__58 



through foramina in its mesial wall and issue from it by their respective 

 openings, while the nervus facialis, after giving off its ramus palatiuus, 

 enters the chamber by the external opening of the canalis facialis 

 (which opening lies close behind the processus basalis) and then issues 

 from the chamber by its facialis opening. I do not find the pituitary 

 vein described, nor is there a vein shown coming from the cranial 

 cavity to join the vena petrosa lateralis as that vein traverses the 

 antrum. The ganglia of the nervus trigeminus apparently lie in a 

 trigeminus recess of the cranial cavity, the ganglia of the facialis and 

 acusticus nerves lying in a separate acustico-facialis recess. The ramus 

 palatinus facialis does not enter the antrum. 



This antrum petrosum laterale of Salamandra is thus evidently 

 the homologue of the trigemino-facialis chamber of my descriptions 

 of fishes, and its mesial wall would seem to correspond to the mesial 

 wall of the chamber in teleosts, and not to that in Amia and Lepi- 

 dosteus (Allis, 1914a). This is in accord with the fact that the 

 ramus palatinus facialis does not enter the chamber in Salamandra. 

 And the further fact that there are apparently still a few lateralis 

 fibers in the nervus facialis of the adult salamander is in accord with, 

 but perhaps not related to, the fact that the nervus facialis still traverses 

 the chamber. For it is to be noted that when the lateralis fibers 

 associated with the trigeminus and facialis nerves are numerous, as 

 in fishes, the nervus facialis traverses the trigemino-facialis chamber, 

 while when those fibers are wholly wanting, as in mammals, the 

 nerve does not enter the chamber at all. 



In Triton taeniatus conditions similar to those in Salamandra are 

 said by Drüner to exist, but the antrum petrosum laterale is less 

 extensive, and the connection of the quadrate with the operculum is 

 not found. In Amblystoma a similar chamber is described by Winslow 

 (1898), and the nervus facialis is there apparently wholly excluded 

 from it, for Winslow says that the processus oticus and processus 

 palato-basalis (basalis) are there so fused with each other that the 

 only line of demarcation between them is a foramen that gives pas- 

 sage to a blood-vessel, this vessel undoubtedly being the vena capitis 

 lateralis although it is not so stated to be. 



It is thus probable that a trigemino-facialis chamber is found in 

 most if not all urodeles, and it is to be noted that in all these animals 

 the processus ascendens palatoquadrati fuses with the trabecular crest 

 (sphenolateral cartilage) before either the oticus or basalis processes 



