464 



CHARLES S. TOMES. 



Gadus, the species seglefiuus, luscus, miuutas, tomcodus^ mer- 

 langus, vireas, and pollachius have been investigated. 



These genera are usually arranged^ with but little deviation 

 on part of the various writers, in the following order (5) : 



1. 

 2. 

 3. 

 4. 

 5. 

 6. 

 7. 

 8. 

 9. 

 10. 



Gad us. 



Merlucius. 



Lotella. 



Uraleptus, 



Phycis. 



Lota. 



Molva. 



Motella. 



Raniceps. 



Brosmius. 



Were we to attempt to classify them by their tooth structure 

 alone we should arrive at something like the following arrange- 

 ment : 



2. Merlucius 



7. Molva 

 10. Brosmius 



6. Lota . . 

 ? 4. Uraleptus 



9. Raniceps . 

 1. Gadus . . 



8. Motella . 



5. Phycis 

 ? 4. Uraleptus 

 3. Lotella . 



^ Highly developed systems of vascular canals, the outer 

 non-vascular portion being sharply marked off by the 

 " bounding canal : " in Brosmius the canal system is 

 reduced somewhat, and in Lota it is disappearing. 



The vascular canals, abundant in G. morrhua, becoming 

 less frequent in other species, but always retaining 

 rounded external loops ; in some species of Gadus 

 they are sparse, and in Motella there are hardly any. 



The vascular loops are so reduced that little can be said 

 as to their arrangement. In Lotella they are quite 

 absent. 



It will be noticed that the teeth of Brosmius (fig. 3, the 

 torsk) partake pretty closely of the structural characters of 

 those of the ling and the hake, whereas in the accepted classi- 

 fication it does not come near to them ; and this is perhaps the 

 most conspicuous instance of incompatibility between the tooth 

 structure and the usually accepted affinities of the creature met 

 with in the family. 



