Proposal (954) to South American Classification Committee

Recognize minuta Bonaparte as the second species in the genus Nyctiprogne

Proposal 673 advocated splitting Nyctiprogne leucopyga into two species, based mainly on the genetic data of Sigurdsson & Cracraft (2014); the name given for the second species was N. latifasciata. The proposal was accepted by SACC. However, the discussions regarding the split disclosed several underlying problems. To begin with, the distributions of the two species were very incompletely known; no clear differences in plumage or morphology were available (beyond the impression that these are “subtle”). Likewise, apparent differences in vocalizations were mentioned but not tied to specific specimen evidence or details on localities. Finally, based on examination of the illustration of N. leucopyga in Spix (1825) and the description of latifasciata by Friedmann (1945), Piacentini concluded that the latter was indistinguishable from nominate leucopyga and therefore is a synonym of leucopyga, thus leaving the second species without a name. Shortly after this, the committee received a communication from Nigel Cleere (in press) that presented an apparently satisfactory nomenclatural resolution. He presented detailed information on morphology and distributions that clarify the recognition of the two species, confirming Piacentini’s judgment that latifasciata is a synonym of leucopyga, and stated that the second species should be known as N. minuta Bonaparte 1850, and that the additional subspecies described since then pertain to minuta, also giving more details on their ranges and distributions. Upon applying Cleere’s criteria to a small series of Nyctiprogne in the collection of the Instituto de Ciencias Naturales of the Universidad Nacional de Colombia, I found that both species occur in Colombia: two specimens from Mitú, Vaupés are clearly leucopyga, while one from Arauca, two from Casanare, plus a recording that I made from western Meta are clearly minuta. However, because the name minuta had been proposed by Bonaparte (1850) with virtually no supporting details, its applicability appeared definitely questionable. Doubts in this regard were voiced by several committee members, such that SACC tabled the acceptance of this proposal pending resolution of the nomenclature.

After several attempts to contact Nigel Cleere for more details, I recently received his reply, which I believe will serve to definitively resolve these doubts. Cleere provided a reference to the typification of Bonaparte´s (1850) Caprimulgus minuta: Holotype RMNH 88413, adult male, Brazil: type specimen specified by van den Hoek Ostende, L.W., Dekker, R.W.R.J. & Keijl, G.O. 1997, p. 163 in: Type-specimens of birds in the National Museum of Natural History, Leiden. Part 1. Non – Passerines. Nationaal Natuurhistorisch Museum Leiden, Technical Bulletin 1, 248 pp. These authors noted that Bonaparte had spent a year in the Leiden museum, putting his new names on many specimens for which no published names existed, although there were various unpublished manuscript names by Temminck and Natterer (part of his rush to complete his Conspectus before dying, as his terminal illness was becoming more severe: the last version of the Conspectus actually ends with an incomplete generic description!) have seen the van den Hoek et al. publication, which validated Bonaparte’s name and established its type. Cleere personally examined this specimen and confirmed its identification as minuta.

I sent voting members the text supplied by Cleere (to appear in his revision of his book on Nightjars of the World, apparently still in press) with further details regarding the two species of Nyctiprogne (for inspection by SACC members only at Cleere’s request).

If the proposal passes, I will do a follow-up proposal on English names.

F. Gary Stiles, December 2022

Comments from Robbins: “YES. Kudos to Gary and Nigel for clarifying all of this. As a result of their efforts, I vote yes for recognizing minuta as a species.”

Comments from Areta: “YES. OK, now this is looking much better than before. I am glad that SACC did not accept the usage of latifascia before for what has now been shown to be Nyctiprogne minuta. It would make me feel much more comfortable if Cleere´s work was published first, in the interest of full public disclosure of the features that distinguish minuta from leucopyga, and with an available photograph of the type of minuta. BUT, given Gary´s successful test of the features in museum specimens based on features in the name-bearing type of minuta, I will accept the notion that N. minuta is the valid name for this nighthawk species."

Comments from Claramunt: “NO. I am still a bit confused. I would like to see the evidence more clearly. Cleere describes leucopyga as having white patches at either side of the throat but the type of leucopyga illustrated by Spix conspicuously lacks such patches:

https://www.biodiversitylibrary.org/page/41278046#page/105/mode/1up

“Can the illustration in Cleere’s book be made available? Can we wait until the evidence is published?”

