Speciation and extinction in the fossil record of North American mammals

Introduction Paleontological data have been used for decades to address a series of very general and intrinsically interesting questions concerning speciation. Many of them are essentially microevolutionary, morphological or both. What is the relative prevalence of anagenesis and cladogensis (Wagner & Erwin 1995)? Do constraints on morphology cause occupation of morphospace to slow down as diversity increases (Foote 1993)? Ismorphological change gradual or punctuated across speciation events (Simpson 1944)? A survey of the analytical paleobiology literature would reveal, however, that interest in all of these questions has waned over the last decade or two. The one topic relating to speciation that remains very popular is the quantification and modelling of turnover rates (Foote 1994b, 2000, 2003; Sepkoski 1998; Newman & Eble 1999; Kirchner andWeil 2000; Allen et al. 2006; Alroy 2008). Coincidentally and fortuitously, the explosion of molecular data sets and great improvements in phylogenetic methods have led to quantifying speciation rates by tracking the accumulation of lineages through time (Nee et al. 1992, 1994; Purvis et al. 1995; Magallon & Sanderson 2001; Roelants et al. 2007). Nonetheless, paleontological research has focused far more strongly on taxonomic diversity than on speciation in recent years (Alroy 1996, 1998b, 2000; Miller & Foote 1996; Sepkoski 1997; Alroy et al. 2001, 2008; Connolly & Miller 2001; Peters & Foote 2001; Smith 2001; Jablonski et al. 2003; Bush et al. 2004; Krug & Patzkowsky 2004; Allen et al. 2006; Crampton et al. 2006). In addition to research on diversity at the global scale, there has been an explosion of studies on diversity at the community scale (Powell & Kowalewski 2002; Bush& Bambach 2004; Olszewski 2004; Peters 2004; Kosnik 2005; Kowalewski et al. 2006; Wagner et al. 2006; and many others). Because the latter kind of work concerns the richness and evenness of individual fossil collections from precise locations and stratigraphic horizons, it has no direct bearing on turnover. When turnover rates do get attention, the paleontological literature has focused on two overriding issues: whether they can be explained using intrinsic dynamic mechanisms, such as either density dependence (Sepkoski 1978, 1979,