a region adjacent to known sites in Bahia and appropriate habitat (caraiba woodland) existed there formerly; the last claimed sighting was of a pair along the Rio Brígida in 1988 (F. B. Pontual in 13).

However, Reiser (25; see also 3) unquestionably observed the species around Lago de Parnaguá, in southern Piauí, in June 1903, but it could not be found there in 1958 (26), and elsewhere in the same state there are reasonably convincing (but still uncertain) reports from Serra da Capivara National Park (in March/April 1975) (20) and near Manoel Emídio, but Roth’s (27, 22, 28) wide-ranging surveys failed to find evidence of this macaw in any part of Piauí. Keller (29) was the last to revive the notion that the species occurs in the state, but his basis was evidently incorrect (13).

For Maranhão, Sick and Teixeira (21) reported that five birds discovered in trade in the late 1970s emanated from the state (and this was repeated by 23 and 30), and local reports suggested that the species might occur along the Rio Parnaibinha, between Morro da França and Fazenda Promissão, but dedicated surveys revealed these sightings to be misidentifications (27, 30, 22, 31). However, Roth (17) himself observed four small and unidentified macaws in the Serra do Itapecurú, which sighting was apparently never pursued (13). Pinto (7, 32) included the upper Rio Parnaíba, which forms part of the frontier between Maranhão and Piauí, in the distribution, but again Roth’s surveys found nothing to confirm this.

Eastern Tocantins (at the time part of the state of Goiás) was first included in the species’ distribution by Ridgely (23, based on reports of hunters to H. Sick and R. S. Ridgely in 1977), but neither Sick and Teixeira (21) nor Sick (33) mentioned the region in reference to Spix's Macaw. Based on trapped birds and information proffered by trappers, both Keller (24) and Roth (30) were confident that at least individuals stemmed from this region, and mentioned the area between Tocantinha and Filadélfia, along with Pedro Afonso and São Miguel do Araguaia, although searches by the latter failed to find additional evidence (22, 30, 28, 31).

In addition to the varying levels of probability attached to the species’ occurrence in some of the states surrounding Bahia, it also was present at further localities in the latter state. Roth (18) postulated, based solely on the extent of caraiba woodland, that the species must originally have occurred 50 km either side of the Rio São Francisco and spanning 150‒200 km between Juazeiro in the west and Santa Maria de Boa Vista in the east, which range Juniper and Yamashita (34) extended even further east, to Abaré, at riacho da Vargem, where Spix's Macaw had reputedly been trapped out recently, at the time. Much farther west, Sick and Teixeira (21, 33) reported seeing seven individuals, in two small flocks, over a buriti palm (Mauritia flexuosa) grove near the Formosa do Rio Preto, Riachão, in December 1974 (year reported incorrectly in 35), and this locality has been generally accepted as correct (13), despite that it lies within the Cerrado domain, and surveys in the following decades found no evidence of the species there, not even a local person who knew it formerly (18, 22, 28, 31, 34).

Distribution of the Spix's Macaw - Range Map
  • Year-round
  • Migration
  • Breeding
  • Non-Breeding
Distribution of the Spix's Macaw

Recommended Citation

Sharpe, C. J., G. M. Kirwan, N. Collar, and P. F. D. Boesman (2022). Spix's Macaw (Cyanopsitta spixii), version 2.1. In Birds of the World (G. M. Kirwan, Editor). Cornell Lab of Ornithology, Ithaca, NY, USA. https://doi.org/10.2173/bow.spimac1.02.1