Five subspecies recognized (but see Related Species):
- torquatus, described as Ramphastos torquatus Gmelin 1788; type locality México, restricted to Veracruz by Brodkorb 1939. Occurs from southern Mexico (except the Yucatan Peninsula) south to extreme northwestern Colombia (Gulf of Urabá). See Detailed Description.
- erythrozonus, described as Pteroglossus torquatus erythrozonus Ridgway 1912; type locality Temax, Yucatán, Mexico. Occurs in southeastern Mexico (Campeche, Yucatán, and Quintana Roo), northern Guatemala (Petén), and Belize. Similar to nominate torquatus, but black breast spot reduced or lacking; size smaller (Short and Horne 2001).
- nuchalis, described as Pteroglossus nuchalis Cabanis 1862; type locality Puerto Cabello, Venezuela. Occurs in northern Colombia and northern Venezuela. Similar to nominate torquatus, but black breast spot averaging larger, slightly paler cinnamon on "thighs", shallower notches on bill, and broader white basal line on bill (Short and Horne 2001).
- sanguineus, described as Pteroglossus sanguineus Gould 1854; no type locality specified, restricted to San José [Valle del Cauca], Colombia by Chapman (1917: 333). Occurs from eastern Panama (eastern Darién) (Ridgely and Gwynne 1989) south to northern Ecuador in northern Esmeraldas and adjacent Imbabura (Ridgely and Greenfield 2001a); forms narrow hybrid zones with torquatus in Panama and in the northern Colombia, near the Gulf of Urabá (Haffer 1967). Similar to nominate torquatus, but the cinnamon rufous nuchal collar; the bill has smaller but more numerous notches on the tomia of the maxilla; the tip of the maxilla, and sometimes of the mandible, is yellowish white; there is a black streak along the maxilla, just above the tomia; and the bare facial skin is black or blue between the eye and the bill.
- erythropygius, described as Pteroglossus erythropygius Gould 1843; no type locality specified, but Ecuador. Occurs in western Ecuador, north to western Esmeraldas (Ridgely and Greenfield 2001a); recently reported from extreme northwestern Peru (Tumbes), where perhaps only a vagrant. Reportedly hybridizes with sanguineus at one site in Pichincha, Ecuador (Short and Horne 2001). Similar to sanguineus, but lacks the black streak along the culmen; and the mandible is mostly yellowish white, with black confined to the tip.
Phylogenetic analyses of genetic data, based both on protein electrophoresis and on DNA sequence data from mitochondrial genes and a nuclear intron, consistently show that Pteroglossus is monophyletic, if Saffron Toucanet, formerly classified in a monotypic genus (Baillonius), is included in Pteroglossus (Hackett and Lehn 1997, Eberhard and Bermingham 2005, Pereira and Wajntal 2008, Patel et al. 2011). Patel et al. (2011) modelled rates of diversification of the Pteroglossus genus in the lowland Neotropics. Diversification of the Pteroglossus radiation predates the Pleistocene (Eberhard and Bermingham 2005, Patel et al. 2011), which has been predicted to have played a pivotal role in diversification in the Amazon rainforest biota. Patel et al. (2011) found a constant rate of diversification in Pteroglossus, and thus no support that events during the Pleistocene caused an increase in diversification.
Within Pteroglossus, torquatus (Collared Aracari) and frantzii (Fiery-billed Aracari) form a clade (Hackett and Lehn 1997, Eberhard and Bermingham 2005, Pereira and Wajntal 2008, Patel et al. 2011). These same genetic surveys, however, also consistently show that torquatus is paraphyletic with respect to frantzii, as frantzii is sister to torquatus sanguineus and torquatus erythropygius rather than to nominate torquatus. Both sanguineus and erythropygius often are considered to be separate species (Peters 1948, Hilty and Brown 1986, Ridgely and Greenfield 2001a, del Hoyo and Collar 2014), under which classification the issue of paraphyly with respect to frantzii is resolved. Haffer (1967, 1974) documented hybridization between torquatus and sanguineus in both eastern Panama and in northwestern Colombia, with no evidence of assortative mating, although in both regions the zone of hybridization seems to be very narrow (20 km or less in width). The taxa sanguineus and erythropygius also must come into contact, although "the area of potential contact in Esmeraldas and extreme nw. Pichincha remains poorly explored ornithologically" (Ridgely and Greenfield 2001a). Haffer (1974) and Ridgely and Greenfield (2001a) could not document contact between sanguineus and erythropygius, but Berg (2001) reported, with little comment, that these two were sympatric at one site (Cotacachi-Cayapas Ecological Reserve, Esmeraldas/Imbabura, Ecuador). Short and Horne (2001) mention a small series of specimens from one site in Pichincha, all of which they determined to be hybrids, and further state that "nearly half of the specimens designated 'erythropygius', and a like proportion of of SW Colombian- NW Ecuador 'sanguineus' converge in the culmen and mandible characters", a signal of rampant introgression that seems to have been missed by other researchers (indeed, Haffer 1974: 234 referred these taxa as "well differentiated subspecies").
Pteroglossus torquatus and Pteroglossus frantzii are largely allopatric, but Short and Horne (2001) mention one apparent hybrid.