A phylogeny of extant penguins (Aves: Sphenisciformes) combining morphology and mitochondrial sequences
Abstract
The phylogenetic relationships among the penguins have received little attention, despite their well‐known anatomy and the conspicuous nature of the group. Previous attempts have included datasets limited to few, mostly osteological characters, and one study was based on integumentary and breeding characters. We developed a morphological matrix comprising 159 morphological characters of osteology (70 characters), myology (15), digestive tract (1), integument (66), and breeding (7 characters), scored in 18 extant forms (all currently recognized species plus one distinct subspecies). A gaviiform was placed at the root, and 11 species of representative procellariiform groups completed the outgroup. A heuristic parsimony analysis under equal weights was performed. We also compiled DNA sequences available in GenBank for the mitochondrial genes 12S rDNA and cytochrome b. We included the two data partitions in a combined analysis under direct optimization. Both analyses recovered the monophyly of Sphenisciformes and all the traditional polytypic genera. Morphological characters performed optimally at the ordinal and generic nodes, also providing resolution and varying degrees of support at supra‐ and intrageneric nodes. The comparison of molecular and morphological results indicated that the most significant problem in the phylogeny of extant penguins is rooting the ingroup. The mutual interaction of molecular and morphological data decreases the ambiguity regarding the placement of the root, and provides a resolved, relatively well‐supported phylogeny of extant penguins. Biogeographical patterns based on breeding ranges and derived from the combined analysis show that the major intercontinental vicariance events detected are consistent with cold marine current patterns of the Southern Hemisphere.
© The Willi Hennig Society 2005.
Penguins (Aves: Sphenisciformes) belong to a homogeneous group of marine birds of the southern oceans that exhibit the most remarkable adaptations for an aquatic life among birds. These include wings transformed into flippers for underwater propulsion, skeletal modifications for upright locomotion on land, and a highly modified plumage for insulation in critically cold conditions. Adaptations of the breeding system in all aspects (morphology, physiology, behavior) are also notable. Penguins are thought to be more closely related to other marine birds, especially the Procellariiformes (petrels and albatrosses; Stresemann, 1934; Simpson, 1946; Ho et al., 1976; Saiff, 1976; Sibley and Ahlquist, 1990; Mey et al., 2002; Mayr and Clarke, 2003; cf. Mayr, 2005) and Gaviiformes (loons or divers; Cracraft, 1981; Olson, 1985; Groth and Barrowclough, 1999). Simpson (1946) classified all extant penguins (17 currently recognized species in 7 genera; Martínez, 1992) in one subfamily, Spheniscinae. Zusi (1975), on the basis of osteology, suggested a division of the Spheniscinae into two main lineages within which other suprageneric groups were nested (Fig. 1A).

(A) Intergeneric relationships proposed in a non‐phylogenetic analysis of osteological data (modified from Zusi, 1975). (B) Network of penguin relationships obtained in a phylogenetic analysis using morphological characters; alternative rootings are indicated by arrows (modified from O'Hara, 1989). (C) Intergeneric relationships obtained in a previous cladistic analysis of integumentary and breeding characters (modified from Giannini and Bertelli, 2004).
Phylogenetic relationships among extant penguins were first addressed, in a cladistic context, by O'Hara (1989). This author used a limited data set of 22 characters from osteology (16 characters), the integument (4), and breeding behavior (2 characters), scored for 14 species of penguins and four outgroups, an albatross Diomedea exulans, a frigatebird Fregata magnificens, a loon Gavia immer, and a grebe Podilymbus podiceps. O'Hara (1989) did not include characters supporting the monophyly of penguins nor characters that could define an unequivocal rooting of the penguin tree. Although the monophyly of penguins never seemed in question, both considering traditional treatises on the group (Beddard, 1898; Huxley, 1867; Sclater, 1880; Gadow, 1892; Pycraft, 1898) and modern authors (e.g., Schreiweis, 1982; Zusi, 1975), when all four outgroups were included in the analysis, penguins were not monophyletic. In addition, some polytypic genera were not recovered in the strict consensus. It is remarkable, however, that when using one outgroup at a time, roughly the same network of penguin relationships emerged, but the position of the root varied with outgroup choice (Fig. 1B). The two main groups proposed by Zusi (1975) were recovered when Gavia was used to root the tree.
Recently, we explored the potential of integumentary and breeding characters for phylogeny reconstruction in penguins (Giannini and Bertelli, 2004). Although these types of characters are widely used in bird taxonomy, they are only rarely employed in phylogenetic studies (e.g., Chu, 1998; Bertelli et al., 2002). Our analysis included 65 integumentary and 5 breeding characters scored for all extant species of penguins plus one subspecies, and 12 outgroup taxa from the Procelariiformes and Gaviiformes. One limitation of this dataset is that the adult plumage in the outgroup taxa is segregated into different tracts (pterylae) that leaves definite areas of the body featherless (apteria), as is typical in birds. Because penguins are uniformly covered by feathers (i.e., they lack apteria), most adult plumage characters (34 characters) were not comparable with respect to the outgroup, thus reducing the number of characters that could provide evidence for ingroup rooting. Our results (Fig. 1C), however, strongly supported the monophyly of penguins, the monophyly of all polytypic genera, suggested that genera may be grouped into two lineages (but not those proposed by Zusi, 1975), and provided evidence of several intrageneric relationships (not shown; see Fig. 1 in Giannini and Bertelli, 2004).
For this study we developed a morphological matrix of 159 characters that adds osteological (70 characters), myological (15), and digestive tract (1) characters, to an expanded integumentary (66) and breeding (7) data set derived from Giannini and Bertelli (2004). These new data were scored for the same ingroup and outgroup taxa of Giannini and Bertelli (2004). We also compiled DNA sequences available in GenBank for the mitochondrial genes 12S rDNA and cytochrome b. We include the two data partitions in a combined analysis under direct optimization (Wheeler, 1996), an approach used, to our knowledge, for the first time in birds. The comparison of molecular and morphological results indicates that the most significant problem in the phylogeny of extant penguins is rooting the ingroup. The mutual interaction of molecular and morphological data decreases the ambiguity regarding the placement of the root and provides a well‐resolved, generally well‐supported phylogeny of extant penguins. On the basis of this phylogeny, we present a reconstruction of the biogeographical history of the group.
Materials and methods
Taxa
Outgroups. We selected representatives from the Gaviiformes and Procellariiformes on the basis of morphological (Cracraft, 1981; Olson, 1985; Mayr and Clarke, 2003) and molecular evidence (Groth and Barrowclough, 1999; Nunn and Stanley, 1998; Kennedy and Page, 2002). We used the same terminal taxa as Giannini and Bertelli (2004). Those are 11 species of representative procellariiforms, namely albatrosses (Dioemedea melanophrys and Phoebetria palpebrata), fulmarine petrels (Macronectes giganteus and Daption capense), gadfly petrels (Pterodroma incerta), shearwaters (Puffinus gravis and Procellaria aequinoctialis), prions (Pachyptila desolata), southern and northern storm petrels (Oceanites oceanicus and Oceanodroma leucorrhoa), and diving petrels (Pelecanoides urinatrix), plus the gaviiform Gavia stellata which was set as the root for all analyses.
Ingroup. We included 18 forms of extant penguins. These are the 17 currently recognized species, namely Aptenodytes forsteri, A. patagonicus, Eudyptes chrysocome, E. chrysolophus, E. pachyrrhynchus, E. schlegeli, E. sclateri, E. robustus, Eudyptula minor, Megadyptes antipodes, Pygoscelis adeliae, P. antarctica, P. papua, Spheniscus demersus, S. humboldti, S. magellanicus, and S. mendiculus (Martínez, 1992), plus Eudyptes chrysocome moseleyi which scored distinctly in some integumentary characters (Giannini and Bertelli, 2004). The specimens examined are listed in Appendix 1.
Characters
Morphological and breeding characters. We included 159 characters from the integument (66 characters), osteology (70), myology (15), breeding behavior (7) and one character of the digestive tract (Appendix 2). We used our published 65 integumentary characters (Giannini and Bertelli, 2004), which consisted of the color and structure of bill and legs, and plumage of adult, immature and downy chick. In that work, we also provided five breeding characters that included egg traits, nesting, and the sociability of immatures. We added one integumentary character (see character 65 in Appendix 2) and two breeding characters (71 and 72 in Appendix 2).
The main sources of osteological characters were Watson (1883), Pycraft (1898, 1899, 1907) and Wilson (1907), whose detailed observations were further elaborated and expanded principally by Simpson (1946), Verheyen (1958), Zusi (1975), Saiff (1976), and O'Hara (1989). In total, we scored 70 osteological characters that included absence and presence, relative development, and the relationships of bony structures in the cranium and postcranium of all ingroup terminals and all but one outgroup terminal (material from Phoebetria palpebrata was not available). All of the character states were scored from museum specimens (Appendix 1). Most of the 70 characters are used in a cladistic context here for the first time, although some of them were already described and discussed previously by other authors in anatomical and/or systematic studies (see comments on character descriptions in Appendix 2). Some characters were defined by the only previous cladistic study available that used osteology: O'Hara (1989). For the myological characters, we based our scorings on Schreiweis (1982). This author made a comprehensive description of the appendicular musculature of all but three of our ingroup terminals, including representatives of all genera (Eudyptes robustus, E. sclateri and E. chrysocome moseleyi missing). We coded each of Schreiweis's distinct descriptive statements for each structure he recognized, including presence and absence, divisions and fusions, and variations in origin and attachment of muscles and tendons. We restricted the use of myological information to cases in which the morphological variation described could be distinctly scored in discrete character states. This resulted in 15 myological characters (143–157 in Appendix 2). For some outgroup taxa, we completed the scoring following McKitrick (1991). The total morphological matrix is provided in Appendix 3.
Finally, we did not include two osteological characters from O'Hara (1989). In our sample, the perforation of the processus craniolateralis of the sternum (O'Hara's 2‐states character 12) varied greatly across specimens, precluding a clear definition of character states. In the case of O'Hara's (1989) character 17, we could not precisely define which structure was the most proximal end of the femur (either the head or the trochanter in O'Hara's character) due to extensive individual variation and ambiguity of anatomical landmarks.
DNA sequences. We assembled a molecular data set of DNA sequences from GenBank for the mitochondrial genes 12S rDNA and cytochrome b (accession numbers and authors given in Appendix 4). Sequence representation varied among taxa (Table 1). The size of the mitochondrial data set is ca. 2.1 kilobases for complete species.
| Species | Base pairs | |
|---|---|---|
| 12S | Cyt‐b | |
| Outgroup | ||
| Daption capense | 377 | 1143 |
| Diomedea melanophrys | — | 1143 |
| Gavia stellata | 979 | 1143 |
| Macronectes giganteus | 376 | 1143 |
| Oceanites oceanicus | — | 1143 |
| Oceanodroma leucorrhoa | — | 1143 |
| Pachyptila desolata | — | 1143 |
| Pelecanoides urinatrix | 379 | 1143 |
| Phoebetria palpebrata | — | 1143 |
| Procellaria aequinoctialis | — | 1143 |
| Pterodroma incerta | — | 1143 |
| Puffinus griseus | 772a | 1143 |
| Ingroup | ||
| Aptenodytes forseri | 375 | 771 |
| Aptenodytes patagonicus | 976 | 1143 |
| Eudyptes c. chrysocome | 976 | 1143 |
| Eudyptes c. moseleyi | — | – |
| Eudyptes chrysolophus | — | 1143 |
| Eudyptes pachyrrhynchus | 768b | – |
| Eudyptes schlegeli | 393 | 771 |
| Eudyptes sclateri | — | – |
| Eudyptes robustus | — | – |
| Eudyptula minor | 975 | 1143 |
| Megadyptes antipodes | 376 | – |
| Pygoscelis adeliae | 977 | 771 |
| Pygoscelis antarctica | 386 | 1143 |
| Pygoscelis papua | 387 | 1143 |
| Spheniscus demersus | — | – |
| Spheniscus humboldti | — | 771 |
| Spheniscus magellanicus | 376 | 771 |
| Spheniscus mendiculus | — | 771 |
- a 375 bp from X82522 and 393 bp from U88007.
- b 377 bp from U74353 and 395 bp from U88024.
Guided by conserved areas in each sequence string, we split the complete sequences into fragments (each included in a separate input matrix) in order to accommodate homologous fragments of incomplete species in the corresponding input matrices. This procedure need not affect homology identification, and is important in speeding‐up searches (Wheeler, 2003). When the procedure does affect homology, this can be identified (and eliminated) provided that a check of tree length with complete sequences is made. The 12S sequences were cut into five fragments. The cyt b sequences were treated as prealigned, as 1143 is the known, fixed number of base pairs in most birds (incomplete fragments were filed with “N”s in the missing regions). Thus a total of six input molecular matrices were used in molecular and combined analysis (available at ftp://ftp.amnh.org/pub/mammalogy).
Cladistic analyses
We analyzed the evidence available (morphology and DNA sequences) separately and in combination. For the morphological analysis, we used the program TNT 1.0 (Goloboff et al., 2003) to search for optimal trees under equal weights. We conducted heuristic, unconstrained searches for optimal trees using Tree Bisection Reconnection (TBR) branch‐swapping in each of 200 replications of random taxon addition sequences, keeping up to five trees per replication. A second TBR round was applied to each of the optimals, to increase the confidence of finding all topologies of minimum length. Zero‐length branches were collapsed and strict consensus trees were generated. Bremer values, both absolute (Bremer, 1994) and relative (Goloboff and Farris, 2001), were used to estimate the support of groups. In order to prevent an over‐estimation of support values, we implemented a strategy for obtaining suboptimal trees in seven successive stages, saving up to 2000 suboptimals in each stage. We first searched for suboptimal trees 1 step longer than the optimals, next saving suboptimals up to 2, 3, 4, 5, 6 and 7 steps longer than the optimals. Finally, as recommended by Goloboff and Farris (2001), relative support values were calculated considering only those trees within the absolute Bremer support for each group.
For the molecular analysis, we combined the sequences from the 12S and cyt b in a single parsimony analysis using a combination of direct optimization (DO; Wheeler, 1996) and iterative pass optimization (IPO; Wheeler, 2003) as implemented in POY 3.0.11 (Wheeler et al., 2003). DO is a heuristic method that considers sequence alignment as an optimization problem, without an intermediate step of multiple alignment. Base‐to‐base correspondences are treated dynamically, so primary homology statements are tree‐dependent—as in all automated alignment approaches—and the optimal (less costly) homology statements are found by means of a tree search. In DO, the tree is generated by Wagner builds followed by branch swapping, and the transformation cost of the sequences across the tree is minimized according to predefined substitutions and indel costs. In our study, we opted for an equal‐cost transformation matrix (the cost ratio of indels:transversions:transitions is 1 : 1 : 1). The tree that minimizes overall molecular transformations is selected as optimal—and this tree directly determines the phylogenetic relationships among taxa.
As in DO, IPO minimizes the total transformation costs of DNA sequences on a given tree. IPO uses DO‐initialized nodal sequences and performs a three‐dimensional type of local cost calculation that takes into account, for a given node, the sequences of the two descendants plus the ancestor (hence 3D). Once all node assignments have been stabilized (consider that modifying one nodal sequence potentially affects assignments in neighboring nodes, and is hence iterative), length is calculated by summing up the cost along all branches (i.e. of all pair‐wise transformation costs between a node and one descendant). IPO explores a much wider space of sequence transformations than DO, and this has two consequences. First, DO provides an upper bound of tree length, but very significant length reductions can be achieved, and tree cost may be much lower than under DO. Therefore, IPO results are preferred on optimality grounds. Second, IPO is highly demanding in calculation time, this may restrict its application in mid‐to‐large data sets. For instance, Giannini and Simmons (2003) applied a two‐phase strategy of replicated swapping + refinements under DO, with only final refinements under IPO. For the present, relatively small dataset, IPO and other refinements were used from the building phase and in all stages of search (using commands ‐iterativepass ‐exact ‐slop 5 ‐checkslop 10). One hundred replications (Wagner builds based on independent random addition sequences of taxa) were each followed by TBR branch swapping. Tree fusing (Goloboff, 1999) was also used after each build and after each round of TBR whenever more than one different tree was reported. Finally, all distinct trees found were collected and submitted to a final round of fusing (using the commands ‐treefuse ‐fuselimit 10000 ‐fusemingroup 2 ‐fusemaxtrees 1000).
As a measure of clade support, we calculated Bremer values using constrained searches as implemented in POY. We checked the accuracy of some values with searches using the command ‐disagree (see Frost et al., 2001 for details).
The combined molecular and morphological analysis was carried out in POY under the same search strategy, adding the morphological matrix to the scripts and weighting morphological and molecular changes equally. During the searches, POY optimizes each morphological character and DNA sequence in each tree and keeps the tree(s) that minimize overall cost across all data matrices. All analyses were run on a 1.8 GHz PC.
Biogeography
On the basis of the phylogenetic results of the combined analysis, we conducted a preliminary analysis of the biogeographical history of the group, as supported by data from the living species, based on simple mapping. Penguins are extraordinarily vagile marine birds, and therefore seemingly unsuitable for biogeographical study. However, they exhibit a remarkable nest‐site fidelity (Martínez, 1992), so the breeding ranges of penguins offer great potential for historical biogeographical reconstruction. The breeding ranges of penguins are located on the shores of islands and continents of the Southern Hemisphere, and are well‐known for all species (see summary information in Harrison, 1985). We used the maps of Harrison (1985) to locate the breeding ranges of each penguin species. A preliminary analysis showed that the breeding ranges can be grouped into natural regions of the Southern Hemisphere (Fig. 5A), which are often home to well‐differentiated subspecies of penguins of wide Austral distribution. Those regions are: the South American coasts (hereafter “sam”), including the shores of the subcontinent and islands within the continental shelf north of Drake passage; the Galapagos Islands (ga); the South African coasts (saf); the isolated Atlantic Ocean Islands of Tristan da Cunha and Gough (tg); the isolated Bouvet Island (bu); the scattered Southern Indian Ocean Islands including Marion, Prince Edwards, Kerguelen, and Heard Islands (io); the Australia–New Zealand region, and nearby islands (az); the Antarctic Peninsula (ap); the islands of the Scotia Arc (S. Orkney, S. Georgia, and S. Sandwich Is.; sa); and the shores of the Antarctic continent (ac). Each region represents a distinct breeding (and environmental) area. For instance, species that regularly breed in the Antarctic Peninsula may not breed in the continental shore (e.g., Pygoscelis papua), and species that breed on islands of the Scotia arc may not breed on the Antarctic Peninsula (e.g., Eudyptes chrysocome). Each one of those regions was considered a character state of a single geographic character (cf. Bremer, 1992), and each species was assigned as many states as required by the distribution of their breeding ranges (Fig. 5A). We optimized the distributions of the breeding regions in the consensus penguin subtree from the combined analysis.

