SKELETON AND LARYNX OF XENOPUS AND PIPA. 113 



nepliros subserving all the requirements of the tadpole until a late 

 larval stage ; and the direct development of the first three pairs 

 of appendages in those Crustacea in which the Nauplius stage is 

 suppressed, — the metamorphosis of the larval appendages is 

 omitted, and when the appendages appear they assume directly 

 the form of the antennule, antenna, and mandible of the adult. 

 On this hypothesis the difference in the mode of origin of the 

 thyrohyals in Xenopus and Pipa is probably not morphologic- 

 ally important. 



In 1881 Parker (34) propounded the view that the four pairs 

 of irregular cartilages (Spicula, Glaupp, 14) at the junction of 

 the branchial arches with the hypobranchial plate in the tadpole 

 were the true ceratobranchials of the fi^sh, and that the arches 

 themselves were but extrabranchials, — accessory structures 

 peculiar to the tadpole of Anura and not represented in the 

 larvae of Urodela. Also, that the fourth pair of these processes 

 became the thyrohyals. This view, although entertained in some 

 degree by Cope * (7. p. 244) and Schulze (38. p. 12), cannot be 

 substantiated by the facts of anatomy and development. Glaupp 

 (14. p. 414) forcibly describes it as " ganz absurd," and other 

 writers, such as Stohr (39) and Naue (31), still adhere to the 

 earlier interpretation. The relation of the vascular tissue to the 

 supporting cartilage in a tadpole is certainly exceptional, but 

 this is due to the enormous size of the pharyngeal cavity and the 

 thinness of its lateral perforated walls. The main branchial 

 blood-vessels lie external to the cartilage in both tadpole and fish, 

 whereas if the spicula are the true arches and the branchialia 

 something external to them, the vessels should run upwards from 

 the spicula to the roof of the pharynx. As it is, even the paired 

 aorta lies external to the branchialia of the tadpole. 



Marshall (24. p. 164) describes a pair of diverticula in the floor 

 of the pharynx, behind the last gill-clefts in Bana tadpoles, which 

 disappear after the metamorphosis, and he takes them to represent 

 in a modified form a fifth pair of branchial clefts. He does not 

 explain their relations to the thyrohyals, but since he says the 

 pits are situated at the sides of the glottis it is evident that they 

 are in their immediate vicinity. The presence of anything that 

 can be regarded as the fifth branchial clef fc lends support to the 

 hypothesis that the fi^fbh branchial arch is also represented. 



* Cope, however, contradicts himself by regarding the arches in PI. 50^ 

 fig. 2 as ceratobranchials. 



LINN. JOUKN. — ZOOLOaT, VOL. XXVI. 8 



