Conference Agenda

According to the molecular clock estimates, cave bears diverged from the arctoid lineage c. 1.2 million years ago. Even if their anatomy is well known, there are many unknown aspects about the evolution of this lineage. U. deningeri and U. spelaeus belong to the same lineage and are considered chronospecies from Middle and Late Pleistocene.

The cave of Santa Isabel (301 m above sea level) is located in the Ranero neighborhood (Karrantza, Biscay, Spain) and currently has two openings and approximately 500 m of development. During the 1970's a speleological group from Catalonia (G.E. Badalona) explored the cave system and found over a hundred fossils which were given to the paleontologist F. Villalta. During the spring of 1985 the Grupo Espeológico Esparta further explored the cave finding new fossil remains, including a complete skull. Due to the fact that the site was at risk from visitors to the cave, an excavation season was performed in 1991 to gather all the fossils present at the site (>2,000 remains), followed by a second visit in 1994 in order to sample the site for ESR dating which has yielded an age of c. 300 ka for the bear sector (the most important of the 4 fossiliferous sectors in the cave).

In order to quantitatively compare the anatomy of the mandibles of the Santa Isabel de Ranero (SIR) bears with other Middle Pleistocene bears as well as cave, brown and American black bears, both traditional and 3D geometric morphometric analyses were applied.

The SIR mandibles were scanned using a CT or a surface scanner and the surface models subsequently landmarked. Comparative bear mandibles were digitised with a Microscribe. The landmarks were chosen based on a compromise between functional morphology and sample size.

A principal component analysis using a complete set of landmarks indicates that the SIR mandible has very similar PC1 scores, but slightly lower PC2 scores than the cave bears. This implies that the SIR mandible had a relatively longer mandibular corpus and the distance between the fulcrum (mandibular joint) and the tooth row was smaller than U. spelaeus. An analysis using only landmarks on the mandibular body in order to include more specimens showed more overlap between the species in morphospace. Middle Pleistocene cave bears overlap strongly with cave and brown bears, but not with black bears. The allometric analysis shows that the SIR mandibles show an intermediate position between U. arctos and U. spelaeus. Our traditional morphometric assessement is consistent with the previous results.

The results show that middle Pleistocene cave bears are morphologically closer to cave bears than to brown or black bears. Our results are consistent with the Santa Isabel de Ranero sample being U. deningeri as previously proposed.

Brown bears (Ursus arctos) originated in Asia and diverged from the spelaeoid line in the late early Pleistocene, but they do not appear in the European fossil record until half a million years ago. Middle and Late Pleistocene bears are referred to in Europe as U. arctos priscus (GOLDFUSS, 1818) and U. arctos (LINNAEUS, 1758), respectively. U. arctos priscus is larger than recent European brown bears. Recent U. arctos comprises three subspecies in Europe and presents a high variability.

In the Iberian Peninsula the oldest brown bears are dated to the Late Middle Pleistocene and their remains are very scarce. The oldest brown bear mandible of the Iberian Peninsula was found in the Middle Pleistocene site of Postes cave, in the Natural Monument of Fuentes de León, Extremadura (Spain). The bearing has yielded 62 bear remains identified as U. arctos sealed by a speleothem dated by U/Th in 192,986 +15,451/-13,837 ka BP.

The purpose of this study is to compare the Postes specimen with European Ursus arctos focusing on mandibular morphology. The sample comprised of Pleistocene brown bears from the Iberian Peninsula, Southern France, Austria and Italy; Holocene (from Neolithic to Iron Age) Iberian bears and recent bears (XX-XXI century) from Iberian Peninsula, Central Europe, Balkans and Caucasus. Comparisons were made using traditional and 3D geometric morphometric analyses. The mandibles were digitized with a 3D scan and traditional measurements were taken with a calliper.

Statistical analyses were carried out to compare the difference in size between European brown bears from different chronologies and test its relationship with shape. A principal component analysis (PCA) with traditional measurements shows that most of the variation is distributed along PC 1, with a minimum overlap between Pleistocene and recent U. arctos.

A geometric morphometric (PCA) shows that Pleistocene U. arctos are grouped together, unlike recent specimens which show a larger degree of morphological variation. This might be caused by the diverse geographic origin of the recent sample and/or by the larger size range of the specimens. Furthermore, Holocene specimens seem to have a different shape from Pleistocene bears. A regression analysis of the Procrustes coordinates onto logarithmic centroid size indicates significant allometry in the sample. A One-way ANOVA of the centroid size shows there are significant differences between the groups. Additionally, a Tukey’s pairwise test shows that the difference lies between the recent specimens and the fossils. The results of the MANOVA are consistent with the previous analysis.

In summary, these results suggest that Pleistocene U. arctos are uniform in size and shape, but present a difference in size with respect to recent specimens. Postes specimen fits within Pleistocene bear variation in both size and shape.

