36 ILLINOIS BIOLOGICAL MONOGRAPHS [36 



perichondrale entstanden Abschnitt ossifiziert auch ein Teil der hautig ge- 

 bliebenen Bettenwand der Orbitotemporal region. Das Alisphenoid schliesst 

 in selbstandige Foramina den N. Trochlearis und den ersten Ast des Trigemin- 

 us ein und begrenzt von oben her das Opticum. " 



The posterior aHsphenoid ossification between the optic and trigeminal 

 nerves of Amiurus has most of these characters given by Gaupp for the alis- 

 phenoid of Salmo, who says little about the nature of the 'hautig' wall and its 

 relation to the trabecula. As I interpret it, this wall is developed at a very 

 early age frorn the perichondrium of the alisphenoid cartilage, since they are 

 intimately connected in the 10 mm. stage. The relations of the ventral part 

 of this region will be taken up later in the discussion of the trabecular region. 

 In Salmo, the ophthalmicus superficialis trigemini has a different course 

 through the alisphenoid cartilage and is closer to the otic capsule than it is in 

 Amiurus. Gaupp does not discuss the relations of the ophthalmicus super- 

 ficialis facialis of Salmo. 



Both alisphenoid and orbitosphenoid ossifications occur very late in the 

 development of the ganoids in approximately the same place in regard to the 

 passage of nerves as in the teleosts. In the Amphibia the orbitosphenoid 

 (sphenethmoid; Parker, 1872) is developed on the anterior end of the 

 alisphenoid-trabecular cartilage and the posterior margin of the ethmoid carti- 

 lage, in approximately the same position that it has in Am.iurus. The 

 alisphenoid ossification does not form. 



As mentioned above, Gaupp (1902) claims that the alisphenoid of the mam- 

 malian cranium is developed from a newly added cartilaginous part, the 'ala 

 temporalis,' but I have adopted a diflferent view, and believe that the cartilage 

 is the same in both Mammals and Ichthyopsida, because of its relation to the 

 trigeminal nerve. Gaupp himself admits that nerves are good landmarks in 

 the establishment of homologies and yet denies this homology. 



The frontal bones first appear developmentally in the Acipenseridae 

 (Parker; 1882) and the lower Siluridae (Hertwig; 1876), as dermal plates, slight- 

 ly separated from the corium and having their origin in this layer. They never 

 form a pair of distinct frontalia as in Polypterus, Amia and Lepidosteus, but 

 remain as groups of plates. They do not touch the cartilage of the tegmen 

 cranii, but are separated from it by connective tissue. Walther (1882) found 

 that the frontalia in Esox develop in the same way, but were nearer the carti- 

 lage and had less relation to the corium. Williamson (1851), Heincke (1867), 

 Hertwig (1876), came to the conclusion that the frontals of the teleosts 

 and ganoids were descended from dermal scales, a view now universally accepted 

 for all osseous vertebrates. 



Vrolik (1873), in his discussion of the development of the frontals in the 

 teleosts, remarked that they were formed for the protection of the lateral line 

 canal which, in the adult of most teleosts, runs along the dorsal surface of the 

 bone. Walther (1882) in his work on Esox denies that such a condition exists, 



