Ariomma geslini, Carnevale & Bannikov, 2006

Carnevale, Giorgio & Bannikov, Alexandre F., 2006, Description of a new stromateoid fish from the Miocene of St. Eugène, Algeria, Acta Palaeontologica Polonica 51 (3), pp. 489-497 : 490-494

publication ID	https://doi.org/ 10.5281/zenodo.13651560
persistent identifier	https://treatment.plazi.org/id/03946631-FF99-FFBC-EF01-FC8FFB01FA35
treatment provided by	Felipe
scientific name	Ariomma geslini
status	sp. nov.

Treatment

Ariomma geslini sp. nov.

Figs. 1 View Fig , 2A View Fig .

Holotype: MNHN ORA1540, incomplete fish from the Messinian of St. Eugène , Oran, Algeria.

Etymology: Specific name in honour of Dr. Léopold Geslin.

Diagnosis.—An Ariomma with pleural ribs robust and slen− der, extending posteriorly on the anteriormost caudal vertebra; posterior parapophyses and anterior haemal spines enlarged antero−posteriorly; proximal shaft of the anteriormost anal−fin pterygiophore bent backward over the four adjacent pterygiophores; sequential arrangement of anal−fin pterygiophores in relation to haemal spines 5_1_1_1_1_1_2_1_2_1; second dorsal−fin and anal−fin formulae (I + 15) and (III + 15) respectively; pectoral fin consisting of 21 elongate rays; body covered with large scales.

Measurements (mm).—Length of incomplete specimen ( Fig. 1 View Fig ): 122; maximum body depth: 51.7; (preserved) total dorsal−fin length: 94.5; soft dorsal−fin length: 61; anal−fin length: 59.

Description.—Most of the head skeleton and caudal region are missing ( Fig. 1 View Fig ). The body is elongate and laterally compressed; its depth measured at the level of the pelvic−fin origin approximately corresponds to the length of 11 vertebrae.

The cranial bones are missing except for some fragments of the posteroventral portion of the opercular complex.

The vertebral column is robust and almost straight. Twenty−one vertebrae are preserved. The separation of vertebrae into precaudals and caudals is rather problematic in ariommatids, and more generally in stromateoid fishes, because of the co−occurrence of a pair of pleural ribs and a haemal spine on one or more caudal vertebrae (caudal ribs sensu Tominaga et al. 1996). Moreover, as pointed out by Ahlstrom et al. (1976), there is a second complicating and obfuscating factor related to the backward protrusion of the organs of the abdominal cavity into the space ordinarily occupied by the anterior haemal spines. Aboussouan (1983) suggested that the separation between precaudal and caudal vertebrae can be defined by the position of the ventral folding of the caudal vein. Based on the description of Aboussouan (1983), the ventral folding of the caudal vein occurs just anterior to the first haemal spine. Because of preservation there is no way to determine this character on MNHN ORA 1540. Ahlstrom et al. (1976) found that the haemal spines never precede the anal fin pterygiophores. Following this criterion, the 11 th preserved vertebra is interpreted herein as the anteriormost caudal element of the vertebral column. The vertebral centra are strongly ossified, subrectangular, longer than high. The dorsal pre− and postzygapophyses and the ventral postzygapophyses are well developed on most of the preserved vertebrae, whereas the ventral prezygapophyses are present on the posterior two precaudal and on the caudal vertebrae ( Fig. 1 View Fig ). The lateral surface of the vertebral centra is extensively pitted. In general, the neural and haemal spines are relatively slender, moderately elongate, slightly curved and pointed, except for the haemal spine of the first caudal vertebra that is irregular in shape, laterally compressed and widely ornamented. The anteriormost 11 preserved vertebrae bear robust slender pleural ribs. The ribs originate on the parapophyses of the five posterior precaudal centra, and on the ventral tip of the haemal spine of the first caudal centrum. Epineural bones are poorly preserved.

