Late Miocene reefs were widespread over the entire Mediterranean. Knowledge of their long-term sedimentary and ecological variability on orbital time-scales and their responses to climatic changes is well established, while understanding of annual to inter-annual dynamics is poor because of inappropriate coral preservation related to the dissolution and cementation of the skeletal aragonite. Exceptionally well-preserved Porites corals retaining the aragonite mineralogy and primary skeletal porosity have been reported by the authors previously from an early Tortonian mixed clastic-carbonate system at Psalidha site (Crete/Greece, Eastern Mediterranean). Here, we present a review of all sub-annually resolved stable isotope chronologies (δ¹⁸O, δ¹³C) and corresponding skeletal growth data (skeletal extension rate and density, calcification rate) available from the Psalidha site, covering a total of 266 years, together with data on sedimentary facies and paleontological inventory. Based on categories of present-day Porites regarding calcification properties and stable isotope systematics, we find average annual SST ~23 °C with mean seasonal SST amplitudes ~7.5 °C. Overall low skeletal growth was limited most during the cool seasons, fully consistent with paleontological inventory (low diversity of reef-building corals) and sedimentary facies, Mediterranean region geographic latitude (subtropics) and biogeography (near northern limit of global reef belt). No evidence for bleaching events has been identified so far. We discuss the growth responses of an important reef-building coral genus in the Late Miocene Mediterranean Sea to environmental variability on seasonal to decadal time-scales in the light of ancient climatic variability and modern reef dynamics during the ongoing reef crisis.

Toothcarps (Cyprinodontiformes) are a highly diverse group of freshwater fish found worldwide. Their fossil record known until now is particularly rich in Eurasia, with few examples from Africa, and spans from the Late Eocene to the Pliocene. Although not as diverse as their extant relatives, approximately 76 fossil species have been described, of which 16 (21%) are based on skeletons with otoliths (“ear stone”) in situ, 24 (32%) are skeletons only, and 35 (46%) are based solely on otoliths. Linking skeletal with otolith-based findings can only be done in rare cases where fossils are found with otoliths in situ and/or when information from extant relatives is available. However, the lack of otolith knowledge from extant families and fragmented data of the fossil record has hindered the integration of skeletal and otolith-based fossil record. By gathering new morphometric, meristic and osteological data from fossil and extant species, together with a literature review, we started to investigate the skeletal and otolith-based fossil species to better understand the evolutionary and biogeographic history of toothcarps. The newly produced and compiled data suggest that the fragmentation of the Tethys Sea and the formation of its related basins (Paratethys) together with the development of various lake systems during the Oligocene and Miocene appear to have promoted the diversification and expansion of Eurasian toothcarps. Furthermore, we found some evidence that representatives of African and East Asian families, whose fossil record is very poor (apart from one family), may also have been present in Eurasia during this time.

Fossil remains of gobies are well known from Miocene freshwater deposits in the North Alpine Foreland Basin (NAFB) and Central Paratethys. Some of those have reportedly large distributions almost akin to extant invasive gobies. One example is “Gobius” (now Eleogobius) brevis (Agassiz, 1839), known from six localities across Western and Central Europe. Given this rather large distribution, the question stands whether this is the actual distribution of a single species or a wastebasket taxon.

Here, we reassess the taxonomic diversity of fossil remains known as Eleogobius brevis by focusing on two Middle Miocene palaeolakes from southern Germany (Ries crater, Oehningen) using historical material from museum collections (97 skeletons, many with otoliths preserved in situ). Based on comparative morphology using both osteological traits and otoliths, we show that the material comprises an unexpected taxonomic diversity of at least five previously unrecognized species.

Notably, none of the Ries gobies correspond to previously described species from the NAFB or Central Paratethys. Given the short time span of life-sustaining environmental conditions in the Ries crater lake and its isolation from major river systems, we thus infer a rapid goby radiation.

In contrast, otoliths preserved in situ in the Oehningen goby species indicate that these species had a broader distribution across the Middle Miocene freshwater systems of the NAFB.

Thus, E. brevis can be considered a wastebasket taxon. A careful re-examination of all specimens that were previously assigned to E. brevis is needed to confirm their taxonomic assignment.

The Late Miocene record of the continental vertebrates has a rich record in the north of the Eastern Paratethys. The territories of Romania, Moldova, Ukraine and Southern Russia provide continental and partially marine deposits rich in marine and terrestrial faunas. Previous extensive studies dated the ages of these deposits from the Sarmatian s.l. to Maeotian (11-6 Ma). In most cases, the ages of sites have broad ranges, significantly limiting paleobiogeographic, -geographic, -environmental, and -climatic reconstructions and analyses.

In the framework of a research project, our team reassessed the ages of the known sites and documented new ones from Moldova and Romania with vertebrate faunas using a broad range of multidisciplinary up-to-date methods of geological sciences. They include sedimentological description, magnetostratigraphy, micropaleontology, mollusc biostratigraphy, dating using autogenic nuclides and vertebrate paleontology. The sections represent mostly fluvial deposits of the Balta Formation, but they also include deposits from the shore and/or deltaic environments.

Our studies allowed to estimate more precisely the ages of the sites ranging from the Bessarabian to Maeotian ages. Further, the dating using autogenic nuclides allowed to establish a geochronological model for our sites independent from the marine biozones and mammalian MN units.

Despite over 250 years of species descriptions since Linnaeus, the challenge of accurately delimiting species remains unresolved – particularly between fossil and living taxa, where data types and availability differ greatly. This has resulted in multiple competing species concepts and complicated biodiversity comparisons across time.

I present preliminary results from a project developing a novel machine learning framework for species delimitation that integrates image data from extant species – where boundaries are molecularly established – with related fossil species. Our study focuses on the renowned Plio-Pleistocene Viviparus beds of paleo-lake Slavonia (SE Europe), which preserve an iconic viviparid species flock. Since the 19th century, over 40 morphospecies have been described from these deposits, many serving as key biostratigraphic markers. The striking morphological diversification spans from small, smooth-shelled forms to large, highly ornamented shells. As a baseline for morphological diversity, we use a modern viviparid species flock from long-lived lakes Yunnan (China), exhibiting a similar range of variation.

To delineate species, we train a Siamese Convolutional Neural Network, which can learn from relatively small, labelled datasets, generalize to previously unseen classes, and perform classification with minimal retraining. A specific aim is to assess how varying degrees of fossil preservation affect model performance, offering potential for broader paleontological applications. Ultimately, we seek to reconstruct biodiversity patterns across geological time, infer diversification dynamics within the Slavonian viviparids, and critically evaluate the biostratigraphic utility of traditional morphospecies. Our novel system should be readily transferable towards a variety of species groups.