Chapter 8
Angiosperm divergence times: congruence
and incongruence between fossils and
sequence divergence estimates
Niklas Wikström, Vincent Savolainen and Mark W.Chase
ABSTRACT
The documentation of derived angiosperm lineages from increasingly older
geological deposits, and growing evidence of considerable diversity in flower, seed,
and pollen morphology in the mid-Cretaceous both imply that the timing of early
angiosperm cladogenesis may be older than our current fossil-based estimates
indicate. An alternative to fossils for calibrating the phylogenetic tree comes from
divergence in DNA sequence data. Here, we report on an analysis using non-
parametric rate smoothing and a three gene dataset covering c. 75 per cent of all
angiosperm families recognized in recent classifications. The results provide an
initial hypothesis of angiosperm diversification times; by using an internal
calibration point, an independent evaluation of angiosperm and eudicot origins is
accomplished. Results are compared with fossil-based estimates of both magnolids
and eudicot divergence times, and possible directions of future analyses are
discussed.IntroductionFlowering plants (angiosperms) have dominated terrestrial ecosystems since the Late
Cretaceous (Crane 1987), and their estimated 250 000–300 000 living species represent
an overwhelming majority of extant land plants. Our understanding of angiosperm origin
and diversification has, however, been hampered by a number of problems. The early fossil
record was, until recently, comparatively poorly understood and insufficiently known
(Crane et al. 1995) and, up until the 1960s, angiosperms were widely held to have
originated in the Late Palaeozoic or possibly early Mesozoic (Axelrod 1952, 1970; Thomas
1957; Eames 1959; Takhtajan 1969). The assignment of early fossil leaves to putatively
derived angiosperm lineages led to this conclusion, but later work, based not only on
leaves but also on pollen, flowers, and fruits, changed this view dramatically, pushing the
angiosperm origin into the Early Cretaceous (Doyle 1969; Doyle and Hickey 1976;
Hughes 1976; Hickey and Doyle 1977; Doyle 1978; Friis and Skarby 1981; Muller 1981;
Friis 1984; Friis et al. 1988). From a neobotanical perspective, resolving the relationships
among extant lineages, as well as establishing a rooting point for the angiosperm clade,
have both been problematic. However, in the late 1980s and early 1990s several landmark