EPAXIAL MUSCLES AND ACANTHOMORPH RELATIONSHIPS 



131 



,SCA 



EPAX 



Fig. 9 Type 2 epaxial musculature as exemplified by: a. Opistognathus maxilloxus (MPM 30098, 98.3 mm SL); b, Ronquilus jordani (MPM 

 394, 133.1 mm SL). Abbreviations and other methods of presentation as in Figs 1, 3. Scale bars = 5 mm. 



Fig. 10 Epaxial insertions via long tendons of Sparisotna rubripinne (MPM 30040, 62.6 mm SL), typical of some Type 2 epaxial muscles. 

 Abbreviations and other methods of presentation as in Figs 1 , 3. Scale bar = 5 mm. 



Fig. 11 Type 3 epaxial musculature as exemplified by Parupeneus multifasciatus (MPM 13530, 79.0 mm SL). In contrast to Types 1 and 2, 

 the dorsal epaxial has direct fibre insertion to only one anterior pterygiophore, and ventral to the articulation with the spine. These anterior 

 fibres merge with what is possibly a modified supracarinalis medius (SCM?), which has a similar anterior insertion and tendonous insertions 

 to a few posterior pterygiophores more dorsally. The epaxial muscle shares only a few fibres with the supracarinalis medius near the 

 posterior end of the first dorsal fin. The supracarinalis medius is contiuous with the supracarinalis posterior. SCM?, possible supracarinalis 

 medius; other abbreviations and methods of presentation as in Figs 1, 3. Scale bar = 5 mm. 



pterygiophore morphologies could help to determine the 

 relationships of some of the incertae sedis genera of the 

 Percoidei as identified by Johnson (1984: table 119). For 

 example, Siniperca has Type 2 musculature, which, although 

 a relatively common morphology, does circumscribe a 

 smaller perciform group from which possible relationships 

 could be initially explored. Johnson (1984) suggested a rela- 

 | tionship between Symphysanodon and Synagrops based on 

 [larval morphology. We find the former taxon to have Type 

 ! and the latter to exhibit Type 2 epaxial morphologies. 

 Although this does not refute a relationship, clearly more 

 work needs to be done. Other orphan percoid genera such as 

 Lateolabrax and Hapalogenys have Type morphology, 



which suggests they are unlikely to be included among Type 1 

 taxa such as the Serranidae and Haemulidae (where each 

 genus, respectively, had been traditionally placed). 



Many percoid families have not had their close relatives 

 identified. Epaxial morphology might limit the search for 

 possible relationships for some of these taxa. For example, 

 the Pholidichthyidae exhibit Type 2 morphology, and their 

 relationships might be narrowed to other taxa with this 

 morphology. Gill & Mooi (1993) summarized evidence sug- 

 gesting a possible relationship of the Notograptidae to acan- 

 thoclinine plesiopids. Notograptids and some acanthoclinines 

 share Type 2 morphology, which is absent in other plesiopids 

 (Table 1), and this perhaps provides additional support for 



