Much attention has been focused on the K-T extinction since the now famous hypothesis of bolide impact was first mooted in 1980. Substantial information has been amassed and used to address issues relating to the causal mechanism(s) of the extinction (e.g., Alvarez et al. 1980, Bohor et al. 1984, Venkatesan and Dahl 1989, Hildebrand and Boynton 1990, Izett et al. 1991), as well as to patterns of extinction and recovery experienced by particular groups of organisms (e.g., Jablonski and Raup 1995, Marshall and Ward 1996, Sheehan et al. 1996, MacLeod et al. 1997, d’Hondt et al. 1998, Jablonski 1998, Smith and Jeffrey 1998). With regard to the latter, biological groups with more utilitarian fossil records (e.g., planktonic foraminifera) or broad scientific and public appeal (e.g., dinosaurs) have received the most attention, whereas relatively little is currently known about extinction patterns in many other groups, including bryozoans.
Bryozoans are a phylum of colonial metazoans with a rich marine fossil record due to the readily fossilisable calcareous skeletons possessed by most species. Two clades of calcareous bryozoans are represented in the Cretaceous and Cenozoic, cyclostomes (Class Stenolaemata) and cheilostomes (Class Gymnolaemata). Both clades were diverse during the Late Cretaceous; however, knowledge of their systematics is woefully inadequate—many species have yet to be formally described and others remain unrevised and are virtually unrecognisable using the original descriptions published in the 19th Century. This lack of knowledge is compounded for the cyclostomes which, in contrast with the skeletally more complex cheilostomes where a close correspondence has been demonstrated between genetically and skeletally determined species (Jackson and Cheetham 1990), have relatively simple and easily confounded morphologies (McKinney and Jackson 1989; Håkansson and Thomsen 1999). Ranges of species in time and space at all scales of resolution, from single stratigraphical sections to globally, are poorly documented, and almost nothing is known about phylogenetic relationships that might aid the interpretation of extinction patterns (e.g., Smith and Jeffery 1998, for echinoids). Despite these problems in documentation, the completeness of the fossil record apparently is rather high for bryozoan genera, with an approximately 70 percent probability of the preservation of a genus in each stratigraphic stage throughout its range (Foote and Sepkoski 1999). With regard to global patterns, the finest level of analysis that can reasonably be attempted at present for bryozoan genera is at the level of the stratigraphical stage/age, as done, for example, by Viskova (1980, 1997).
Alternative approaches to understanding the historical record of bryozoans have looked at changes in species assemblage diversity and in relative skeletal biomass around and across the K-T boundary (Lidgard et al. 1993; McKinney et al. 1998, McKinney et al. in press). The only bed-by-bed study of bryozoan species distributions across the K-T boundary is that of Håkansson and Thomsen (1999) at Nye Kløv in Denmark. In this paper, we track the history of originations and extinctions of Late Cretaceous through Cenozoic bryozoans at the genus level. Therefore, we utilise primarily the global record of temporal distribution of the genera rather than local species assemblage diversities or skeletal biomass.
The existence of two distinct clades of bryozoans crossing the K-T boundary provides a replicate data-set of extinction and origination patterns, thereby lessening potential problems caused by factors specific to one clade. Cyclostomes date back in the fossil record to the Lower Ordovician, cheilostomes to the Upper Jurassic (Taylor 1993). Each group is thought to have acquired its mineralised skeleton independently from a bryozoan of ctenostome-grade and to share a common ancestor no younger than Early Ordovician (Todd 2000). They show important differences in zooidal baupläne, which apparently are linked with the greater rate of suspension feeding, growth rate, and – in tropical and temperate waters – ability to compete for substrate space in cheilostomes (McKinney 1993, 1995; but see Barnes and Dick 2000 for reversal of competitive interactions in high latitude waters). Species belonging to the two clades commonly co-exist and compete with one another in the same habitats, and a comparable range of colony forms is developed in the two clades. Cyclostomes and cheilostomes, therefore, occupy broadly similar ‘ecospace’; however, some cheilostomes have invaded brackish waters and a few others have evolved free-living colonies, while cyclostomes have done neither. Other than these two categories of exceptions, the two clades might be expected to show parallel responses to the majority of environmental changes.
To put our consideration of extinction at the K-T boundary into a broader evolutionary context for the phylum and to identify other times of enhanced generic turnover during the past 100 million years—whether resulting from changes in extinction or origination rates—we analysed a new database of uppermost Lower Cretaceous to Recent (Albian-Recent) bryozoans containing genus ranges at stage/age-level precision. The three principal questions we set out to address initially were
Did generic extinction of Albian-Recent bryozoans peak in the Maastrichtian, corresponding with the K-T event?
Are temporal patterns of generic extinction and origination intensities the same for cheilostomes and cyclostomes?
Do generic origination rates support Voigt's (1981) notion of the Danian as a stage of low 'creativity'?