The brown bear (Ursus arctos) is one of the closest living relatives to the extinct Late Pleistocene cave bear (Ursus spelaeus). According to a preliminary analysis late Pleistocene bear skull fragments found in caves in Belgium show intermediate phenotypes between U. arctos and U. spelaeus. To analyse a potential hybridisation between these two species a 3D geometric morphometric analysis was performed. In total eight skull fragments were analysed: three cranial fragments and five mandible fragments. The fragments were digitized using a CT scan and then a 3D model was created using segmentation. Landmarks were chosen corresponding to existing studies on U. spelaeus and an existing dataset was used to compare the potential hybrids to existing landmark data of U. spelaeus and U. arctos. The existing dataset was combined with the potential hybrid data and a generalized Procrustes superimposition was used to project the landmarks onto tangent space by orthogonal projection. A linear regression was performed on the Procrustes coordinates onto log centroid size to remove the effect of allometry. A principal components analysis was then performed using the regression residuals. Depending on the landmarks chosen and the principal components displayed, some specimen showed to be in an intermediate space between the two reference groups. Some specimen were plotted as belonging to U. arctos although they possess the dental formula of U. spelaeus. These results could point to a potential hybridisation between U. arctos and U. spelaeus in the Late Pleistocene.

Variation in the skeletal morphology within the Carnivora has long been demonstrated as a predictor of a species’ ecological habits. The elbow joint in particular is a clear indicator of predation strategy[1]. Members of the Ursidae display significant diversity of locomotory and foraging habits and fall into three distinct subfamilies, thus proving to be an excellent group in which to study the interplay of ecology and phylogeny on elbow joint morphology. Previous work using 2D Geometric Morphometrics has revealed an effect of both phylogeny and ecology yet did not discriminate between all locomotory groups[2]. However, given the complexity of the elbow joint, 2D Geometric Morphometrics may be insufficient to fully capture the complexity of this feature.

A 3D Geometric Morphometric approach was applied to thirty Ursid specimens with the aim of exploring how ecology and phylogeny can explain morphological variation at the elbow joint. Morphological variation was a good predictor of both locomotory and foraging habits. However, the relative importance of these two ecological factors cannot be easily separated. Furthermore, phylogeny also contributes to variation in the shape of the elbow joint although to a lesser extent than the above-mentioned ecological factors.

In contrast to previous work, we demonstrate morphological differences between all ursid groups based on locomotory and foraging habits. This confirms elbow joint morphology as both a clear adaptation to the many ecological niches the Ursidae has occupied since its Oligocene emergence and an excellent illustration of the adaptive radiation of mammalian carnivores throughout the Cenozoic.

1. Andersson, K. I. (2004) ‘Elbow‐joint morphology as a guide to forearm function and foraging behaviour in mammalian carnivores-’, Zoological Journal of the Linnean Society, 142, pp. 91–104. https://doi.org/10.1111/j.1096-3642.2004.00129.x
2. Meloro, C. and Marques de Oliveira, A. (2017) ‘Elbow Joint Geometry in Bears (Ursidae, Carnivora): a Tool to Infer Paleobiology and Functional Adaptations of Quaternary Fossils’, Journal of Mammalian Evolution. Journal of Mammalian Evolution, pp. 1–14. doi: 10.1007/s10914-017-9413-x

Drastic morphological changes often occur within mammalian species during prolonged isolation on islands. The depauperate nature of insular ecosystems is often associated with phenomena known as insular dwarfism and gigantism, with such changes in size particularly well-known in the fossil taxa of Mediterranean islands. Insular gigantism has been noted in rodents, lagomorphs and insectivores. Within Rodentia, dormice appear to increase body size more frequently during isolation than other rodents. At least eight different Mediterranean islands have been occupied by giant dormice from the onset of the Miocene. We studied the cranial and mandibular morphology of three fossil giants: Hypnomys onicensis, H. morpheus and Leithia melitensis. All three species are believed to be related to the garden dormouse Eliomys quercinus, a species still present on some Mediterranean islands, including a giant population on Formentera. We evaluated to what extent the shape of giant dormice is explained by allometry, or by non-allometric processes. A predicted size over size (PSOS) model was created to assess the cranial and mandibular shape. It appears that gigantism in dormice is not simply an extrapolation of the allometric trajectory; instead, both cranium and mandible show individualistic evolutionary shape changes, with some morphologies suggesting adaptations to the feeding apparatus. Our results suggest that insularity can lead to context-dependent phenotypic modifications, resulting in highly idiosyncratic fauna characteristic of islands.

Fossil insular mammalian dwarfs have variably been reported to show rapid ontogenetic development, slow development, as well as truncation of growth. These three modes of dwarfing result from different life history strategies of the animals under consideration and have implications for their morphological plasticity.