(A) Regions of the breeding ranges of penguins in the Southern Hemisphere. Abbreviations: ac, shores of the Antartic continent; ap, Antarctic Peninsula; az, Australia–New Zealand region; bu, Bouvet Islands; io, Southern Ocean Islands; ga, Galapagos Islands; sa, islands of the Scotia Arc; saf, South African coasts; sam, South American coasts; tg, Islands of Tristan da Cunha and Gough; 1, Circumpolar Antarctic current; 2, Humboldt current; 3, West wind drift and Brazil‐Benguela system. (B) Reconstructions of the most relevant biogeographical events based on optimization of breeding ranges (see Fig. 5A) on the results of combined analysis.
Results
Analysis of morphological and breeding characters
Searches yielded four optimal trees of 394 steps, a length obtained in 189 replications. Support values were calculated on the basis of a sample of 10 908 suboptimal trees and are shown in Fig. 2. The strict consensus tree (Fig. 2) was fully resolved except for the trichotomy in Spheniscus. Procellariiformes and Sphenisciformes were recovered with high support. Relationships in the procellariiforms (not shown) were poorly supported, as expected given that we did not include enough characters to deal specifically with those relationships.

Morphological analysis, strict consensus of four most parsimonious tree at 394 steps. Values of absolute and relative Bremer support are given above and below branches, respectively. The most likely point of attachment of the root suggested by the molecular analysis (see Fig. 3) is shown with an arrow (conflicting branches are in gray). Clades A to D are commented in the text.
Relationships within Sphenisciformes included monophyly of all currently recognized (polytypic) genera recovered with moderate‐to‐high support; the weakest support value is found in Pygoscelis (BS = 3). As in the integumentary analysis (Fig. 1C; Giannini and Bertelli, 2004), two suprageneric groups were recovered: Spheniscus + Eudyptula (clade A in Fig. 2) versus a group including the remaining genera (clade B). Clades A and B were weakly supported. Megadyptes + Eudyptes (clade D) were sister taxa. In the current results, Aptenodytes and Pygoscelis grouped together (clade C), although with minimal support. Within Spheniscus, S. humboldti and S. mendiculus formed a group but with minimal support. Within Pygoscelis, P. antarctica was sister to the other two species; the relationship of P. papua and P. adeliae is as supported as the genus. Within Eudyptes, E. pachyrrhynchus was the sister of the other species. Next, E. robustus and E. sclateri formed a group that was sister to two other clades (E. chrysolophus + E. schlegeli) and (E. c. chrysocome + E. c. moseleyi). Again, this pattern is identical to the integumentary analysis. Support values ranged from 1 to 3, but were low especially toward the basal groups within the genus. Overall, relative support values followed the pattern of absolute values.
Analysis of mitochondrial sequences
Searches under IPO + the fusing of 12S and cyt b sequences produced three optimal trees of length 1919. This analysis did not recover Procelariiformes as monophyletic, with Oceanitidae forming a basal trichotomy with the remainder of the Procellariiformes and Sphesniciformes. Sphenisciformes were recovered as a well‐supported group. Three of the four polytypic genera were recovered (the exception was Pygoscelis). Support of the genera varied from 5 (Spheniscus and Eudyptes) to 19 (Aptenodytes). The available sequences did not resolve the base of the penguin subtree, with a tetrachotomy consisting of Aptenodytes, two Pygoscelis groupings, and a clade formed by the remainder of the genera. Aptenodytes and (Pygoscelis antarctica + P. papua) showed the highest support values within penguins. In turn, a large group including the four remaining genera (clade E in Fig. 3) was well supported. Two suprageneric groups (clades A and D) were recovered with low support. Within Spheniscus, a sister relationship between S. humboldti and S. mendiculus was only poorly supported; this subtree is compatible with the morphological results. The relationships within Eudyptes were minimally supported (Fig. 3), and incompatible with the relationships favored by the morphological analysis in the placement of E. pachyrrhynchus as sister to E. chrysocome (cf. Figs 2 and 3).

Molecular analysis, strict consensus of three optimal trees at 1919 steps (iterative pass optimization). Absolute Bremer support values are given above branches. The placement of the root suggested by the morphological analysis (see Fig. 2) is indicated by arrow (conflicting branches are in gray). Clades A, D and E are commented in the text.
Combined analysis
Searches combining the morphological matrix and the two mitochondrial sequences under IPO + fusing produced two optimal trees of length 2331. As in the morphological analysis, the consensus is fully resolved, except for the same trichotomy in Spheniscus(Fig. 4). The order Procellariiformes was recovered as monophyletic and its internal relationships were resolved (Fig. 4).

Combined analysis, strict consensus of 2 optimal trees at 2331 (diagnosed under iterative pass optimization). Numbers above branches are Bremer support values. Clade P is Procellariiformes; clades A, D and E are commented in the text.
The monophyly of Sphenisciformes was recovered with very high support. All polytypic genera were also recovered, with Bremer values varying from 8 to 28 (Fig. 4). Aptenodytes was found to be sister to a poorly supported clade including the remainder of the genera. A monophyletic Pygoscelis was sister to a well‐supported clade E. Clades A and D were recovered with higher support as compared with the previous separate analyses. Within Pygoscelis, P. antarctica and P. papua formed a group sister to P. adeliae. Within Spheniscus, only one group was recovered (S. humboldti + S. mendiculus). The relationships within Eudyptes were less resolved than in the morphological consensus, with comparable Bremer values (1 ≤ BS ≤ 3). A reconstruction of the biogeographical history of the penguin subtree is shown in Fig. 5B (see Discussion).
Discussion
Morphology
One of our goals was to provide a strong morphological phylogeny of penguins. We believe that we have provided a comprehensive data set, at least insofar as the ratio of characters to taxa is 5 : 1. The morphological analysis, including integumentary, osteological, myological, and breeding characters, recovered a tree structure similar to the integumentary data set alone (Fig. 1c) from Giannini and Bertelli (2004). Overall, we added twice as many characters from the osteology and myology, with two effects. First, absolute and relative support increased in three of the polytypic genera. Second, suprageneric clades maintained or decreased their support values, for instance in Eudyptula + Spheniscus (clade A), and Megadyptes + Eudyptes (clade D). This means that the new data brought a considerable amount of conflict, in addition to evidence for well‐established genera.
The present analysis recovered an additional suprageneric clade formed by Aptenodytes + Pygoscelis (clade C). However, the support for this clade is minimal. This is one of the groups proposed by Zusi (1975) on the basis of osteology, and was also recovered by O'Hara (1989) when the root is determined using Gavia, as in our study.
In summary, the morphological data set supported the monophyly of Procellariiformes (and some groups nested within, not shown), Sphenisciformes, polytypic genera, and three suprageneric groups, one of which was moderately supported (Megadyptes + Eudyptes, clade D in Fig. 2).
Molecular analysis
The analysis using mitochondrial sequences did not resolve the base of the penguin subtree, although it suggested that either Aptenodytes or members of Pygoscelis, or a group formed by the two genera, were sister to clade E (Fig. 3). This is in contrast with the morphological results, in which the clade formed by Eudyptula + Spheniscus (clade A) was sister to a group containing all the other genera (clade B). The molecular data set also failed to recover Pygoscelis, but strongly indicated that P. antarctica and P. papua were more closely related than P. adeliae and P. papua, as suggested by morphology. In other groupings, the morphology and the mitochondrial data were in agreement (Fig. 4). A minor exception was the partially incompatible relationships within Eudyptes, but the alternatives were poorly supported in both analyses (cf. Figs 2 and 3).
Combined analysis
In the combined analysis (Fig. 4), the most relevant contribution of the mitochondrial data set was the recovery of the strongly supported clade E, contradicted by the morphological analysis. This group, however, was proposed by Zusi (1975) on the basis of osteology alone (Fig. 1A). This was the most influential suprageneric grouping in the combined analysis. Here all polytypic genera were monophyletic, and Aptenodytes was the sister to all other genera.
The chief difference between the combined and morphological analyses was the position of the root of the ingroup subtree. The same network structure attached to clade A in the morphology and to Aptenodytes in the combined analysis. This resulted mainly from the strong mitochondrial support for clade E. Note that, in the combined analysis, the absolute support was greater than in the sequence analysis alone, suggesting that some morphological changes contributed to supporting the clade as well (see below). All other relationships remained basically the same as in the morphology, with the exception of the grouping within Pygoscelis, in which the mitochondrial data predominated over the morphological hypothesis.
Character analysis
Integumentary and breeding characters are infrequently used for phylogeny reconstruction in ornithology (see Chu, 1998; Bertelli et al., 2002). The current analysis showed, however, that integumentary characters provided synapomorphies for every node resolved in the ingroup subtree, including seven synapomorphies for Sphenisciformes (see Appendix 5). This confirmed the value of this complex organ system as a source of useful characters. Furthermore, breeding characters, despite comprising a minor fraction of the total character set (7 out of 159 characters), were important, as they provided synapomorphies for 5 of the 13 nodes resolved, including two for Sphenisciformes.
Clade E, not recovered in the morphological analysis, was supported by six morphological changes in the combined analysis from three different character sources, specifically, the integument, osteology, and myology (Appendix 5). This explains the increased Bremer value in the combined analysis, compared with the separate molecular analysis, which was primarily responsible for the recovery of clade E.
We could not identify a specific hierarchical level at which non‐molecular characters (integument, osteology, myology, behavior) behaved optimally. Indeed, each morphological partition provided synapomorphies at several levels (Appendix 5). Trends were apparent, however (Fig. 6). The integumentary set was more important than other sets at lower nodes (within recognized genera), as these nodes were supported mainly by integumentary changes. In turn, there were no myological synapomorphies at the intrageneric level. Osteology was important at most nodes, and was the main source of synapomorphies for the order: three‐quarters of the 35 morphological changes supporting the monophyly of Sphenisciformes were osteological (Fig. 6). However, a potentially high number of myological traits that are known to be unique to penguins (Schreiweis, 1982) could not be included in the current analysis due to invariant states within penguins and a lack of information in the specific outgroup species that we included. We believe that a different choice of outgroups would not have changed this situation too much, as few species of gaviiforms and procellariiforms (or birds, for that matter) are actually examined at the level necessary for appropriate comparisons with Schreiweis's (1982) study. Only a part of the myological variation may be correlated with the skeletal structure already coded here.

