REVISED CLASSIFICATION FOR CERTAIN CIRRHITOID GENERA 



The cirrhitoids as a monophyletic lineage 



On the basis of several apparently synapomorphic character- 

 istics (see below) the cirrhitoids would seem to be a mono- 

 phyletic lineage, a conclusion implied by both Gill (1862) and 

 Regan (1911) who described the group as a 'natural' one but 

 gave no reasons for that conclusion. The derived characters 

 on which I would base an hypothesised monophyly of the 

 cirrhitoids are, taken in conjunction, a reduced number (15) 

 of principal caudal fin rays, the unbranched lowermost five to 

 nine rays in the pectoral fin (usually with their tips produced 

 beyond the fin membrane), the lower part of each cleithrum 

 greatly expanded anteroposteriorly and meeting its antimere 

 in a deep, carinate symphysis, an increased number of 

 vertebrae relative to other percoids (26-35, comprising 10-16 

 abdominal and 15-21 caudal elements), and the presence, 

 ventrally, in subadults of a peculiar, lipoid-filled sac (Fig. 3), 

 free from the overlying hypaxial muscles, and extending from 

 the urohyal, to which it is attached, to the anus, with the 

 lipoidal material apparently contained in hexagonal compart- 

 ments. 



To the best of my knowledge, this lipoid sac has not previ- 

 ously been noted as a feature of subadult cirrhitoid fishes, nor 

 indeed of any other perciforms except the stichaeid Lumpenus 

 maculatus (see Falk-Petersen et al., 1984). I first observed it in 

 small specimens (the so-called 'paperfish' stage) of Cheilodacty- 

 lus pixi ca 43 to 44 mm standard length (Fig. 3), where its 

 presence results in the 'pouter-pigeon'-like ventral profile of the 

 paperfish stage in this and other cirrhitoid species (see photo- 

 graphs in Whitley, 1957; Allen & Heemstra, 1976, and Nielsen, 

 1963). Subsequent dissections revealed a lipoid sac in members 



of all but two of the cirrhitoid families I have dissected (see 

 p. 2). The exceptions are a chironemid, Chironemus marmora- 

 tus Giinther, 1860 (160 mm standard length) and an aplodac- 

 tylid, Aplodactylus lophodon Giinther, 1859 (180 mm S.L.). 

 Since, however, the sac is a juvenile {i.e. sub-adult) feature in 

 the other taxa, and the exceptional specimens were, to judge 

 from their gonadial development, young adults, I suspect that it 

 would also be present in smaller specimens of these species. In 

 Cheilodactylus pixi, for example, the sac is well-developed in a 

 specimen of 44 mm standard length, but has disappeared in one 

 of 46 mm. Likewise, in Chirodactylus brachydactylus (Cuv., 

 1830), it is present in a fish of 38 mm standard length, but absent 

 in one of 42 mm. The presence of a lipoid sac in specimens from 

 128 to 243 mm standard length of other species (referred in the 

 current literature to the Latridae and Cheilodactylidae) whose 

 maximum adult lengths are from 50 cm to one metre, suggests 

 that the size at which the lipoid sac disappears is positively 

 correlated with that at which members of a species become 

 adult. This supposition is borne out by the presence of the sac in 

 a juvenile Chirodactylus grandis (Giinther, 1860) of 57 mm 

 standard length, a species whose adults reach a length of one 

 metre, whereas it has disappeared, at a length of 42 mm, in 

 young Chirodactylus brachydactylus, whose adults reach a 

 length of 40 cm. Again, it is present in a specimen of Acanthola- 

 tris monodactylus 243 mm S.L.; adults of this species attain a 

 standard length of at least 65 cm. Thus, the sac's apparent 

 absence in chironemids and aplodactylids could be artefactual, 

 and related to the size-range of the specimens I was able to 

 examine. 



It is hoped to carry out a more detailed examination of the 

 lipoid sac when specimens suitably fixed for detailed histo- 



Fig. 3 Chirodactylus pixi, 49 mm standard length (RUSI 19842) in right lateral view; partially dissected, and with the greater part of the 

 pectoral and the entire pelvic fin removed. The large anterior portion of the lipoid sac (LS) is clearly visible; part of its posterior portion is 

 also visible (x). Throughout its length, the wall of the sac, unlike the muscles above it, is heavily peppered with melanophores. 



