674 NATURAL SCIENCE. j,ov., 



phenomena observed in the process of compHcation of sharks' teeth. 

 There must be some other means by which cusps can be occasionally 

 added and suppressed. In some modern Lamnidas, for example, the 

 development of lateral denticles in the teeth is very variable and 

 uncertain ; and the palasontological history of the dentition of the 

 Notidanidae (7) is especially difficult to comprehend, except on the 

 theory that development sometimes leads to the multiplication of 

 similar parts when they occur in series. 



The evolution of the teeth in the Notidanidae is, indeed, so note- 

 worthy, that the principal stages have been illustrated in succession 

 in the accompanying Figs. g-12. The earliest known tooth closely 

 approaching that of Notidaniis in shape is one from the Lower Oolite 



Teeth of NoTiDANiD.t. 

 Fig. g. " Hybodits polyprioit." Fig. lo. Notidantts daviesi. Fig. ii. N. microdoit. 



Fig. 12. N. gigas. 



(Fig. g) : it is commonly referred to Hybodiis polypyion, and only diflfers 

 from the tooth of the ordinary Hyhodus in the remarkable reduction of 

 the cusps at one end. The next form of tooth, in the Upper Oolite, 

 referred to Notidaniis muensteri and N. daviesi (Fig. 10), has the reduced 

 cusps absent or mere serrations, and the cusps at the other side of 

 the principal cone somewhat enlarged ; but the root or base is still 

 depressed, as in Hybodns. In the Upper Cretaceous species (Fig. 11) 

 the root is compressed and deepened, and the cusps behind the 

 principal cone are as many as seven in number ; while in some 

 Pliocene species (Fig. 12) the corresponding teeth have not less than 

 ten cusps behind the principal cone, thus attaining the maximum 

 complication. A few of the simpler types of teeth, of course, persist 

 as usual until periods later than those at which they were dominant ; 

 but the more complex teeth are invariably later than the earliest 

 known examples of the simpler teeth. 



To the present writer the evolution of the multi-cuspidate teeth 

 of the Notidanidae thus described appears analogous to that observed 

 among mammalian teeth in those of the elephants. The principle 

 that has admitted of the gradual multipHcation of the plates (or 

 ridges) in the grinders of elephants as they have evolved from the 

 stage of the Miocene Dinothevium to that of the surviving species of 

 India, may be assumed to have also governed the multiplication of 

 the cusps in the sharks under consideration. In neither case is there 

 any scope for a process of fusion of once-detached cusps or cones ; 

 in both cases it is obvious that changes have taken place in the 

 formative tissues. 