Comments from Bret Whitney (who has Remsen’s vote): “NO. Sigurdsson and Cracraft (2014) stumbled into the realization that there are two genetically distinct clades of nighthawks masquerading under the single name Nyctiprogne leucopyga. Knowing little about these birds in life, or their distributions beyond collecting localities of a few tissues, they applied, thanks to the fabulous work of Friedmann (1945), the name leucopyga to one, and latifascia to the other. Nomenclature, actual application of those names… well, that was 100% “shot in the dark". Going back 20+ years, Whitney and Cohn-Haft had independently identified two distinct song types inhabiting the same spaces on river islands of the middle-lower Rio Negro in Brazil during the low-water season when island substrates become accessible to these ground-nesting birds. However, no definitive resolution of the nomenclature of these syntopic, morphologically weakly differentiated species could be presented without first making voice recordings of at least several individuals (what are levels of vocal and plumage variation?) before collecting the birds, enabling a clearer understanding of the identifications and widespread (and they are both remarkably widespread) distributions of the two vocal types. That work has been happening, in spurts, for years now, but there are some “loose ends” that those of us involved with the work feel ought to be tidied up before presenting a hypothesis of species limits and biogeographical relationships. Truth be told, no genetic analysis was necessary for a two-species split based on recorded specimens that could then be diagnosed morphologically, and matched to named forms, or shown to be unnamed. Well… maybe not. Because now, as the type of N. leucopyga is lost, we are left with an illustration accompanying the original description on which to base the name — and the illustration certainly appears to better match Friedmann’s (1945) N. l. latifascia than any other, earlier name. The type locality “Rio Amazonas, Brazil” is useless. Accepting these realities, the name latifascia becomes a synonym of leucopyga, and the problem shifts, once again, to what name should be applied to “the other” Nyctiprogne.

“Cleere (in press?) recommends that Caprimulgus minutus (Bonaparte, 1850) is that name. Although we do not yet have sufficient data and analysis available for independent consideration, we do have some material to work with. For example, in 2016, Thiago V. V. Costa sent me photos of the type of minutus; it’s clearly a Nyctiprogne, and the tail-band appears to be more medially situated than is typical of leucopyga, which we now understand (from the illustration of the type of leucopyga) to be situated more proximally, nearer the undertail coverts, along with other minor distinctions. Bonaparte’s minutus has no locality beyond “Bresil”, which is useless, but inconsequential, because the only result we really need to see is which clade that specimen aligns with: “trillers" (= nominate) or “chunk-chewinkers”. If it’s the latter, I would be accepting of the name minutus. If it’s in the triller clade, it’s likely best considered another synonym of leucopyga, and we have to dig deeper (more on that if it should come to that).

“So, where does that leave us, today? Not materially different from where it all started 30 years ago: we have two species-level Nyctiprogne nighthawks, the names and distributions of which are incompletely understood. How is this situation resolved? It is most adequately, and elegantly resolved by a combination of a small amount of highly focused vocal + collecting fieldwork (a great deal has already been done), followed by genetic analysis of all specimens collected after voice recording, together with — ideally — analysis of toe-scrapes of all extant type specimens in the genus. At the very least, it’s critically important to obtain a sufficient DNA yield from the type of minutus to confidently place it in the phylogeny.

“To those who wring their hands that, by following the above, recommended course of actions, we would be choosing to wait for an adequate (but who-knows-when? soon, we expect!) analysis of the situation rather than move ahead with recognition of two species and fix (the almost inevitable) problems with nomenclature down the line... I would offer that it’s best to calm down and just pretend you don’t know there are two species of Nyctiprogne, so there’s nothing to worry about. After all, that’s pretty much exactly the way things would be at the moment if Sigurdsson and Cracraft had not been supplied with tissues that glaringly revealed the paraphyly of Nyctiprogne.”

Additional comments from Stiles: “Cleere provided a diagnosis of minuta vs. leucopyga and a reference to the typification of minuta with an explanation of Bonaparte´s names bestowed in Leiden, and examined this type and identified it as minuta. I applied Cleere´s criteria for distinguishing minuta from leucopyga and found that it works, permitting me to identify both species in Colombia. So, it´s not quite true that nothing has changed in the last 30 years ago, as Bret laments! Looking through the literature, I found that minuta had not been applied to any other Nyctiprogne - and if Nyctiprogne remains a recognized genus, anything like a Caprimulgus minuta becomes a non-starter.”

Comments from Del-Rio: “YES. Thanks Gary and Nigel!”

Comments from Bonaccorso: “NO. It would be important to revisit these arguments based on published data. If we accept this proposal, as it is, we would then also have to accept, for example, Nacho´s claims about not recognizing S. iberaensis because it might be S. melanops.”

Comments from Lane: “NO. I don’t understand the need to deal with this before the publications in question are even published; this is a recipe for error. Bret has made some comments above that seem to suggest we pump the brakes here a bit, and I’m inclined to agree. From what I can understand from the proposal and from follow-up comments here, I see that two names are applied to two genetic lineages, but what does that mean? We have no distributional handle for them, nor do we know which vocalization type is represented by which name. This complex is a confusing one, and it would make me feel better if the situation was laid out plainly so that there was no question which name is applied to which group. So between these concerns, I believe we should just wait for the Cleere publication to be published, and perhaps for better definitions of the taxa to which these names are applied.”