Distribution of changes per morphological partition (symbols) in the penguin subtree from the combined analysis. Number of changes of each main partition at each node are indicated above the symbols.
Biogeography
Based on our mapping (Fig. 5B), we interpreted selected nodes that seemed to demonstrate the most relevant biogeographical events. The ancestral penguin bred in the Subantarctic, specifically Australia–New Zealand (az) and the Antarctic Peninsula (ap). A major vicariance event appeared in the next node up, separating the breeding ranges of Pygoscelis in the Antarctic Peninsula from the ranges in the Australian region, so the ancestor of Clade E penguins (the best supported suprageneric group) was hypothesized to be unambiguously Australian. This implies that many speciation events occurred within the region, which is not surprising as this region includes numerous islands of different ages. Following Ronquist's (1997) interpretation, the Australian ancestor of Clade A dispersed to South America (sam), then vicariance occurred and the lineage differentiated into Spheniscus in South America and Eudyptula in the Australia–New Zealand region. Given that no Indian or Atlantic Ocean ranges were reconstructed as ancestral in Clade A, the likely dispersal route was trans‐Pacific (Fig. 5A). From South America, an ancestral Spheniscus form reached Africa, likely a trans‐Atlantic dispersal. The African descendant of the vicariance event that followed was S. demersus. A comparable event can be hypothesized for the Galapagos penguin: A South American ancestor reached the Galapagos Islands (ga), then a vicariance event produced the sister species S. mendiculus in Galapagos and S. humboldti in the Pacific South American coast (Atacama). It is significant that these three major, intercontinental vicariance events involving clade A are consistent with the dominant west‐east direction of cold marine currents, and their northern upturn when they encounter the west coasts of South America and South Africa (Fig. 5A). Specifically, the West Wind Drift or Circumpolar Antarctic Current (in the Eudyptula–Spheniscus vicariance event), the West Wind Drift and the Brazil–Benguela system (in the S. demersus vicariance event), and the Humboldt Current (in the S. mendiculus vicariance event).
The reconstructions in the polytypic genera Eudyptes and Pygoscelis, whose species breed in several different ranges, indicate several post‐speciation dispersal events. It is noteworthy that none of the territories that are more recent in geological time (the volcanoes of Tristan, Gough, Bouvet and the Indian Ocean Islands) are reconstructed as ancestral breeding ranges in deep internal nodes, adding a degree of confidence to the reconstruction in general. It is also noteworthy that fossil penguins are especially abundant in New Zealand, the Antarctic Peninsula (Seymour Island) and Patagonia (see Simpson, 1972; Clarke et al., 2003). Our reconstruction suggests Subantarctic ancestors for penguins, present in New Zealand and the Antarctic Peninsula, but not in South America. According to our combined analysis, Patagonian penguins (in the genus Spheniscus) belong to a relatively modern clade. This is not a major discrepancy, however, as South America and the Antarctic Continent were connected through the Antarctic Peninsula during the Eocene via the Scotia Arc.
Conclusion
The morphology of penguins has been investigated in great detail over the last two centuries, from Huxley (1867) and Sclater (1880) to Zusi (1975) and Schreiweis (1982). However, this information was not placed within a phylogenetic perspective. Previous attempts used a small character set that assumed the monophyly of penguins (O'Hara, 1989), or were explorations on the performance of a particular organ system (the integument; Giannini and Bertelli, 2004). The present study provides a more comprehensive morphological data set that incorporates the sources of morphological information translated into scorings, along with many newly developed characters. The addition of mitochondrial sequences allowed the discovery of the clade E (all genera except Aptenodytes and Pygoscelis), which in turn suggests a different rooting of the penguin subtree. The molecular data, however, are insufficient to recover the monophyly of all genera and to resolve basal relationships. It was the combination of the morphological and molecular data sets that resolved the internal branching of the penguin subtree and decreased the ambiguity in the placement of the root. It also improved the support of most groups, including the key clades A, D, and E (Fig. 4). The combined analysis allowed a preliminary biogeographical reconstruction in which major vicariance events, as well as the trans‐oceanic dispersals of ancestors, are evident.
Direct optimization has not previously been used in ornithology. The optimization of molecular and morphological data in a single, combined analysis has provided a strong phylogenetic hypothesis for all extant species of an avian order, Sphenisciformes. This was accomplished in a reasonable amount of time (less than one hour per replication) in spite of the aggressive searches implemented.
Future research may improve on both morphology and molecules. For instance, 11 potential myological synapomorphies of penguins were described by Schreiweis (1982, p. 20–30) but could not be included here because they were invariant in the ingroup and were not studied in our outgroup taxa (absence of the M. scapulohumeralis cranialis, M. pronator superficialis, M. pronator profundus, M. entepicondylo‐ulnaris, M. abductor alulae, M. flexor alulae, M. adductor alulae, M. extensor brevis alula, M. adductor digiti II, M. adductor digiti IV, and the Pars metapagialis of M. serratus superficialis). In our study, taxonomic sampling was not an issue because we included all extant species. However, a number of fossils have been described (see Clarke et al., 2003), and a significant problem arises with the fossil taxa. Issues of comparability become crucial because many taxa were described using non‐overlapping skeletal elements and very few taxa (and very few specimens of those taxa) are reasonably complete. The addition of fossils is needed to understand the evolution of the diversity of penguins as a group, and may well change our perception of the biogeographical history of the group as reconstructed from living taxa.
Acknowledgements
For permitting access to materials, we thank P. A. Tubaro (Museo Argentino de Ciencias Naturales, Buenos Aires, Argentina); E. Alabarce (Colección Ornitológica Lillo, Tucumán, Argentina); P. Sweet, J. Cracraft, F. Vuilleumier, and G. Barrowclough (Department of Ornithology, American Museum of Natural History, New York, USA); M. Adams and R. Prys‐Jones (Bird Division, The Natural History Museum, Tring, UK); D. Willard, J. Bates and S. Hackett (Field Museum of Natural History, Chicago, USA); and K. Garrett (Natural History Museum of Los Angeles County, Los Angeles, USA). We thank J. Faivovich, K. Pickett, J. Cracraft, G. Mayr, E. Bourdon, U. Göhlich and P. Sweet for comments and suggestions. This project was supported by the Chapman postdoctoral fellowship (to S.B.) and Coleman and Vernay postdoctoral fellowships (to N.P.G.) at the AMNH.
Appendices
Appendix 1
List of specimens examined. Abbreviations of Institutions: AMNH, American Museum of Natural History, New York, USA; The BMNH, Natural History Museum, Tring, UK; COL, Colección Ornitológica Lillo, Facultad de Ciencias Naturales e IML, Tucumán, Argentina; FMNH, Field Museum of Natural History, Chicago, USA; LACM, Natural History Museum of Los Angeles County, Los Angeles, USA; MACN, Museo Argentino de Ciencias Naturales “Bernardino Rivadavia”, Buenos Aires, Argentina. Taxa are listed alphabetically.
1. Integumentary specimens
Aptenodytes forsteri AMNH 196281–2, 265389, 435602–7, 435609–11, 435613–4, 525862, 525867–71, 793545, LACM 69986; Aptenodytes patagonicus AMNH 132452–3, 132456, 132459, 132460–1, 269635, 428536, 435830, 435832, 443456, 442457, 525873–7, 525879–84, 525886, 525888–9, COL 10988; Daption capense AMNH 445493–4, COL 8089–93, 8095; Eudyptes chrysocome chrysocome AMNH 196414, 211923, 211991–2, 211994, 211996–7, 211999, 212002, 212009, 215732, 215759, 445213–20, 445222–6, 525731, 525733, 525736, 525738, 525740, 525743, 525745–52, 525754–8, 525760, 525762–3, 525765–73,COL 10986, 582–263; Eudyptes chrysolophus AMNH 196167–71, 349606, 445227–8, 5257926, BMNH 1905.12.30.164 [holotype], COL 10987; Eudyptes pachyrrhynchus AMNH 29898, 525703–4, 525707–8, 525713, 525726, BMNH 1845.1.13.29 [holotype]; Eudyptes schlegeli AMNH 211984, 525798–810, 525812; Eudyptes sclateri AMNH 196413, 212001, 212003–5, 212007, 212010–1, 525714, 525722, 525777, 525780–5, 525787, 525789, 525791, BMNH 1889.4.7.1 [holotype]; Eudyptes robustus AMNH 525720–1, 525723, 748427; Eudyptula minor albosignata AMNH 212016, 212018–9, 212021–5, 220909, 225701, 525702; Eudyptula minor minor AMNH 168804, 220910, 212013–15, 525528, 525599–603, 525605–21, 525623–4, 525627, 525629–30, 5256324, 525636–7, 525651, 525685, LACM 23909–11; Diomedea melanophrys AMNH 445399, 445410, 445417–8, 445420, COL 5936–9, 598–2729; Gavia stellata AMNH 525949, 525951, 525987, 703321, BMNH 1935.9.9.36, 1937.10.17.157, 1937.10.17.158, 1965.m.134; Macronectes giganteus AMNH 211703, 749313, COL 600, 5940–1; Megadyptes antipodes AMNH 461127–8, 525843–59, 525861–2, 5258646, LACM 101235; Oceanites oceanicus COL 8121, 8123, 8125–9; Oceanodroma leucorrhoa AMNH 70249, 70252, 79397, 349391, 528706; Pachyptila desolata AMNH 132519, 269679, 269682, 269285, 407676, COL 12831; Pelecanoides urinatrix AMNH 212088, 334637, 528738, 528760, 748433, COL 12445; Phoebetria palpebrata AMNH 26714, 132538, 196430, 211408, 527084, 749332, COL 10992; Procellaria aequinoctialis AMNH 211612, 211616, 527317, COL 10095–6; Pterodroma incerta AMNH 132491–2, 269656, 269659, 527987; Puffinus griseus AMNH 211986–7, 748419, 749268, 527606, COL 4219, 5893–7; Pygoscelis adeliae AMNH 196161–3, 325834, 325245, 325247–9, 435615–6, 442407–8, 442410–1, 525833, 525835–9, LACM 54430, 63897, COL 8138; Pygoscelis antarctica AMNH 196159, 132460, 168780, 168806, 190158, 196156, 196160, 349607, 442419–21, 525840–2, 775711, COL 8136; Pygoscelis papua AMNH 132462–9, 196165–6, 196415, 269638–9, 435821–9, 442412–7, 442458–62, 445204–12, 525819–20, 525822, 525824–31, 775712, COL 8019, 8137, 14043, 15837–9, two specimens uncataloged; Spheniscus demersus AMNH 139880, 300427, 525902–3, 525905, 525908–11, 525913–6, BM 1911.12.18.13, 1939.5.11.1, 1939.5.11.2, 1987.2.4.191, 1987.24.211, 1987.24.212, LACM 20531; Spheniscus humboldti AMNH 147869, 196395, 305680, 305682–5, 308767, 357489, 44522931, 525937, 828554, BM 1913.10.11.112, LACM 18007, 18473, 18498, 25439–40, 50001–2, 84448, COL uncataloged; Spheniscus magellanicus AMNH 445240–1, 525890, LACM 25436–8, 69988, COL 4206, 7217, 7235–8, 12378, 16001, 581–2642, two specimens uncataloged; Spheniscus mendiculus AMNH 178170, 196305–6, 266348, 292163, 292233–4, 292636, 299219, 407655, 442424, 525917, 525919–34, 525936, 804313–5, LACM 18649, 18653, 20056, 30295–8, 35882, 84451.
2. Osteological specimens
Aptenodytes forsteri AMNH 3745, 3725, 3727, 3767, 4856, 8110–2, 11634, 12002, FMNH 106829, 339515, LACM 99854; Aptenodytes patagonicus AMNH 83, 2611, 4382, 26471–2, FMNH 351211, 106475, 107781, LACM 86301, 111055; Daption capense AMNH 3126, 8868, 8915, LACM 102367, 102496, 102534, 102571, 102579, 102597; Diomedea melanophrys AMNH 3135, 23506, 23564–5, MACN 54409–10; Eudyptes chrysocome chrysocome AMNH 5398, 5972, 9173, 26477, LACM 99775; Eudyptes chrysocome moseleyi FMNH 291231–3, 312902, 314652, 345115–8; Eudyptes chrysolophus AMNH 5964, 26478, FMNH 431779, 339520, LACM 104403, 110995; Eudyptes pachyrhynchus AMNH 26509; Eudyptes robustus AMNH 27678, Eudyptes schlegeli AMNH 5399; Eudyptes sclateri BMNH 1952.1.38; Eudyptula minor AMNH 5314, 5620, 6957, 11561, FMNH 339521, 106434, 106492, 106505–6, 106997, LACM 90093, 102347–9, 102351, 102353–4, 103167–8, 103938; Gavia stellata AMNH 4974, 5971, 12905, 24132, LACM 86316, 90036, 99695, 100736, 101089–90, 112288, 112745, 112747–8, 112751; Macronectes giganteus AMNH 1634, 5400, LACM 102362–3, 103782, 104882, MACN 1655a, 4474a; Megadyptes antipodes AMNH 5613, 5615; Oceanites oceanicus AMNH 1357, 16599, LACM 102519, 102553, 102590, 103972, 104105, 104108–9, 104610; Oceanodroma leucorrhoa AMNH 22019, 22032, LACM 18127, 18129, 20081, 86402, 104314, 104316, 107468, 107969; Pachyptila desolata AMNH 3119, 4145, LACM 102501–2, 102510, 102522, 102532, 104299, 106728–9; Pelecanoides urinatrix AMNH 16043, FMNH 339530; Phoebetria palpebrata LACM 102361; Procellaria aequinoctialis AMNH 1638, 8929, LACM 101822, 102261, 102381–2, 103767, 104102, 110997; Pterodroma incerta AMNH 3120, 4693; Puffinus griseus AMNH 2902, 23554, LACM 86374, 103648, 103786–7, MACN 68380; Pygoscelis adeliae AMNH 26162–3, 26474–6, FMNH 96173, 104041, 339519, 378439, 104213–4, 106733, LACM 101721, 101722; Pygoscelis antarctica AMNH 26158–9; 26160–1, FMNH 104215–7, 390994; Pygoscelis papua AMNH 3191, 4361, 5766, 22679, 26164, FMNH 315111, 330150–4, 339516–8, 344864, 345119; Spheniscus demersus AMNH 1310, 1625, 3875, 4069, 10898, 10926, 12782, 22678, FMNH 289829, 291343, 290227, 291234, 375397, 378685, 378754, 104173,104434, LACM 111061; Spheniscus humboldti AMNH 4920, 26165, FMNH 330157, 339522–5, 339527–9, LACM 18473, 18498, 88834; Spheniscus magellanicus AMNH 823, 8826, 22677, 26479, 26481, FMNH 379141, LACM 90632, 101723, MACN 54412, 54682–3, 54685; Spheniscus mendiculus AMNH 3648, 3772, FMNH 105189, 105390, 105405, 105586, LACM 16055, 18653, 86303–4, 89906.
3. Oological specimens
Aptenodytes patagonicus AMNH 5843, 13389–91, 13547–8, 15536; Daption capense AMNH 6326, 13444–5; Eudyptes chrysocome AMNH 13392, 13412–5; Eudyptes chrysolophus AMNH 17297, 17301; Eudyptes robustus AMNH 17217; Eudyptula minor AMNH 5846; Gavia stellata AMNH 39, 42, 44, 46, 6288; Macronectes giganteus AMNH 6130, 6136, 13438, 13440, 13555; Oceanites oceanicus AMNH 13497 [type]; Oceanodroma leucorrhoa AMNH 6329, 15537; Pachyptila desolata AMNH 13464, 14572–3, 14576; Pelecanoides urinatrix AMNH 13513–4; Phoebetria palpebrata AMNH 6709–10, 13437; Procellaria aequinoctialis AMNH 13456, 13461–2; Puffinus griseus AMNH 3034, 5858, 6308, 13489–90; Pygoscelis antarctica AMNH 13411; Pygoscelis adeliae AMNH 5844, 13410; Pygoscelis papua AMNH 13393–6, 13400, 13549–51; Spheniscus demersus AMNH 17477; Spheniscus magellanicus AMNH 13416–9, 13604; Spheniscus humboldti AMNH 13603.
Appendix 2
Description of morphological characters.
Integumentary characters
Bill (rostrum)
- 0
Tip of maxilla (rostrum maxillare): pointed (0); hooked (1).
- 1
Tip of mandible (rostrum mandibulare): pointed (0); slightly truncated (1); strongly truncated, squared off (2); procellariiform‐like (3). In procellariiforms, the bill tip is squared but with a rounded margin in lateral view. Modified from O'Hara (1989, character 7). Our coding differs from O'Hara in that he distinguished only “decurved” (= pointed) from “recurved” (the alternative condition), coding outgroups as non‐comparable. We added a state 3 for some of the outgroup taxa and drawed a distinction within the non‐pointed mandible tips (states 1 and 2).
- 2
Longitudinal grooves on the base of the culmen: absent (0); present (1).
- 3
Longitudinal grooves on the base of latericorn and ramicorn: absent (0); present (1); Dabbene (1920).
- 4
Feathering of maxilla (rostrum maxillare), extent: totally unfeathered (0); slightly feathered, less than half the length of maxilla (1); feathering that reaches half the length of maxilla (2); feathering surpassing half the length of maxilla (3).
- 5
Ramicorn, inner groove at tip: absent (0); present and single (1); present and double (2). In frontal view, the mandible tip presents none, one or two folds between the outer borders of the tip.
- 6
