738 



it runs forwards along tlie auditory capsule and is separated from 

 it by the jugular vein and the facial nerve '). 



If we trace the diveriiciiluin from the base rostrally, we see 

 it after 8 sections already changed into a flat and narrow duct 

 with a lateral and medial wall. The duct is prolonged over 4 

 sections, and then with nearly no change of lumen, passes over 

 into the top part of the diverticulum, which is perceptahle on 9 

 sections. The medial wall of this part has over its whole length 

 a neuroinast, whose posterior end is clearly defined. Near the rostral 

 end (the blind top) of the diverticulum the branch of the ram. 

 oticus unites with the neuromast. 



We may now, proceeding from the anterior margin of the spiracle, 

 distinguish three parts, seen resp. on 8, 4 and 9 sections which we 

 shall call vestibulum of the spiracle, excretory duct and corpus of 

 the sense organ bladder. 



Excretory duct and corpus aie partners, but the vestibulum is 

 nothing more than an ordinary diverticulum of the anterior wall 

 of a visceral pouch, and disappears later, in consequence of the 

 enlargement of the external opening of the spiracle. 



The vestibulum is still present in an embryo 69 m.m. long, but 

 in embryos of 78 m.m. or more, it has disappeared. We then only 

 see on a section, passing posterior to the anterior margin of tiie 

 spiracle, the opening, which meanwhile has become very minute, of 

 the excretory duct. Then the condition of the sense organ bladder 

 principally corresponds to that of the organ which occurs in the 

 adult animal. It then forms an appendix of the spiracle. The descrip- 

 tion by VAN Bemmelen of the Galeoidei and rays also applies to the 

 sense organ of Acanthias. 



Probably these bladders are homologous in all the Selachians and 

 of ectodermal origin. They have in some forms sunk somewhat 

 deeper into the spiracle, than in others. We shall still examine the 

 little bladder sowewhat closer in a series of cross sections of the 

 Acanthias embryo 98 m.m. long. 



The very minute opening in the anterior wall of the spiracle is 

 only to be seen in one section. From here the organ passes rostral- 

 wards over 50 sections. It runs along the auditory organ from 



') During the translation of this paper I prepared a series of sagittal sections, 

 stained with haematoxylin and eosin, of a 22 m.n) long embryo of Torpedo 

 marmorata. I found the deep neuromast at tlie inner wall of the spiracle innervated 

 by a branch of the ram. oticus, crossing the outer side of the vena jugiilaris, just 

 as in Acanthias. 



