 |
| Fig. 4. Extinct lineages of Bythitidae and Dinematichthyidae inserted on the calibrated molecular phylogeny, numbers are discussed in the text.
All drawings are prepared by S.W. Knudsen. Drawings are not to scale. DOI: 10.1016/j.ympev.2016.04.008
|
Highlights
• A new family in Ophidiiformes, Dinematichthyidae, is established.
• Two major lineages, the viviparous Bythitoidei and the oviparous Ophidioidei, are supported.
• The subfamily Brosmophycinae is polyphyletic.
• Molecular age estimates and fossil data suggest the Ophdiiformes arose in the Cretaceous.
• Aphyonids are considered a part of Bythitidae.
Abstract
The order Ophidiiformes is a large but not very well known group of fishes, unique among teleosts for showing high diversity in both deep sea and shallow reef habitats. The current classification includes more than 500 species, 115 genera and four families, based primarily on mode of reproduction: viviparous Aphyonidae and Bythitidae vs oviparous Carapidae and Ophidiidae. Since 2004 we revised the bythitid tribe Dinematichthyini, described more than 100 new species and noticed that this group has unique morphological characters, perhaps supporting a higher level of classification than the current status. Here we study the viviparous families phylogenetically with partial mitochondrial (nd4, 16s) and nuclear (Rag1) DNA sequences (2194 bp). We use a fossil calibration of otolith-based taxa to calibrate the age of the clade comprising bythtitid and dinematicththyid representatives, together with fossil calibrations adopted from previous phylogenetic studies. The separation of the order into two major lineages, the viviparous Bythitoidei and the oviparous Ophidioidei is confirmed. At the familial level, however, a new classification is presented for the viviparous clades, placing Aphyonidae as a derived, pedomorphic member of Bythitidae (new diagnosis provided, 33 genera and 118 species). The current subfamily Brosmophycinae is considered polyphyletic and we propose a new family, Dinematichthyidae (25 genera, 114 species), supported by unique, morphological synapomorphic characters in the male copulatory apparatus. Previous use of the caudal fin separation or fusion with vertical fins is ambiguous. Age estimates based on calibrated molecular phylogeny agrees with fossil data, giving an origin within the Cretaceous (between 84-104 mya) for a common ancestor to Ophidiiformes.
Keywords: Bythitinae; Aphyonidae; Pedomorphism; Coral reef fishes; deep-sea fishes; Cave fishes
Family Dinematichthyidae Møller, Knudsen, Schwarzhans, Nielsen, 2016
Diagnosis
Male copulatory organ with penis and 1-2 (rarely 3) pairs of pseudoclaspers in cavity of ventral body wall covered by fleshy hood. Caudal fin free, not fused with dorsal and anal fins. First anal fin pterygiophore slightly to strongly elongated. Head pore system generally not reduced, 6 mandibular, 2-4 preopercular, 5-7 infraorbital and 3-4 supraorbital pores, including supraorbital pore above opercular spine. Posteriormost supraorbital head pore tubular. Vertebrates 10-25+25-65 = 36-87, dorsal fin rays 56-191, anal fin rays 40-124, pectoral fin rays 14-28, pelvic fin rays 1, elongate gill rakers 0-7, maximum lengths 22-197 mm SL.
The new family currently contains 25 genera, 114 species (see appendix B). An extraordinary large number of species and genera belonging to this group of cryptic reef fishes has been described in recent years (e.g. Møller et al., 2004, Møller et al., 2005b, Schwarzhans et al., 2005, Schwarzhans and Møller, 2007 and Schwarzhans and Møller, 2011), so a better understanding of the internal phylogeny of the family will have to await a few more alpha-taxonomic publications and more sampling of material suitable for molecular work. Worth mentioning from the present topologies is that the East Pacific and Atlantic genera (Ogilbichthys, Ogilbia and Typhliasina) form a monophyletic clade sister to a clade of Indo West Pacific taxa Diancistrus and Dinematichthys. The cave fish Typhliasina is sister to a clade with both Ogilbichthys and Ogilbia, supporting the validity of Typhliasina as an independent genus, and not assignment to Ogilbia as suggested by e.g., Nielsen et al. (1999) and Proudlove et al. (2001). Our results also confirm that the other cave-fish Diancistrus typhlops included here is not closely related to Typhliasina.
 |
| Fig. 4. Extinct lineages of Bythitidae and Dinematichthyidae inserted on the calibrated molecular phylogeny, numbers are discussed in the text.
All drawings are prepared by S.W. Knudsen. Drawings are not to scale.
|
Peter Rask Møller, Steen Wilhelm Knudsen, Werner Schwarzhans and Jørgen G. Nielsen. 2016. A New Classification of Viviparous Brotulas (Bythitidae) – with the Establishment of A New Family Dinematichthyidae – Based on Molecular, Morphological and Fossil Data. Molecular Phylogenetics and Evolution. DOI: 10.1016/j.ympev.2016.04.008
