LEIS AND RICHARDS: ACANTHUROIDEI 



551 



rangids at early stages. Leiognathid larvae are similar to siganid 

 larvae in many respects (head spination, fin spine development, 

 silvery gut— see G. D. Johnson, this volume and Leis and Gold- 

 man, 1 983), and we suggest the leiognathids should be evaluated 

 as a potential primitive sister group of the acanthuroids. There 

 is little evidence from ELH characters to support the notion 

 that the acanthuroid fishes are the primitive sister group of the 

 tetraodontiform fishes (Leis, this volume). 



Intra-ordinal relationships of acanthuroid fishes as suggested 

 by ELH characters fully support those based on adult characters. 

 The siganids are distinguished from the other acanthuroids by 

 the following derived characters: demersal egg, two spines in 

 pelvic fin, and seven spines in the anal fin. Larvae of acanthurids, 

 luvarids, and zanclids have the following derived characters: no 

 elongate preopercular spines; kite-shaped body; elongate snout; 

 extremely elongate dorsal and pelvic spines; early-forming spe- 



cialized scales; and reduced number of dorsal fin spines. Thus 

 the siganids appear to be the primitive sister group of the other 

 acanthuroids. Interrelationships of the acanthurids, zanclids, 

 and luvarids cannot be clarified given the current knowledge of 

 zanclid and luvarid ELH characters. Larval zanclids have an 

 extremely elongate dorsal spine and a retrose preorbital spine. 

 Acanthurids have caudal peduncle armature, and luvarids have 

 ontogenetic reduction in fin elements, no anal spines, and a very 

 squared head. None of these specializations are shared by any 

 two of the families, so they shed no light on interrelationships. 



(J.M.L.) The Australian Museum, 6-8 College Street, 

 Sydney 2000, Australia; (W.J.R.) National Marine 

 Fisheries Service, Southeast Fisheries Center, 75 Vir- 

 ginia Beach Drive, Miami, Florida 33149. 



Blennioidei: Introduction 

 R. H. Rosenblatt 



THE modem concept of the perciform suborder Blennioidei 

 dates from the paper of Regan (1912b), who defined and 

 delimited the group as "Percomorphous Teleosts with the pelvic 

 fins jugular or mental, each of a spine and four soft rays or still 

 further reduced, with the dorsal and anal rays typically corre- 

 sponding in number to the vertebrae, each basal bone attached 

 to its own neural or haemal spine (rays more numerous in Ophi- 

 diiformes) with well developed wings of the parasphenoid as- 

 cending in front of the prootics, and with all or most of the ribs 

 inserted on strong parapophyses." 



As Regan himself indicated this definition encompasses a 

 heterogeneous group, and his series "Ophidiiformes" has now 

 been removed from the Perciformes. Subsequent to Regan sev- 

 eral widely differing classifications have been proposed, with 

 groups often being added or removed without comment. Jordan 

 (1923) proposed the most radical arrangement. He placed in the 

 order Jugulares almost all spiny rayed fishes with advanced 

 pelvic fins. Jordan's Jugulares was divided by him into 1 2 series, 

 comparable to suborders, and no less than 62 families. Jordan, 

 in his magisterial fashion, provided an outline classification, 

 without substantiation by characters. 



Berg in his 1940 classification rationalized the classification 

 of the Blennioidei. He restricted the suborder mainly to Regan's 

 series "Blenniformes" and "Cliniformes," and redistributed the 

 remainder of the Jugulares, either to the Percoidei or to the 

 suborders Ophidioidei (equivalent to Regan's series Ophidi- 

 iformes), Ammodytoidei, or Callionymoidei. Some indication 

 of relationships is perhaps implicit in Berg's placement of por- 

 tions of Jugulares auctorum immediately preceding the Blen- 

 nioidei. 



Although a number of works on various blennioid groups 

 have appeared (see particularly Hubbs, 1952; Makushok, 1958, 

 and the papers of V. Springer) the only subsequent attempt to 

 characterize and deal with the group as a whole is that of Gosline 

 (1968). The classification given by Nelson (1976) differs from 



that of Gosline as well as the outline classification of Greenwood 

 et al. (1966). The discussion of larval forms given here mostly 

 accords with Nelson's Blennioidei as a convenience, regardless 

 of the eventual disposition of the taxa. The only major departure 

 from the arrangement of Nelson is that the family Zoarcidae is 

 treated here, although Nelson included it in the Gadiformes (see 

 Anderson, this volume). 



The reasons for the varying treatment of these fishes are not 

 difficult to find. The unraveling of phyletic lines within the 

 Perciformes is made difficult by the sheer number of species 

 and genera. One is faced with the choice of mining a narrow 

 vein for nuggets of knowledge which lie isolated, or engaging in 

 a strip mining operation which reveals broad patterns at the 

 expense of ignoring contradictory details. In other terms, in- 

 sufficient knowledge of morphological variation within the Per- 

 ciformes precludes at this time either identification of unequiv- 

 ocal synapomorphies or the determination of polarity of a number 

 of characters within almost any presumed lineage. 



A number of features taken to characterize, if not to define, 

 the Blennioidei may be the product of convergent or parallel 

 evolution, correlated with the assumption of benthic life. 



As pointed out by Gosline ( 1 968) the blennioids, as compared 

 with percoids, have less deep bodies, with a short trunk and a 

 relatively attenuated caudal region. The dorsal and anal are long 

 and low, terminating near the caudal, and the pectoral and usu- 

 ally the caudal fins are rounded. There is an exact correspon- 

 dence in number between dorsal and posterior anal soft-rays 

 and vertebrae supporting them. The pelvic fins are inserted in 

 advance of the pectoral fins, and the number of rays is generally 

 reduced; the spine often rudimentary or splint-like, and the soft- 

 rays three or fewer. 



The deep, relatively compact body of a generalized perciform 

 is that of a fish which hovers, probably near the substrate, but 

 which makes rapid bursts either in feeding or predator avoid- 

 ance, or both. The body shape is adapted for slow swimming. 



