DEVELOPMENT OF THE EAR IN THE COMMON EROG. 171 



faces, i.e., those looking towards and away from the brain respec- 

 tively. The outer wall is thickened and bears sensory cells, and it 

 is strongly curved, with the concavity towards the brain. The 

 opposite or inner wall is flat and exceedingly thin, consisting of a 

 layer of very flat pavement cells, and against this fiat part abuts the 

 perilymphatic canal. This canal can be traced downwards and back- 

 wards as far as a foramen in the inner and hinder part of the floor 

 of the capsule. Through this foramen it passes, and lies between 

 the pharynx and the base of the skull. Here it expands somewhat 

 (saccus perilymphaticus), though it is spacious throughout its course, 

 and from this point it extends backwards some distance, even 

 behind the foramen of the glossopharyngeal nerve. Just in front of 

 this nerve a duct, somewhat narrower than any part yet described, 

 passes up from the perilymphatic space through the foramen along 

 with the nerve, and enters the skull. Its connections within the 

 skull will be described below. The second canal has relations with 

 the pars neglecta of the cochlea, just as the first has with the 

 lagena. It is applied to the flattened wall of this pouch from 

 below ; besides this part in immediate connection with the cochlea, 

 there is a spacious continuation of it within the capsule winding past 

 the posterior ampulla, and lying above the hinder part of the utriculus. 

 The remainder of this canal passes through a foramen in the lower, 

 inner, and hinder part of the wall of the capsule (foramen rotundum), 

 and extends back within the brain case till it reaches and com- 

 municates with the previously described canal. The part within 

 the skull is of considerable size, occupying the lower angle on each 

 side, and thence it extends up the side wall for some distance, and 

 along the floor nearly to the median line. 



The course of these two canals will perhaps be rendered intelligible 

 by Fig. 19. This figure is diagrammatic, and compounded from a 

 large series of sections. By this method the whole course of the 

 system is shown, though it is foreshortened, so that structures are 

 brought into view that could not be cut in one section. Turning 

 now to the other forms in which I have found these structures, I will 

 how that two very similar canals exist in all, though their mutual 

 relatione vary. 



The toad 1 have only examined in a young stage just after the 

 tail is lost., hut I have no reason to doubt that at this time the adult 



