298 Mr. G. A. Boulenger on the 



any very close relationship between these two aberrant groups 

 lias been abandoned by liim *. 



In spite of the absence of spines in the fins, the Gadoids 

 offer a combination of cliaracters — closed air-bladder, jugular 

 ventral fins, reduced parietal bones, maxillary excluded from 

 the border of tlie mouth — which, taken together, indicate 

 descent from the Acanthopterygians, and not from the lower 

 Teleosteans, a conclusion furtiier supported by their relation- 

 ship to the Blenniids and Trachinids. This being admitted, 

 it follows that the characters which serve to define them as a 

 group are the result of specialization, not primitive. These 

 characters are : — 



1. The diphycercal or isocercal t termination of the verte- 

 bral column. This has often been regarded as a primitive 

 character ; but if we accept, as I do, the conclusions of Dollo 

 in his remarkable discussion of the Dipneusti \, we cannot 

 liesitatetolay down as an axiom that all Teleosteans are origi- 

 nally descended from heterocercal forms. But the caudal tin 

 may become reduced or disappear, as in the series Mormi/rops-^ 

 Oymnarchns, UrencIielt/s-Murcetia^ Thyrsites-Tnchiurus, 

 Pleuronectes-Cynoglossus, to mention only examples in which 

 the direction of the line of evolution does not seem open to 

 controversy ; and if it should reajjpear, it cannot be again in 

 the specially modified condition known as homocercy. Such a 

 form of secondary caudal fin is exemplified among the Crosso- 

 pterygians by the Ccelacanthidje §. 1 have reason to believe 

 that the Gadoids must have been derived from such a group 

 as the Berycidfe, through forms of which the Macruridaj, 

 with thoracic ventral fins composed of 7 to 12 rays, are the 

 nearest known examples, and in which the caudal fin had 

 entirely vanished. 1 regard the isocercal condition of the 

 Gadidffi as the result of the formation of a new caudal fin, 

 the homocercal extremity of the vertebral column having 

 been lost b}' the direct ancestors of these fishes. 



2. The relations of the bones supporting the pectoral fin, 

 which differ considerably from those of the earlier Acantho- 

 pterygians. The scapular bone is imperforate and the fenestra 

 is situated between it and the coracoid. Of the basalia or 



* It is not improbable that the Trachypteridae have branched off from 

 the hypothecial primitive Acanthopterygians out of which the Berycidae, 

 Zeidse, and Macriiridae may be derived. 



f For definitions of these terms, cf. Boulenger, Poiss. Bass, du Congo, 

 p. 7 (1901). 



t Mem. Soc. Beige G^ol. ix. 1895, p. 79. 



§ Vf. Bashlbrd JJean, ' i'ishes Living and Fossil," p. 153, figs, loo and 

 15(3 (1895). 