Orange or pink plate on ramicorn: absent (0); present (1).
- 7
Plates of rhamphotheca, inflated aspect: absent (0); present (1). In most penguins and outgroup taxa, the horny plates of the rhamphotheca are flat and closely follow the underlying bone. In Eudyptes (and in Macronectes among the outgroup taxa), the surface of the rhamphothecal plates are more decurved in shape, suggesting the inflated aspect, especially at the base of latericorn (Dabbene, 1920).
- 8
Gape (rima oris), aspect: not fleshy (0); margin narrowly fleshy (1); margin markedly fleshy (2).
- 9
Ramicorn color pattern: black (0); reddish (1); pink (2); yellowish (3); orange (4); yellowish (5); green (6); blue (7).
- 10
Latericorn and ramicorn, light distal mark: absent (0); present (1). A vertical light mark crosses the dark bill near the tip in Spheniscus.
- 11
Latericorn color: black (0); red (1); orange (2); yellow (3); green (4); blue (5).
- 12
Culminicorn color: black (0); red (1); orange (2).
- 13
Maxillary and mandibulary unguis, color: black (0); red (1); yellow (2); green (3); bluish gray (4).
- 14
Bill of downy chick, color: dark (0); reddish (1); pale, variably horn to yellowish (2); bluish (3).
- 15
Bill of immature, color: dark (0); bicolored reddish and black (1); red (2); yellowish (3); grayish (4).
- 16
Nostril tubes: absent (0); present in chick (1); present in adult (2). Nostril tubes are typical of Procellariiformes. Kinsky (1960) described the development and resorption of tube‐like structures in Eudyptula. Following Sibley and Ahlquist (1990), we assume the homology of the tubes of Procelariiformes and Eudyptula, but we scored them differently so as to reflect the degree of ontogenetic development of the structure.
- 17
External nares: present (0); absent (1). Variation in the presence or absence of the external rhamphothecal nares in penguins was reported by Zusi (1975). After O'Hara (1989, character 2).
Iris (regio orbitalis: oculus [organa sensuum])
- 18
Iris color: dark (0); reddish‐brown (1); claret red (2); yellowish (3); white (4); silvery gray (5).
Feathers (Pennae)
- 19
Scale‐like feathers: absent (0); present (1). This type of feathers, with an extremely specialized structure, is unique to penguins, and are especially distinct over the wings. This is a small, hard and lanceolate feather, slightly curved with long and flattened rhachis and hyporhachisrhachis that provide extra thickness.
- 20
Rhachis of contour feathers: cylindrical (0); flat and broad (1). State 1 is a sphenisciform synapomorphy described in detail by Chandler (1916).
- 21
Rectrices: forming a fan functional for steering (0); not (1). The tail in penguins lacks the overlapping structure that enables it to act as a rudder in air (state 0 in most birds). Instead, the tail is a set of loosely arranged quills (state 1, typical of penguins), mainly used to assist the bird while standing upright.
- 22
Remiges: differentiated from contour feathers and specialized for flight propulsion (0); indistinct from contour feathers (1). State 1 is a sphenisciform synapomorphy whose homology issues were addressed by Chandler (1916).
- 23
Apteria: present (0); absent (1).
- 24
Molt of contour feathers gradual (0); simultaneous (1). Penguins molt all of the contour feathers in a single process that preclude them from swimming during the molting period.
Adult plumage
- 25
Yellow pigmentation in crown feathers (pileum): absent (0); present (1). This character accounts for the presence of yellow‐colored feathers around the edge of the crown in Megadyptes and Eudyptes.
- 26
Head plumes (crista pennae): absent (0); present (1). After O'Hara (1989, character 1).
- 27
Head plumes (crista pennae), aspect: compact (0); sparse (1). Species that lack head plumes were assigned a non‐comparable state (–) in this and the next three characters.
- 28
Head plumes (crista pennae), aspect: heading upward (0); heading backward, not drooping (1); heading backward, drooping (2).
- 29
Head plumes (crista pennae), position of origin: at base of bill close to gape (0); on the recess between latericorn and culminicorn (1); on forehead (2).
- 30
Head plumes (crista pennae), color: yellowish (0); orange (1).
- 31
Nape (occiput), crest development: absent (0); slightly distinct (1); distinct (2).
- 32
Periocular area (regio orbitalis), color: black (0); white (1); yellow (2); bluish gray (3).
- 33
Fleshy eyering (regio orbitalis): absent (0); present (1).
- 34
White eyering (regio orbitalis): absent (0); present (1).
- 35
White eyebrow (regio orbitalis, supercilium): absent (0); narrow, from postocular area (1); narrow, from preocular area (2); wide, from preocular area (3). This character encodes two aspects of the white eyebrow of Spheniscus (its origin and thickness) that can be arranged in a single transformation series and, hence, be treated as a single character.
- 36
Loreal area (lorum), aspect: feathered (0); with spot of bare skin in the recess between latericorn and culminicorn (1); with spot of bare skin contacting eye (2); bare skin extending to the base of bill (3).
- 37
Auricular patch (regio auricularis): absent (0); present (1).
- 38
Throat pattern (jugulum): black (0); white (1); yellowish (2); irregularly streaked (3); with chinstrap (4).
- 39
Collar: absent (0); at most slight notch present (1); present, diffusely demarcated (2); black, strongly demarcated (3).
- 40
Breast (pectus), golden in color: absent (0); present (1).
- 41
Dorsum color: black (0); dark bluish gray (1); light bluish gray (2).
- 42
Black marginal edge of dorsum between lateral collar and axillary patch (axilla), contrasting with dorsum: absent (0); present (1).
- 43
Black dots irregularly distributed over white belly (venter): absent (0); present (1).
- 44
Flanks (ilia), dark lateral band reaching the breast (pectus): absent (0); present (1).
- 45
Distinct dark axillary patch of triangular shape (axila): absent (0); present (1).
- 46
Flanks (ilia), extent of dorsal dark cover into the leg: incomplete, not reaching tarsus (0); complete, reaching tarsus (1).
- 47
Rump (pyga): indistinct from dorsum (0); distinctly whitish (1). After O'Hara (1989, character 3).
- 48
Tail length (cauda): short, the quills barely emerge from the rump (0); medium, the quills are distinctly developed but do not surpass the feet extended caudally (1); long, the quills surpass the feet extended caudally (2).
- 49
Outer rectrices, color: same as inner rectrices (0); lighter than inner rectrices (1). The outer rectrices of Pygoscelis antarctica and P. adeliae are light in color; in contrast, outer rectrices are indistinguishable from inner rectrices in all other penguins.
- 50
White line connecting leading edge of flipper with white belly (venter): absent (0); present (1).
- 51
Flipper (ala[membrum thoracicum]), upperside, light notch at base: absent (0); present (1). In some Spheniscus, the shoulder has a small irregular mark that contrasts with the dark dorsal surface of flipper.
- 52
Leading edge of flipper (ala[membrum thoracicum]), pattern of upperside: black (0); white (1).
- 53
Leading edge of flipper (ala[membrum thoracicum]), pattern of underside: white (0); incompletely dark (1); completely dark and wide (2).
- 54
Flipper (ala[membrum thoracicum]), underside, dark elbow patch: absent (0); present (1).
- 55
Flipper (ala[membrum thoracicum]), underside, tip pattern: immaculate (0); patchy, in variable extent (1); small circular dot present (2).
Immature plumage
- 56
White eyebrow (supercilium): absent (0); present (1).
- 57
Throat pattern (jugulum): black (0); mottled (1); white (2); brown (3).
- 58
Flanks (ilia), dark lateral band: absent (0); present (1).
Natal plumage (neossoptilus)
- 59
Chicks hatch almost naked: no (0); yes (1). Newly hatched chicks are only thinly covered by sparse down in Aptenodytes (state 1), while in all other penguins the chick hatches densely covered by thick down (Watson, 1975; Martínez, 1992).
- 60
Dominant color pattern of first down: pale gray (0); distinctly brown (1); bicolored, dark above and whitish bellow (2); uniformly blackish gray (3).
- 61
Dominant color pattern of second down: pale gray (0); distinctly brown (1); bicolored, dark above and whitish bellow (2); uniformly blackish gray (3). This character overlaps only partially with the previous one, suggesting that the pattern of each downy coat is ontogenetically distinct.
- 62
Chick, second down, collar: absent (0); present (1). This is a special trait of some Spheniscus species. For us, it merits a separate scoring given the restricted body area that it represents, instead of being an extra state of the previous character.
Feet (podoteca)
- 63
Feet (pedes), dorsal color: dark (0); pinkish (1); orange (2); white‐flesh (3); blue (4).
- 64
Feet (pedes), dark soles: absent (0); present (1). Pygoscelis adeliae and Eudyptes species present a dark undersurface of feet that vividly contrast against the ground color of feet. After O'Hara (1989, character 4).
- 65
Feet (pedes), unguis digiti: flat (0); compressed (1). Character from Watson (1883) that we added to the integumentary dataset of Giannini and Bertelli (2004).
Breeding characters
- 66
Clutch size: two eggs (0); one egg (1). After O'Hara (1989, character 21).
- 67
Incubatory sac: absent (0); present (1). Aptenodytes species incubate their single egg between their feet and a marsupium‐shaped fold of skin in the belly.
- 68
Nest: no nest, incubation over the feet (0); nest placed underground, either burrowed on sand or inside natural hollow or crack (1); open nest, a shallow depression on bare ground or in midst of vegetation (2). After O'Hara (1989, character 22).
- 69
Size of first egg relative to the second egg: similar (0); dissimilar, first smaller (1); dissimilar, second smaller (2). In Eudyptes, the size difference between the first and second egg (20–30%) is the greatest found among all birds, and typically leads to the death of the first hatched chick (Martínez, 1992). In Pygoscelis papua and P. adeliae, the second egg is the smaller one, but by a difference much less marked than in Eudyptes (Watson, 1975). The second chick is comparatively smaller and it usually does not survive. Non‐comparable in one‐egged Aptenodytes.
- 70
Crèche: absent (0); small, 3–6 birds (1); typical, conformed by dozens to hundreds of immatures (2). The large crèches are aggregations of immatures that occur in Aptenodytes, Pygoscelis and Eudyptes, in response to the cold or predation. In other genera this behavior is much less accentuated, and crèches are only modest groupings (state 1). Finally, the solitary habits of Megadyptes preclude contact between chicks, so crèches are totally absent (Martínez, 1992).
- 71
Eggs, shape: oval (0); conical (1); spherical (2). Our terminology follows Romanoff and Romanoff (1949, p. 60).
- 72
Ecstatic display: absence (0); presence (1). This particular behavior ocurrs in male penguins to attract a mate or state their territory. These birds shake their heads using vocalizations or “ecstatic calls” in order to attract a female (Davis and Darby, 1990). Our coding differs from Paterson et al. (1995, character 34) in that the authors distinguished an extra state (a minor variant of the display pattern) in Eudyptula minor.
Osteological characters
Cranium
- 73
Basioccipital, subcondylar fossa (os basioccipitale, fossa subcondylaris) (Fig. 7): absent to shallow (0); deep (1). O'Hara (1989) discarded this character because he found no variation within penguins. We agree, but we include it here because it provides structure to our outgroups. The presence of a deep subcondylar fossa in some procellariforms was noted by Pycraft (1899, as “precondylar”).
- 74
Supraoccipital, paired grooves for the exit of the venaeoccipitalis externae (os supraoccipitale,sulci venae occipitalis externae) (Fig. 8): poorly developed (0); deeply excavated (1).
- 75
Frontal, salt‐gland fossa (os frontale, fossa glandulae nasalis), lateral supraorbital shelf of bone (Fig. 9): absent (0); present (1). After O'Hara (1989, character 10). We observed some variation in the degree of development among the specimens possessing a shelf, but do not consider this to be justification for an additional character state.
- 76
Squamosal, temporal fossa (os squamosum, fossa temporalis), size (Fig. 9): less extensive, both fossae separated by considerable cranial surface (at least the width of the cerebellar prominence) (0); more extensive, fossae meeting or nearly meeting at midline of the skull (1). Zusi (1975, p. 72) noted that the condition coded in state 1 is related to the greatest development of the m. adductor mandibulae externus.
- 77
Squamosal, temporal fossa (os squamosum, fossa temporalis), depth of posterior region (Fig. 10): flat (0); shallow (1); greatly deepened (2). Pycraft (1898, p. 963) discussed the degree of development of the crista temporalis (squamoso‐parietal wings of Pycraft) which forms the posterior wall of the fossa.
- 78
Squamosal (os squamosum), development of the lateral foramen rami occipitalis arteriae ophthalmicae externae in the caudoventral area of the fossa temporalis (near the crista nuchalis) (Fig. 10): small or vestigial (0); well‐developed (1).
- 79
Orbit (orbita), fonticuli orbitocraniales (Fig. 10): small or vestigial (0); broad and conspicuous openings (1). These apertures in the caudal wall of the orbit communicate with the cranial cavity. O'Hara (1989) stated that the size of the f. orbitocraniales varies with age and across species and that further study is required, and so did not included this character in his 1989 study. However, we note that in Aptenodytes these openings are invariably present and are definitely larger than in any other penguin or outgroup taxon (our state 1).
- 80
Ectethmoid (os ectethmoidale): absent (0); weakly developed, widely separate from the lacrimal (1); well developed, contacting or fused to the lacrimal (2). Variation of the ectethmoid in outgroups and penguins were described by Cracraft (1968).
- 81
Lacrimal (os lacrimale) (Fig. 11): unperforated (0); perforated (1). References: O'Hara (1989, character 11). Perforated lacrimals in penguins were described by Beddard (1898) and Lowe (1933) and by Cracraft (1968) in Diomedea.
- 82
Lacrimal (os lacrimale): reduced, concealed in dorsal view (0); exposed in dorsal view, without a prominent orbital process (proc. orbitalis) (1); highly exposed dorsally, with a prominent orbital process (2). Pycraft (1899, p. 387) described condition 2 in procellariiforms. O'Hara (1989) discarded a two‐state character dealing with the dorsal exposure of the lacrimal, arguing a possibly continuous variation. When considering our outgroup taxa, it becomes clear that the variation fits into three different conditions (our coding).
- 83
Lacrimal, dorsal border: closely applied to the frontal (0); separated by a wide split from the frontal (1). Pycraft (1899, p. 404) described condition (1) in outgroup taxa (Oceanitidae).
- 84
Nasal cavity, external naris (cavum nasi, apertura nasi ossea), caudal margin (Fig. 11): extended caudal to the rostral margin of the fenestra antorbitalis (0); not extended caudal to the rostral margin of the fenestra antorbitalis (1). O'Hara's (1989) character 5.
- 85
Nasal cavity (cavum nasi, pila supranasalis; Fig. 12): slender, slightly constricted laterally (0); wide throughout its length (1). O'Hara's (1989) character 6.
- 86
‘‘Basitemporal plate’’ (lamina parasphenoidalis), dorsoventral position with respect to the occipital condyle: ventral to the level of the condyle (0); at the level of the condyle (1); dorsal to the level of the condyle, surface depressed (2).
- 87
Basipterygoid process (Fig. 7): absent (0); vestigial or poorly developed (1); well developed (2). Huxley (1867, p. 430) and Pycraft (1899, p. 403) described the variation of this character in the outgroup taxa.
- 88
