132 



R.D. MOOI AND AC. GILL 



a 



EPAX 



Fig. 12 Epaxial muscle morphology in: a, Lates niloticus (ROM 28524, 80.8 mm SL); b, Psammoperca waigiensis (ROM 46627, 91.2 mm 

 SL). Note the insertions on to the second pterygiophore just ventral to the spine/pterygiophore articulation. Abbreviations and other 

 methods of presentation as in Fig. 1. Scale bars = 5 mm. 



Fig. 13 Type 1 epaxial musculature in Niphon spinosus (USNM 59739, 128 mm SL). Note the separate slip of muscle inserting on to the 

 second dorsal-fin pterygiophore and insertions to the 2nd through 8th pterygiophore, as in Epinephelus (Fig. 2). A separate bundle of fibres 

 originates tendonously from the 10th pterygiophore to merge with those from the main epaxial muscle body. Abbreviations and other 

 methods of presentation as in Fig. 1. Scale bar = 10 mm. 



their relationship, or at least does not contradict such a 

 conclusion. 



Variation within families exhibiting a particular morpho- 

 type has considerable potential for exploring internal rela- 

 tionships. Among serranids, the anthiines Hypoplectrodes, 

 Acanthistius , and Plectranthias all have very similar epaxial 

 morphologies (Fig. 15), in which a short and not highly 

 differentiated slip of muscle inserts on to the second pterygio- 

 phore, and a weak tendon extends from the myoseptum to 

 the first pterygiophore. This differs notably from the condi- 

 tion in more typical anthiines, such as Pseudanthias , where a 

 completely separate slip of epaxial muscle extends from 

 below the fifth pterygiophore to insert on to the first through 

 fourth pterygiophores (Fig. 16). These differences could 

 provide evidence to unite members of one or another of these 

 anthiine groups. If epinephelines are the sister group of 

 anthiines as implied by Johnson (1988) and supported by 



Baldwin & Johnson (1993), decisions concerning homology 

 and character definition become crucial; primitive epi- 

 nephelines {Niphon, Epinephelus) have a separate slip of 

 muscle inserting on to the second pterygiophore, but no weak 

 tendon to the first pterygiophore, a combination of features 

 found in the two anthiine groups (cf. Figs 2, 13, 15, 16). 



Variation in morphology of epaxial musculature might 

 prove useful in other taxonomic groups. Insertion patterns of 

 epaxial fibres to pterygiophores, the portions of the pterygio- 

 phore involved in the insertion, the degree of separation of 

 the involved musculature from the main body of the epaxial, 

 and the relationship of the muscle with the supracarinalis all 

 vary. Among the haemulids examined, Anisotremus has a 

 limited number of attachments involving only the fourth and 

 fifth pterygiophores, Conodon exhibits a more robust con- 

 tinuous series of insertions extending from the third to 

 seventh pterygiophores more typical of Type 1, and Haemu- 



