226 



OSTEICHTHYES 



and there is nothing in Polypterus to support it. According to 

 Sagemehl [379], the bladder, originally ventral and bilobed, has 

 shifted round the right side in the Dipnoi, and the left side in the 

 Actinopterygii. The median dorsal sac of the latter would represent 

 the left lobe only, and such genera as Erythrinus would show an 

 intermediate stage in the shifting. This theory is inconsistent 

 with the structure of Arnia, where not only are the bladder and its 

 opening median and dorsal, but the bladder receives a right and a 

 left afferent vessel directly from the last branchial arch (sixth of 

 the embryonic series), and the nerve -supply is also paired and 

 symmetrical. It is true that the right and left efferent vessels join 

 and, passing to the left of the oesophagus, enter the left ductus 



Fig. 199. 



A, air-bladder of Lebiqsina limaeiilata, opened to show internal cells. B, stomach of 

 Tchthyoborus niloticus, showing entrance of pneumatic duct, ab, anterior division ; ce, cellular 

 wall ; oe, oesophagus ; pb, posterior division ; p.d, pneumatic duct ; «.;>, pyloric end of stomach ; 

 st, stomach. (After Rowntree.) 



Cuvieri (Fig. 197, C) ; but it is easier to suppose that the return 

 stream of the blood has been diverted to the left side, than that 

 the arterial and nervous supply have been entirely readjusted after 

 the shifting of the air-bladder from a ventral to a dorsal position. 

 In all the Actinopterygii (excepting Amid) the blood-supply of the 

 air-bladder resembles that of the other abdominal viscera (Figs. 197, 

 351). As a rule, there appears to be a double arterial supply, from 

 the coeliac artery in front and from the dorsal aorta behind ; the 

 blood is returned to the cardinal veins from behind, and the portal 

 vein from in front (Corning). 



A third possibility remains to be considered. According to 

 this the most plausible theory, the air-bladder may be derived from 

 paired outgrowths, possibly a posterior pair of modified gill-pouches 

 (Spengel [415], Goette [170]). The ventral union of such out- 



