Split the Sharp-beaked Ground-Finch (Geospiza difficilis) and the Large Cactus-Finch (Geospiza conirostris) into multiple species

Proposal (676) to South American Classification Committee

Split the Sharp-beaked Ground-Finch (Geospiza difficilis) and the Large Cactus-Finch (Geospiza conirostris) into multiple species

Background

Although work on speciation, hybridization, beak size evolution, and many other topics have been studied in depth, and sometimes in novel ways in the Darwin’s Finches, it is only recently that a modern phylogeny was produced for this group of tanagers (Petren et al. 2005). It also has been clarified that the Darwin’s Finches are part of the “dome nest clade” which includes Tiaris and various other largely Caribbean taxa (Burns et al. 2002). However, even in recent molecular phylogenies, populations from different islands were not sampled, leaving in question whether morphologically unique populations were in fact also genetically unique. Further, these molecular phylogenies have been based on mtDNA and a few microsatellite loci.

New Information

Lamichhaney et al. (2015) have produced a new molecular phylogeny, including samples from multiple islands for various species of Darwin’s Finches. This study is based on a whole-genome re-sequencing of 120 individuals. They discovered evidence for widespread historical gene flow between various populations of Darwin’s Finches, and that genetic diversity is higher than would be expected in small insular populations, likely through gene flow from other species. Their results upheld the general phylogenies proposed recently; however, they differed in finding that various populations of Geospiza difficilis (Sharp-beaked Ground-Finch) and Geospiza conirostris (Large Cactus-Finch) are not sisters to each other. They also discovered a locus associated with beak shape, and discussed this in terms of ecology and evolution of members of this group. They concluded that the group is approximately 1 million years old, and that certain branches, such as the tree finches are rather recent (200,000 years).

Below is the suggested phylogeny as well as a map of populations sampled.

[Note from Remsen: In the tree, difficilis G[enovesa]= acutirostris; difficilis F, S, P = nominate difficilis; difficilis D[arwin], W[olf] = septenttionalis; G. conirostris G[enovesa] = propinqua; G. conirostris E[spañola] = nominate conirostris.

Evidence for species splits

1) Geospiza difficilis, Sharp-beaked Ground-Finch. Results confirm that there are three taxa widely separated in the phylogeny. Three species level taxa are recommended to be recognize:

G. acutirostris Ridgway; found on Genovesa.

G. difficilis Sharpe; found on Pinta, Fernandina, and Santiago.

G. septentrionalis Rothschild and Hartert; found on Wolf and Darwin.

These were formerly recognized as species, based on differences in size and bill shape. The species difficilis has a straight culmen and is truly sharp beaked, whereas septentrionalis has a curved culmen. These populations also differ in song (Grant et al. 2000). The species acutirostris is very much smaller in mass than the other two; in many ways it resembles a Small Ground-Finch (G. fuliginosa), and in fact it is genetically much closer to fuliginosa and fortis (Medium Ground-Finch) than it is to true Sharp-beaked Ground-Finch. Curiously song is more similar to septentrionalis (Grant et al. 2000). Lamichhaney et al. (20150 suggested that acutirostris may be a species derived from mixed ancestry, i.e. of hybrid origin, but that it is a distinct and separate entity (species).

2) Geospiza conirostris Large Cactus-Finch. The two populations, one on Genovesa and one on Española, are each genetically more similar to other species than they are to each other. The Genovesa population (propinqua) is sister to the Common Cactus-Finch (G. scandens), whereas the nominate forma is sister to the Large Ground-Finch (G. magnirostris). This mirrors morphology, both are big and large-billed, but propinqua has a long bill like a Cactus-Finch, whereas conirostris has a deep bill like a Large Ground-Finch. The two forms of conirostris differ in song, and song playbacks on Genovesa found only a weak attraction effect when playing back songs from Española (Ratcliffe et al. 1985). Note that there was absolutely no response from playback of Large Ground-Finch songs on conirostris. The playback results between propinqua and Common Cactus-Finch are more complex. The form propinqua reacts strongly to one song type of Common Cactus-Finch from Daphne Major, but Common Cactus-Finch does not react to songs of propinqua. Furthermore male Common Cactus-Finch discriminated strongly against propinqua when tested with a pair of museum specimen models in female plumage, one of propinqua and one of local scandens (Common Cactus): the reciprocal experiment on Genovesa was not completed (Ratcliffe et al. 1983). Putting this together, the evidence suggests that Common Cactus-Finch and propinqua, its sister species, would rarely if ever interbreed if they came into contact. Note that there are old records of Large Cactus-Finch from Darwin and Wolf islands. However, these are based on only a few records and specimens. The current status of a population on Darwin Island is unclear, and there are none currently on Wolf Island. The subspecies darwini has been named, although there is no evidence that there is a recent population from Darwin Is., and in fact there has been confusion as it is thought that these may have been stray Large Ground-Finches (Wiedenfeld 2006).

Recommendation

I recommend a Yes vote to separate these species, raising the number of Darwin’s Finch species from 15 to 18. Note that geographically the existence of a Large Cactus-Finch on Darwin and Wolf islands is very unlikely, unless it also was a separate and unique population. There is no recent evidence of a long-term sustaining population on Wolf. It is unclear if the species is present and common on Darwin Island, and indeed these may have been either stray Large Ground-Finch, or perhaps a population of Large Cactus-Finch. Therefore, I think it is best to delay any decision on what to do with darwini (if indeed it still exists), until genetic data from specimens is studied. In the past darwini has been lumped with propinqua.

English Names

G. acutirostris Ridgway; found on Genovesa. – Restricted to Genovesa, why not Genovesa Ground-Finch? It is small and similar to Small Ground-Finch, so a size or even a bill shape name does not jump out based on its morphology.

