The apex of amphicyonid hypercarnivory: solving the riddle of Agnotherium antiquum Kaup, 1833 (Mammalia, Carnivora)

ABSTRACT

Since its first description, only a single undisputed specimen, an isolated, highly abraded m1, was known from the enigmatic amphicyonid Agnotherium antiquum Kaup, 1833. Here, we describe the second specimen of A. antiquum from the type locality of Eppelsheim (early late Miocene, MN 9/10 of Germany), unearthed 184 years after the first description of A. antiquum. The specimen consists of a partial left juvenile mandible with dp3, m1 in eruption, i3, canine, p4, and a highly reduced m2 still in crypt. The p1–3 and m3 are completely reduced, as are the metaconids of m1 and m2. We confirm previous assignments of late Miocene specimens from the localities of Charmoille, Switzerland, and Pedregueras 2A, Spain, to A. antiquum, whereas a maxilla from Bled Douarah, Tunisia, can only tentatively be assigned to the species. We confirm that the monotypic genus Agnotherium is a thaumastocyonine amphicyonid, discuss other supposed thaumastocyonine specimens, and briefly review the evolutionary history of the subfamily. Agnotherium was a 275 kg, extremely powerful, strictly carnivorous ambush hunter that represents the apex of amphicyonid evolution toward carnivory. Its tooth eruption pattern resembles recent Ursus more than recent Canis, but it shows a considerable delay of eruption timing like in the ursid Agriotherium. Agnotherium antiquum is one of six carnivoran species from the Eppelsheim Formation that is heavier than 200 kg. The co-occurrence of Agnotherium (or its relatives Tomocyon and Thaumastocyon), Machairodus, and a smaller saber-toothed carnivoran is typical for early late Miocene faunas of Europe and tentatively North Africa.

ACKNOWLEDGMENTS

We thank L. Werdelin for providing information on the Tunisian material described by Kurtén (1978), S. Mayda (Ege University) for providing information on the lost Turkish specimen and pictures of comparative material, L. Costeur (NMB) for generously providing the material from Charmoille, E. Robert (UCBL-FSL) for verifying collection numbers of the hypodigm of Tomocyon grivense, and J. Morales (MNCN) and E. Heizmann (SMNS) for fruitful discussions and providing access to comparative material in their care. J. Morales provided access to specimens from Pedregueras 2A. We are grateful to the reviewers J. Morales and R. M. Hunt for improving the quality of the manuscript with their comments. The Paul-Ungerer-Stiftung, Frankfurt, supported this research by bearing the travel costs to Madrid. We are thankful for the financial support granted by the state of Rhineland-Palatinate that provided the Naturhistorisches Museum Mainz/Landessammlung für Naturkunde Rheinland-Pfalz and the Generaldirektion Kulturelles Erbe Rheinland-Pfalz, Direktion Landesarchäologie, the resources required to support the long-term scientific excavations in Eppelsheim. Finally, we also thank the capital of Mainz for enabling and supporting our scientific activities. This is ‘Fossilfundstätte Urrhein-Ablagerungen bei Eppelsheim’ no. 32.

ORCID

Michael Morlo http://orcid.org/0000-0003-3956-7912

Katharina Bastl http://orcid.org/0000-0002-8200-9708

Doris Nagel http://orcid.org/0000-0002-8940-6351

Note to Proofs:

During production of this contribution, Morales et al. (2019) published the name Peignecyon felinoides for the taxon from Tuchořice and place the new genus into Thaumastocyonini, close to Thaumastocyon and the re-instated name Tomocyon. Based on a phylogenetic analysis, the authors assigned the genera Crassidia and polyphyletic Ysengrinia as part of a separated tribe Ysengrini into Thaumastocyoninae and verify Y. valentiana as part of Thaumastocyonini. Morales et al. (2019) did not include Agnotherium (as well as Amphicyonopsis from Steinheim and the taxon from Baigneaux-en-Beauce) into Thaumastocyoninae, but as all of these taxa are no part of the phylogenetic analysis, the reason for this decision is not discussed.