There are two scarcely separated dorsal fins ( Fig. 1B View Fig ). The anterior portion of the spinous dorsal fin and the supraneurals are missing. Of the spinous dorsal fin, six fragile spines are preserved. Based on the meristic values commonly observed in extant congenerics ( Table 1), it is possible to estimate that the anterior three or four spines have been lost. The second dorsal fin is nearly complete. This fin consists of a single spine and 14 soft rays; an additional 15 th soft ray is not preserved being represented only by its pterygiophore ( Fig. 1 View Fig ). The spine associated with the second dorsal fin is longer than the terminal spine of the first dorsal fin. Each spine and ray of the dorsal fins has a non−serial arrangement with the pterygiophore immediately posterior to it (see Ahlstrom et al. 1976) in addition to regular serial association. The dorsal−fin pterygiophores are irregular in shape. Distinctly separated distal pterygiophores are present from the sixth soft ray posteriorly (= from the 12 th preserved element backward).

The anal fin inserts slightly posterior to the second dorsal fin origin. There are three spines and 13 soft rays, although 15 soft elements were certainly present, as indicated by the presence of a posterior rayless pterygiophore ( Fig. 1 View Fig ). The anal−fin soft rays are widely spaced. The spines progressively increase in size; the two anterior spines are in supernumerary association with the first pterygiophore. The third spine, as well as the following soft rays, articulate in a non−serial arrangement with the pterygiophore immediately posterior to them, in addition to serial articulation to the preceding pterygiophore. The anteriormost pterygiophore is greatly enlarged antero−posteriorly ( Figs. 1 View Fig , 2A View Fig ). It has an irregular shape and is ornamented on its lateral surface. The proximal shaft of this pterygiophore is bent backward over the four adjacent pterygiophores. The anal−fin pterygiophores are similar to those of the dorsal fin. A distinct separation between proximal + middle and distal pterygiophores occurs from the sixth pterygiophore posteriorly; similar to the condition observed in the dorsal−fin skeleton, such a separation occurs from the pterygiophore lying in the interhaemal space between the second and the third caudal vertebrae. The sequential arrangement of anal−fin pterygiophores in relation to haemal spines is as follows: 5_1_1_1_1_1_2_1_2_1. The five anterior pterygiophores are interpreted herein as lying in the interhaemal space between the first and the second caudal vertebrae ( Fig. 1 View Fig ).

The pectoral girdle is poorly preserved. The lower portion of the cleithrum is partially exposed; its upper portion was probably curved slightly forward. A robust elongate postcleithrum also is present. This bone is obliquely inclined and posteroventrally oriented. The scapula and coracoid are partially preserved. A median foramen appears to be present on the scapula. The pectoral fin inserts low on the flank of the body. The fin is relatively elongate and has at least 21 rays ( Fig. 1A View Fig ).

The pelvic fin origin is situated just behind the pectoral−fin base. The pelvic fin consists of one spine and five soft rays. The basipterygium is relatively long and narrow, wedge−like; an anterior process (see Stiassny and Moore 1992) appears to be present along the ventral margin of this bone.

Large deciduous cycloid scales cover most of the preserved body. A few lateral−line scales also are present; these scales are placed relatively high along the flanks and approximately follow the dorsal profile of the body. The tubes of these scales are branched (see Fig. 1B View Fig ).

