In articulation these bones fit together as follows: the forward pro longation of the vomer fits into the posterior and dorsally placed cavity between the maxillaries; the backwardly projecting processes of the premaxillaries fit into the spaces between the shoulders of the vomer and the forwardly projecting processes of the palatines; while the united nasals lie dorsal to the frontal process of the vomer and between the forwardly projecting horns of the palatines; and lastly the flattened and rounded ventral surfaces of these same horns fit into the grooves excavated into the anterior dorsal surfaces of the maxillaries. Thus there is formed a beautiful joint allowing little motion laterally but considerable in the vertical plane. However, the union of the right and left halves of both maxillaries and premaxillaries by ligaments and cartilage allows a widening of the upper jaw at the angle of the mouth, as will be shown below.
The rami of the lower jaw are likewise united in front by cartilage and connective tissue, and behind they are hinged onto the quadrates. These in turn are immovably attached by suture joints to the pala tines. These latter bones are hinged like a door or the lid to a chest, posteriorly to the prefrontals and anteriorly to lateral projections of the vomer. In widening the mouth, the upper and lower jaws spread apart behind and the quadrates and palatines swing outward and upward. This takes place at the same time that the lower jaw drops downward while the tip of the upper jaw is raised. All this results in giving the barracuda an enormous gape, a necessity for a fish which preys on other large fish and which swallows its food in large fragments. These points are well shown in figures 5, plate n; 7, plate in; 11, 12, and 13, plate iv.
The lower jaw is filled with knife like canine teeth set palisade fashion just inside the lip-like membrane.
These are small in front but larger in the middle and posterior parts. • . •• backward, point sugnuy nacxwara, especially those in rear. Their numbers for my five specimens appear in the accompanying table, the dried heads being taken in the ascending order of size. The + teeth are the small ones at the front of each ramus of the jaw; those marked +e are in the act of erupting. Two factors seem to account for this variability. First the size of the fish; the larger the fish the greater the number of teeth. The fresh head was unfortunately not measured, but a study of the teeth and comparative measurements of the bones show that it was about the size of the dried head No. m, and possibly a little larger. The second factor relates to the number of teeth that have been broken off with out sufficient time having elapsed for their being replaced. This is
particularly true of dried head No. iv, where on the right side several teeth have been broken off. Figure 13, plate Iv, from a photograph of the lower jaw of the fresh specimen, will make these points clear.
In front, at the symphysis of the rami of the lower jaw, is a huge fang, bowed on the front edge and slightly hooked backward. This was single in each of my five specimens, on the right side in four speci mens and on the left in one. Two of these teeth might be expected but one seems to be always broken off. This matter will be discussed later, but it should be noted here that this tooth is always inclined toward the vertical median plane of the head and jaws—that on the left side towards the right and that in the right vice versa.
The upper jaw has in front at the apex, under the snout, a group of huge knife-shaped fangs, two, three, or four on each side, not two pairs as Jordan and Evermann state.
The two anterior of these generally point about straight downward, while the pos terior ones are sharply hooked backwards.
All are convexly bowed on the front edge.
These teeth may be seen in nearly every figure of the fish and of its head given in this paper. The accompanying table shows their numbers in my five specimens.
Implanted on the palatines in palisade fashion are a number of large, pointed, lancet-shaped teeth, in structure and arrangement very simi lar to the mandibular teeth but much larger. These are largest in front and decrease in size backward, becoming very small at the rear. The most anterior ones are but little if any smaller than the great fangs and on each side the series runs backward in line with these fangs.
These teeth have, however, no phys ical connection with each other, there being a wide hiatus between them, as may be seen in figure 12, plate rv. The accompanying table gives the numbers of these teeth for each specimen. The + teeth are the small ones at the rear, while +e indicates teeth just erupting. Here again, it will be noted that the larger the specimen the larger and more numerous the palatine teeth.
Closely set in palisade fashion, beginning about opposite the space between the first and second great fangs, are the small knife-like teeth of the premaxillaries. These teeth, pointing slightly backward and extending clear to the angle of the mouth, give the upper jaw a very saw-like appearance, as may be seen in figure 8, plate III. The data tabulated herewith would seem to show that the number of these teeth is fairly constant, the discrepancies being chiefly due to teeth broken off and not yet replaced.