Proposal (511) to South American Classification Committee
A). Split Eudyptes moseleyi from E. chrysocome
B). Add E. moseleyi to the main SACC species list
Crested penguins breeding chiefly at the remote subtropical “maritime” Tristan da Cunha and Gough Islands (ca. 85% of the population), along with small numbers at Amsterdam and St. Paul Islands (taxon moseleyi) have, until fairly recently, been treated as a subspecies of the well-studied subantarctic Rockhopper Penguin E. chrysocome. A growing number of recently published papers have induced most authorities to accept moseleyi as a separate species. Subspecies filholi of subantarctic Heard Island and Macquarie Island has not been recorded in the SACC region, and its taxonomic position is controversial. This proposal is necessary due to the post-1995 appearance of documented vagrant Eudyptes [chrysocome] moseleyi penguins at the Falkland Islands.
A NO vote for part a) disqualifies a vote for part b). If added to the SACC list, a decision on an English vernacular name will also be required, because at least five different names are available.
(A). Split Eudyptes moseleyi from E. chrysocome.
Jouventin (1982) provided sonograms of moseleyi and chrysocome showing that the former has considerably lower-pitched frequency calls, and first drew attention to differences in behaviour, noting also the better known distinctions in morphology (moseleyi is larger) and plumage (moseleyi has exceedingly long laterally splayed head plumes, unlike the short, narrow plumes in chrysocome and several other Eudyptes penguins); visually these penguins look very different. He suggested that moseleyi could be elevated to species rank.
A more thorough study (Jouventin et al. 2006) included DNA sequences of the ND2 gene and mitochondrial control region, breeding cycle, further studies of voice and of head ornamentation; the conclusion reached was that moseleyi warrants species rank, and also considering that “divergence in mating signals …. may have been enough to isolate these taxa without the need for morphological differentiation.” Jouventin et al. (2006) synthesized the vocal and morphological evidence as follows: “The northern populations have a much lower-pitched nuptial call (2-3 kHz for the maximum frequency vs. 6-7 kHz in the southern populations) with a completely different structure (longer modulated phrases and fewer syllables), and longer (79-81 mm vs. 66-68 mm) and denser yellow crest feathers than the southern form.” Biogeographically, the main breeding areas of moseleyi (subtropical) and chrysocome (subantarctic) are separated by the Subtropical Convergence, which divides subantarctic and subtropical waters (see the especially synthetic Fig. 1 in Jouventin et al. 2006).
Thereafter, Banks et al. (2006) analysed 1441 nucleotides of three mitochondrial genes to 1) reconstruct the phylogenetic relationships, and 2) to compare the genetic distances between the Rockhopper taxa and compared this with the uncorrected pairwise distances between other sister penguin species. Phylogenetically, all three taxa (moseleyi, chrysocome, filholi) were monophyletic with the latter two representing a sister position to one another. With regard to the genetic distances, in the first gene (ribosomal), chrysocome–moseleyi was more divergent than Spheniscus mendiculus–S. humboldti (two species on the SACC list), and S. magellanicus–S. demursus, both of which showed a similar divergence as chrysocome – filholi. The cytochrome b analysis showed a similar genetic distance between chrysocome–moseleyi compared to filholi–moseleyi but slightly less than E. pachyrhynchus–robustus and mendiculus–humboldti. In the final gene test (cytochrome oxidase), chrysocome–moseleyi came out as being more divergent than mendiculus–humboldti, than E. chrysolophus–schlegeli and chrysocome-filholi, each of which showed a similar divergence. In synthesis, genetic (mtDNA) data supports the split of moseleyi from chrysocome, but the position of filholi is more open to alternative interpretation.
Demogin et al. (2010) studied the reversed hatching asynchrony comparing the egg size dimorphism in the three Rockhopper taxa, finding that chrysocome has considerably larger volume eggs and higher breeding success than either moseleyi or filholi.
In addition to the above, there are also published data on host-exclusive and non-exclusive lice in the different Rockhopper taxa if anyone needs more evidence.
