Notes on the Systematic Status of the Eels Neenchelys and Myroconger GARETH J. NELSONl

N eencbelys . Xenocongridae, Dysomminidae, and Muraeni­ BOHLKE (1960) SUGGESTED that eels of the dae (N elson, 1966b) . In completely lacking genus N eenchelys possibly have overlapping basibranchials, the arches of M yroconger differ branchiostegal rays and that, if they did,they from those of H eterencbelys, Anguilla, and should be assigned to the family Ophichthidae. M oringua, but resemble those of xenocongrids, Nelson (1966a) described the osteology of Dysommina, and muraenids. In lacking a second N eenchelys buitendijki, confirming the presence pharyngobranchial they are unlike xenocon­ of overlapping branchiostegal rays, and for this grids , but resemble Dysommina and muraenids. and other reasons referred the genus N eenchelys Like that of Dysommina the fourth arch of to the family Ophichthidae, subfamily Eche­ M yroconger is not appreciably enlarged and linae. The present report, based on an ex­ "pharyngeal jaws" like those of muraenids do amination of the holotype of Neenchelys micro­ not occur. Thus, the arches of M yrocol1ger are tretus, confirms the presence of overlapping most like those of Dysommina. The most nota­ branchiostegal rays in the type species of N een­ ble differences include the presence in M yro­ chelys. Like those of N . buitendijki (Nelson, conger of third hypobranchials (a primitive 1966a, fig. 2A), those of N . microtretus in­ feature) and the apparent absence of fifth clude six rays articulating with the dorsal por­ ceratobranchials (an advanced one) . tion of the ceratohyal and more than 25 others What could be learned of the pharyngeal widely overlapping in the midline. musculature also suggests a relationship with the more advanced eels of the anguilloid lineage, for a subpharyngealis occurs, as it does at least M yroconger in M oril1 gua, Kaupichthys, and muraenids, and Thi s genus and the family it represents ap­ retractor muscles have a small area of origin on parently are known only from the holotype of the vertebral column, foreshadowing the large M yroconger compressus. The specimen had area of origin in some muraenids (N elson, been partly dissected, leaving the gill arches 1967) . exposed, which allowed the following observa­ These observations of M yroconger complete tions to be made: third and fourth upper a review of gill arch structure for the families ph aryngeal tooth plates separate; first and sec­ of anguilloid eels (N elson, 1966b). Within ond pharyngobranchials absent, the third sup­ this group, on the basis of gill arch structure porting the tooth plates; basibranchials absent; there seem to be three main lines of specializa­ independent rodlike hypobranchials in arches tion, each characterized by reduction of the gill one-three, those of the third cartilaginous; arch skeleton : one leads toward the Morin­ fourth ceratobranchials not extended anteriorly, guidae, another toward the Muraenidae, the not separating the third arches of either side; other toward the Cyemidae. If the anguilloid fifth ceratobranchials apparently absent; ventral eels are given the status of a suborder, these parts of the arches not meeting in the midline. lines of specialization could be given the status M yroconger has the frontal bones separated of superfamilies. However, on the basis of gill by a suture and therefore belongs to the anguil­ arch structure alone it is difficult to distinguish loid lineage of Regan (1912) , including the between generalized members of these different Heterenchelidae, Anguillidae, Moringuidae, lines, or to decide which if any Recent forms can be considered generalized muraenoids. 1 Department of Paleozoology, Swedish Museum of Consequently, the following synopsis is offered N atural H istory, Stockholm 50, Sweden. Manuscript received January 20, 1967. Present address: Ameri­ more as a working hypothesis than as a final can Museum of N atural History, Ne w York, N .Y. classification: 562 Neencbelys and Myroconger-NELsoN 563

Caudal fin continuous with dorsal and anal; at the British Museum (Natural History) while no pelvic fins; frontal usually paired . I was on an NSF postdoctoral fellowship...... suborder Anguilloidei a. Jaws not produced ; gill arch skeleton in­ REFERENCES cluding at least rudimentary basibranchials BOHLKE, J. E. 1960. A New Ophichthid Eel of · ...... superfamily Anguilloidae the Genus Pses domyropbis from the Gulf (including families Heterenchelidae, An­ of Mexico. N otulae Naturae 329, 8 pp., 2 guill idae, Moringuidae) figs. b. Jaws not produced; gill arch skeleton with- NELSON, G. J. 1966a. Osteology and relation­ out basibranchials . ships of the eel, N eencbelys buitendijki . · ...... superfamily Muraen oidae Copeia 1966(2) :321-324, 2 figs. (including Xenocongridae, Dysomminidae, --- 1966b. Gill arches of teleostean fishes Myrocongridae, Muraenidae) of the order Angui lliformes. Pacific Sci. c. Jaws produced. "gill arch skeleton with or 20(4 ) :391-408, 58 figs. without basibranchials . - - - 1967 . Branchial muscles in repre­ · superfamily Nemichthyoidae sentatives of five eel families. Pacific Sci. (including Serrivomeridae, Nem ichthyidae, 21 (3) :348-363. Cyemidae) REGAN, C. T. 1912 . The osteology and classi­ fication of the teleostean fishes of the order Observations on type specimens were made Apodes. Ann . Mag. Nat. Hist., Ser. 8, 10: through the courtesy of Dr. P. H . Greenwood 377-387, 2 figs.