SPHENISCIFORMES (PENGUINS)

Ksepka, DT & Clarke, JA, 2015, Phylogenetically vetted and stratigraphically constrained fossil calibrations within Aves, Palaeontologia Electronica (Miami, Fla.) 18, pp. 1-25 : 6-8

SPHENISCIFORMES (PENGUINS)

Node Calibrated. Divergence between Sphenisciformes and Procellariiformes .

Fossil Taxon. Waimanu manneringi Ando, Jones, and Fordyce, 2006 (in Slack et al., 2006)

Specimen. CM zfa35 (Canterbury Museum , Christchurch, NZ), holotype of Waimanu manneringi : thoracic vertebrae, caudal vertebrae, pelvis, femur, tibiotarsus, and tarsometatarsus .

Phylogenetic Justification. Waimanu was recovered as the most basal penguin taxon by multiple analyses using morphological data ( Slack et al., 2006) and combined (morphological plus molecular) data ( Ksepka et al., 2006, 2012; Clarke et al., 2007, 2010; Ksepka and Clarke, 2010a). No alternative hypothesis has been proposed linking Waimanu to any group other than Sphenisciformes .

Minimum Age. 60.5 Ma.

Age Justification. The only described specimen of Waimanu manneringi was collected from the basal Waipara Greensand ( Slack et al., 2006). The top of the Waipara Greensand marks the Paleocene-Eocene boundary, placing the fossil well within the Paleocene. Calcareous nannofossils further constrain this locality’s age. The last appearance datum for Hornibrookina teuriensis is no younger than early Selandian (60.5 Ma), and the overlap between the ranges of Hornibrookina teuriensis and Chaismolithus bidens spans 60.5-61.6 Ma ( Cooper, 2004; Slack et al., 2006; Ogg et al., 2008). The youngest possible age (60.5 Ma) is used as a hard minimum age.

Phylogenetic Position of Sphenisciformes . A sister group-relationship between Sphenisciformes and Procellariiformes (Austrodyptornithes of Yuri et al., 2013) is supported by the largest published morphological ( Livezey and Zusi, 2006, 2007) and molecular ( Hackett et al., 2008) datasets for avian phylogeny ( Figure 3 ). Alternative hypotheses include a sister group relationship between Sphenisciformes and a core clade of 'Pelecaniformes', supported by analysis of sequence data from the beta-fibrinogen gene ( Fain and Houde, 2004) and morphological characters ( Mayr, 2005b); a sister group relationships between Sphenisciformes and Podicipediformes + Gaviiformes recovered by analysis of morphological data ( Smith, 2010); and a sister group relationship between Sphenisciformes and Ciconiiformes recovered from some analyses of mitochondrial data ( Slack et al., 2003; Harrison et al., 2004; Watanabe et al., 2006). Notably, all of these taxa are supported as part of a large waterbird clade by most recent analyses ( Ericson et al., 2006; Livezey and Zusi, 2006, 2007; Hackett et al., 2008; Morgan-Richards et al., 2008; Pratt et al., 2009). We consider a sister group relationship between Sphenisciformes and Procellariiformes to be the most well-supported hypothesis. We note, however, that the position of Waimanu manneringi on the penguin stem lineage does not hinge upon this topology. That is, the polarization of the many synapomorphies that place this species along the penguin stem lineage do not rely on the assumption that Procellariiformes are the closest living relative of penguins.

Fossil Record of Pan-Sphenisciformes. Although Waimanu manneringi is known only from the holotype specimen, the slightly younger (~58- 60 Ma) Waimanu tuatahi is represented by multiple specimens that together comprise almost the entire skeleton, further supporting the presence of Sphenisciformes in the Paleocene ( Slack et al., 2006). An additional late Paleocene (~56 Ma) record is provided by a few elements described as Crossvallia unienwillia ( Tambussi et al., 2005; Jadwiszczak et al., 2013). The Early-Middle Eocene record of penguins includes several dozen isolated elements dated to ~49 Ma from Seymour Island, Antarctica ( Myrcha et al., 2002; Jadwiszczak, 2006a, 2006b; Jadwiszczak et al. 2013; Tambussi et al., 2006), a humerus from the Waipwan-Mangaorapan (50- 56.5 Ma) of New Zealand ( Fordyce and Thomas, 2011) and several partial skeletons and individual bones from the Middle Eocene of South America ( Clarke et al., 2003, 2007). By the Late Eocene, penguins are distributed throughout much of the Southern Hemisphere and abundant in many localities (reviewed by Fordyce and Jones, 1990; Ksepka and Clarke, 2010a).

The absence of substantial gaps in the early record of penguins, their relatively high preservation potential (due to dense bones and occurrence in shallow marine environments), and support for an area of origin in New Zealand ( Slack et al., 2006; Thomas et al., 2011) all suggest that the appearance of penguins in the fossil record closely corresponds at least to the timing of their shift to a flightless marine ecology, if not to their split from their sister taxon Procellariiformes ( Figure 4 ).

Fossil Record of Related Clades. Procellariiformes have a shallower and more sparsely sampled fossil record than Sphenisciformes . Possible Paleogene records of Procellariiformes have been put forward, but these were based on partial limb bones and considered tentative even by the original describers ( Olson and Parris, 1987; Hope, 2002). One of these fossils, Tytthostonyx glauconiticus , was later considered to represent a possible stem tropic bird by Bourdon et al. (2008). The Paleogene Eopuffinus kazachstanensis is known only from a cranial fragment ( Nessov, 1992). Though more material is required to confirm this identification, it is not directly relevant to the Sphenisciformes-Procellariiformes minimum age calibration because Waimanu manneringi is older. Eocene records of Procellariiformes are likewise scrappy, limited to a few isolated bones ( Tambussi and Tonni, 1988; Feduccia and McPherson, 1993; Mayr and Smith, 2012). Early Oligocene records include the first good specimens, among them complete skeletons of the extinct clade Diomedeoididae ( Fischer, 1985; Olson, 1985; Cheneval, 1995; Peters and Hamedani, 2000; De Pietri et al., 2010). Given the extensive nature of the penguin record and the higher preservation potential of thick-boned diving birds, longer gaps in the fossil record of Procellariiformes should be of little concern.

Figures

FIGURE 3. Phylogenetic tree showing position of Waimanu manneringi, illustrating the divergence calibrated by CM zfa35.

FIGURE 4. Stratigraphic distribution of fossil Sphenisciformes in the Paleogene. The minimum possible age of the calibrating specimen of Waimanu manneringi is presented, and midpoints of age ranges are shown for the ages of other Paleogene penguin specimens.