A synopsis of Ordovician trilobite distribution and diversity

Ordovician trilobites are reviewed based on a new species-level relational database. The stratigraphical ranges of all 56 families with occurrences in the Ordovician are documented and the content, phylogenetic status, diversity and Ordovician distribution by major palaeocontinent/terrane are discussed. Aspects of higher classification are also dealt with. Global sampling is heavily biased towards a small number of highly sampled areas. Much of the world has a very limited record of formally named trilobite species. Even within heavily sampled units, sampling is patchy by environment and time. Genus endemism was at a peak in Laurentia, Baltic, and Avalonia in the Floian and declined more-or-less steadily through the remainder of the Ordovician. Ordovician trilobite distribution was addressed in a classic paper by Whittington & Hughes (1972). This work featured one of the earliest applications of nonmetric multidimensional scaling in palaeontology. The faunal provinces they discriminated – the Laurentian Bathyurid province, the Baltic Asaphid province and the Gondwanan Selenopeltis and Asaphopsis provinces – remain valid today. A follow-up work on the Tremadoc (Whittington & Hughes 1974) was marred by the global miscorrelation with the upper Cambrian prevalent at the time. Since this work, there have been a number of excellent continentalor regional-scale synthetic studies (e.g. Fortey & Barnes 1977; Shaw & Fortey 1977; Fortey & Cocks 1988; Zhou et al. 2007, 2009; Zhou & Zhen 2008; Benedetto et al. 2009; Harper et al. 2009). Trilobites provided some of the Ordovician data for a magisterial series of papers by Cocks & Torsvik (2002, 2005, 2007, 2011); Torsvik & Cocks (2004, 2009; also Cocks & Fortey 2009). Probably the most important recent paper on Ordovician trilobite distributions is the excellent global review by Fortey & Cocks (2003). Despite these advances, the paper by Whittington & Hughes (1972) remains the only attempt ever published at a global numerical synthesis of Ordovician trilobite biogeography. Serious progress in resolving Ordovician biogeographical problems will require the application of component-based cladistic biogeography (see, e.g. Ebach & Edgecombe (2001) for a review using trilobites). For example, there has been a great deal of debate over the derivation and position through time of the Argentine Precordillera Terrane (see Benedetto et al. (2009) for a review). Trilobite data have been applied to the problem, but thus far only in a phenetic way based on the presence of higher taxa (genera and subfamilies). There is a clear influx of Gondwanan taxa in the Darriwilian and Sandbian, while the bulk of the fauna remains of Laurentian aspect. Much more information could be gleaned if more species-level phylogenetic analyses were available (see Chatterton et al. (1997) and Edgecombe et al. (1997, 1998) for those that have been carried out). Separate undescribed species from the Darriwilian Table Cove Formation of western Newfoundland (Edgecombe et al. 1999, p. 1144) have been identified as putative sister species of the cheirurid Macrogrammus rafi Edgecombe, Chatterton, Vaccari & Waisfled, 1999, and the telephinid Telephina problematica Chatterton, Edgecombe, Vaccari & Waisfeld, 1999, from the Darriwilian and Sandbian of the Precordillera. If these and similar relationships were confirmed in a modern analytical framework, the resulting phylogenies could formally test hypotheses of isolation v. faunal exchange between the terrane and Laurentia. There is excellent potential for multiple tests and corroboration as many separate Darriwilian–Sandbian groups have been described from the Precordillera on the basis of rich silicified faunas. Similar work could contribute much to unravelling the histories of other complex areas (e.g. the Kazakh terranes) beyond simple occurrence data. That said, perusal of the family-level summaries presented below will demonstrate that modern phylogenetic work is not widely available for Ordovician trilobites. Many large and important families have received no analysis whatsoever, and even those taxa (such as encrinurids and dimeropygids) for which multiple studies have appeared still lack anything approaching a phylogenetic hypothesis of their overall structure. There is much work to be done. Given that, it seems there may be value in a fresh look at global distributional data, taking into account the advances in understanding of the last 40 years. I have completed a hierarchical taxonomic database of all trilobite species, which can form the foundation for new analyses. The database at time of writing includes 21,161 valid trilobite species, of which 5460 are Ordovician. Initially (in collaboration with A. W. Owen and R. A. Fortey), I attempted ordination and clustering analyses similar in scope to those of Whittington & Hughes (1972). It soon became apparent, however, that sampling problems (see below) made analyses using only formally named species less than satisfactory. The record from many parts of the world and various times during the Ordovician is much sparser than one might expect. Much of the diversity in these cases tends to be reported in open nomenclature, or sometimes (usually incorrectly) as established species from other regions or continents. I have added functionality to the database to incorporate these data, but have only begun to compile the necessary information for the Ordovician. In anticipation of such analyses, here I survey what is known of the diversity and distribution of Ordovician trilobites, mostly on the basis of their formally named species. The goals of the work are to present the first species-level assessments of global and continental/terrane-level taxonomic richness and endemism, and to document the diversity history, phylogenetic status and distribution of each family present in the Ordovician. I also take the opportunity to expand on several phylogenetic questions, mainly stemming from a recently published revised classification of trilobites (Adrain 2011) and in anticipation of preparation of further revised volumes of the Treatise on Invertebrate Paleontology, Part O (Trilobita, Revised). Summed diversity, sampling and endemism As classified below, there are 56 families of Ordovician trilobites, and the stratigraphical range of each is shown in Figure 20.1. Total global species richness is plotted in Figure 20.2, and divided by palaeocontinent/terrane. No attempt was made to account for From: Harper, D. A. T. & Servais, T. (eds) 2013. Early Palaeozoic Biogeography and Palaeogeography. Geological Society, London, Memoirs, 38, 297–336. http://dx.doi.org/10.1144/M38.20 # The Geological Society of London 2013. Publishing disclaimer: www.geolsoc.org.uk/pub_ethics by guest on November 27, 2013 http://mem.lyellcollection.org/ Downloaded from