MAMMALS 85 



heredity afterwards. Also by the operation of heredity the 

 loss of the skin might become more and more early and 

 regular until it took place as soon as fighting commenced, 

 and then reached the present condition in which it takes 

 place before the fighting, and the skin is voluntarily rubbed 

 off by the stag. One advantage of this hypothesis is that it 

 enables us to attribute the progressive size and branching of 

 the antlers to an original progressive development of per- 

 manent antlers in the same individual in consequence of 

 continued irritation. 



On the whole, however, I am inclined to think that this 

 hypothesis does not agree with the facts. In the earliest 

 fossil forms already mentioned, Dremotherium or Palceo- 

 meryx, there is at least an incipient burr, and we have no 

 evidence of the existence of any antler in ancestral deer 

 more complicated than a simple fork, which was not 

 periodically shed. 



The truth, therefore, seems to lie somewhere between the 

 two hypotheses I have outlined. The earliest stag probably 

 had a simple undivided antler, the point of which at some 

 stage of life was partly denuded of skin. A laceration of 

 skin from the point downwards caused the formation of a 

 nodule of bone at the base of the laceration, and this was 

 the commencement of the burr. Such destruction of skin 

 and the underlying periosteum, whether caused by fighting 

 or by voluntary effort, ultimately caused the shedding of the 

 dead bone ; but this may not have happened at first till late 

 in life. When it had happened, the stag fought with the 

 pedicle and irritated the point of it to greater hypertrophy 

 than before. Thus arose the forked antler, at first a simple 

 fork, afterwards a more complicated branching. In this 

 way, the progress in the size and complexity of the antler 

 went on equally with the increase in the regularity of the 

 shedding. There is, however, a difficulty in the fact that 