Cyprus, in the Pleistocene, was extremely isolated from a geological and biogeographical point of view. Only two macromammals successfully colonised the island before the Holocene: Elephas cypriotes, approximately 1.4 metres tall at the withers, and Phanourios minor. The latter is the smallest dwarfed hippo ever found; it stood 70 centimetres at the withers and weighed an estimated 200 kilograms, approximately 10% of its mainland ancestor’s weight.

The main objective of this study was to determine the mode of dwarfing in hippopotami from Cyprus using bone histology. Radii and several other long bones of P. minor, and their normal sized relatives (Hippopotamus amphibius), were thin-sectioned and studied under a polarizing optical microscope. Type of bone matrix and counts of lines of arrested growth (LAGs) served as proxies for development rate and time.

Histological analyses suggest that Cypriot dwarfed hippos had slower postnatal growth rates than their normal-sized relatives. In fact, Cypriot dwarfed hippos would have needed at least two decades to reach adult size as demonstrated by the radius. Slow growth might have been selected for in a resource-limited but predator-free environment.

The tribosphenic molar presents phylotypic stage of the mammalian dental evolution. A study of its enamel microarchitecture in several clades (Chiroptera, Afrosoricida, Eulipotyphla, Didelphomorhia, Cetartiodactyla) revealed, apart of considerable taxon-specific variation, a set of structural (i-iii) and developmental (iv-x) characteristics common to all: (i) patterning crown shape by shearing crests of major cusps, the basic structural modules of tribosphenic design, (ii) heterotopy of enamel thickness at individual modules (thick enamel on the convex vs. its lack on concave sides of shearing crests), (iii) radial prismatic enamel (PE) with interprismatic matrix (IPM) and aprismatic surface enamel (AE), (iv) heterochrony of amelogenetic processes (establishing PE scaffold as primary amelogenetic product of the secretory stage, infilling the between-prism spaces by IPM at late maturation stage, termination of enamel maturation with AE related to amelotin secretion prior to eruption), (v) initiation of PE by intervention of odontoblastic processes, (vi) gradual growth of PE crystallites with abrupt drop of microstrain and increase of mechanical resistance at late maturation, (vii) extensive growth of the tooth during PE formation with considerable heterotopies and heterochronies among particular structural modules (as revealed by different patterns of prism curvatures), (viii) prolonged perieruptional stage enabling (ix) terminal enlargement of tooth and (x) refinement of shearing effect by growth heterotopy under control of first occlusion efforts of matured crests.

The observed taxon-specific variations were mostly related to different dynamics of tooth enlargement (vii) and patterns and spatial variation in IPM formation. In Afrosoricida an extensive growth of paracone wall producing the zalamdodont design is associated with a transversal shift of PE. In shrews (Soricidae) we observed a lamelar PE pattern and crystalization disorder during late secretory stage with metasomatic alternation of hydroxyapatite by iron mineralisation in some clades.

The results suggest that the developmental processes producing tribosphenic molars include a complex set of amelogenetic apomorphies prefiguring the potential of tremendous dental evolution already at early onset of mammal history.

The Giant Deer Megaloceros giganteus (Blumenbach, 1799) is an icon of the Ice Age due to their famous antlers, although there has been little focus on isotopic analysis of these animals. migrational behaviour in these animals has been suggested, mainly due to this behaviour being common in deer species today. Migrational behaviour and seasonal movement is analysed in three Giant Deer from Ireland dating back to the Pleistocene late glacial through isotopic analysis. Analysis is performed using sequential intra-tooth sampling for ⁸⁷Sr/⁸⁶Sr ratios acquired from TIMS and LA-MC-ICPMS on the Giant Deer’s tooth enamel. A comparison between TIMS and LA-MC-ICPMS is performed to try and improve the precision and accuracy of LA-MC-ICPMS for strontium isotopic analysis. This research is conducted as part of a Utrecht University joint MSc thesis.

The reasons for migrational behaviour is explored. As a deer species, Giant Deer required a high nutrient uptake for antler growth and were not as well adapted for the cold-dry steppe environments of the Pleistocene glacials. These stresses would have made migrational behaviour beneficial for maximising nutrient uptake during the warmer months, and for avoiding the harshest winter conditions of the glacials by moving to sheltered, low lying areas. Giant Deer from interglacials of the Pleistocene would have not been under the same stresses, as the closed environments of that period allowed for the mix-feeding strategies they likely employed. Irish Giant Deer may have been pushed to migrational behaviour especially by the onset of the Younger Dryas cold stage as the preferred lime-rich grasses and forbs they depended on disappeared from the island. Therefore, migrational behaviour and seasonal mobility strategies may have been paramount for Giant Deer during cold periods of the Pleistocene.