224 H. S. Harrison, 



labyrinth wall, are clearlj^ in a position to transmit vibrations from peri- 

 lymph to enclolymph. I shall therefore speak of them tympanal areas. 



In discussing the results of previous observers, I shall be concer- 

 ned mainly with their descriptions of the ductus perilymphaticus and 

 its diverticula. By most of them the ductus was supposed to open 

 into the large space they believed to surround the whole labyrinth. 

 Villy however (1. c), described the ductus as ending blindly in the 

 perilymphatic tissue. 



As regards methods, I have relied mainly on the examination of 

 serial sections of specimens in various stages of growth. I have how- 

 ever, in several instances, made use of the wax -model method of 

 Born, or rather, the modification of this method pursued in the Ana- 

 tomisches Institut, Freiburg, The admirable arrangements existing in 

 the Institut render the preparation of wax -models a comparatively 

 simple task, the large projection-apparatus in particular contributing 

 greatly to the ease and rapidity with which drawings of a series of 

 sections can be prepared. 



I wish here to express my heartiest thanks to Professor Wieders- 

 heim for his kindness in placing the resources of the Anatomisches 

 Institut at my disposal. I am also under the greatest obligation to 

 Professor Gaupp, who suggested this research, and allowed me the 

 use of his many models and series of sections. In addition I have to 

 thank him for his ready assistance in various directions, and for the 

 interest he has taken in the progress of the investigation. 



II. The Perilymphatic System in the Urodela. 



As might be anticipated, the simplest condition of the perilym- 

 phatic spaces is to be found in the Urodela, although we can even 

 here trace the origin of the greater complexity existing in the Aniira 

 (see Diagrams 1, 2 and 3). Speaking generally, we recognise in the Uro- 

 dela, as the most extensive portion of the perilymphatic system, a large 

 space lying in the main laterally and externally to the sacculus, and 

 in close contact with the membrane closing the fenestra vestibuli 

 (fenestia ovalis). For this portion I shall use the term siuitium sac- 

 culare (see Diagram 1 and fig. 1 , *S'. 8.). From its cavity a duct. 