Eustachian tubes (tuba auditiva) (Fig. 7): open or very little bony covering near the posterior end of the tube (0); mostly enclosed by bone (1). O'Hara (1989) discarded this character, arguing that it is variable intraspecifically and also attributing variation to preservation. Our scoring is based on specimens that did not present evidence of damage. Therefore, we considered this variation as phylogenetically informative and included the character.
- 89
Pterygoid (os pterygoideum), shape (Fig. 7): elongated (0); broad, triangular‐shaped (1). O'Hara (1989) reported that a broad and flat pterygoid in penguins is a unique condition among birds, but did not include this character because it is invariant within penguins. We include this character as it represents a well‐known synapomorphy of Sphenisciformes (Watson, 1883; Pycraft, 1898).
- 90
Palatine (os palatinum), lamella choanalis(Fig. 13): curved and smooth plate, sligthly differentiated from main palatine blade (0); ridged, distinct from main blade by a low keel (1); extended vertically ventrally forming the crista ventralis (2).
- 91
Vomer (vomer) (Fig. 13): laterally compressed, vertical laminae and free from palatines (0); ankylosed with palatines (1). Pycraft, 1898, p. 973, 1899, p. 391).
- 92
Facial foramen (ossa otica, fossa acustica interna, foramen n. facialis): absent (0); present (1). Saiff (1974, 1976) described the presence of the facial foramen in procellariiforms (1974, p. 220) and Aptenodytes among penguins (1976, p. 758). We agree with Saiff's observations for these taxa (i.e., the foramen is present in all procellariiforms and is absent in all penguins save Aptenodytes), and also find this foramen to be present in Gavia.
- 93
Jugal arch (arcus jugalis), bar shape in lateral view (Fig. 14): straight (0); slightly curved (1); ventrally bowed (2); strongly curved, sigmoid shape (3). O'Hara (1989) discarded this character in view of its possibly continuous nature, after attempting to accommodate all variation into two states. However, Zusi (1975, p. 70) discriminated three states of jugal curvature in penguins—conditions that we were able to corroborate in our sample. A third condition, straight (0), is typical of the outgroup taxa.
- 94
Jugal arch (arcus jugalis), dorsal process: absent (0); present (1). This pointed process is located on the caudal end of the jugal, adjacent to the condyle for articulation with the quadrate.
- 95
Premaxilla (os premaxillare), naso‐premaxillary suture (Fig. 12): visible (0); obliterated (1). Our character refers to the naso‐premaxillary suture and should not be confused with O'Hara's (1989, p. 108) character 10 (discarded by the author), which refers to the interpremaxillary suture.
- 96
Quadrate, otic process (os quadratum, proc. oticus), ventral border, process for attachment of the M. adductor mandibulae externus, pars profunda (Hofer, 1950) (Fig. 15): absent (0); present, as a ridge (1); presence, as a tubercle (2). The process is located on the rostrolateral surface of the otic process of the quadrate, immediately ventral to the squamosal articulation (capitulum squamosum).
- 97
Tomial edge (crista tomialis): plane of tomial edge approximately at the level of the ‘‘basitemporal plate’’ (lamina parasphenoidalis) (0); dorsal to the level of the basitemporal plate (1). This character was identified by Zusi (1975, p. 69), who interpreted it to differences in feeding habits.
- 98
Mandible, coronoid process (mandibula, processus coronoideus), position on the dorsal margin of the mandible with respect to the caudal fenestra (fenestra caudalis mandibulae) (Fig. 16): markedly anterior (0); on the anterior end of the fenestra (1); posterior (2). O'Hara (1989, p. 108) discarded this allegedly continuous character. We found that all but one species, Eudyptes chrysocome, could be clearly scored. Specimens of E. chrysocome showed a coronoid process varying in position between states 1 and 2 and was therefore scored as polymorphic.
- 99
Mandible, rostral fenestra (mandibula, fenestra rostralis mandibulae) (Fig. 16): imperforated or small opening (0); well developed (1). Our scoring differs from O'Hara's (1989) character 8, in which small openings were considered the same condition as the large and distinct fenestrae (e.g., in Aptenodytes). In our study, we considered that small openings and the imperforated condition are often indistinguishable or highly variable (probably in relation to age or individual size), whereas the large fenestra found in taxa exhibiting state 1 are always very conspicuous.
- 100
Mandible, caudal fenestra (mandibula, fenestra caudalis mandibulae) (Fig. 16): open, can be seen through from the medial or lateral aspects (0); nearly or completely concealed by the os spleniale medially, i.e., fenestra not visible in the medial aspect (1). After O'Hara (1989, character 9).
- 101
Mandible, mandibular ramus (mandibula, ramus mandibulae) (Fig. 16): depth subequal over entire ramus (0); noticably deepening at midpoint (1). Zusi (1975, p. 69) described this character, considering two states in Eudyptes. We found no substantial difference among Eudyptes, so we scored all species (plus Pygoscelis adeliae, as in Zusi, 1975) as having state 1.
- 102
Mandible, dentary (mandibula, os dentale), posterior border divided into (Fig. 16): one limb (0); two limbs (1). The presence in Eudyptes of a dentary with pronounced sinuous border was noted by Pycraft (1899, p. 965).
- 103
Mandible, dentary, length of dorsal edge (ossa mandibulae, os dentale, margo dorsalis) relative to mandibular ramus length in lateral view: markedly more than half the length of ramus (0); approximately half the length of ramus (1).
- 104
Mandible, articular, medial process (os articulare, proc. medialis mandibulae) (Fig. 17): not hooked (0); hooked (1). The tip of the process is curved rostromedially in penguins. This character was scored as nonapplicable for Gavia because the articular does not project medially to form a medial process in these birds.
- 105
Mandible, angular, retroarticular process (mandibula, os angulare, proc. retroarticularis), aspect in dorsal view in relation to the articular area with the quadrate (area between the lateral condyle [condylus lateralis] and medial condyle [condylus medialis]) (Fig. 17): broad, approximately equal to the articular area (0); moderately long, narrower than the articular area (1); very long, longer and narrower than the articular area (2). Coded from Zusi (1975, p. 61). We code this character non‐comparable in outgroups because the angular does not project caudally to form a retroarticular process (see character 106).
- 106
Mandible, angular (mandibula, os angulare), aspect in dorsal view (Fig. 17): sharply truncated caudally (0); caudally projected, forming the retroarticular process (proc. retroarticularis) (1). The condition in the outgroup (state 0) was described by Pycraft (1899, p. 393).
- 107
Mandible, caudal fossa (mandibula, fossa caudalis): shallow (0); deep (1).
- 108
Atlas (atlas), proc. ventralis corporis(Fig. 18): absent or sligthly developed (0); well developed, high and prominent ridge on the ventral surface of the corpus atlantis (1).
- 109
Transition to free cervicothoracic ribs (costae incompletae) arrives at: 13th cervical vertebrae: (0); 14th cervical vertebrae (1); 15th cervical vertebrae (2). An alternative way to describe this variation was used by Pycraft (1899, p. 406), who stated that procellariiforms bear three “cervico‐dorsal vertebrae” (rather informal terminology by Newton, 1896), as opposed to penguins and gaviiforms that bear two. In fact, all these taxa have 15 cervical vertebrae, with the last two (state 0) or three (state 1) provided with a free rib.
- 110
Cervical vertebrae (vertebrae cervicales), elongated dorsal process (processus spinosus) on the sixth cervical vertebra: absent (0); present (1). The cervical vertebrae bear an elongated dorsal process from the second vertebra to the fifth (state 0) or sixth (state 1) vertebrae. Elongated is defined here as the process being more dorsoventrally developed than craniocaudally developed.
- 111
Cervical vertebrae (vertebrae cervicales), transverse processes (processus transversus vertebrae) in last five cervical vertebrae: not elongated laterally (0); greatly elongated laterally (1).
- 112
Cervical vertebrae (vertebrae cervicales), transverse processes (processus transversus vertebrae) of vertebrae 12–13: laterally oriented (0); deflected dorsally (1).
- 113
Caudal vertebrae (vertebrae caudales): seven (0), eight (1), nine (2). Watson (1883, p. 19) reported the variation in the number of caudal vertebrae in penguins. In our coding, we include the free caudal vertebrae (vertebrae caudales liberae) and pygostile (pygostylus); we consider the pygostyle as one caudal element.
- 114
Ribs, uncinate processes (costae, proc. uncinatus) (Fig. 19): elongate, narrow (0), wide, spatulate (1), wide, bifurcated (2).
- 115
Sternum, external spine (sternum, spina externa rostri) (Fig. 20): present (0); absent (1). This character was described by Watson (1883) who referred to it as “the episternum”, Zusi (1975) who referred to it as the ventral manubrial spine and by O'Hara (1989, character 13).
- 116
Sternum, furcular facet (sternum, facies articularis furculae) projects as a distinctive process (Fig. 20): absent (0); present (1). The carina (carina sterni) of procellariiforms exhibit a furcular facet developed as a wide or stubby process.
- 117
Furcula, furcular process (furcula, apophysis furculae): absent (0); projecting blade (1); knob‐like process (2); long process (3).
- 118
Scapula, blade, caudal half (scapula, corpus scapulae, extremitas caudalis): flattened and usually blade‐like (0); paddle‐shaped (1). A wide paddle‐shaped scapula in penguins has been described by Baumel and Witmer (1993, p. 96).
- 119
Coracoid, medial margin (coracoideum, margo medialis) fenestrate lamella (O'Hara, 1989) (Fig. 21): complete (0); incomplete (1); absent (2). This character was described by O'Hara (1989, character 14) as the fenestrate lamella. Zusi (1975, p. 61) also described this character for penguins, with the same distribution as O'Hara (1989) and the present study. In some penguins (genus Aptenodytes, Pygoscelis, Eudyptula), the medial margin of the coracoid forms an incomplete ridge that does not reach the procoracoid process dorsally, leaving a wide notch between the medial margin ridge and the procoracoid process (state 1). In most penguins, the medial margin projects dorsally reaching the procoracoid process so that a large fenestra is enclosed in the substance of the medial ridge (state 0).
- 120
Coracoid, sternal end (coracoideum, extremitas sternalis coracoidei) (Fig. 21): wide (0); narrow (1). This conditions are related to the development of the lateral process (processus lateralis).
- 121
Forelimbs (ossa alae), strongly flattened: absent (0); present (1). The presence in penguins of a flattened wing has been reported as a synapomorphy of the group by O'Hara (1989).
- 122
Humerus, head (humerus, caput humeri), very developed and reniform, ventrally directed (Mayr, 2005, character 36): absent (0); present (1).
- 123
Humerus, pneumatic fossa (humerus, fossa pneumotricipitalis ventralis), aspect: small with pneumatic foramina (0); without pneumatic opening, moderate size (1); without pneumatic opening, great size and deep fossa (2).
- 124
Humerus, pneumatic fossa (humerus, fossa pneumotricipitalis ventralis), subdivided into cavities (Fig. 22): single (0); partially divided (1); bipartite (2).
- 125
Humerus, deltoid crest, impression for attachment of pectoral muscle (humerus, crista deltopectoralis, impressio M. pectoralis): superficial or shallow groove (0); deep oblong fossa (1).
- 126
Humerus, development of dorsal supracondylar process (humerus, proc. supracondylaris dorsalis): absent (0); compact tubercle (1); very long process (2).
- 127
Humerus (humerus), distal end, ventral border with ‘‘trochlear process’’ (caudal‐most processus‐like crest at the epicondylus ventralis caudally bordering the sulcus humerotricipitalis) (Fig. 23): absent (0); present (1). This feature is discussed by O'Hara (1989, character 15 (and Mayr (2005, character 37).
- 128
Humerus (humerus), proximal‐most ‘‘trochlear process’’ (Fig. 23): extends beyond the humeral shaft (0); does not extend beyond the humeral shaft (1). This feature is discussed by O'Hara (1989, character 15).
- 129
Phalanges of manus, free pollex (ossa digitorum manus, phalanx digiti alulae): absence (0); presence (1). According to Watson (1883, p. 38) and Pycraft (1907, pp. 19–20, figs 8, 9), the pollex is present in penguins but completely fused to the proximal phalanx of the digiti II (phalanx proximalis digiti majoris). This condition has been also described for Sphenisciformes by O'Hara (1989).
- 130
Phalanges of manus, phalange digit III (ossa digitorum manus, phalanx digiti minoris), proximal process: absent (0); present (1).
- 131
Phalanges of manus (ossa digitorum manus, phalanxs): short (0); long (1).
- 132
Pelvis (pelvis et os coxae), size of ilio‐ischiatic fenestra (foramen ilioischiadicum) in relation to acetabulum (foramen acetabuli) (Fig. 24): smaller (0); similar to larger (1). This feature is discussed by O'Hara (1989, character 16).
- 133
Pelvis, ischio‐pubic fenestra (pelvis et os coxae, fenestra isquiopubica): very wide and closed at its caudal end (0); slit‐like and open at its caudal end (1).
- 134
Ischium (ischium), most caudal extent in relation to postacetabular ilium (ala postacetabularis illi) (Fig. 24): ischium shorter than postilium (0); ischium projects slightly beyond the postilium (1); ischium produced far backward postilium (2). Pycraft (1898, 1899) described the variation in the postacetabular ilium in Sphenisciformes (1899, p. 399) and Procellariiformes (1898, p. 978).
- 135
Femur, patella (femur, patella), sulcus m. ambiens(Fig. 25): shallow groove (0); deep groove (1); perforated (2). The tendon of m. ambiens may groove the patella superficially (state 0), deeply (state 1), or actually perforate its cranial surface (state 2) (Baumel and Witmer, 1993, p. 109). We consider this character non‐comparable for our outgroups because the patella is fused to the patellar crest (crista patellaris) of the tibiotarsus. According to Barnett and Lewis (1958), an elongated patellar crest in Pelecanoides represents fusion of the femoral patella with tibiotarsus. This character is modified from O'Hara's (1989) and character 18 and Mayr's (2005) character 43 (we distinguished three instead of two states).
- 136
Tibiotarsus, patellar crest (tibiotarsus, cristae cnemiales): greatly enlarged (0); slightly developed (1). Pycraft (1899) and Mayr (2005, character 44) described this condition for procellariiforms.
- 137
Tibiotarsus (tibiotarsus), tuberositas poplitea, extents to the medial cotyla (cotyla medialis): absent (0); present (1).
- 138
Tarsometatarsus (tarsometatarsus): slender, proximodistal length much greater than mediolateral width (0); very stout, mediolateral width nearly equal to proximodistal length.
- 139
Tarsometatarsus (tarsometatarsus), blood vessel foramen located on fossa para hypotarsalis medialis(Fig. 26): absent (0); present (1). This opening is continuous with the inner proximal foramen (foramina vascularia proximalia). This character is modified from O'Hara's (1989) character 20. Zusi also (1975, p. 61) discussed the variable presence of an equivalent accessory medial foramen in penguins.
- 140
Tarsometatarsus, proximal foramen (tarsometatarsus, foramina vascularia proximalia) opens lateral to medial crest (crista medialis hypotarsi) (Fig. 27): absent (0); present (1). This character is modified from O'Hara (1989, character 19). Zusi (1975, p. 61) described this opening as the inner proximal foramen.
- 141
Tarsometatarsus, hypotarsus, tendinal canals (tarsometatarsus, hypotarsus, canales hypotarsi): present (0); absent (1).
- 142
Tarsometatarsus (tarsometatarsus), tuberositas tibialis: flat (0); raised (1).

