Ophidium to Anarrhichas, Achiri for the genera Channa 

 to Syngnathus. Thus, Batistes, along with the 

 mastacembelids, was one of the Apodes Hexapterygii, 

 while the other plectognaths, along with Channa and the 

 syngnathids, were Achiri Pentapterygii. Such a system 

 could scarcely be more artificial, but several new species 

 were described and the use of Orthagoriscus for molids 

 set a generic usage to be followed for the next hundred 

 years. 



One of the widest separations of the balistoids from the 

 other plectognaths was that made by Rafinesque (1810) 

 in his "Indice d'lttiologia Siciliana." He recognized two 

 subclasses of fishes, the Pomniodi (branchial apparatus 

 with both an operculum and branchiostegal membrane) 

 and the Atelini (branchial apparatus incomplete). The 

 Sotto-Classe Pomniodi contained four divisions based on 

 pelvic fin position; Guigulari, Toracici, Abdominali, and 

 Apodi. Each of these four "Divisione" was subdivided 

 into a variable number of "Sezione," which in turn con- 

 tained the orders. Tetrodon, Diodon, and Orthagoriscus 

 were placed in the Ordine Gli Odontini (Rafinesque 

 1810:40; all but the pelvic fins present; jaws in the form 

 of a bony beak), while Ostracion was placed in the Or- 

 dine Ostracidi (p. 39; body covered by a cuirass; all but 

 pelvic fins present; jaws with teeth not formed as a bony 

 peak). Both of these orders were in the Sezione 

 Branchiosomi (body short; spherical or elliptical) of the 

 Divisione Apodi of the Pomniodi. Batistes, however, was 

 elevated as the Ordine Balistini of the Divisione Gli Om- 

 nanchidi (branchial apparatus with a branchiostegal 

 membrane but without an operculum) of the Atelini. 

 Rafinesque's treatment of the plectognaths is best 

 remembered as a nomenclatural curiosity. 



The name of Cuvier looms large, of course, in the study 

 of plectognaths, and not only because it was he who es- 

 tablished them as a natural group at the ordinal level. In 

 his "Legons d'Anatomie Comparee," Cuvier (1805) fol- 

 lowed a classification (folding sheet at end of volume 1) 

 that borrowed heavily from previous systems. Fishes 

 were divided into those with "a squelette cartilagineux" 

 and those with "a squelette osseux," with the bony fishes 

 subdivided into Apodes, Jugulares,Thoraciques,and Ab- 

 dominaux. The cartilaginous fishes were subdivided into 

 two groups of Chondropterygiens and six groups of 

 Branchiosteges, of which two were the "bouche au bout 

 du museau; des dents" for Batistes and Ostracion, and 

 the "os des machoires tenant lieu de dents" for Tet- 

 rodon, Ouoides, Mota, and Diodon. 



The plectognaths were thus associated with much the 

 same genera as in the 12th edition of Linnaeus' (1766) 

 "Systema Naturae." An important improvement was 

 made, however, by the fact that Batistes and Ostracion 

 were placed together as a subgroup distinct from the 

 other plectognaths. This was the first time that such a 

 distinction had been made so clearly. Cuvier further con- 

 tributed to the study of plectognaths in this work, for 

 throughout all five volumes numerous anatomical notes 

 were given for this group. 



Cuvier (1817) put this anatomical information to good 

 use in his "Le Regne Animal," in which the Order Plec- 



tognathi was established, as one of the six orders of Pois- 

 sons Osseux (in contrast to Chondropterygiens), as fol- 

 lows; 



Ordre Plectognathes. 

 La premiere famille, ou les Gymnodontes. 

 La deuxieme famille, ou les Sclerodermes. 



Cuvier placed the Plectognathi as the first order of 

 Poissons Osseux, just after the Chondropterygiens, 

 because (1817:144) "il se rapproche un peu par I'imper- 

 fection des machoires, et par le durcissement tardif du 

 squelette; cependant ce squelette est fibreux, et en 

 general toute sa structure est celle des poissons or- 

 dinaries." He continued with the diagnosis of the order, 

 of which "Le principal caractere distinctif tient a ce que 

 I'os maxillaire est soude ou attache fixement sur le cote 

 de I'intermaxillaire qui forme seul la machoire, et a ce 

 que I'arcade palatine s'engrene par suture avec le crane, 

 et n'a par consequent aucune mobilite." He added that 

 the operculum and branchiostegal rays were hidden un- 

 der the thick skin and that the branchial aperture was 

 restricted. He further stated that there were only ves- 

 tiges of ribs, no pelvic fins, a large intestine without 

 caeca and usually a large swim bladder. The Gymno- 

 dontes (p. 145) were defined as those Plectognathi which 

 "A, au lieu de dents apparentes, les machoires gamies 

 d'une substance d'ivoire, divisee interieurement en 

 lames," while the operculum was small and there were 

 five branchiostegal rays. Three genera were de- 

 scribed; Diodon, Tetrodon, and Orthagoriscus. The 

 Sclerodermes (p. 149-150) were defined as being dis- 

 tinguished by "le museau conique ou pyramidal prolonge 

 depuis les yeux, termine par une petite bouche armee de 

 dents distinctes en petit nombre a chaque machoire. 

 Leur peau est generalement apre ou revetue d'ecailles 

 dures; leur vessie natatoire ovale, grande et robuste." 

 Five generic groups were recognized, three of them for 

 the first time: Batistes, Les Monacanthes, Les 

 Aluteres, Les Triacanthes, and Ostracion. 



This, then, was the basic anatomical classification of 

 the Plectognathi upon which all subsequent work on the 

 group rests. It is perhaps unfortunate that the order was 

 not more carefully diagnosed, for subsequent workers 

 found it easy to erode the structure that Cuvier had 

 erected. For instance, the principal character used by 

 Cuvier, the suturing (or at least immovable attachment) 

 of the maxillary to the premaxillary is true for all plec- 

 tognaths, except for the triacanthoids. But since Cuvier 

 had examined Triacanthus biaculeatus he might have 

 been expected to have noticed that his principal diag- 

 nostic character of the Plectognathi did not apply to one 

 of the genera. The definition of the order can also be 

 criticized because some nonplectognaths also have the 

 two bones of the upper jaw intimately connected. These 

 criticisms, of course, are valid, but by far the more im- 

 portant point is that it was Cuvier who clearly saw in the 

 early 19th century the naturalness of the plectognaths as 

 a group, elevated them to ordinal rank, and defined the 



