Diversity-Dependent Cladogenesis and Trait Evolution in the Adaptive Radiation of the Auks (Aves: Alcidae)

Abstract

Through the course of an adaptive radiation, the evolutionary speed of cladogenesis and ecologically relevant trait evolution are expected to slow as species diversity increases, niches become occupied, and ecological opportunity declines. We develop new likelihood-based models to test diversity-dependent evolution in the auks, one of only a few families of seabirds adapted to underwater "flight," and which exhibit a large variety of bill sizes and shapes. Consistent with the expectations of adaptive radiation, we find both a decline in rates of cladogenesis (a sixfold decline) and bill shape (a 64-fold decline) evolution as diversity increased. Bill shape diverged into two clades at the basal cladogenesis event with one clade possessing mostly long, narrow bills used to forage primarily on fish, and the other with short thick bills used to forage primarily on plankton. Following this initial divergence in bill shape, size, a known correlate of both prey size and maximum diving depth, diverged rapidly within each of these clades. These results suggest that adaptive radiation in foraging traits underwent initial divergence in bill shape to occupy different food resources, followed by size differentiation to subdivide each niche along the depth axis of the water column.

Sixteen vetted fossil calibrations for divergence dating of Charadriiformes (Aves, Neognathae)

“…Moreover, nearly every calibration employed by Pereira and Baker (2008) and Weir and Mursleen (2012) was incorrectly dated (e.g., a date of 11.6 Ma assigned to 'Uria antiqua'), incorrectly taxonomically assigned (e.g., Cerorhinca dubia; see discussion of node 8 below), or arguably inappropriate for use as a calibration (i.e., external calibrations). The variability of calibrations used in those three analyses is evident in the range of resulting node age estimates.…”

“…Discussion. This fossil has been used previously to date the minimum time of divergence between Pan-Alcidae and its sister taxon Stercorariidae (Smith, 2011a;Weir and Mursleen, 2012;Smith and Clarke, in press). GCVP 5690 displays the apomorphic dorsoventrally flattened shaft and proximally extended dorsal supracondylar process that distinguishes the humeri of pan-alcids from all other charadriiforms (see Smith, 2011a, figure 8.2, table 8.5).…”

“…The variability of calibrations used in those three analyses is evident in the range of resulting node age estimates. For example, the age of the basal split among crown Alcidae was estimated at 19.8 Ma (Weir and Mursleen, 2012), 33.83 (Smith, 2011a), and 54.9 Ma (Pereira and Baker, 2008). Although some of the discrepancy between those estimates is likely due to utilization of different divergence estimation software [i.e., Multidivtime (Thorne and Kishino, 2002) versus BEAST Drummond et al, 2012)], this example highlights the importance of vetting fossils to be used as calibrations and the effect that calibration choice can have on results of divergence time analyses.…”

“…Of the three previous divergence estimates that focused specifically on alcids (Pereira and Baker, 2008;Smith, 2011a;Weir and Mursleen, 2012) only that of Smith (2011a) provided justification for the calibrations used (i.e., apomorphybased specimen referrals and review of age determinations). Moreover, nearly every calibration employed by Pereira and Baker (2008) and Weir and Mursleen (2012) was incorrectly dated (e.g., a date of 11.6 Ma assigned to 'Uria antiqua'), incorrectly taxonomically assigned (e.g., Cerorhinca dubia; see discussion of node 8 below), or arguably inappropriate for use as a calibration (i.e., external calibrations).…”

“…In part because charadriiforms were previously considered a basal neornithine lineage and were influential in the development of the largely refuted 'transitional shorebird' hypothesis (Olson, 1985;Feduccia, 1996), dating cladogenetic events in Charadrii-SMITH: CHARADRIIFORMES CALIBRATION 2 formes has remained a topic of considerable interest. Consequently, there have been at least seven divergence time estimates that have included or focused on this clade (or charadriiform subclades) in recent years (Paton et al, 2002;Paton et al, 2003;Pereira and Baker, 2006;Ericson et al 2006;Brown et al, 2007;Baker et al, 2007;Weir and Mursleen, 2012). Those analyses have utilized different data (i.e., a variety of mitochondrial and nuclear genes), varied in the choice, age assignments, and types of calibrations applied (e.g., secondary external and/or internal fossil calibrations), and have used multiple types of dating software (reviewed by Smith, 2011a).…”