Closeups of the palate and basicranium of Macronectes giganteus and Spheniscus humboldti. In Macronectes the eustachian tubes (et) are open (char. 88 [0]), whereas in Spheniscus the tubes are enclosed by bone (char. 88 [1]). In Macronectes, a deep subcondylar fossa is present (sf; char. 73 [1]). In Spheniscus, the pterygoids (pt) are broad (char. 89 [1]), whereas in Macronectes the pterygoids are elongated (char. 89 [0]). The basipterygoid process (bp) is present in Macronectes (char. 87 [2]), whereas it is absent in Spheniscus (char. [0]).

Occipital region of skull of Spheniscus humboldti and Megadyptes antipodes showing the absence (Megadyptes) or presence (Spheniscus) of a deep sulcus venaeoccipitalis externae (voe; char. 74 states [0] and [1], respectively).

Skull of Pygoscelis antarctica and Spheniscus magellanicus in dorsal view. The temporal fossae (ft) are widely separated in P. antarctica (char. 76 [0]) and are more extensive in S. magellanicus (char. 76 [1]). P. antarctica shows a supraorbital shelf of bones lateral to the f. glandulae nasalis (fgn; char. 75 [1]).

Closeups of the caudoventral area of the cranium of Spheniscus demersus and Aptenodytes forsteri in lateral view. S. demersus displays a deep temporal fossa (ft; char. 77 [2]). A. forsteri has a well‐developed foramen for the rami occipitalis arteriae ophthalmicae externae (aoe; char. 78 [1]), and a broad fonticulus orbitocranialis (fo; char. 79 [1]).

Closeups of the caudal rostrum of Spheniscus demersus and Pygoscelis antarctica in lateral view. In Spheniscus, the apertura nasi ossea (an) is located rostrally (char. 84 [1]) with respect to the fenestra antorbitalis (ho). The lacrimal of Pygoscelis antarctica is imperforated (la; char. 81 [0]).

Rostrum of Pygoscelis antarctica and Spheniscus magellanicus in dorsal view. S. magellanicus has a wide pila supranasalis (ps, char. 85 [1]) which is slender in P. antarctica (char. 85 [0]). P. antarctica exhibits a visible naso‐premaxillary suture (nps; char. 95 [0]).

Palate of Spheniscus humboldti and Macronectes giganteus. The palatine (pa) of Macronectes giganteus shows a developed crista ventralis (cv; char. 90 [2]). The vomer (v) is free in S. humboldti (char. 91 [0]), and ankylosed with palatines in M. giganteus (char. 91 [1]).

Skull of Daption capense, Spheniscus demersus, Aptenodytes forsteri, and Eudyptes pachyrrhynchus in lateral view, showing the shapes of the jugal bar (aj): straigth (char. 93 [0]), sligthly curved (char. 93 [1]), ventrally bowed (char. 93 [2]), and strongly curved (char. 93 [0]).

Quadrates in lateral view. In Eudyptes chrysocome, the quadrate displays a tubercle (p, char. 96 [0]) on the otic process (op), whereas in Megadyptes antipodes this process (p) is present as a ridge (char. 96 [1]).

Mandible of Aptenodytes forsteri and Eudyptes pachyrrhynchus in lateral view. In Aptenodytes forsteri, the coronoid process (cp) is located on the anterior end of the caudal fenestra (cf; char. 98 [1]), whereas in Eudyptes pachyrrhynchus cp is posterior (char. 98 [2]). Aptenodytes has a perforated rostral fenestra (rf; char. 99 [1]). The posterior border of the dentary (d) is divided in one limb in Aptenodytes (char. 102 [0]), whereas in Eudyptes has two limbs (char. 102 [1]). In Eudyptes, the mandibular ramus is deep at midpoint (char. 101 [1]).

Closeups of the articular area of right mandible in dorsal view. In Puffinus griseus, the medial process (mp) is not hooked (char. 104 [0]), and the retroarticular process (rp) is absent (char. 106 [0]). In Aptenodytes patagonicus, a hooked medial process is present (char. 104 [1]). The retroarticular process (rp) is broad in Aptenodytes patagonicus (char. 105 [0]), moderately long in Eudyptes pachyrrhynchus (char. 105 [1]), and very long in Spheniscus humboldti (char. 105 [2]). Additional references: cm, medial condyle; cl, lateral condyle.

Atlas, cranial view, showing a well developed ventral process (pv) in Aptenodytes forsteri (char. 108 [1]), and a slightly developed process in Spheniscus humboldti (char. 108 [0]).

Ribs of Diomedea melanophris, Eudyptes robustus and Pygoscelis papua in lateral view, showing elongated (char. 114 [0]), wide (char. 114 [1]), and bifurcated (char. 114 [1]) uncinate process (up).

Sternum of Daption capense and Aptenodytes forsteri in cranial view. In Daption, the sternum has a furcular facet (ff) developed as a wide process (char. 116 [1]), and the external spine is present (es; char. 115 [0]).

Right coracoids of Pelecanoides urinatrix, Megadyptes antipodes and Eudyptula minor in ventral view. The coracoidal fenestra (cf) is open in Eudyptula (char. 119 [1]), and enclosed by fenestrate lamella (f1) in Megadyptes (char. 119 [0]). Pelecanoides has a wide sternal end (lp: lateral process; char. 120 [0]), in contrast to the narrow condition in Megadyptes and Eudyptula (char. 120 [1]). Fs: foramen nervi supracoracoidei (Mayr, 2005).

Closeups of proximal end of the left humerus of Aptenodytes forsteri and Pelecanoides urinatrix in caudal view. The humerus of Aptenodytes has a pneumatic fossa (pf) subdivided into well‐developed cavities (char. 124 [2]).

Distal end of right humerus of Spheniscus magellanicus and Pygoscelis adeliae in caudal view. In Pygoscelis, the proximal trochlear process (ptp) extends beyond the humeral shaft (char. 128 [0]).

Pelvis of Eudyptula minor and Aptenodytes forsteri in lateral view. In Eudyptula, the ilio‐ischiatic fenestra (fi) is wide (char. 132 [1]) in relation to the acetabulum (fa); in contrast the fenestra in Aptenodytes is relatively small (char. 132 [0]). The ischium (is) of Eudyptula is shorter (char. 134 [0]) than the postacetabular ilium (po), and it projects beyond in Aptenodytes (char. 134 [1]).

Patella of Pygoscelis antarctica, Eudyptes schlegeli, and Aptenodytes forsteri in lateral view, showing different grades of development of the sulcus m. ambiens (sa). This groove is shallow in Aptenodytes (char. 135 [0]), deep in Eudyptes (char. 135 [1]), and perforated in Pygoscelis (char. 135 [2]).

Left tarsometatarsus of Spheniscus magellanicus and Pygoscelis adeliae in medial view. In Spheniscus, a proximal foramen (f) is present close to the fossa parahypotarsalis medialis (mf) (char. 139 [1]).

