DEVELOPMENT OF THE SKULL IN THE CEOCODILIA. 281 



" cartilaginous annulus tympanicus " {(i-ty), with long cells, however, like those of 

 the stapedial base of the columella. This is quite like that found in Clielom viridis 

 ("Turtle's Skull," op. cit. pi. 10. figs. 10, a.ty). 



The main part of the hyoid arch, the ceratohyal [c.hy), is still perfectly continuous 

 with the mandible, and immediately in front of its confluence, on the outside, a diver- 

 ticulum of the tympanic cavity is growing downwards into the substance of the arti- 

 cular cartilage ; this is the " siphonium " {sph), forming, already, the large air-cell of the 

 lower jaw. There is still a joint between the ceratohyal and epihyal {e.hy), below; 

 and between that small, curved, normal segment and the extrastapedial [e.st), above. 

 That part of the columella is still distinct from the mediostapedial (m.sf), which, 

 contrary to rule, gives off the suprastapedial stem. This latter process (s.sf^) is a flat 

 triangular blade, with its broad end outwards, and is, now, confluent with the supra- 

 stapedial segment (s.st) ; the seam is still visible. The suprastapedial is a bilobate 

 thick flap, notched above, and its outer edge ribbed. The extrastapedial (e.st) is a large 

 falcate plate with a thickly ribbed edge above. The mediostapedial and this thick edge 

 is turned almost directly outward, so as to make this part seem like a mere rod. But this 

 is only the back of the upper part, for the plate grows inwards and backwards, and ends 

 below and behind in a notched angle ; to this posterior notch the epihyal is articulated ; 

 the long, convex, postero-internal part of the extrastapedial has a neatly thickened 

 selvedge ; the top of the extrastapedial is somewhat dilated. 



At present the mediostapedial shaft, only, is ossified, not the base ; the bony tract 

 stops, distally, where the suprastapedial is given off: this bone corresponds to the 

 "hyomandibular" of a Teleostean or a Ganoid fish; the unossified proximal end is 

 always a separate cartilage [stapes) in the Amphibia ; it is never differentiated in 

 Fishes so as to fit into a " fenestra," although the hyomandibular pushes far into the 

 side of the auditory capsule in the Sharks (see Gegenbaur, plates 11 and 12). 



5th Stage. Embryos o/ Alligator mississipensis, 4^ inches long, and oyCrocodilus 



palustris, 4J and 5 inches long. 



a. Endocranium. 



Again, before describing the endocranium, I may point out the progress made iu 

 the development of the tympano-Eustachian cavities. In the palatal view, with the 

 inferior arches removed (PI. LXV. fig. 5), we see that the seam along the middle of 

 the palate is just becoming perfect ; it is open, however, behind, and the posterior 

 nares {p.n) open, still, scarcely behind the middle of the head ; but in the back of the 

 recess into which they open there is a neat round hole with a swollen edge, before, 

 and at the sides. Looking well within this hole we see three lesser passages, one in 

 front and one right and left. A bristle is shown passing from one of the side holes 



