Nouvelles recherches sur l'ichthyofaune oligo-miocène du Puy de Mur (Puy-de-Dôme)

Abstract

The Puy de Mur massif, near Vertaizon (Puy-de-Dôme, France) consists of Oligocene calcareous marl alternating with intercalations of volcanic tuff. In this area, two fossil localities have yielded fossil fishes. The first one is an outcrop of diatomite overlying tuff near the Jallat pass, on the southem slope of the Puy de Mur. According to available radiometric data, the tuff of the Puy de Mur massif has an age of 22 ± 4 Ma. Consequently,the age of the fossiliferous diatomite may be either Early Miocene or Late Oligocene. The fossil fishes collected in this level were first referred by Priem (in Giraud, 1902, and Priem, 1914) to the Percoid genus Smerdis Agassiz (nec Leach). Later, they were considered by Piton (1936a) as Cyprinodontids and described as Haplochilus giraudi Piton. The second locality is at « Fontaine Saint-Blaise », near Dallet on the west side of the Puy de Mur massif. The fishes are fossilized in a single horizon, included in a 40 cm thick silicified bed overlying a tuff breccia. According to palynological analysis of the marl overlying the fossiliferous silicified bed, the age of these rocks probably is early Late Oligocene (Gorin, 1974). The Oligocene Cyprinids from Dallet - Family Cyprinidae - Genre Rutilus Rafinesque Rutilus rudeli (Piton) (Fig. 2-3; P1. I, Fig. 1-5). Anatomical study of the newly collected fossils from Dallet has shown that, contrary to Piton’s opinion, these Cyprinid fishes do not really belong to the genus Barbus Cuvier. In fact, these fishes share several characteristics with the Miocene species of the genus Pa!aeoleuciscus Obrhelova: - a short, oblique, terminal mouth; a broad skull roof; - a dentary bearing a strongly prominent coronoid process; - a vertebral column including less than 40 vertebrae; - a widened operculum, the maximum width of which surpasses 3/4 of the length of its anterior edge. However, contrary to what is known in Palaeoleuciscus, the operculum of some specimens exhibits a prominent postero-dorsal angle, like in the Upper Miocene species Rutilus pachecoi (Royo Gômez) which has, additionally, almost identical meristic characters. For these reasons, the Late Oligocene Cyprinids from Dallet, that slightly differ from Rutilus pachecoi (Royo Gômez), are described in the present paper as Ruti!us rudeli (Piton). This species may be defined by the following diagnosis: « Small Cyprinids, the standard length of which exceptionally surpasses 35 mm. Body rather stout, the maximum height of which, exceeds 1/4 of standard length. Operculum exhibiting a prominent posterà-dorsal angle. Vertebral column made of 37 vertebrae, divided into 2 1-22 abdominal and 15-16 postabdominal. Caudal fin rather large, forked; 10-12+I+9/8+1+8-9 rays. Dorsal fin beginning slightly behind the middle of body; ii+I+7 rays endoskeleton made of 8pterygiophores. Anal fin small ii+I+6 rays endoskeleton with 7 pterygiophores. Pectoral fins moderate; roundly 15 rays. Pelvic fins either opposite to the origin of the dorsal fin, or beginning slightly behind; 8-9 rays.». Neotype: Specimen M.N.H.N.P. 1976-13D + G (A. Rudel’s collection), kept in Paris in the Palaeontological Institute of the National Museum of Natural History. Locus typicus: « Fontaine Saint-Blaise », near Dallet (Puy-de-Dôme, France). Stratum typicum: Late Oligocene (Early Chattian ?). The fossil fishes from the « Randannites » of the Puy de Mur. Family Atherinidae - Genus Palaeoatherina Gaudant - Palaeoatherina giraudi (Piton) - (Fig. 4-6; P1. II, fig. 1-5). The reexamination of the original material figured by Piton and Théobald (1939) and the study of newly collected material have enabled a reassessment of the taxonomic status of these fishes, which belong really to the Atherinids and, more precisely, to the genus Pa!aeoatherina Gaudant, already known in France from several Late Eocene and Oligocene localities. This is demonstrated by the occurrence of a separate spinous anterior dorsal fin, originating slightly behind the middle of the body and above, or just behind, the base of the pelvic fins which have a subabdominal position. Nevertheless, these fishes differ from the four already known species of the genus Palaeoatherina Gaudant. For this reason, they are described in the present paper as Pa!aeoatherina giraudi (Piton) which may be defined by the following diagnosis: « Small sized Palaeoatherina, the standard length of which does not exceed 45 mm. Body slender: maximum height included 4 to 5 times in the standard length. Head moderate: length around 1/4 of standard length. Caudal fin moderately forked, length approximately 1/5 of standard length. Anterior dorsal fin beginning generally slightly behind the middle of the body; 7-9 slender spines. Posterior dorsal fin opposite to anal fin; I+8-9 rays. Anal fin generally beginning slightly in front of the posterior dorsal fin; I+10-11 rays. Pelvic fins subabdominal, placed either slightly in front of or opposite to the origin of the anterior dorsal fin; I+5 rays. Pectoral fins attached against the lower third of the flank; roundly 10 (?) rays. Vertebral column: 36-37 vertebrae, 18-19 being postabdominal. First haemal arches without expanded base. Scales cycloid ». Lectotype: Specimen M.N.H.N.P. 1977-5-1 (L. Piton’s collection), kept in Paris, in the Palaeontological Institute of the National Museum of Natural History. Locus typicus: Diatomite outcropping near the Jallat pass, near Chauriat (Puy-de-Dôme). Stratum typicum: Late Oligocene or Early Miocene (?). The revision of the fossil fishes found in two localities of the Puy de Mur massif confirms that the Late Oligocene (and, possibly, Early Miocene?) fish-fauna of this region differs from that of the Stampian of Limagne. First, the occurrence at Dallet, in the lower part of the Late Oligocene, of a Cyprinid belonging to the genus Rutilus Rafinesque, bears witness of the lacustrine conditions which prevailed at that time in this area, as already suggested by some other finds of Cyprinid fishes (Gaudant, 1984). Second, the occurrence of the genus Pa!aeotherina Gaudant in the Oligo-Miocene diatomite outcropping around the Jallat pass is, by itself, of little significance from a palaeoecological point of view, although this genus is, to date, only known from continental sediments. The identification of halobiont diatoms in the fossiliferous layer bears witness of brackish conditions, although the abundance of Botryococcus demonstrates that this diatomite was generated in lacustrine or slightly brackish waters (H. Méon, in litt., 29/5/1991). A fluctuation of the salt content may be expected from the great variability of polyhalobous diatom percentages, observed by Gasse and Gouhier (1971) in the diatomite layer.