not possess a large amount of cartilage in the skeleton, a 

 characteristic he considered to be of great importance. 

 Rather, he felt that the plectognaths might possibly best 

 be placed in association with the Lophobranchii and 

 Lophius. If one wishes to place such an emphasis on the 

 amount of cartilage present in the skeleton of some plec- 

 tognaths, then one could just as well relate the order to 

 Cycloptenis as well as to Lophius, and return to the 

 Artedian Branchiostegi. 



In his "Studien iiber die Flossenmuskulatur der 

 Teleostier," Grenholm (1923:245) supported Rosen's 

 view that "Die alte Zweiteilung der Plectognathen in 

 Gymnodontes und Sclerodermi ist nach meiner Ansicht 

 . . . schwer aufrecht zu erhalten," while at the same time 

 admitting that the Plectognathi were a distinct group 

 related through the Acanthuridae to the other 

 acanthopterygians. It is of interest that Grenholm's 

 research on the fin muscles and Rosen's on the soft 

 anatomy lead to the same conclusion, a conclusion not 

 supported by the great majority of osteologically based 

 works, nor by Winterbottom's (1974) general myology of 

 the plectognaths. 



Whenever one wishes to identify most plectognath 

 genera, there is only one place to which one usually 

 turns, and that is to the long series of revisionary papers 

 by Eraser- Brunner (1935a, b, 1940a, b, c, 1941a, b, c, 

 1943, 1950, 1951). For species identification the 

 story is somewhat different. Fraser-Brunner 

 simply listed under the keyed-out genus the species 

 which he found to be recognizable. Only for the molids 

 and several genera of monacanthids were keys to the 

 species provided. However, since Fraser-Brunner 

 typically drew his generic lines rather finely, there are 

 usually only a few species within any one genus and the 

 practical matter of species identification is not as dif- 

 ficult as it might first appear. The exemplary service that 

 Fraser-Brunner performed will be appreciated by anyone 

 who attempts to identify plectognaths, especially those 

 from the Indo-Pacific. 



Throughout the series Fraser-Brunner promised even- 

 tually to publish a comprehensive monograph on the 

 order, but, since that has not yet appeared, there is 

 presented below his tentative classification as brought 

 together from his various publications. No genera will be 

 listed by name, for with Fraser-Brunner the number 

 usually accepted was greatly increased: 



Order Plectognathi 

 Suborder Balistoidea 

 Division Triacanthiformes 



Family Triacanthodidae (10 genera) 

 Family Triacanthidae (2 genera) 

 Division Balistiformes 



Family Balistidae (13 genera) 

 Family Aluteridae 

 Subfamily Aluterinae (21 genera) 

 Subfamily Anacanthinae (1 genus) 

 Suborder Ostraciontoidea 



Family Aracanidae (6 genera) 



Family Ostraciontidae 



Subfamily Ostraciontinae (3 genera) 

 Subfamily Lactophrysinae (3 genera) 

 Suborder Tetraodontoidea 

 Division Moliformes 

 Family Molidae 

 Subfamily Ranzaniinae (1 genus) 

 Subfamily Molinae (2 genera) 

 Division Tetraodontiformes 

 Subdivision Tetraodontines 

 Family Canthigasteridae (1 genus) 

 Family Lagocephalidae 

 Subfamily Lagocephalinae (1 genus) 

 Subfamily Sphaeroidinae (3 genera) 

 Family Colomesidae (1 genus) 

 Family Tetraodontidae 

 Subfamily Tetraodontinae (2 genera) 

 Subfamily Arothroninae (1 genus) 

 Family Chonerhinidae (2 genera) 

 Subdivision Diodontines 

 Family Diodontidae (3 genera). 



The above is obviously not complete, for Fraser-Brun- 

 ner has not yet assigned a place to Triodon, although he 

 (1943:4) did say that the "Tetraodontoidea are almost 

 certainly derived from the Triacanthiformes, a some- 

 what fragile connection between the two being provided 

 by the existing Triodon." 



Fraser-Brunner's classification is basically similar to 

 that of Bleeker's, with, of course, the refinements made 

 possible by Fraser-Brunner's own skill and by the results 

 of the researchers who intervened between these two. It is 

 nevertheless a compliment to Bleeker that Fraser- 

 Brunner should, nearly a hundred years later, arrive at 

 essentially the same classificatory scheme — although 

 Fraser-Brunner's diagnoses are, naturally, incomparably 

 better. Fraser-Brunner relied heavily on Regan's os- 

 teological observations and pointed out (1943:1), e.g., 

 that Regan's diagnosis of the Gymnodontes "here re- 

 garded as the Suborder Tetraodontoidea, cannot be 

 surpassed, and its salient features are given below, prac- 

 tically unaltered." This is not to imply that Fraser-Brun- 

 ner relied only on previous diagnoses of the subgroups, 

 for he contributed many original and significant obser- 

 vations on the osteology of various families, notably in 

 the tetraodontids, triacanthodids, and ostracioids. His 

 strong point was his combination of his own and 

 previously published osteological observations with his 

 knowledge of the relative importance of external charac- 

 teristics in arriving at his familial categories. This is par- 

 ticularly evident in his diagnoses of the tetraodontid sub- 

 groups, in which osteology (albeit only the top of the 

 cranium in most cases) is uniquely combined with the 

 structure of the nostril and lateral line, two systems 

 notoriously misleading in this group when used by them- 

 selves. 



In one of his more recent papers, Fraser-Brunner 

 (1950) speculated about the basic phylogeny within the 

 Plectognathi. There is little doubt today about the 