Left tarsometatarsus of Eudyptes robustus and Pygoscelis papua, plantar aspect. Pygoscelis has a proximal foramen (f) lateral to the medial crest (cm) (char. 140 [1]).
Myological characters
- 143
M. latissimus dorsi, pars cranialis, accessory slip: absent (0); present (1).
- 144
M. latissimus dorsi, pars cranialis and pars caudalis: separated (0); fused (1).
- 145
M. latissimus dorsi, pars metapagialis, development: wide (0); intermediate (1); narrow (2).
- 146
M. serratus profundus, cranial fascicle: absent (0); present (1).
- 147
M. deltoideus, pars propatagialis, subdivision in superficial and deep layers: undivided (0); divided (1).
- 148
M. deltoideus, pars major: triangular or fan‐shaped (0); strap‐shaped (1).
- 149
M. deltoideus, pars major, caput caudale: short (0); intermediate (1); long (2).
- 150
M. deltoideus, pars minor, origin on the clavicular articulation of the coracoid: absent (0); present (1).
- 151
M. ulnometacarpalis ventralis: absent (0); present (1).
- 152
M. iliotrochantericus caudalis: narrow (0); wide (1).
- 153
M. iliofemoralis, origin: tendinuous (0); mostly tendinuous (1); mostly fleshy (2); totally fleshy (3).
- 154
M. flexor perforatus digitis IV, rami II–III: free (0); fused (1).
- 155
M. flexor perforatus digitis IV, rami I–IV: free (0); fused (1).
- 156
M. flexor perforatus digitis IV, insertion of middle rami: on phalanx 3 (0); on phalanx 4 (1).
- 157
M. latissimus dorsi, pars caudalis, additional origin from dorsal process of vertebrae (spinous process of Schreiweis, 1982): absent (0); present (1).
Digestive tract
- 158
Mouth, oral mucosa (bucca, tunica mucosa oris), papillae on the medial surface of the lower jaw (ramus mandibularis) at the level of the rictus: absent (0); present (1). We follow Watson (1883), who identified a group of dorsocaudally directed buccal papillae and described its variation within penguins.
Appendix 3
Morphological data matrix. Note: polymorphic entries: a = (01), b = (03), c = (12). Inapplicable data = “–” and missing data = ”?”.
|
External anatomy and behavioral data matrix (Integument–Breeding) |
||||||||
|---|---|---|---|---|---|---|---|---|
| 1 | 2 | 3 | 4 | 5 | 6 | 7 | ||
| 0123456789 | 0123456789 | 0123456789 | 0123456789 | 0123456789 | 0123456789 | 0123456789 | 012 | |
| G. stellata | 1000100000 | 0000b00020 | 00000‐ ‐ ‐ ‐ | ‐ ‐ ‐ ‐ ‐‐ ‐ ‐ ‐ | ‐ ‐ ‐ ‐ ‐‐ ‐ ‐ ‐ | ‐ ‐ ‐ ‐ ‐ ‐ ‐ 0 | 1100000020 | 000 |
| D. melanophrys | 0100000004 | 0221002000 | 00000‐ ‐ ‐ ‐ | ‐ ‐ ‐ ‐ ‐‐ ‐ ‐ ‐ | ‐ ‐ ‐ ‐ ‐‐ ‐ ‐ ‐ | ‐ ‐ ‐ ‐ ‐ ‐ ‐ 0 | 000300102‐ | 000 |
| P. palpebrata | 0100000000 | 0000002000 | 00000‐ ‐ ‐ ‐ | ‐ ‐ ‐ ‐ ‐‐ ‐ ‐ ‐ | ‐ ‐ ‐ ‐ ‐‐ ‐ ‐ ‐ | ‐ ‐ ‐ ‐ ‐ ‐ ‐ 0 | 000300102‐ | 000 |
| M. giganteus | 0100100005 | 03‐2232050 | 00000‐ ‐ ‐ ‐ | ‐ ‐ ‐ ‐ ‐‐ ‐ ‐ ‐ | ‐ ‐ ‐ ‐ ‐‐ ‐ ‐ ‐ | ‐ ‐ ‐ ‐ ‐ ‐ ‐ 0 | 330b00102‐ | 000 |
| D. capense | 0100100000 | 0000002000 | 00000‐ ‐ ‐ ‐ | ‐ ‐ ‐ ‐ ‐‐ ‐ ‐ ‐ | ‐ ‐ ‐ ‐ ‐‐ ‐ ‐ ‐ | ‐ ‐ ‐ ‐ ‐ ‐ ‐ 0 | 000001102‐ | 000 |
| P. aequinoctialis | 0100000006 | 0403?02000 | 00000‐ ‐ ‐ ‐ | ‐ ‐ ‐ ‐ ‐‐ ‐ ‐ ‐ | ‐ ‐ ‐ ‐ ‐‐ ‐ ‐ ‐ | ‐ ‐ ‐ ‐ ‐ ‐ ‐ 0 | 110001101‐ | 000 |
| P. griseus | 0100000000 | 0000002000 | 00000‐ ‐ ‐ ‐ | ‐ ‐ ‐ ‐ ‐‐ ‐ ‐ ‐ | ‐ ‐ ‐ ‐ ‐‐ ‐ ‐ ‐ | ‐ ‐ ‐ ‐ ‐ ‐ ‐ 0 | 000101101‐ | 000 |
| P. incerta | 0100000000 | 0000002000 | 00000‐ ‐ ‐ ‐ | ‐ ‐ ‐ ‐ ‐‐ ‐ ‐ ‐ | ‐ ‐ ‐ ‐ ‐‐ ‐ ‐ ‐ | ‐ ‐ ‐ ‐ ‐ ‐ ‐ 0 | ??0101101‐ | 000 |
| O. leucorrhoa | 0100000000 | 0000002000 | 00000‐ ‐ ‐ ‐ | ‐ ‐ ‐ ‐ ‐‐ ‐ ‐ ‐ | ‐ ‐ ‐ ‐ ‐‐ ‐ ‐ ‐ | ‐ ‐ ‐ ‐ ‐ ‐ ‐ 0 | 000001101‐ | 000 |
| O. oceanicus | 0100000000 | 0000002000 | 00000‐ ‐ ‐ ‐ | ‐ ‐ ‐ ‐ ‐‐ ‐ ‐ ‐ | ‐ ‐ ‐ ‐ ‐‐ ‐ ‐ ‐ | ‐ ‐ ‐ ‐ ‐ ‐ ‐ 0 | 110001101‐ | 000 |
| P. desolata | 0100000007 | 0504342000 | 00000‐ ‐ ‐ ‐ | ‐ ‐ ‐ ‐ ‐‐ ‐ ‐ ‐ | ‐ ‐ ‐ ‐ ‐‐ ‐ ‐ ‐ | ‐ ‐ ‐ ‐ ‐ ‐ ‐ 0 | a00401101‐ | 000 |
| P. urinatrix | 0100000000 | 00‐0002000 | 00000‐ ‐ ‐ ‐ | ‐ ‐ ‐ ‐ ‐‐ ‐ ‐ ‐ | ‐ ‐ ‐ ‐ ‐‐ ‐ ‐ ‐ | ‐ ‐ ‐ ‐ ‐ ‐ ‐ 0 | 000401101‐ | 000 |
| E. minor | 0110120000 | 0000001051 | 1111100 ‐ ‐ | ‐030000010 | 0200001000 | 1010020200 | 1213010010 | 101 |
| A. forsteri | 1000201000 | 0000000101 | 1111100 ‐ ‐ | ‐000000100 | 1110001010 | 0000000001 | 000000110‐ | 211 |
| A. patagonicus | 1000101000 | 0000000101 | 1111100 ‐ ‐ | ‐000000100 | 1110001010 | 0001110001 | 110000110‐ | 211 |
| P. antartica | 0100210000 | 0000000121 | 1111100 ‐ ‐ | ‐010000040 | 0000010011 | 0001110100 | 0001000020 | 221 |
| P. papua | 0100210001 | 0100110111 | 1111100 ‐ ‐ | ‐000100000 | 0000010111 | 0011010000 | 0202000022 | 221 |
| P. adeliae | 0100310001 | 0100110141 | 1111100 ‐ ‐ | ‐100100000 | 0000010010 | 0001010200 | 0101100022 | 221 |
| M. antipodes | 0100120002 | 0111000131 | 1111110 ‐ ‐ | ‐021000020 | 0000001010 | 1010100200 | 1101010020 | 001 |
| E. c. moseleyi | 0100120101 | 0111020121 | 1111111121 | 0200000000 | 0000001010 | 0001110100 | 2201110021 | 201 |
| E. c. chrysocome | 0100120111 | 0111020121 | 1111111121 | 0200000000 | 0000001010 | 0001110100 | 2201110021 | 201 |
| E. chrysolophus | 0100120121 | 0111020121 | 1111111122 | 1000000000 | 0000001110 | 0001110100 | 2201110021 | 201 |
| E. schlegeli | 0100120121 | 0111020121 | 1111111122 | 1010000010 | 0000001110 | 0001110100 | 2201110021 | 201 |
| E. pachyrrhynchus | 0100120101 | 0111020121 | 1111111011 | 0000000000 | 0000001010 | 0001110200 | 2201110021 | 201 |
| E. robustus | 0100120121 | 0111020121 | 1111111011 | 0000000000 | 0000001010 | 0002110200 | 2201110021 | 201 |
| E. sclateri | 0100120121 | 0111020101 | 1111111000 | 0000000000 | 0000001110 | 0002110200 | 2201110021 | 201 |
| S. demersus | 0211010000 | 1000000001 | 1111100 ‐ ‐ | ‐001032001 | 0001101000 | 0001111100 | 2210010010 | 101 |
| S. humboldti | 0211010000 | 1000000011 | 1111100 ‐ ‐ | ‐001023030 | 0001101000 | 0101110300 | ?210010010 | ?01 |
| S. magellanicus | 0211010000 | 1000000001 | 1111100 ‐ ‐ | ‐001021003 | 0001101000 | 0100111110 | ?210010010 | 201 |
| S. mendiculus | 0211010003 | 1000000001 | 1111100 ‐ ‐ | ‐001013032 | 0001101000 | 0001110310 | ???0010010 | ?01 |
|
Internal anatomy data matrix (Osteology–Myology–Digestive tract) |
|||||||||
|---|---|---|---|---|---|---|---|---|---|
| 7 | 8 | 9 | 10 | 11 | 12 | 13 | 14 | 15 | |
| 3456789 | 0123456789 | 0123456789 | 0123456789 | 0123456789 | 0123456789 | 0123456789 | 0123456789 | 012345678 | |
| G. stellata | 1111100 | 0020001000 | 2010010000 | 0000‐‐0111 | 1000000202 | 00010010‐1 | 00100‐0001 | 001?021?1? | 01??????? |
| D. melanophrys | 1000000 | 2120110200 | 2110000020 | 00000‐0000 | 0001011202 | 00000020‐1 | 01102‐0001 | 000??????? | ????????? |
| P. palpebrata | 1000000 | 2120112200 | 2110000020 | 00000‐000? | ???1011202 | 00000020‐1 | 001?2‐0001 | 000??????? | ????????? |
| M. giganteus | 1101101 | 2120100200 | 2110000000 | 00000‐0100 | 0001011102 | 00012020‐1 | 01102‐0001 | 001??????? | ????????? |
| D. capense | 1101101 | 2120000200 | 2110000000 | 00000‐0000 | 0001011302 | 00010020‐1 | 01102‐0001 | 001??????? | ????????? |
| P. aequinoctialis | 1101111 | 2120000200 | 2110000000 | 10000‐0000 | 0001011302 | 00010020‐1 | 01102‐0001 | 001??????? | ????????? |
| P. griseus | 1101111 | 2120000200 | 2110000000 | a0000‐0000 | 1001011302 | 00010020‐1 | 01102‐0001 | 001??????? | ????????? |
| P. incerta | 0101101 | 2120000200 | 2110000000 | 10000‐000? | ?????????? | ?????????? | ?????????? | ?????????? | ????????? |
| O. leucorrhoa | 0100001 | 2121001100 | 21100000?0 | 10000‐0000 | 0001011302 | 00010020‐1 | 01102‐0001 | 001??????? | ????????? |
| O. oceanicus | 0100001 | 2121001100 | 21100000?0 | 10000‐0000 | 0001011302 | 00010020‐1 | 01102‐0001 | 001??????? | ????????? |
| P. desolata | 1101101 | 2120001200 | 2110000000 | 10000‐0000 | 0001011302 | 00010020‐1 | 01102‐0001 | 001??????? | ????????? |
| P. urinatrix | 0101100 | 2120001200 | 2110000010 | 10000‐000? | ?????????? | ?????????? | ?????????? | ?????????? | ????????? |
| E. minor | 0100200 | 1110002011 | 1001011010 | 0000111000 | 0102110011 | 1112110110 | 1011111011 | 0110021001 | 001000001 |
| A. forsteri | 0100111 | 1010002011 | 0012a01111 | 0000101111 | 1111100011 | 1112210100 | 1001101111 | 1101101101 | 10120011? |
| A. patagonicus | 0100111 | 1010002011 | 0012001111 | 0000101111 | 1111100011 | 1112210100 | 1001101111 | 1101101101 | 101200111 |
| P. antartica | 0110100 | 1010002001 | 0003001120 | 1001101101 | 0111210011 | 1112210100 | 1001121010 | 1111121002 | 11100111? |
| P. papua | 0010110 | 11100020?1 | 0002a01110 | 0001101101 | 0111210011 | 1112210100 | 1001111110 | 1111121102 | 111001111 |
| P. adeliae | 0110110 | 1010002011 | 0003001120 | 0101101001 | 0111110011 | 1112210100 | 1001121010 | 1111121002 | 11100111? |
| M. antipodes | 0010200 | 1110002011 | 0002101020 | 1001111001 | 1101110010 | 1112210100 | 1001111011 | 1110010102 | 11001110? |
| E. c. moseleyi | 0010200 | 1110002011 | 0003102020 | 0111111001 | 0101110a10 | 1112210100 | 1001111111 | 011??????? | ????????1 |
| E. c. chrysocome | 0010200 | 1110002011 | 00031020c0 | 0111111001 | 0101110010 | 1112210100 | 1001111111 | 0110011100 | 000111001 |
| E. chrysolophus | 0010200 | 1110002011 | 0003102020 | 1111111000 | 0101110010 | 1112210100 | 1001111111 | 0110011100 | 000111001 |
| E. schlegeli | 0010200 | 1110002011 | 0003102020 | 1111111001 | 0101110110 | 1112210100 | 1001111111 | 0110011100 | 00011100? |
| E. pachyrrhynchus | 00a0200 | 1110002011 | 0003102020 | 0111111002 | 010111011a | 1112210100 | 1001111111 | 0110011a00 | 00011100? |
| E. robustus | 0010200 | 1110002011 | 1003102020 | 0111111000 | 0101110010 | 1112210100 | 1001111111 | 011??????? | ????????? |
| E. sclateri | 0010200 | 1110002011 | 0003102020 | 111111100? | ???1110010 | 1112210100 | 1001111111 | 011??????? | ????????? |
| S. demersus | 0101200 | 1100112011 | 1001012000 | 0001121101 | 1102110010 | 1112210100 | 1011111111 | 0110010110 | 110311001 |
| S. humboldti | 0101200 | 1100112011 | 1001012000 | 0001121101 | 1102110010 | 1112210110 | 1011111111 | 0110010110 | 11031100? |
| S. magellanicus | 0101200 | 1100112011 | 1001012000 | 0001121101 | 1102110010 | 1112210110 | 1011111111 | 0110010110 | 110311001 |
