Trends in size changes in the coccolithophorids, calcareous nannoplankton, during the Mesozoic: A pilot study Marie-Pierre Aubry 1 , David Bord 2 , Luc Beaufort 3 , Alicia Kahn 4 and Scott Boyd 1 1, 2, 4 Department of Geological Sciences, Rutgers University, 610 Taylor Road, Piscataway, NJ 08854-8066 email: aubry@rci.rutgers.edu, davebord@yahoo.com, kahn@rci.rutgers.edu 3 CNRS-Université Aix-Marseille III, CEREGE, Europôle de l’Arbois, BP 80, 13545 Aix-en-Provence Cedex 04, France email: beaufort@cerege.fr ABSTRACT: We have conducted a preliminary analysis of the history of size change of coccoliths through the Jurassic and Cretaceous. This is essentially based on a compilation of literature data. The results demonstrate that the average size of coccoliths has increased from early Jurassic through the Santonian, stabilized until the Campanian and decreased during the Maastrichtian. Remarkably, this size his- tory parallels the diversity (species richness) history of the Mesozoic Coccolithophorids, and constitutes an illustration of Cope’s rule . The amplitude of change of the average size through time appears to have remained small, which may result from competition by other, larger, contemporaneous calcareous nannoplanktonic groups. INTRODUCTION The coccolithophorids, a group of essentially marine Hapto- phyceae (Edwardsen et al. 2000) occupy a special position in today’s biosphere. They share with the opal-secreting diatoms and the organic-walled dinoflagellates the dominant role of pri- mary producers, while also being the most productive calcify- ing organisms on earth today (e.g., Rost and Riebesell 2004). The group appeared in the Late Triassic and diversified rapidly during the Early Jurassic (Bown, this volume), but it is not until the Cretaceous that it became a major contributor to deep sea sedimentation (Hay 2004). The coccolithophorids were strongly affected by the Late Cretaceous event, as first pointed out by Bramlette and Martini (1964). Their Paleogene radiations are easily rooted in Mesozoic lineages (Perch-Nielsen 1981a, b; Aubry 1998). Yet, the Mesozoic and Cenozoic coccolitho- phorids show little commonality with regard to the morpho- structures of the coccoliths they secreted (Aubry unpublished data). A current view is that the coccolithophorids reached greater diversity (measured as species richness) during the Me- sozoic than thereafter, and have been declining progressively through the Cenozoic (Bown et al. 2004; Bown this volume; Katz et al. 2005). However, this view is entirely dependent on the means used to establish diversity. Fundamental questions are: How do morphotypes (i.e., the paleontological species) re- late to biological species? How reliable is a history of paleodiversity that is established through counts of morpho- types? Would the same conclusions be reached using other di- versity proxy than species richness? Studies aimed at reconstituting the evolutionary history of the coccolithophorids have been essentially concerned with the ra- diation and extinction of lineages and evolutionary rates within lineages (e.g., Haq 1973; Roth 1987; Aubry 1992 1998; Bown et al. 1992 2004; Bown this volume). This means that changes in species richness through time have been determinant in iden- tifying the critical moments in the evolution of the group. One problem with this approach, however, is that the concept of paleontological species in this group of phytoplanktonic organ- isms is far removed from the biological concept (Aubry 1989; Young et al., this volume). The paleontologic concept relies essentially on individual skeletal parts (the coccoliths) that orig- inally constituted a protective armor (the coccosphere) around the living cell. Extant species characteristically have mono- morphic, dimorphic, or varimorphic coccospheres or still ex- hibit monocathetism or dicathetism (for an introduction to the living coccolithophorids, the reader is referred to Young et al., this volume). They also produce dissimilar coccoliths during their complex life cycle in which a diploid, heterococcolith-pro- ducing stage alternates with (in most taxa) a holococcolith-pro- ducing stage (e.g., Cros et al. 2001; Geisen et al. 2002, 2004 Triantaphyllou et al. 2004). Perhaps more importantly, the dis- covery of cryptic species among the extant coccolithophorids (Saez et al. 2003; De Vargas et al. 2004; Saez and Lozano 2005) limits considerably the reliability of estimates of past diversity based on species richness. Morphologic differences that are cur- rently interpreted as intraspecific variations may well be spe- cies-specific (see Geisen et al. 2004). Taxonomy below the genus level is therefore essentially pragmatic. Disagreements among authors as to species concepts for fossil coccoliths (e.g., Bralower et al. 1989; Aubry in press) constitute a mere demon- stration of the inherent precarious nature of specific taxonomy in nannopaleontology. The discovery of a selection pressure leading to time-specific, global morphologic strategies among the coccolithophorids (Aubry submitted) adds to the difficulty of using consistent cri- teria for differentiating species among fossil communities. The overall morphology of coccoliths is controlled across lineages by external, abiotic forcing that narrows the spectrum of possi- ble morphological differentiation, resulting in selection for sim- ilar morphologies in different lineages (i.e., for widespread homoplasy). If the intensity of selection pressure on morphol- ogy has varied through time, morphologic differences between species will likely be smaller when selection pressure is stron- ger, and larger when it is weaker. This in turn may result in un- derestimating species richness during time of strong selection pressure, as during the Cenozoic (Aubry unpublished data). Di- versity in the calcareous nannoplankton can be measured by micropaleontology, vol. 51, no. 4, pp. 309-318, text-figures 1-6, tables 1-2, 2005 309