AQUATIC MAMMALS 



which the tips are applied to some hard surface — usually the ground. 

 If such habits be entirely relinquished it is only natural that some de- 

 velopment should follow which one would never expect to encounter 

 in the normal mammal. This should be further stimulated, or com- 

 plicated, by the tendency for the disappearance of the nails, a decreased 

 rate of ossification which seems to be characteristic of the mammalian 

 flipper, trend toward the alteration in the character of the manual joints, 

 and an unknown number of other factors. Certainly the reduction of 

 the nails has been followed by an irregularity in the form of the ungual 

 phalanges not only in the Sirenia, but the Otariidae as well. It does not 

 seem at all remarkable to me that this should be followed in the former 

 group by the addition of a cartilaginous, predigital nodule. The con- 

 dition that has brought it into existence will doubtless stimulate it to 

 further differentiation, accompanying what may be termed the simplifica- 

 tion of the true phalanges: so there is doubtless a tendency for a conver- 

 gence in the characteristics of the ungual phalanx and the predigital 

 element. What can happen once can happen again, and a second pre- 

 digital nodule could be added in course of time, as the first increased in 

 size, and this would continue to occur as long as there was an activating 

 stimulus. As a matter of course there would eventually appear a center 

 of ossification in each nodule as it attained sufficient size. Presumably 

 the start of this process occurred at a sufficiently remote period so that 

 the transition in form between the true ungual phalanx and the first 

 accessory element is now almost insensibly gradual. The growth of the 

 distal digital elements is therefore very similar to the successive distal 

 caudal elements, as they become defined after birth in a rodent or other 

 mammal with unusually long tail. 



But there can be nothing to this theory unless it be upheld by em- 

 bryological evidence. This should consist of a condition in which the 

 digital bones are not all laid down at once, but after the embryological 

 differentiation of the first four phalanges (including metacarpus), there 

 should be a gradual addition, one after the other, of digital cartilaginous 

 segments, in each of which ossification is slightly less pronounced than 

 in the segment next proximad. And as this process progresses the flip- 

 per length should gradually grow distad from the tip of the original 

 limb bud. This, apparently, is exactly what happens, and this theory 

 to account for hyperphalangy is the only one yet advanced to which 

 the known embryological evidence supplies any confirmation. 



Presumably the difference between this sort of addition of cartilaginous 

 nodules, one by one, to the digits, and the predigital cartilage of otariids, 

 [266] 



