Monotypic.
While formerly considered a subspecies of Herald Petrel, differences in plumage, morphology, and possibly vocalizations argue for elevation to full-species status. There are rumors that Storrs Olson of the Smithsonian Institution and David Lee of the North Carolina Museum of Natural Sciences have prepared a paper outlining the arguments for the split. Named after one of the two known breeding localities for this 'new' species, a small group of islets (more correctly spelt 'Trindade') off northeastern Brazil. According to Ned Brinkley (in turn quoting Storrs Olsen) 'Trindade' sounds something like "Treen-DA-djze" in local dialect.
 

Identification

[Note that much of the published identification literature (e.g. Harrison, 1993) discusses nominate Herald Petrel (Pacific form) rather than Trinidade Petrel.]

Comprises three color morphs - pale, intermediate and dark. Relative ratios are uncertain. The majority of birds seen/photographed in the Gulf Stream off North America have been classified as dark-morph.

Dark morph: Potentially mistaken for a Sooty Shearwater or even a dark morph jaeger, especially in low winds. In stronger winds (in excess of 15 knots) flies with steep and high arcs. Makes 3-4 deep flaps between arcs. In silhouette shows a distinctive profile, with narrow wings and a very long 'hand'. Notice also the heavy bill. In higher winds adopts a more curved profile, often compared to an 'archers bow'. Uppersurface, dark brown with slightly darker 'M' mark that may be visible in certain lights, particularly in younger birds or those in fresh plumage.

Pale morph: Browish-gray head and upperbreast contrasting with white lower breast and belly and whitish throat. Much more extensive white on the underwing than the dark morph, particularly on the secondaries and secondary coverts. Examination of specimens suggest that most Trinidade Petrels (and those from Reunion) have entirely dark legs and feet, while Herald Petrels from the Pacific have bicolored legs and feet (black webs and the rest of the foot and tarsus pale pink) (Brinkley and Patteson, 1998).

Intermediate morph: Possibly just pale or dark variants of the above? Diffuse mottling on belly with flecks of pale feathering on throat.

Brinkley and Patteson (1998), address the essentially unexplored but very interesting question of how Herald Petrel might be separated from Trinidade Petrel in the field. The critera raised include: leg/foot color, presence or absence of white lesser (marginal) underwing-coverts, extent of pale feathering surrounding the base of the bill. Much more work is needed, and matters are complicated by regional variation within Herald Petrel.
 

Where and When

Breeds on Trindade island (note spelling), and the islet of Pedro Segundo on the nearby Martin Vaz archipelago (c. 40 miles from Trindad Island). Population estimated at 5000 birds on Trindad. Under serious threat from habitat destruction by humans and their introduced goats and pigs.

Range at sea poorly known. Now regularly observed in very small numbers (max. 1-2/day) off North Carolina from late May - late September. It is unclear whether birds observed off the Azores (Dubois and Seitre, 1997).
 

Literature

Brinkley, E.S. (1996) Secrets of the Deep. Birdwatchers Digest 18(6): p66-72

Brinkley, E. S. and Patteson, J. B. (1998) Gadfly Petrels in the western North Atlantic. Birding World 11(9): p341-354.

Dubois, P. and Seitre, R (1997) Herald Petrel new to the Western Palearctic. Birding World 10: 456-459.

Gochfeld, M., Burger, J., Sliva, J., and Gochfled, D. (1988) Herald Petrel new to the West Indies. American Birds 43(3): p1254-1258

Lee, D. S. (1979) Second record of the South Trinidad Petrel (Pterodroma arminjoniana) for North America. American Birds 33: 138-139.