FAHAY AND MARKLE: GADIFORMES 



283 



on this structure since an undescribed ontogenetic sequence is 

 involved. This character is an important part of our current 

 concept of Merlucciidae, thus descriptions of its ontogeny could 

 contribute to a better understanding of this family's interrela- 

 tionships. 



Genilal papilla.— A. genital papilla develops precociously in most 

 gadiforms. It is most pronounced in morids. macrourids and 

 Steindachneria (see figures), but we also could find it in gadids 

 and Merluccius. 



Mental barbels. — Mental barbels usually develop late in the lar- 

 val or early in the pelagic juvenile period. They are found in 

 most gadids (being lost in some of the secondarily pelagic forms 

 such as Pollachhis pollachiiis and Micromesistiiis). most ma- 

 crourids, muraenolepidids and morids. Additional fleshy, snout 

 barbels are found in phycine rocklings. The propensity to de- 

 velop snout and mental barbels seems widespread in gadiforms 

 and can also be found in some ophidiiform fishes; it appears to 

 have a strong ecological component and we are unable to attach 

 phylogenetic significance to its presence or absence. 



Gadiform Phylogeny 



A framework of interrelationships of the eight gadiform fam- 

 ilies has developed from among others, Marshall ( 1 966b), Gos- 

 line (1968), Rosen and Patterson (1969), Okamura (1970b) and 

 Marshall and Cohen (1973). A concensus on minor as well as 

 some major points does not exist, and we therefore follow a 

 modification of Rosen and Patterson (1969) and Cohen (this 

 volume). In this framework muraenolepidids are the most prim- 

 itive group, showing no obvious relationships, and are the pre- 

 sumed sister group to all other gadiforms. Based on fossil 

 evidence (Danil'chenko, I960), bregmacerotids are thought to 

 be related to a group composed of morids and melanonids. 

 These three families are the sister group of macrourids and 

 together form a principal gadiform lineage. Steindachneria and 

 merlucciids are sister groups and with gadids form the other 

 principal gadiform lineage. 



On the basis of available data, we can identify the following 

 early developmental characters, their denved states, and known 

 distribution in the order. In many cases the "holes" in our data 

 severely reduce the weight of our arguments. (1) Oil globule in 

 egg— lost — gadines; (2) Chorion ornamentation — honey- 

 combed— macrourines; (3) Lateral premaxillary spines— pres- 

 ent— muraenolepidids; (4) Pterotic spines— present— some phy- 

 cines; (5a) Sequence of fin formation— caudal first— gadines and 

 Merluccius; (5b) Sequence of fin formation — pel vies last— gad- 

 ines; (6) Pelvic fin ontogeny— reduction in ray number— phycine 

 hakes and morids; and (7) Larval pectoral fin— pedunculate— 

 macrourids and Steindachneria. To this list we can add onto- 

 genetically persistent characters taken in part from Rosen and 

 Patterson ( 1 969), Marshall and Cohen (1973) and Markle (1982). 

 (8) X and Y bones— loss in forms with tails— melanonids, gad- 

 ines and lotines; (9) Total caudal fin rays— over 50 — melan- 

 onids, gadines and lotines; (10a) Anterior dorsal fin rays to 

 centra ratio — 7:1 —phycine rocklings; (10b) Anterior dorsal fin 

 rays to centra ratio— ca. 1:1— gadines, morids?, macrourids and 

 merlucciids: (1 1) Precaudal vertebrae— counts greater than 20— 

 gadines, lotines, merlucciids and muraenolepidids; (12) Hypur- 

 als— fusion into two plates— muraenolepidids, bregmacerotids, 

 gadids and meriucciids; (13) Otophysic connection — present— 

 morids; and (14) Fin diflferentiation— three dorsals and two 

 anals— gadines, some morids (Merluccius and bregmacerotids 

 to a lesser degree). 



These characters generally do not support the above hypoth- 

 eses of relationships. Notable discrepancies and areas for ad- 

 ditional investigation are: ( I ) whether gadids are monophyletic, 

 specifically whether phycines belong in and Merluccius belongs 

 out; (2) relationship, if any, of melanonids to gadines; and (3) 

 relationships of Steindachneria. 



(M.P.F.) National Marine Fisheries Service, Northeast 

 Fisheries Center, Sandy Hook Laboratory, High- 

 lands, New Jersey 07732; (D.F.M.) Huntsman Marine 

 Laboratory, Brandy Cove, St. Andrews, New 

 Brunswick EGG 2X0 Canada. 



Gadidae: Development and Relationships 

 J. R. Dunn and A. C. Matarese 



LARVAE of the fishes of the family Gadidae have received 

 a great deal of study through the years and because of the 

 economic value of the family, the larvae are taxonomically as 

 well known as those of most families of teleosts. Svetovidov's 

 (1948) classic work on the systematics of adult gadid fishes is 

 the benchmark of knowledge of the family. He considered 22 

 genera (including Merluccius). examined osteological characters 

 of representatives of all genera, and based his classification 

 scheme mainly on the structure and number of median fins (see 

 also Svetovidov, 1956). Subsequent workers (Mujib, 1967, 1969; 

 Marshall and Cohen, 1973) have extended our understanding 



of the relationships of certain members of the family, but a 

 comprehensive study of Gadidae, including early life history 

 stages, has not yet been accomplished. Recently Markle (1982) 

 examined larval and adult representatives of all gadoid families 

 which led him to recognize three gadid subfamilies: Phycinae, 

 Lotinae, and Gadinae. 



Our purpose here is to summarize available knowledge of the 

 taxonomy of eggs and larvae of the family Gadidae. We include 

 observations on eggs, larval morphology and pigment patterns, 

 and developmental osteology. Included are illustrations of lar- 

 vae of representatives of all currently recognized gadid genera 



