MOEPHOLOGY OF CKAXIAL MUSCLES IN SOME YE IM'EBliATES. 239 



trapezius is formed from tlie fourth levator, and jDcrsists ; 

 it is the muscle described by Allis as the " tiftli externus" 

 levator, " fouud iu 40 mm. fishes as a part of the fourth 

 levator." The portion of the fifth bi-anchial myotome above 

 the coraco-branchialis Y does not divide into levatoi- and 

 (dorsal portion of) obliquus ventralis ; it forms two muscles — 

 the second obliquus dorsalis and second adductor of Allis, 

 passing from the fourth epi- and cerato-branchial to the tiftli 

 cerato-branchi al . 



The development of the branchial muscles of Lepidosteus 

 is similar to that of Amia, the only exceptions being (1) the 

 coraco-branchialis V does not divide into pharyngo-clavi- 

 cularis externus and internus (Text-fig. 25) ^ ; (2) the inter- 

 arcualis ventralis IV is not developed. 



The differences between the Ijiauchial muscles of Salmo 

 and Amia are that in tlie former (1) the interarcualis ven- 

 tralis IV retains tlie primitive condition of a longitudinal 

 muscle between the fourth and third bars, and does not, as 

 in Amia, secondarily extend back to the fifth bar. (2) The 

 third levator arcuum branchialiuin, as well as the first two, 

 divides into external and inrcrnal portions, of whicli the 

 externus is inserted into the third epibrancliial, and the 

 internus into the fourth pharyngo-branchial. (3) The portion 

 of the fifth myotome above the coraco-branchialis V forms 

 one muscle only (obliquus dorsalis of Vetter), passing from 

 the fourth epibrancliial to the fifth cerato-branchial. (4) No 

 adductor is formed in the fourtli arch. 



It is noteworthy that the trapezius persists iu Salmo, as in 

 Menidia (Herrick) ; in Esox, Cyprinus, and Perca it is absent 

 in the adult (Vetter). In some Teleostei there are a greater 

 number of interarcuales ventrales present than in Salmo, 

 e.g. in Cyprinus an interarcualis ventralis I, and in Esox 

 an interarcualis ventralis III are additionally present (Vetter). 



The muscles attached to the ventral ends of the branchial 

 bars of Polypterus, apparently, are very different iu various 



' According to Fiirljriuger. " Lepidosteus hat kein conico-braucliialis 



