- The Paleontological Society
Very different questions are involved when we attempt to assess modern versus ancient global biodiversity. Because of the megabiases of taphonomy, eustasy, and tectonics, our estimates of paleobiodiversity can never be absolute; whether or not we can accurately estimate total diversity in the modern world is an open question, but we certainly cannot in the fossil record. The issues are whether we have any way of studying relative change in biodiversity through time, and how best this might be accomplished. That is, can we meaningfully estimate the shape of a global temporal diversity curve?
Diversity can be studied at different scales (e.g., Sepkoski 1988), and patterns retrieved from one level of analysis may not be similar to those from another (see also Miller 2000). As recognized 30 years ago (Valentine 1973), processes that regulate diversity at local scales differ from those that control global patterns. For example, field collections of trilobites show no change in within-habitat diversity between the Late Cambrian and mid-Silurian (Westrop and Adrain 1998; Adrain et al. 2000). In contrast, literature-based compilations show that global generic richness declined sharply in the mid- to Late Ordovician to a minimum at the end-Ordovician extinction (Adrain et al. 1998; Adrain and Westrop 2000). The difference between these two compilations is informative, and suggests that the drop in global diversity is not driven by local processes. Rather, global taxic diversity seems decoupled from local or regional ecologic diversity. This result undermines the notion that patterns in local species richness can lend support to patterns of global taxic diversity in the Phanerozoic (cf. Bambach 1977; Sepkoski et al. 1981).
In the case of Ordovician-Silurian trilobites, much of the global signal likely reflects declining provinciality as Avalonia and Baltica approached Laurentia (Adrain et al. 2000). This interpretation could not …