| S. mendiculus | 0101200 | 1100112011 | 1001012000 | 0001121101 | 1102110010 | 1112210110 | 1011111111 | 0110010110 | 110311001 |
Appendix 4
Accession numbers and authorship of DNA sequences.
1. GenBank accession numbers and authorship of sequences used
12S rDNA: Cooper and Penny (1997): U88006‐7, U88024; Emslie (2003): AY234841, AY236364; García‐Moreno et al. (unpublished): AY139621, AY139623, AY139630; Paterson et al. (1995): X82517‐8, X82520, X82522‐3, X82527, X82533, X82536; Slack et al. (2003): NC_004538; van Tuinen et al. (2000): AF173573, AF173578.
cytochrome b: Horner et al. (unpublished): AF338587, AF338597, AF338601, AF338606, AF338608‐9; Nunn et al. (1996): U48943, U48955; Nunn and Stanley (1998): AF076051‐2, AF076046, AF076060, AF076062, AF076064, AF076068, AF076076, AF076089‐90, U74335, U74350, U74353; Slack et al. (2003): NC_004538; Stanley and Harrison (1999): AF158250.
2. GenBank accession numbers by species and locus. Taxa are listed alphabetically
Aptenodytes forseri: 12S: X82520; cyt‐b: AF338608; Aptenodytes patagonicus: 12S: AY139221; cyt‐b: AY139623; Daption capense: 12S: X82517; cyt‐b: AF076046; Eudyptes chrysocome: 12S: AY139630; cyt‐b: AF076051; Eudyptes chrysolophus: cyt‐b: AF076052; Eudyptes pachyrrhynchus: 12S: U88007, X82522; Eudyptes schlegeli: 12S: U88006; cyt‐b: AF338609; Eudyptula minor: 12S and cyt‐b: NC_004538; Diomedea melanophrys: cyt‐b: U48955; Gavia stellata: 12S: AF173578; cyt‐b: AF158250; Macronectes giganteus: 12S: X82523; cyt‐b: AF076060; Megadyptes antipodes: 12S: X82536; Oceanites oceanicus: cyt‐b: AF076062; Oceanodroma leucorrhoa: cyt‐b: AF076064; Pachyptila desolata: cyt‐b: AF076068; Pelecanoides urinatrix: 12S: X82518; cyt‐b: AF076076; Phoebetria palpebrata: cyt‐b: U48943; Procellaria aequinoctialis: cyt‐b: U74350; Pterodroma incerta: cyt‐b: U74335; Puffinus griseus: 12S: X82533, U88024; cyt‐b: U74353; Pygoscelis adeliae: 12S: AF173573; cyt‐b: AF338587; Pygoscelis antarctica: 12S: AY236364; cyt‐b: AF076089; Pygoscelis papua: 12S: AY234841; cyt‐b: AF076090; Spheniscus humboldti: cyt‐b: AF338597; Spheniscus magellanicus: 12S: X82527; cyt‐b: AF338601; Spheniscus mendiculus: cyt‐b: AF338606.
Appendix 5
List of non‐molecular changes: integument (I), breeding (B), osteology (O), myology (M).
| Groups | Characters | Change |
|---|---|---|
| Sphenisciformes | I External nares (17) | 0 → 1 |
| I Scale‐like feathers (19) | 0 → 1 | |
| I Rhachis contour feathers (20) | 0 → 1 | |
| I Rectrices forming fan (2) | 0 → 1 | |
| I Remiges‐contour feathers different. (22) | 0 → 1 | |
| I Apteria (23) | 0 → 1 | |
| I Molt contour feathers (24) | 0 → 1 | |
| B Crèche (70) | 0 → 12 | |
| B Ecstatic display (72) | 0 → 1 | |
| O Lacrimal expos. in dorsal view (82) | 2 → 1 | |
| O Parasphenoid plate pos. relat. occ. cond. (86) | 1 → 2 | |
| O Eustachian tubes (88) | 0 → 1 | |
| O Pterygoid shape (89) | 0 → 1 | |
| O Palatine interal keel (90) | 2 → 0 | |
| O Jugal arch shape (93) | 0 → 2 | |
| O Quadrate process otic p. (96) | 0 → 1 | |
| O Mandible angular aspect (106) | 0 → 1 | |
| O Ribs uncinate process (114) | 0 → 1 | |
| O Scapular blade (118) | 0 → 1 | |
| O Coracoidal fenestra (119) | 2 → 1 | |
| O Coracoid sternal end (120) | 0 → 1 | |
| O Forelimbs flattened (121) | 0 → 1 | |
| O Humerus head very developed (122) | 0 → 1 | |
| O Humerus pneumatic fossa aspect (123) | 1 → 2 | |
| O Humerus pneumatic fossa subdivided (124) | 0 → 2 | |
| O Humerus deltoid crest m. pect. impres. (125) | 0 → 1 | |
| O Humerus dorsal suprac. proc. develop. (126) | 1 → 0 | |
| O Humerus trochlear process (127) | 0 → 1 | |
| O Phalanges manus free pollex (129) | 1 → 0 | |
| O Phalange digit III (130) | 0 → 1 | |
| O Pelvis ilio‐ischia.fenestra acetab. size (132) | 1 → 0 | |
| O Pelvis ischio‐pubic fenestra (133) | 0 → 1 | |
| O Tibiotarus patellar crest (136) | 0 → 1 | |
| O Tarsometat. width‐length nearly great (138) | 0 → 1 | |
| O Tarsometatarsus tendinal canals (141) | 0 → 1 | |
| Aptenodytes | I Orange‐pink ramicorn (6) | 0 → 1 |
| I Chicks hatch almost naked (59) | 0 → 1 | |
| B Incubatory sac (67) | 0 → 1 | |
| B Nest incubation (68) | 2 → 0 | |
| B Eggs shape (71) | 0 → 1 | |
| O Foramen temporal fossa (78) | 0 → 1 | |
| O Mandible rostral fenestra (99) | 0 → 1 | |
| O Tibiotarus tub. poplitea (137) | 0 → 1 | |
| O Tarsometatarsus tub. tibialis (142) | 1 → 0 | |
| M M. latissimus dorsi metapagialis (145) | 12 → 0 | |
| M M. ulnometacarpalis ventralis (151) | 1 → 0 | |
| Pygoscelis + Eudyptula + Spheniscus+ Megadyptes + Eudyptes Pygoscelis | I Ramicorn inner groove (5) | 0 → 1 |
| O Facial foramen (92) | 1 → 0 | |
| O Mandible dentary length (103) | 0 → 1 | |
| I Feathering maxilla (4) | 1 → 2 | |
| I Axillary patch (45) | 0 → 1 | |
| I Flanks dark pattern into leg (46) | 1 → 0 | |
| B Eggs shape (71) | 0 → 2 | |
| O Tarsometat. for. foss. hipotars. med. (139) | 1 → 0 | |
| M M. deltoideus major caudalis (149) | 1 → 2 | |
| P. papua + P. adeliae | I Ramicorn color (9) | 0 → 1 |
| I Latericorn color (11) | 0 → 1 | |
| I Downy chick bill color (14) | 0 → 1 | |
| I Inmature bill color (15) | 0 → 1 | |
| I White eyering (34) | 0 → 1 | |
| I Flipper underside elbow patch (54) | 1 → 0 | |
| B First‐second egg relative size (69) | 0 → 2 | |
| O Squamosal foramen temporal fossa (78) | 0 → 1 | |
| Eudyptula + Spheniscus + Megadyptes + Eudyptes | I Feet unguis digiti (65) | 0 → 1 |
| O Temporal fossa (77) | 1 → 2 | |
| O Lacrimal perforation (81) | 0 → 1 | |
| O Mandible retroarticular process (105) | 0 → 1 | |
| M Latissimus dorsi cranialis slip (143) | 1 → 0 | |
| M Latissimus dorsi caudalis add. origin (157) | 1 → 0 | |
| Eudyptula + Spheniscus | I Longitudinal grooves culmen (2) | 0 → 1 |
| I External nares (17) | 1 → 0 | |
| I Tail length (48) | 1 → 0 | |
| I Chick second down collar (62) | 0 → 1 | |
| B Nest incubation (68) | 2 → 1 | |
| O Palatine internal keel (90) | 0 → 1 | |
| O Jugal arch dorsal process (93) | 2 → 1 | |
| O Premaxilla naso‐premaxillary suture (95) | 0 → 1 | |
| O Caudal vertebrae (113) | 1 → 2 | |
| O Humerus proximal trochlear process (128) | 0 → 1 | |
| O Pelvis ilio‐ischia.fenestra acetab. size (132) | 0 → 1 | |
| Eudyptula | I Nostril tubes (16) | 0 → 1 |
| I Periocular area color (32) | 0 → 3 | |
| I Throat pattern (38) | 0 → 1 | |
| I Dorsum color (41) | 0 → 2 | |
| I Line connect. leading flipper belly (50) | 0 → 1 | |
| I Leading flipper upper. pattern (52) | 0 → 1 | |
| I Leading flipper under. pattern (53) | 1 → 0 | |
| I Flipper under. elbow patch (54) | 1 → 0 | |
| I Flipper under. tip pattern (55) | 1 → 2 | |
| O Mandible dentary length (103) | 1 → 0 | |
| O Free rib attach. 13 cervical rib (109) | 1 → 0 | |
| O Humerus pneumatic fossa (124) | 2 → 1 | |
| M M. deltoideus propatag. layers subd. (147) | 1 → 0 | |
| M M. deltoideus minor origin (150) | 1 → 0 | |
| M M. ulnometacarpalis ventralis (151) | 1 → 0 | |
| M M. flexor perforatus dig. IV (155) | 1 → 0 | |
| Spheniscus | I Tip mandible (1) | 1 → 2 |
| I Grooves latericorn ramicorn (3) | 0 → 1 | |
| I Feathering maxilla (4) | 1 → 0 | |
| I Latericorn ramicorn ligth mark (10) | 0 → 1 | |
| I Fleshy eyering (33) | 0 → 1 | |
| I White eyebrow (35) | 0 → 2 | |
| I Loreal area aspect (36) | 0 → 12 | |
| I Collar (39) | 0 → 1 | |
| I Back dots over belly (43) | 0 → 1 | |
| I Flanks dark band reach. breast (44) | 0 → 1 | |
| I Throat pattern (57) | 2 → 1 | |
| O Temporal fossa size (76) | 0 → 1 | |
| O Lacrimal exposition in dorsal view (82) | 1 → 0 | |
| O Ext. naris extens. relative hiatus orbit. (84) | 0 → 1 | |
| O Pila supranasalis (85) | 0 → 1 | |
| O Quadrate process otic p. (96) | 1 → 2 | |
| O Mandible coronoid proc. position (98) | 1 → 0 | |
| O Mandible retroarticular proc. aspect (105) | 1 → 2 | |
| O Coracoidal fenestra (119) | 1 → 0 | |
| O Tibiotar. tuberositas poplitea (137) | 0 → 1 | |
| M M. serratus profundus (146) | 1 → 0 | |
| M M. deltoideus major (148) | 0 → 1 | |
| M M. deltoideus major caudale (149) | 1 → 0 | |
| M M. iliofemoralis (153) | 01 → 3 | |
| S. humboldti + S. mendiculus | I Loreal area aspect (36) | 12 → 3 |
| I Throat pattern (38) | 0 → 3 | |
| I Immature throat pattern (57) | 1 → 3 | |
| Megadyptes + Eudyptes | I Latericorn color (11) | 0 → 1 |
| I Culminicorn color (12) | 0 → 1 | |
| I Maxillary mandibulary unguis (13) | 0 → 1 | |
| I Crown feathers yellow pigmentation (25) | 0 → 1 | |
| O Supraoccipital v.occipit. exter. groove (74) | 1 → 0 | |
| O Jugal arch dorsal process (94) | 0 → 1 | |
| O Coracoidal fenestra (119) | 1 → 0 | |
| Megadyptes | I Color periocular area (32) | 0 → 2 |
| I Fleshy eyering (33) | 0 → 1 | |
| I Throat pattern (38) | 0 → 2 | |
| I Line connect. leading flipper belly (50) | 0 → 1 | |
| I Leading flipper upper. pattern (52) | 0 → 1 | |
| I Leading flipper under. pattern (53) | 1 → 0 | |
| I Flipper elbow patch (55) | 1 → 0 | |
| B Crèche (70) | 12 → 0 | |
| O Mandible caudal fenestra (100) | 0 → 1 | |
| M M. serratus profundus (146) | 1 → 0 | |
| M M. deltoideus major caud. (149) | 1 → 2 | |
| Eudyptes | I Rhamphotheca inflated aspect (7) | 0 → 1 |
| I Gape aspect (8) | 0 → 12 | |
| I Inmature bill color (15) | 0 → 2 | |
| I Head plumes (26) | 0 → 1 | |
| I Feet dark soles (64) | 0 → 1 | |
| B First‐second egg relative size (69) | 0 → 1 | |
| O Jugal arch shape (93) | 2 → 3 | |
| O Quadrate process otic p. (96) | 1 → 2 | |
| O Mandibular ramus depth (101) | 0 → 1 | |
| O Mandible dentary limbs (102) | 0 → 1 | |
| O Tibiotar. tuberositas poplitea (137) | 0 → 1 | |
| M M. deltoideus major caudale (149) | 1 → 0 | |
| M M. deltoideus minor origin (150) | 1 → 0 | |
| M M. ulnometacarpalis ventralis (151) | 1 → 0 | |
| E. c. moseleyi + E. c. chrysocome + E. chrysolophus + E. schlegeli | I Head plumes aspect (27) | 0 → 1 |
| I Head plumes heading (28) | 1 → 2 | |
| I Throat pattern (57) | 2 → 1 | |
| E. c. moseleyi + E. c. chrysocome | I Nape crest development (31) | 0 → 2 |
| E. chrysolophus + E. schlegeli | I Head plumes position (29) | 1 → 2 |
| I Head plumes color (30) | 0 → 1 | |
| I Rump (47) | 0 → 1 | |
| O Mandible caudal fenestra (100) | 0 → 1 |