Remarks.—Despite its incompleteness, the specimen examined herein shows several characters that clearly support a well−defined taxonomic placement. Its general appearance could justify its inclusion within the Stromateoidei . As stressed by Haedrich (1967), “…there is no mistaking the stromateoid look…, these fishes nonetheless have a physiognomy that nine times out of ten says stromateoid!…”. The relevance of the external morphology as a diagnostic feature of these fishes also was discussed by Regan (1902). However, most of the external features indicated by these authors refer to the head that is almost completely absent in MNHN ORA 1540. Fishes of the suborder Stromateoidei have a peculiar modification of the anterior portion of the gut; just behind the last gill arch there are several saccular outgrowths in the gullet that are equipped with numerous small teeth ( Haedrich 1967). These structures, called pharyngeal sacs, are present in all stromateoids except for the Amarsipidae (see Haedrich 1969). Based on the absence of the pharyngeal sacs and on the branchial muscles pattern, the placement of the Amarsipidae within the suborder Stromateoidei recently has been considered as questionable by Springer and Johnson (2004) who considered such a monotypic family to be incertae sedis among the percomorphs. The teeth of the pharyngeal sacs occasionally can be observed in fossil fishes ( Bannikov 1995), although they are usually obscured by the overlying bones of the opercular series and shoulder girdle. There are no traces of such teeth on the specimen described herein. Horn (1975) described a further synapomorphy of the stromateoid fishes, namely the possession of a relatively small, euphysoclistous swimbladder which forms in preflexion larvae and regresses before maturity. This character cannot be determined on the specimen, since it refers to a soft structure clearly not prone to the fossilization processes. In a subsequent paper, Horn (1984) considered the following three features as diagnostic of the Stromateoidei : (1) possession of cycloid scales, (2) preopercle scaled, and (3) presence of six hypurals. Of these characters only the first, presence of cycloid scales, can be observed on the specimen, while the two others cannot be seen because of the incompleteness of the material. However, two additional osteological features allow a well−supported inclusion of the specimen within the Straomateoidei: presence of a median scapular foramen (Regan 1902) and anterior prezygapophyses present on posteriormost precaudal vertebrae as well as on the caudal vertebrae ( Doiuchi et al. 2004). MNHN ORA 1540 also shares with selected stromateoids the posterior extension of pleural ribs on the caudal vertebrae (shared with Amarsipus , Ariomma , Cubiceps , Nomeus , Pampus , Peprilus , Psenes , Stromateus , Tetragonurus ; Doiuchi et al. 2004), the possession of two distinct dorsal fins (shared with Amarsipus , Ariomma , Cubiceps , Nomeus , Psenes ; Doiuchi et al. 2004) and the continued presence of pelvic fins in adult (shared by all the Stromateoidei except for the members of the family Stromateidae ; Doiuchi et al. 2004). According to Doiuchi et al. (2004), the family Ariommatidae is defined by several autapomorphies that mostly apply to features not exposed or not preserved on the specimen described herein (e.g., basisphenoid slender and rod−like in lateral view; dorsal end of the basisphenoid articulated only with the prootic; second infraorbital enlarged and slightly separated anteriorly from the first infraorbital; subocular shelf present; adductor arcuus palatini did not occupy the anteroventral part of the orbit; dorsal and ventral hypohyals articulating with one another through an interdigitating suture on the medial aspect; obliquus dorsalis inserts only onto the fourth epibranchial; procurrent spur absent; scapula and coracoid articulating with one another with an interdigitating suture; opercle not scaled). Nevertheless, the assignment of MNHN ORA 1540 to the family Ariommatidae , and more precisely to the genus Ariomma , is justified by the combination of a number of characters, including: pelvic fins present, dorsal fins scarcely separated, 15 soft rays present in the anal fin, presence of three anal spines not separated from the rays, soft dorsal and anal fins approximately the same length, scales thin cycloid and deciduous, lateral line high following the dorsal profile of the body, tubes in the lateral line scales branched, presence of 21 pectoral−fin rays, anal fin origin behind second dorsal fin origin, and pelvic fin origin behind the pectoral fin base ( Haedrich 1967; 1968; Nelson 1994; Doiuchi et al. 