Bull. nat. Hist. Mus. Lond. (Zool.) 61(2): 121-137 



Issued 30 November 1995 



Association of epaxial musculature with 

 dorsal-fin pterygiophores in acanthomorph 

 fishes, and its phylogenetic significance 



RANDALL D. MOOI _ (^ 



Milwaukee Public Museum, 800 West Wells St., Milwaukee, WI, U.S.A. 53233-1478 



ANTHONY C. GILL 



Department of Zoology, The Natural History Museum, Cromwell Road, London SW7 5 BD 



Synopsis. A survey of acanthomorphs reveals that epaxialis attachments to distal radials or the distal tips of 

 proximal-middle pterygiophores have a relatively restricted distribution. Four basic morphotypes are recognized: Type 0- 

 no distal insertions of epaxialis (lampridiforms, polymixiiforms, basal paracanthopterygians, zeiforms, beryciforms, 

 smegmamorphs, pleuronectiforms and many perciforms); Type 1 - partially separate epaxialis slip(s) inserting on to 

 dorsoposterior and dorsolateral processes of proximal-middle and/or distal radials (batrachoidids [Paracanthopterygii], 

 scorpaeniforms, and among perciforms in blennioids, most cirrhitoids, apogonids, centrogeniids, latine centropomids, 

 grammatids, haemulids, percids, serranids, champsodontids and cheimarrhichthyids); Type 2 - insertions of epaxialis to 

 distal portions of pterygiophores without separate slips (possibly basal tetraodontiforms, various perciform taxa including 

 callionymoids, notothenioids, zoarcoids, and some cirrhitids, labrids, percoids and trachinoids); Type 3 - completely 

 separate slip of muscle dorsal to the main body of the epaxialis inserting on to anterior pterygiophore shaft with dorsal 

 insertions on to more posterior spine-bearing pterygiophores, and the first ray-bearing pterygiophore, then becoming 

 continuous with the supracarinalis posterior (percoid family Mullidae). Type is considered to be plesiomorphic, and the 

 remaining morphologies apomorphic. Their phylogenetic significance is discussed in the context of other characters. 

 Among our conclusions, the Scorpaeniformes is awarded subordinal status within the Perciformes, and the centropomid 

 Latinae is given full familial status. 



INTRODUCTION 



Within the last five years, there has been renewed interest in 

 higher relationships among acanthomorphs. The recent pub- 

 lication of the Symposium on Phytogeny of Percomorpha 

 (Johnson & Anderson, 1993) and other contributions 

 (Stiassny, 1990; Stiassny & Moore, 1992) have shifted the 

 focus somewhat from phylogenetic work on individual fami- 

 lies to broader studies involving interrelationships of subor- 

 ders and orders. Such studies are hampered by the difficulties 

 inherent in examining large numbers of taxa, determining 

 appropriate character complexes, and interpreting homolo- 

 gies among the variation within those complexes. In many 

 instances, characters are too complex or difficult to survey 

 resulting in an incomplete understanding of their distribution 

 within the included groups. During the course of investiga- 

 tions on the relationships among pseudochromoids (sensu 

 Mooi, 1990), we began surveying the relation of dorsal 

 epaxial myology to the dorsal-fin pterygiophores. Dorsal 

 epaxial myology appears to exhibit limited but sufficient 

 variation over a broad range of taxa and the character states 

 are simple enough to suggest it to be of high potential for 

 phylogenetic analysis of higher relationships among acantho- 

 morphs. 



Epaxial muscles, the dorsal component of the body muscu- 

 lature, have received little attention from fish systematists. 

 Although some studies have used variation in the anterior 

 insertions of epaxial slips on to the head (e.g., Mooi, in press; 



Stiassny, 1990), few workers using myological features have 

 surveyed this muscle group (Winterbottom, 1974a for a 

 review). Mok et al. (1990) were the first to report variation in 

 the relationship of the epaxial musculature with the dorsal-fin 

 pterygiophores. They found that in two percoid families, the 

 Grammatidae and Opistognathidae, the epaxial muscles 

 insert on to the distal portions of anterior dorsal-fin pterygio- 

 phores, and interpreted this as evidence for uniting the two 

 taxa as sister groups. 



Our continuing studies on the phylogenetic positions of the 

 Grammatidae, Opistognathidae and other pseudochromoid 

 families have failed to provide corroborating evidence for a 

 sister-group relationship between the Grammatidae and 

 Opistognathidae. Moreover, a preliminary survey of epaxial 

 morphology in perciforms revealed that the reportedly 

 unique association of epaxial musculature with dorsal-fin 

 pterygiophores described by Mok et al. (1990) is more widely 

 distributed (Gill & Mooi, 1993: 333). Here we present an 

 extensive survey of acanthomorph taxa, and show that, 

 despite having a wider distribution than indicated by Mok et 

 al. , epaxial muscle/dorsal-fin pterygiophore associations nev- 

 ertheless appear to be relatively restricted within acantho- 

 morphs, and exhibit a number of recognizable morphologies. 

 We explore the possible phylogenetic significance of the 

 distribution of epaxial muscle insertions to dorsal-fin ptery- 

 giophores and their homology. 



)The Natural History Museum, 1995 



£^ ^M*^\ 



