A 
A.J. CHARIG AND A.C. MILNER 
Fig. 47 Lepidotes scales, scanning electron micrographs. A, from rib-cage region (Block 44) of the Baryonyx skeleton, showing acid-etching of the 
enamel, x 8; B, from the same horizon and locality but not associated with the Baryonyx, showing smooth unetched enamel, x 11. 
Conversely, Baryonyx possesses characters that suggest different 
methods of acquiring its food: 
1. The jaws are long and narrow, not unlike a pair of forceps. They 
could have been dipped, either into water to seize relatively small 
fish, or into the body cavity of a large dead animal to seize the 
entrails. 
2. The external nostrils were lateral and far posterior in position. 
This would have enabled the animal to continue breathing com- 
fortably even while its snout was still deep in the water or in the 
body cavity of a carcase. 
3. The large and presumably heavy head would have limited the 
mobility of the neck. 
4. The ends of the cervical centra are not “offset”; it is therefore 
unlikely that the neck would have been held in a sigmoid curve. 
5. The cervical vertebrae have well developed epipophyses, but 
their neural spines are low and lack spine tables. This suggests 
that the intervertebral muscles were strong and the neck highly 
mobile, which again would have helped its feeding activities. 
6. The humerus is very robust, with both ends broadly expanded. 
The proximal end in particular is expanded anteriorly into an 
unusually long and high deltopectoral crest and posteriorly into a 
well-developed internal tuberosity (compare with other theropods, 
including Jorvosaurus). The pectoral girdle too is well developed 
and there was also an ossified sternum. All this would have 
facilitated the powerful adduction, abduction (and perhaps rota- 
tion?) of the fore-limb as a whole. 
7. The radius is relatively short and that too is robust and powerful. 
The same applies to the ulna without the olecranon, but the 
olecranon itself is remarkably long; thus the ratio ‘length of 
olecranon/length of ulnar shaft’ is exceptionally high. This pro- 
duced a mechanical advantage (1.e., leverage) when the fore-arm 
was extended. 
8. One manual ungual is enormous; it is of typical theropod form, 
though somewhat more slender than in Al/osaurus, and it is not 
modified into a “sickle” like the enlarged pedal ungual of 
Deinonychus (Ostrom 1969). This was clearly an extremely 
powerful offensive weapon. 
The characters of the fore-limb and manus suggest that the fore- | ~ 
limbs of Baryonyx were exceptionally powerful, the fore-arm being | 
capable of exerting great force at the wrist when extended. By | ~ 
activating the enormous claw on the thumb, this would have ena- | 
bled the animal not only to catch and kill its prey (if necessary), but |” 
also to rip and tear it to pieces. In short, the active predation of 
larger animals and the breaking up of its food were probably \ 
performed by the fore-limbs and claws rather than by the jaws and | 
teeth. The enlarged claws could also have been used for ‘gaffing’, | 
i.e. hooking or flipping fishes out of the water as is done today by | ~ 
grizzly bears. 
The fine tooth denticles of Baryonyx do not seem to be suitable for 
the ‘rip and grip’ cutting action demonstrated by Abler (1992) for 
more coarsely serrate theropod teeth. Farlow, Brinkman, Abler & |) 
Currie (1991) suggested that theropod lateral teeth with very fine 
denticles might function in a manner indistinguishable from that of | 
smooth-bladed teeth; this might be true of Baryonyx. Farlow et al. | 
also showed that the lateral blade-like teeth of most theropods | 
displayed a consistent relationship between tooth size and denticle , 
size. They reasoned that a serrated blade might inflict as much | 
damage as a smooth-edged thinner blade but would be less likely to 
break. Since the blade-like teeth of Baryonyx are less compressed 
than those of most other theropods, the reduction in the size of their 
denticles might be correlated with their greater robustness. 
If we consider the function of the dentition as a whole, all the teeth | 
seem suitable for piercing prey and cutting it smoothly. The long, 4 
comparatively straight teeth in the terminal rosette and expanded tip | 
of the dentary also possessed a stabbing function. 
Those teeth on the lower jaw that lie below the preserved portion |_, 
of the maxilla were (on the evidence of the alveoli) smaller, more} \ 
crowded together, and more than twice as numerous per unit length | 
of jaw as the maxillary teeth that opposed them. A somewhat similar 
discrepancy is known not only in Troodon, which has 15—20 maxil- | . 
lary teeth opposed by 35 in the dentary (Currie et al. 1990), but has | 
also been observed in the strange new ornithomimid Pelecanimimus 
(Pérez-Moreno et al. 1994). This discordance between the number f 
of teeth in the upper jaw and the corresponding number in the lower 
is very unusual and, as far as we know, has no analogue among living | 