2004). The family Ariommatidae consists of two genera, the extant Ariomma and the fossil Isurichthys . The genus Isurichthys from the Lower Oligocene deposits of Canton Glarus, Switzerland traditionally has been interpreted to be related to the Scombridae (see Agassiz 1833 –1844; Wettstein 1886; Woodward 1901). Bannikov (1993) and Baciu and Bannikov (2004) tentatively referred it to the Ariommatidae based on several osteological, morphometric, and meristic features. Isurichthys clearly differs from Ariomma in having a shorter and deeper body. Thus, although a more detailed osteological analysis of Isurichthys is necessary in order to understand its phylogenetic placement, there are no substantive characters that would suggest a close relationship between this genus and MNHN ORA 1540. The anatomical investigation of MNHN ORA 1540 has identified features that strongly support its placement within the genus Ariomma ( Haedrich 1967; 1968; Doiuchi et al. 2004). Two groups of species can be easily recognized within the genus Ariomma : elongate species, with the maximum body depth less than 30% standard length (SL), and deep−bodied species, with the maximum body depth greater than 40% SL. The specimen described herein evidently belongs to the elongate species group. Five extant species constitute the elongate group: A. bondi Fowler, 1930 , A. brevimanum ( Klunzinger, 1884) , A. lurida Jordan and Snyder, 1904, A. melanum ( Ginsburg, 1954) , and A. parini Piotrovskiy, 1987 [see Karrer (1984) for synonymies]. According to Haedrich (1967), the problem of delineating the species of Ariomma is one of the most perplexing of the entire suborder, because almost all have the same or overlapping morphometric and meristic features ( Table 1; see also McKenney 1961; Lamkin 1997). Moreover, the small number of distinguishing features and the widespread and often disjunct distribution of the populations have made the determination of species limits a problematic matter ( Horn 1972). Karrer (1984) suggested that the morphology of the anterior anal−fin pterygiophores represents a character useful for the separation of the Pacific species of the genus Ariomma . The results of our investigation on extant representatives of the genus support the arguments proposed by Karrer (1984) ( Fig. 2 View Fig ). The analysis of the anatomical variability of the anal fin within the family indicates the diagnostic value of the orientation of the anteriormost pterygiophores and their interdigitation with the overlying haemal spines. The species of the elongate group have two different morphological patterns of the anterior sector of the anal fin ( Fig. 2 View Fig ). The first pattern is characterized by anteriorly oriented pterygiophores distinctly separated from each other ( Fig. 2C View Fig ). The second pattern is characterized by a completely different arrangement of the anteriormost anal−fin pterygiophores, which are posteriorly oriented and with the proximal shaft of the first pterygiophore bent backward over the adjacent pterygiophores. The second pattern is shared by A. bondi , A. melanum , and MNHN ORA 1540 ( Fig. 2A, B View Fig ). Thus, the specimen described herein seems to be closely related to A. bondi and A. melanum . Ariomma bondi and A. melanum are rather similar in body shape. Horn (1972) documented consistent recognizable differences between these species, related to the dimensions of the scales, number of lateral line scales, interorbital squamation, relative development of the cephalic lateral line, distribution of peritoneal melanophores, intestine length, and otolith morphology (see also Schmidt 1968). None these characters, except the dimensions of the scales, can be observed in MNHN ORA 1540. The development of relatively large scales (see description above) is indicative of some affinity between MNHN ORA 1540 and A. bondi . Ariomma melanum is characterized by small scales ( Horn 1972). However, A. bondi and A. melanum differ from MNHN ORA 1540 in that the pleural ribs are considerably thinner and more slender, the posterior parapophyses and the anterior haemal spines are not enlarged antero−posteriorly, the pleural ribs occur on two caudal vertebrae, the abdominal cavity protrudes further posteriorly into the space usually occupied by the anterior haemal spines, the proximal shaft of the first anal pterygiophore is bent posteriorly over the five posterior pterygiophores and the sequential arrangement of anal−fin pterygiophores in relation to the haemal spines is 6_1_1_1_1_2_1_1_2.

In summary, our study of MNHN ORA 1540 has identified several features that unequivocally support its placement as a new species of the genus Ariomma .