Feeding systems, jaw protrusion and ram ventilationThe feeding system derived for all Acipenseri-
formes is based on a jaw projection system very dif-
ferent from that of late Paleozoic and early Meso-
zoic paleonisciforms. As exemplified by outgroup
taxa such as †Cheirolepis,†Mimia,Polypterusand
Lepisosteus, the premaxilla and maxilla were prim-
itively sutured into the bones of the cheek, so that
forward projection of the jaws was impossible
(Schaeffer & Roscn 1961). Although the maxilla is
not sutured to the cheek in bowfins, jaw protrusion
inAmia calvais still very limited (Lauder 1980).
Head lift is also an essential aspect of jaw opening
and feeding in outgroup actinopterygians (Lauder
1980) and sarcopterygians (Bemis & Lauder 1986,
Bemis 1987a). The importance of head lift and limi-
tations to jaw protrusion are most parsimoniously
interpreted as plesiomorphic conditions in Acti-
nopterygii.
Living sturgeons and primitive paddlefishes (i.e.,
polyodontids other thanPolyodon) stand in sharp
contrast to all of these outgroup taxa. Their jaws are
highly mobile, so that the upper jaw can be ‘project-
ed’ far out to capture prey, and head lift plays no
role in feeding at all. This is illustrated in the case of
a white sturgeon,Acipenser transmontanus, in Fig-
ure 22. Within acipenserids, progressive modifica-
tions of the jaw projection system allow even
greater specialization on benthic prey (Findeis
1997).Psephurus gladius can also project its jaws, as
shown in Figure 7 (also see Grande & Bemis 1991,
figures 28C and 33B). This ability allows sturgeons
andPsephurusdietary variety, including both fast
moving (fish) and benthic (mollusk and other inver-
tebrate) prey. Interestingly, within Polyodontidae,
Polyodonexhibits secondary reduction in mobility
of its uppcr jaw, a specialization associated with fil-
ter feeding in this genus. In the top of Figure 23, a
young juvenilePolyodonis shown; at this size, it is
incapable of filter feeding both because of hydrody-
namic considerations and becausc its gill rakers are
not yet long enough to function as a filtering mecha-
nism (Rosen & Hales 1981). Note that the upperjaw
is firmly attached to the head. In the bottom four
photographs of Figure 23, the maximum extent of
mouth opening during filter feeding is shown.
Again, the upper jaw remains firmly attached to the
neurocranium. A photograph of filled gill rakers is
shown in Figure 9b. Thus, sturgeons andPolyodon
specialize the feeding apparatus in different direc-
tions, one primarily towards benthic prey and the
other toward pelagic prey (see Yakovlev 1977, for
much the same idea).
Another functional anatomical aspect is the pro-
tractor hyomandibularis muscle, which is not pre-
served in fossils, but which is probably a synapo-
morphy of all Acipenseriformes. InAcipenser,this
muscle develops as a slip of mandibular arch muscu-
lature which migrates to its new attachment sites on
the neurocranium and hyomandibula (Sewertzoff
1928). The protractor hyomandibularis muscle is
large in all acipenseriforms (e.g., Danforth 1913,
Luther 1913). It plays an important role in protrud-
ing the jaw, and, via a simple mechanical linkage,
also facilitates the opening of the jaws. Amongst all
vertebrates, it is very unusual for a mandibular arch
muscle to function in jaw opening (Bemis 1987b).
Many unresolved issues concern the morpholog-
ical transformation from a typical paleonisciform
feeding system to the acipenseriform system. The
seeming absence of clearly intermediate morpho-
types and questions about the functional adequacy
of such an intermediate form has provoked several
speculations, perhaps most comprehensively by Ya-
kovlev (1977), Yakovlev (1977) interpreted that the
transformation in jaw morphology and function
must have occurred very rapidly, and that it prob-
ably took place in the Triassic in northern Asia. Ya-
kovlev (1977) further suggested that transformation
of the feeding system was initially driven by paedo-
morphosis, and that subsequent stabilization of the
new morphology occurred because it offered a
feeding specialization unique among early Meso-
zoic fishes--benthophagy.
Our approach to studying such transformations is
different: instead of focusing on the absence of mor-
photypes, or any other type of negative data, we
stress the importance of producing a phylogenetic
hypothesis or cladogram on which to map and con-
struct putative functional systems. By turning our
focus away from ancestor speculation to the com-
parative analysis of outgroups, we think headway
will be made in understanding many aspects of cra-