The distributions of trilobite species were controlled by a combination of habitat preferences and paleogeographic constraints, which tend to limit their extent. The Lower Ordovician trilobite Carolinites genacinaca Ross, 1951, however, had a remarkable range unequaled among polymerid trilobites; it has been recognized on all Ordovician paleocontinents. Its distribution has been explained by an epipelagic mode of life, based on evidence from functional morphology, analogy with modern pelagic crustaceans, and geological occurrence. In such a case, morphological identity throughout its range might be anticipated, if all occurrences can be postulated to be members of a single pandemic population. Rotational superimposition has been used to compare variation within samples drawn from Alberta, Spitsbergen, and Australia with a benchmark population from the western United States. All are morphometrically similar. By any criterion, specimens identical to the benchmark population are found within the Alberta, Spitsbergen and Australia samples, which represent the extremes of the species' geographic range. A lone cranidium from France, previously referred to Carolinites vizcainoi, may be a juvenile of C. genacinaca or C. tasmaniensis; its differences are consistent with ontogeny. A small number of specimens from Siberia and central China show differences in cranidial proportions from the Utah specimens that may be the result of preservational factors and/or photographic technique, or may represent genuine morphological disparity; this could be clarified if more specimens were to become available. This study suggests that C. genacinaca was ubiquitous in the epipelagic environment in a belt that encircled the planet between paleolatitudes of approximately 30°N and 30°S.

ABSTRACT. We use geometric morphometrics to test a claim that the Ordovician trilobite Carolinites exhibits gradualistic evolution. We follow a previously proposed de®nition of gradualism, and de®ne the criteria an ideal microevolutionary case study should satisfy. We consider the Lower±Middle Ordovician succession at Ibex, western Utah to meet these criteria. We discovered examples of: (1) morphometric characters which¯uctuate with little or no net change; (2) characters which show abrupt`step' change; (3) characters which show transitional change through intermediate states. Examples belonging to (2) and (3) exhibit reversals. The transitional characters were tested against a null hypothesis of symmetrical random walk. The tests indicated that they were not changing under sustained directional selection. Two alternative interpretations are possible. (1) The characters are responding to random causes (genetic drift or rapidly¯uctuating selection pressures) or to causes that interact in so complex a way that they appear random. This observation may be applicable to most claimed cases of gradualistic evolution in the literature. (2) Sampling was at too poor a resolution to allow meaningful testing against the random walk. If so, then this situation is likely to apply in most evolutionary case studies involving Palaeozoic macrofossils.

Evidence that can be used to interpret the life habits of extinct organisms usually takes three forms: functional analysis, analogy with living organisms, and geological evidence. Independent quantitative tests for habit are rarely available. A theory of optimum eye design originally derived for living aquatic arthropods provides quantitative data that are used to test previous suppositions about the life habits of two Ordovician pelagic trilobites: that the telephinid trilobite Carolinites was epipelagic while the cyclopygid Pricyclopyge was mesopelagic. Optimum compound eye design theory uses measured lens diameters and interommatidial angles to determine the “eye parameter” (p), which can be used to gauge approximate optimum level of illumination for the eyes of these trilobites. The eye parameter provides an independent test for their relative paleobathymetry. Values of the eye parameter measured in the dorso-ventral direction across two eyes of Carolinites killaryensis utahensis were found to have medians of 2.13 and 3.24. Values measured in the antero-posterior direction have medians of 3.17 and 4.86 for the two eyes. Values measured in the dorso-ventral direction across two eyes of Pricyclopyge binodosa have medians of 4.23 and 4.98, while values measured in the antero-posterior direction have medians of 7.06 and 8.31. Eye parameters are higher in Pricyclopyge than in Carolinites, the difference statistically significant at p = 0.05. The eyes of Pricyclopyge are optimally designed for lower levels of illumination than are those of Carolinites. This accords with the previous interpretations of the former trilobite as mesopelagic and the latter as epipelagic.

The Ordovician Period saw the most sustained, steep rise in marine biodiversity in the history of life on Earth and set the ecological pattern for the rest of the Palaeozoic Era. The long history of research, wide variety of depositional settings and juxtaposition of terranes with very different palaeobiogeographical histories makes the Ordovician of the British Isles an excellent laboratory in which to study this change. A database approach to analysis of trilobite biodiversity change is described using a simple compilation of stratigraphical ranges of 663 species of Avalonian trilobites, and a much more sophisticated information‐rich relational database based on the fundamental information of the fossil record–species at localities. The latter includes 2001 occurrence records of 617 trilobite species at 508 localities in the Welsh Basin. The first ever species‐level diversity curves for the whole Ordovician of an entire region are presented. Both databases reveal an overall increase in species‐ and genus‐level diversity through the Ordovician. Random resampling tests and environmental information are used to remove sampling effects from the Welsh Basin diversity curves. The main features of trilobite biodiversity change in Avalonia are the following: a late Arenig–early Llanvirn increase, during which time the highest species‐to‐genus ratios occur, is contemporaneous with a rise in global diversity and rifting of Avalonia from Gondwana; a late Abereiddian dip is followed by recovery during the Llandeilian–early Caradoc; decline during the late Caradoc–early Ashgill is at least partly attributable to lack of preserved rock and restriction of most of the preserved faunas to deep‐water environments. The greatest diversity occurs in the palaeoenvironmentally differentiated Rawtheyan and is followed by the diversity crash seen in the Hirnantian throughout the world. Copyright © 2001 John Wiley & Sons, Ltd.