282 GENERAL ORNITHOLOGY part ii 



lowest of all, though little if any of it reaches farther backward than 

 the great loop of {(<) ; it is the second in size ; in shape it is quite 

 circular, — rather more than a half-circle. Its upper limb joins the 

 lower limb of (rt), as in man, and the two open by one orifice in the 

 vestibule ; but it is not simple union, for the two limbs, before 

 forming a common tube, twine half-round each other (like two 

 fingers of one hand crossed). The loop of (i) reaches very near the 

 back of the skull (outside). The canal (f) is the smallest, and, as it 

 were, set within the loop of (/*), though its plane is nearly the oppo- 

 site of the plane of (h) ; and the cavities of (i) and (c) intercom- 

 municate at or near the point of their greatest convexity, farthest 

 from the vestibule. This decussation of {h) and (c), like the twin- 

 ing inosculation of (a) and (6), is well known. It may not be so 

 generally understood that there is (in the eagle if not in birds 

 generally) a third, extra-vestibular communication of the canals. My 

 sections show this perfectly. The great loop of («), sweejiing past 

 the decussating-place of (jj) and (c), is thrown into a cavity common 

 to all three. Bristles threaded either Avay through each of the three 

 canals can all three be seen in contact, crossing each other through 

 this curious extra-vestibular chamber, which may be named the 

 trivium, or " three-way " place. (The arrangement I make out does 

 not agree well with the figiire of the owl's labyrinth given by 

 Owen, Anat. Vert. ii. 134. The trivium is at the place where, in Fig. 

 84 or 85, the three membranous canals cross one another. It does 

 not follow, however, that these contained membranous canals inter- 

 communicate, and it appears from Ibsen's figures that they do not. 

 Study of these admii-able illustrations, with the explanations given 

 under them, should make the details perfectly clear to the reader.) 



All that precedes relates to the hony labyrinth, — the scrolled 

 cavity of the periotic bone. The membranous labyrinth is a sac lying 

 loosely in the hollow of the bone, and shaped just like it, lining the 

 hollow of the vestibule and tubes of the semicircular canals. With- 

 drawn intact, it Avould be a perfect " cast " of the labyrinth. Ori- 

 ginally this sac is also continuous with one in the cavity of the 

 cochlea, called the membranoiLS cochlea, which afterward becomes 

 shut off from the main sac. This shut-off cochlear part lies between 

 the scala tympani below and the scala vestibuli above ; its interior 

 is the scala media. If demonstrable in birds, it must be quite as 

 rudimentary as the other scalre. The membrane is not attached to 

 the bony walls of the labyrinth, but is separated by a space con- 

 taining fluid, the 'perilymph, which also occupies the scala vesti- 

 Ijuli and scala tympani. A similar fluid, the endolymph, is contained 

 in the cavity of the membranous labyrinth, and scala media of the 

 cochlea ; in it are found concretions, or otoliths, of the same charac- 

 ter as the great " ear-stones " so conspicuous in many fishes. This 



