220 THE SWIM-BLADDER OF FISHES 



in the physoclistous perches, of which Perca flavescens is a 

 familiar example, and in the Squamipinnes (the brilliantly 

 tinted chaetodons, &c.)) it is likewise simple in form and 

 structure often showing a trace of anterior bifidity, (Plate 20, 

 fig. 2). It may still preserve its simple sac-like form, but be 

 paired as in Gymnotiis, the electric eel, or completely bifurcate 

 as in the phj^soclistous Brotulidaj and Acanthurida3. In the 

 simple swim-bladder of the cod or haddock, (Plate 20, fig. 1.), 

 a number of lateral processes fringe the upper margin of the 

 organ on each side, corresponding with intervertebral spaces. 

 while a pair of cyclindrical solid prolongations pass forward to 

 the auditory region ; but in the Sciaenidas, excepting Larimus, 

 which has a simple sac, the fringe of lateral processes becomes 

 elaborate. In Pogonias, the Drum-fish, they are broad, leaf- 

 like, and laterally unite on each side in a tube which opens into 

 the posterior end of the organ, (Plate 22, fig. 4.). In Johnius 

 they form a complicated fringe of ramose pectinate processes 

 and in the example of a Scifenid swim-bladder figured by Dr. 

 Gunther over a hundred of these pectinate, much subdivided, 

 processes occur, (Plate 22, fig. 3.). Another type of swim- 

 bladder is represented in this family by such a form as that 

 assumed in Corvina, (Plate 22, fig. 2.) There seems, indeed, 

 to be no limit to the variation in the details of this organ in 

 various groups. The internal cellular or sacculated character 

 found in the swim-bladder of the Notojjterida' seems to presage 

 the elaborate lung-like structure in some Ganoids and in the 

 Dipnoans ; but I have stated the grounds for not regarding this 

 latter organ as the homologue of the swim-bladder. It may be^ 

 and for the reasons set forth in this paper I do not doubt that 

 it is the case, that the ventral air-bladder or suboesophageal sac 

 even in Teleosteans such as the Gymnodonta, extending beneath 

 the skin of the abdomen and opening by a duct, with a sphincter 

 muscle, into the gullet, is a new structure and not homologous 

 with the dorsally connected swim-bladder of Teleostei generally. 

 If the modern sharks have lost the swim-bladder, or show 



