594 EDWARD PHELPS ALLIS, JR. 
the pharyngohyal, either articulate with or are attached by liga- 
ment to the neurocranium. In this specimen both these ele- 
ments lie against the internal surface of the suspensorium, but 
separated from it by a delicate and highly pigmented membrane. 
The lateral dorsal aorta lies mesial to their dorso-mesial ends, 
the vena jugularis dorsal to them, and the nervus hyomandibu- 
laris facialis dorsal and anterior to them, as in selachians. 
In all living elasmobranchs it is thus seen that the dorsal ends 
of both the pharyngeal and epal elements of the hyal arch always 
lie definitely dorsal (lateral) to the lateral dorsal aorta, ventral 
(internal) to the vena jugularis, and internal (ventro-postero- 
mesial) to the nervus hyomandibularis facialis, and I assume, as 
in part stated in the opening paragraph of this paper, that this 
relation of the inner cartilaginous bar of this arch to the artery, 
vein and nerve is invariable in all of the gnathostome fishes. 
In the Dipneusti (Ceratodus) and Teleostomi there is, however, a 
marked departure from the elasmobranchian type in that the 
definitive cartilaginous bar of this arch always either fuses or 
articulates with the neurocranium dorsal to the vena jugularis. 
It will be best to consider first the conditions in Ceratodus, be- 
cause the development of this fish has recently been very care- 
fully and fully described by several authors. 
In the adult Ceratodus, Huxley (’76) described a caxtilaee 
which he considered to be ‘‘the homologue of the hyomandibular 
element of the hyoidean arch of other fishes;’”’ this expression, in 
no way qualified by him, quite conclusively showing that he con- 
sidered the hyomandibulae of these other fishes to all be homo- 
logous structures. He thought this cartilage in Ceratodus was 
probably the suspensorial tubercule of Giinther’s (’71) earlier de- 
scriptions, and it is said by him to lic internal to the operculum and 
to be firmly attached, by its anterior edge, to the skullat the point 
where the cranium proper passes into the suspensorium. The 
nervus hyomandibularis facialis is said to issue across its anterior 
edge. Van Wijhe (’82). confirms this relation of the nerve to 
the so-called hyomandibula, but adds that it seems to him not 
impossible that this latter cartilage may be an interhyal. Ride- 
wood (94) finds both hyomandibula and suspensorial tubercule as 
