1902. 



PLECTOGXATHOUS FISHES. 



291 



vertebrae, not elongate, subequal in length except the three 

 preceding the square hypural, which are extremely shortened. 



3. Lactophrys Swainson. — Body 3-sided ; carapace with three 

 prominent ridges, a dorsal and two ventro-lateral. Vertebrae 

 fourteen, the first eight elongate, the four preceding the oblong 

 hypural shortened. In other characters like Ostracion. 



Division 2. Gymnodontes. 



Supraclavicle oblique, sometimes nearly horizontal ; lower three 

 pterygials enlarged and immovably united to the coraco-scapular 

 cartilage ; upper pterygial small, suturally united to the scapula 

 (see text-fig. 57). Anterior vertebrae with bifid divergent neural 

 spines. Basis cranii simple. Suture between dentary and articu- 

 lare evident. Pelvis absent. No spinous dorsal ; no anal spines ; 

 no ventrals ; caudal rays, if present, ten to twelve in number. 



Three families : Tetrodontidse, Diodontidee, Molidse. 



I cannot find that the true interpretation of the bones of the 

 pectoral arch in these fishes has been previously published. On 

 a superficial examination there appears to be no scapula, and the 

 pectoral fin to be supported by a series of four enlarged pterygials. 

 In fact, the united upper pterygial and scapula together resemble 

 one of the enlarged pterygials, both in size and shape, whilst the 



Text-fiff. 57. 



Riglit half (inner side) of pectoral ai'clies of (A) Diodoii jpunctidatus and 

 (B) Balistes verrucosus, 

 seh, supraclavicle ; cl., clavicle ; pel., postclavicle ; sc, scapula ; 

 cor., coracoid ; p^j"., pterygials. 



scaj)ular foramen corresponds to one of the series of interspaces 

 between them. The feature of the pterygials being immovably 

 attached to the scapula and coi"acoid, either directly or synchon- 

 drosially, is worth notice. The pectoral arch is so strikingly 

 similar in all three families, that Siebenrock's suggestion that in 

 Mola the bone which attaches the clavicle to the skull is the post- 

 temporal, and not the supraclavicle as in Tetrodontidae, cannot 

 be accepted. 



19* 



