1904.] OSTEOLOGY OF THE ELOPIB^ ANB ALBULID^. 69 



It might reasonably be urged that the interopercular is a bone 

 of the true opercular series which has been squeezed out of its 

 primitive position between the subopercular and the last branchi- 

 ostegal ray, in much the same way as the penultimate member of 

 the branchiostegal -opercular series has been reduced and forced 

 out of its position in the Palteoniscid fishes. The bone in question 

 was originally regarded by Traquair (" Ganoid Fishes Brit. Oarb. 

 Form.," Palseontogr. Soc. 1877, p. 20) as the subopercular, but 

 subsequently {ihid. 1901, p. 62) as an accessory plate. Smith 

 Woodward appears to regard it as the interopercular (Brit. Mus. 

 Cat. Foss. Fish. ii. 1891, p. 487). But, whatever may be the 

 homology of the bone, the evidence that it has been excalated 

 and finally lost is tolerably clear. In Teleosteans, however, there 

 is no evidence available to show that any such displacement of 

 the interopercular has taken place ; even in Pholidophorus, the 

 lowest of the Teleostean sei-ies, the interopercular already occupies 

 its definitive position in advance of the subopercular. 



In that veiy aberrant genus Phractolcenms, the preopercular iS' 

 small and the interopei-cular remarkably large. The latter bone 

 is situated below and anterior to the preopercular, and receives 

 from it the sensory canal that descends from the squamosal. 

 The interopercular is thus here performing the function of the 

 missing horizontal limb of the preopercular. In no other case 

 have I seen the interopercular conveying a sensory canal ; in the 

 vast majority of cases the interopercular is a thin lamina of bone 

 which is almost entii'ely concealed by the lower pai-t of the pi-e- 

 opercular. The evidence for putting the interopercular with the 

 preopercular is unsatisfactory, but owing to the perfect manner 

 in which the branchiostegal rays grade ofi" into the subopercular 

 and opercular, there is no justification for including it with these 

 last as a constituent of the skeleton of the gill- cover. 



The interopercular is of very regular occurrence, but it is said 

 to be wanting in Pantodon. As regards the preopercular, it may 

 be taken as a general rule, not however without exception, that 

 the horizontal limb of this bone is most developed in those forms 

 with a greatly reduced mouth, e. g. Gonorhynchus and Chanos, 

 and absent in those with a very large gape, e. g. Engraulis and 

 Coilia. In Albula and Alejjocephalus the vertical and horizontal 

 limbs are nearly equal. 



Circumorhital mid Nasal Series. — No results of any great 

 morphological impoi'tance are to be expected from a comparative 

 study of the bones of the circumorhital series. They are probably 

 more subject to variation — generic, specific, and individual — than 

 any other bones of the skull ; and the number of the bones 

 surrounding the orbit is not infrequently found to difier on the 

 right and left sides of the same skull. 



In Arapaima, Heterotis, and Osteoglossuni the enlargement of 

 the nasal bones, their meeting in a median siiture, and their rigid 

 union with the cranial bones, are features which are not en- 

 countered in any of the other genera under examination. The 



