Fishes of the Western North Atlantic 65 



at least) it is connected by a ligament with the hyomandibular only, at the upper end of the 

 latter, and is entirely separate from the lower jaw, hence it does not take a direct part in the 

 suspension of the lower jaw. The cranium is also much less intimately connected to the 

 vertebral column in sharks than in rays. It has been stated that while the upper jaw (palato- 

 quadrate cartilage) bears a transverse process by which it is attached to the cranium by a 

 ligament in sharks, this is not true in rays. Actually, however, the two groups intergrade in 

 this respect, for the ligamentary attachment (but not the transverse process) is present in 

 some skates at least,^ while the transverse process may be represented by an articular area 

 only, in some sharks (Heterodontidae), or altered to a rounded knob in others (Isuridae). 



Replacement of Teeth. The number of series of teeth that are in actual use at any 

 given time varies from one to four or even five in different sharks, and in different parts of 

 the jaw of a given shark. There are also one to several additional reserve series lying in a 

 reversed position (points up in the upper jaw, points down in the lower) against the inner 

 surfaces of the gums, new series being developed in a deep dental groove along the inner 

 margin of the jaw and covered over by a fold of the mucous membrane. As functional 

 teeth are lost, whether by accident or by orderly migration to the outer anterior edge 

 of the jaw, those of the next younger series move forward to replace them. This process 

 of replacing older and smaller teeth by younger and larger ones continues throughout life, 

 there being as many reserve rows in adult specimens of a given species as in the young. It is 

 this process that provides for the increase in the size of the teeth, which accompanies the 

 growth of the shark. 



Among the majority of galeoid sharks the loss of older teeth is irregular, in part acci- 

 dental, the older teeth being lost and replaced by younger ones individually. Thus, in 

 Carcharias^ two days to one week are required for a directly observable tooth to become 

 detached} sometimes one may be seen dangling from one of the outer corners of its base; 

 as a rule, too; the teeth are lost singly and not by entire series simultaneously. Evidently this 

 last feature applies equally to the various carcharhinids, to the White Shark {Carcharo- 

 don), and to the Hammerheads, for their jaws commonly show various stages in the 

 progression of teeth. There is no reason to doubt that the replacement is correspondingly 

 irregular in those squaloids in which the teeth are slender, raptorial and spaced along the 

 jaw. But in others of that group, in which they form a continuous cutting edge (Figs. 88 A, 

 97), the process of replacement involves a revolution of the younger series as a unit 

 from the reversed to the erect position, otherwise gaps would occur in the series in use, 

 which is seldom, if ever, the case.'' The teeth of the older series, which are being replaced, 

 do not loosen and fall out forthwith, but continue for some time attached to the outer side 

 of the gum to which they have moved, although standing meantime at a somewhat lower 



2. This is recorded and well illustrated by Parker (Trans, zool. Soc. Lond., ro, 1879: 223, pi. 41, fig. 4) for Raja 

 clavata. 



3. Breder, Copeia, 1924:42. 



4. We have never detected such a gap in numerous specimens of Squalus acanthias except as a result of mutilation by 

 the hook at time of capture; nor among specimens of Centroscymnus, Dalatias, Isistius or Somniosus that we have 

 examined. 



