OKJYAMA: MYCTOPHIFORMES 



259 



logenetic association between Aulopidae and these families has 

 been suggested on the basis of larval characters and the similar 

 mode of anal fin support (Okiyama, 1974b, 1979b). To these 

 can be added the peculiar structures on the chorion surface of 

 the extremely transparent eggs, the pigmentation patterns in the 

 newly hatched larvae, and the mode of reproduction shared by 

 these families, characters which favor their close association. 



Bathysauridae is distinguished from other families of this 

 suborder by some trenchant differences in the peritoneal pig- 

 ment sections and the mode of reproduction, while two derived 

 states are shared by all families. The phylogenetic relationship 

 of these families depends on whether the above mentioned dif- 

 ferences are due to divergence. Larval stages of Bathysauridae 

 are surely highly specialized, adapting to a prolonged pelagic 

 life, but larval dentition described in detail by Rosen (1971) 

 and Johnson (1974) and the character state of the axial skeleton, 

 including the mode of anal fin support (Okiyama, 1976b) are 

 of particular interest in showing the pattern common to Ipno- 

 pidae. 



Alepisauroidei: Paralepididae, 



Anotopteridae, Evermannellidae, 



Omosudidae, Alepisauridae 



The similarity matrix provides certain indication of the co- 

 hesiveness of this suborder. Most remarkable is their common 

 sharing of the derived state of character 8. Regarding the per- 

 itoneal pigment sections dividing five families into two groups, 

 some comments are warranted for Alepisauridae. As discussed 

 by Johnson ( 1 982), this character state is very tentatively defined 

 due to the inadequate state of available material. Even so, a 

 distinct family pair of Alepisauridae and Omosudidae can be 

 readily separated from the remaining families by the many de- 

 rived character states shared by them. Although the possibility 

 of their convergence cannot be fully rejected in view of the clear 

 contrast in the ontogenetic aspects of the pectoral fins, the close 

 similarity between Alepisaurus ferox and Omosudis lowei (trop- 

 ical western Pacific specimen) (see my Myctophiformes: De- 

 velopment, Fig. 1 1 2B, E, F, this volume), in head armature and 

 pigment pattern is extremely striking. 



An association between the Anotopteridae and Paralepididae, 

 particularly the more elongated paralepidids such as Stemo- 

 nosudis and Macroparalepis (Rofen, 1 966a, c), can be seen from 

 the larval standpoint. In addition to their shared derived char- 

 acter states (character 7 and 8), a fleshy projection on the lower 

 jaw tip peculiar to Anotopteridae and Stemonosudis macrura. 

 and the similar larval dentition (huge canines) may substantiate 

 the above association. Their disagreement in the character of 

 the peritoneal pigment sections is probably associated with the 

 odd systematic position of Anotopteridae lying at "an extreme 

 specialized end-point of the paralepidid line" (Rofen, 1966a, c). 



On the basis of the larval characters, two subfamilies of Par- 

 alepididae are well separated. As compared with the relative 

 constancy of conservative characters in larval Paralepidiinae, 

 the many derived character states of larval Sudinae are too 

 specialized to be consistent with the accepted subfamilial level. 

 The latter may be an earlier offshoot preceding the remarkable 

 paralepidine radiation. The complete lack of intermediate forms 

 between them offer strong support for this suggestion. 



As in Scopelarchidae (Johnson, 1974b), the systematics of 

 Evermannelidae were studied in detail using a large character 

 suite, including larval aspects (Johnson, 1982). So far as the 

 present analysis is concerned, this family seems variously as- 

 sociated with families of Alepisauroidei such as Paralepididae, 

 Alepisauridae and Omosudidae, besides Scopelarchidae. It is of 

 interest that limited character states shared by Evermannellidae 

 and Alepisauridae are restricted to derived ones, probably sug- 

 gesting their close association. Perhaps, an Evermannellidae and 

 Scopelarchidae linkage is much more loose, if valid. 



Concerning the possible three main lineages in this order, the 

 larval evidence is less promising. However, additional larval 

 evidence regarding developmental sequences, including osteol- 

 ogy as well as internal morphology, would provide much more 

 fruitful information for elucidating the phylogeny of this inter- 

 esting group. 



Ocean Research Institlite, University of Tokyo, 1-15-1, 

 MiNAMiDAi, Nakano-ku, Tokvo 164, Japan. 



Gadiformes: Overview 

 D. M. Cohen 



GADIFORMES is a particularly interesting order with which 

 to work because it encompasses a high degree of diversity 

 that suggests the existence of several lineages, apparent conver- 

 gence and reductive trends to trap the unwary, a useful fossil 

 record that allows a consideration of the distribution in time of 

 some important taxa and character states, and new suites of 

 characters based on the study of ELH stages. 



Although study of the classification of gadiforms dates from 

 pre-Linnean times, there is still insufficient properly evaluated 

 data available to derive a phyletic classification. In fact, there 

 is not even agreement as to what should be included. Berg (1947) 



restricted the order to the muraenolepids, bregmacerotids, mor- 

 ids, and gadids (including merlucciids) and excluded the mac- 

 rourids. He noted primitive and advanced characters in his 

 gadiforms and suggested derivation from primitive fishes. Rosen 

 and Patterson (1969) revived an expanded Gadiformes dating 

 at least from the time of Gill, which included not only gadoids 

 and macrouroids but also ophidioids and zoarcoids, and which 

 they placed in a supraorder Paracanthopterygii, postulated as 

 being, "in many ways more primitive than the acanthoptery- 

 gians" and representing "a spiny-finned radiation more or less 

 comparable morphologically with that of the Acanthopterygii" 



