FISHES. 385 



of this order on a false hypothesis, namely, that these fishes had no 

 rays to their branchial membrane* : whilst every single one was really 

 furnished with them ; and Artedi himself even actually described them 

 in the lump-fishf . 



Linnaeus, after having placed the chondropterygians in the reptiles. 

 in his tenth edition, to which, by a combination, also without any 

 foundation, he joined lophius ; and, after having placed with the 

 branchiostegal fishes of Artedi, the mormyri and syngnathes, and 

 given them all the character of deficiency, not merely of rays in the 

 gills, but also of opercula, which, so far at least as many are concerned, 

 is at variance with the slightest observation, he united, in his twelfth 

 edition, the chondropterygians and the branchiostegal fishes in a single 

 order of reptiles (amphibia nantes) on a character still more strongly 

 opposed to nature, as they possessed, at the same time, gills and 

 lungs. 



Gmelin restored the two orders of Artedi, still attributing to the 

 branchiostegals the absence of the raj^s. Gouan characterizes them 

 solely by incomplete branchiae ; an expression very vague and ex- 

 ceedingly debateable in almost all genera. Pennant unites them to 

 the chondropterygians by the name of cartilaginous, a denomination 

 adopted by Lacepede, and of which we have already seen the impro- 

 priety. In fact, it is not admissible, in either a positive or negative 

 sense. Nobody can say, with any justice, that the skeleton of balistes is 

 cartilaginous ; and, in the number of fishes which Pennant and his 

 followers allow amongst the osseous fishes, there are some which, as 

 leptocephalus, have scarcely the appearance of a skeleton. 



I felt it necessary to occupy myself in the first place with separating 

 from these fishes, which are in some respect anomalous, those which 

 are so remote from the type of ordinary fishes as deserve to be so 

 removed, and then to find out their exact characters so as to clearly 

 explain them in words. 



This examination satisfied me that it was wrong to se])arate from the 

 whole mass of ordinary fishes, the lophius, lump-fishes, centurus, 

 mormynj, and macrorhynchi, which differed in no essential point from 

 ordinary fishes ; but I have found that the syngnathes, whose form 

 and economy are so singular, can be distinguished by their branchiae 

 in the shape of tufts concealed beneath an operculum, which allows for 

 the exit of the water only, by a small opening in the nucha, and that the 

 diodons, tetrodons, coffres, and balistes, independently of the incom- 

 pleteness of their skeleton, and the singularity of their form, have the 

 jaws, and in general, the whole skeleton of the head, a little different 

 from the common fishes ; that their upper jaw and their ])alatine l)ones 

 are articulated to each other, and with the vomer by immoveable 

 sutures, which leave them full freedom to open or close the mouth. 



* It is thus, at least, that the definition given of it is explained, — Geu. pise, 

 titul. vers. : branchiis osseis, ossibus destitutis, et p. 85, branchiostegi in brancbiis 

 nulla ossicula gerunt. 



t Gen. pise. p. 62, membrana branchiostega ossicula sex pracilia continet. 



I The macroihynsus of Lacepede or the silvery sygnathe of Bonneterre, is noth:n j 

 more than a lepidopus imperfectly described by Osbeck. 



VOL. II. c r. 