G. difficilis Sharpe; found on Pinta, Fernandina and Santiago. – This is the most widespread, and perhaps the archetype “Sharp-beaked Ground-Finch” so I would suggest letting it retain the name Sharp-beaked Ground-Finch.

G. septentrionalis Rothschild and Hartert; found on Wolf and Darwin. Vampire Ground-Finch, based on its well-known habit of feeding on booby blood. The colloquial Vampire Finch has been in use for some time, but to be consistent I think we would need to use Ground-Finch.

G. conirostris; found on Española. This huge-billed bird is most similar to the Large Ground-Finch, which it is sister to. I don’t know if one can come up with a morphological based name, such as Thick-billed Ground-Finch that mentions anything unique? Perhaps Española Ground-Finch would be the best name, because it is endemic to that island.

G. propinqua; found on Genovesa. This is sister to the Common Cactus-Finch, so perhaps it should keep the name Large Cactus-Finch? Although this may be confusing as it is not any more widespread or easily found than conirostris. Genovesa Cactus-Finch would be another possible name, noting that above we already have a Genovesa Ground-Finch.

VOTING:

Subproposal A – Accept separation of G. difficilis into three species.

Subproposal B – Accept separation of G. conirostris into two species.

If these above proposals pass, then we will consider proposals on English Names.

Literature Cited

Burns, K. J., S.J. Hackett and N. K. Klein. (2002). Phylogenetic relationships and morphological diversity in Darwin’s Finches and their relatives. Evolution 56(6): 1240–1252

Grant, B. R., Grant, P. R. & Petren, K. (2000). The allopatric phase of speciation: the sharp beaked ground finch (Geospiza difficilis) on the Galapagos islands. Biol. J. Linn. Soc. 69, 287–317.

Lamichhaney, S., J. Berglund, M. Sällman Almén, M. Khurram, M. Grabherr, A. Martinez-Barrio, M. Promerová, C.-J. Rubin, C. Wang, N. Zamani, B. R. Grant, P. R. Grant, M. T. Webster and L. Andersson (2015). Evolution of Darwin’s finches and their beaks revealed by genome sequencing. Nature 518: 371–375.

Petren, K., P.R. Grant, B. R. Grant and L. F. Keller (2005). Comparative landscape genetics and the adaptive radiation of Darwin’s finches: the role of peripheral isolation. Molecular Ecology 14: 2943–2957

Ratcliffe, L. M. & Grant, P. R. (1983). Species recognition in Darwin’s finches (Geospiza, Gould). II. Geographic variation in mate preference. Anim. Behav. 31, 1154-1165.

Ratcliffe, L. M. & Grant, P. R. (1985). Species recognition in Darwin’s finches (Geospiza, Gould). III. Male responses to playback of different song types, dialects and heterospecific songs. Anim. Behav. 31, 290-307.

Wiedenfeld, D. A. (2006). Aves, The Galapagos Islands, Ecuador. Check List 2(2): 1-27.

Alvaro Jaramillo, July 2015

=========================================================

Comments from Zimmer: “YES to Subproposal A and YES to Subproposal B, for reasons detailed by Alvaro in the Proposal.”

Comments from Stiles:

“A) YES. No problem here.

“B) YES, given the more complete genetic sampling (all this makes me wonder about splitting up the small, compact clade of Camarhynchus (?) in green into all those species; presumably vocal evidence).”

Comments from Pacheco: “YES to A and YES to B. I am backing Alvaro’s two recommendations from molecular phylogeny with samples geographically well distributed.”

Comments from Robbins: “NO. Having read the McKay and Zink (2015) paper on Geospiza, in which they suggest there is only a single species, one needs to pause and rethink things. So, we want to add yet more species to a highly controversial Geospiza species limits definition? For now, I vote no, if for no other reason than to bring to the attention that the committee needs to be aware of the latest perspective on this group.”

Additional comments from Robbins: “With regard to the Geospiza proposal. The abstract of the Lamichhaney et al. (2015) paper states the following:

‘Phylogenetic analysis reveals important discrepancies with the phenotype-based taxonomy. We find extensive evidence for interspecific gene flow throughout the radiation. Hybridization has given rise to species of mixed ancestry.’

“Additionally, in the text on the first page, ‘Extensive sharing of genetic variation among populations was evident, particularly among ground and tree finches, with almost no fixed differences between species in each group (Extended Data Fig. 2).’ Finally, note that sampling is certainly not extensive, so hybridization may well be underrepresented. All of this should cause one to pause and think about the McKay and Zink perspective. As we have appreciated throughout our careers, this a very complex situation and it would be an understatement to say attempting to apply traditional species limit views is difficult.

“I certainly would welcome input from those who have a deep understanding of the genetics and analyses of the Lamichhaney et al. (2015) results.”

Comments from Remsen: “YES to A and B. Darwin’s finches continue to yield insights into the complexity of the early stages of speciation. These populations have all diverged to the point in terms of voice and behavior consistent with their treatment as BSC species. In my view, that their mtDNA has not yet sorted out reflects the lag time between these neutral loci versus those under selection, or a degree of ongoing hybridization that does not conflict with species rank (as in all hybridizing taxa that are also treated as species under BSC).”

Comments from Cadena: “YES. The genomic data clearly point to problems with species delimitation, and Alvaro has nicely summarized evidence showing the genomic data match well with geography, morphology, vocalizations, and behavior. The paper mentioned by Mark is indeed intriguing, but I did not find the analyses of morphological data in that study entirely satisfying. In fact, with Felipe Zapata and Iván Jiménez, we are working on re-analyses of morphological data using quantitative approaches, and the results we get do not quite agree with the conclusions of McKay and Zink based on their PCAs. We hope to write up a note describing this soon.”