Most recent authorities have accepted the moseleyi split and sometimes also the filholi split, based principally on the work of Banks et al. 2006. The one exception is Christidis and Boles (2008), who stated “Genetic distances between the northern [moseleyi] and southern [chrysocome] forms were about half that between E. chrysocome (sensu lato) and E. chrysolophus.’’ Note that I was unable to find this data in Banks et al. 2006, or in any other paper; indeed this statement actually contradicts the study of Banks et al. 2006. They went on to state… “The observed level of difference between northern and southern forms of E. chrysocome is comparable to those between the pairs E. chrysolophus-–schlegeli and E. pachyrhynchus–robustus (Ritchie 2001), which are here treated as conspecific.” Oddly, the Ritchie (2001) paper deals with DNA markers relating to Pygoscelis adeliae. I am at a loss to explain or qualify any of the comments made by Christidis and Boles (2008), which appear to be erroneous.
I would suggest that the collective evidence strongly favours a YES vote that would recognise Eudyptes moseleyi as a valid species. A NO vote would imply that SACC has to review the validity of various penguin species on its main list and Hypothetical list, with a view to lumping, in particular Spheniscus mendiculus into S. humboldti, S. magellanicus into S. demursus, E. robustus into E. pachyrhynchus (there is a reasonable case for this anyway), and E. schlegeli into E. chrysolophus.
(B). Add Eudyptes moseleyi to the main SACC list.
Matias et al. (2009) documented the first four records of Eudyptes moseleyi at the Falkland Islands, coming from East Falkland from late November to late December 1995, and again in late November 1996, at Kidney Island in mid December 2004, and at New Island in late November 2004 with two published photographs from New Island and McBride Head, East Falkland. The diagnostic, long splayed and drooping head plumes and blunt-tipped supercilium in both photographs leave no doubt regarding the identification while the ventral flipper pattern, visible in one photograph, provides yet more diagnostic features. Subsequently, another bird was photographed on the Falklands in December 2009, and another in December 2010-January 2011; photographs available on Internet.
Recommendation: I recommend a YES vote to add E. moseleyi to the SACC list as a confirmed vagrant (V) to the Falkland Islands. A NO vote would refute the identification of the published records.
Banks, J., Van Buren, A., Cherel, Y. & Whitfield, J.B. 2006. Genetic evidence for three species of Rockhopper Penguins, Eudyptes chrysocome. Polar Biology 30: 61–67.
Christidis, L. & Boles, W.E. 2008. Systematics and taxonomy of Australian birds. CSIRO Publishing, Collingwood, Australia.
Demongin, L., Poisbleau, M., Raya Rey, A., Schiavini, A., Quillfeldt, P. Eens, M. & Strange, I.J. 2010. Geographical variation in egg size dimorphism in rockhopper penguins. Polar Biology 33:469–476.
Jouventin, P. 1982. Visual and vocal signals in penguins, their evolution and adaptive characters. Advances in Ethology 24: 1-149.
Jouventin, P., Cuthbert, R. & Ottvall, R. 2006. Genetic isolation and divergence in sexual traits: evidence for the Northern Rockhopper Penguin Eudyptes moseleyi being a sibling species. Molecular Ecology 158: 3413–3423.
Matias, R., Catry, P., Pearman, M. & Morrison, M. 2009. Vagrancy of Northern Rockhopper Penguins Eudyptes moseleyi to the Falkland Islands. Marine Ornithology 37: 287–289.
Ritchie, P.A. 2001. The evolution of the mitochondrial DNA control region in the Adelie Penguins of Antarctica. Massey Univ., Palmerston North, New Zealand.
Mark Pearman, October 2011
Comments from Robbins: “YES, to elevating moseleyi to species level. All data suggest that this is a split long overdue. Also, yes to adding this to the SACC list.”
Comments from Stiles: “YES. There seems to be good evidence from a variety of sources for this split, along with adding moseleyi to the SACC list.”\
Comments from Pérez: “YES. Complete evidence supports this split, including morphological, genetic, and behavioral information. Christidis and Boles (2008) were against the split, based on genetic distances, but this seems to be a mistake because Jouventin et al. (2006) showed the complete opposite pattern. Also YES to adding this species to the SACC list. Observations of both taxa (E. chrysocome and E. moseleyi) in Falkland Islands, assuming lack of morphological overlap (as suggested by evidence given by Mark in the proposal), support species status. However, it would be interesting to do a population genetic study in Falkland Islands to confirm there is no genetic introgression between these taxa, as I don’t think published evidence discard this possibility (no more than five E. chrysocome individuals (no moseleyi for Falkland Islands) sequenced; Banks et al (2006)).”