diverged prior to the end of the Mesozoic (Clarke and Chiappe 2001; Chiappe and Dyke
2002). However, this inference provides little information regarding temporal divergences
among extant lineages. Strong debates remain, but these centre around the question of how
deep the extant lineages of birds (Figure 12.2) can be extended into the Mesozoic
(Cracraft 2001; Chiappe and Dyke 2002). If one accepts hypotheses presented by
proponents of ‘molecular clocks’ (e.g. Hedges et al. 1996; Cooper and Penny 1997; van
Tuinen et al. 1998, 2000) then the neornithine clades extend far below the K-T boundary
and the known fossil record of these taxa is hopelessly incomplete (Dyke 2001a)—
contrasting with the fossil record of other small vertebrates from the Cretaceous (Benton
1999; Fara and Benton 2000).
What use are fossils?The use of ‘molecular clocks’ to provide estimates for divergence times within the
Neornithes has required the consideration of fossil taxa to provide either (or both) internal
and external calibration points for clades (Cooper and Penny 1997; Table 12.1), even
though specimens are incomplete and lack phylogenetic control. The use of internal
calibrations for such a ‘clock’ estimate is preferable since the age of a fossil will provide an
estimate closer to the true divergence time of the clade(s) in question and hence will
require less extrapolation and associated error (van Tuinen and Hedges 2001). But there
are problems inherent to previous uses of the fossil record of Neornithes in this way: (1)
proposed systematic positions of fossil taxa are often cited uncritically, in the absence of
phylogenetic control; and (2) no attempts have been made to distinguish between the
placement of fossils with respect to the stem-or crown-groups of the clades in question.
The single most important issue pertinent to the use of fossils for the calibration of
‘molecular clock’ (or other) estimates of divergences is the position of taxa within the
phylogeny of the group in question, problematic because the relationship of a given fossil
to the crown- or stem-group of a clade can only be known through phylogenetic analysis
(Figure 12.3). Although estimates for the divergence times of the major clades of extant
birds are becoming more and more numerous in the literature, this important distinction
has yet to be made for the vast majority of the fossil record of Neornithes.
In this chapter, I outline osteological evidence for the phylogenetic relationships of a
number of well-preserved fossil neornithine taxa within the context of the relationships of
clades at the base of Neornithes. This review draws on a number of recent cladistic
Table 12.1 The described fossil records used by Cooper and Penny (1997) to estimate the
divergence times of modern bird clades
a Ages of fossils as listed by Cooper and Penny (1997).
272 GARETH J.DYKE