237 
of a-bisected skull of this same fish (c. Pl. 5, Fig. 2) lies decidedly 
dorsal to that line. A large trigemino-facialis foramen occupied, 
in my specimen of Mustelus, the place of the trigeminus foramen of 
GEGENBAUR’S figure, and was separated, by tough connective tissue 
bars, into three parts; a ventro-anterior one for the nervus trigeminus, 
a dorsal one for the nervus facialis and a small ventro-posterior one 
for the ramus palatinus facialis. The nervus profundus issued by a 
separate foramen, as already described for this fish by both me (1901) 
and Trestne (1895). The nervus abducens traversed a canal which 
began immediately mesial to the large trigemino-facialis foramen, 
and running antero-laterally issued in the orbit by a foramen that 
lay immediately anterior to the ventral portion of the trigemino- 
facialis foramen, between it and the orbital opening of the pituitary 
canal. ‘There was in this fish no definite acustico-trigemino-facialis 
recess in the lateral wall of the cranial cavity and no trigemino- 
pituitary fossa at the hind end of the orbit. ‘The external carotid artery 
perforated the subocular shelf by a relatively large foramen which 
had the position of the foramen 3 of GEGENBAUR’S figure of Galeus. 
The hyomandibular, in my specimen, inclined postero-latero-ven- 
trally, instead of lying in a somewhat vertical position, as GEGENBAUR 
shows it (l. e. Pl. 11, Fig. 3), and the superior postspiracular ligament 
arose, as I have somewhat insufficiently described it in an earlier 
work (1901, p. 193), from the dorsal portion of the extreme posterior 
portion of the orbital wall, practically beneath the postorbital process. 
From there the ligament ran at first downward until it reached the 
proximal end of the hyomandibular, where it was strongly attached 
by tissue both to that cartilage and to the adjacent regions of the 
chondrocranium. ‘The ligament then bent sharply backward and 
continued along the antero-lateral edge of the hyomandibular to its 
point of attachment on that element. Between the lateral wall of 
the chondrocranium and that short portion of the ligament that lay 
dorsal to the bend, both the truncus hyoideo-mandibularis facialis 
and the internal jugular vein ran bacleward, the ligament forming the 
lateral wall of an imperfect canal for the nerve and vein. The ramus 
palatinus facialis ran downward and backward along the dorsal sur- 
face of the subocular shelf, without piercing that shelf at’ any point. 
In Acanthias blainvillii I find the pituitary fossa similar to that 
shown by GEGENBAUR in Acanthias vulgaris, but with the dorsal 
edge of the postclinoid wall extended forward so as to form a deep 
