412 INTRODUCTION TO EVOLUTION 



Returning to the subject of the evolution of the preorbital portion of the 

 horse skull (Fig. 18.3), we note that the increase in relative length as the 

 horses increased in size can be explained by a tendency of the face to in- 

 crease in length faster than does the cranium. Up to a certain size, small 

 horses have short faces, larger horses have disproportionately longer ones. 

 This is true today; it was true in prehistoric times. The observed evolu- 

 tionary trend through at least the larger Mesohippus (Fig. 18.3) can be 

 explained without assuming any change in the genes concerned. The larger 

 the horse the longer the face. Thus, if by evolutionary change we mean 

 change in the genes, genetic change, we can not say that the horses under- 

 went evolutionary change in facial length until the Merychippus stage was 

 reached. Before that, increase in length of face had been an automatic ac- 

 companiment of increase in body size. Genetic change was evidently intro- 

 duced with Merychippus, however, connected with development of hypso- 

 dont teeth (p. 200). 



The changes in the feet of horses resulting in the one-toed condition 

 characteristic of modern horses presents a somewhat similar situation. We 

 recall (Fig. 10.7, p. 204) that there were many lines of three-toed horses, 

 one of which eventually gave rise to the one-toed genus. It has been shown 

 (work of Robb, summarized by Simpson, 1944) that the relationship be- 

 tween the growth rate of the lateral digits and that of the cannon bone re- 

 mained constant among these three-toed horses {y — \.5x'^' '" ■'\ in which 

 y is the length of the lateral digits, x is the length of the cannon bone). 

 There was no evolutionary, i.e., genetic, change among them; changes in 

 proportionate length of digits were the automatic accompaniment of 

 whatever changes in size of foot (typified by changes in size of cannon 

 bone) occurred. But in the one-toed horse line there was an abrupt 

 change in proportion, as compared to the proportions of three-toed an- 

 cestors. This change took the form of a sudden relative decrease in the 

 length of lateral digits as compared to the length of the cannon bone. The 

 change is expressed in the allometry formula by a reduction of the constant 

 b to about half its former value: y = .76r^'' '" ^ "°. Once inaugurated, 

 this changed value has continued and is found to apply to modern adult 

 horses of various sizes, and to size changes encountered in the ontogeny 

 of modern horses. 



Viewed in the light of allometry, horse evolution assumes quite a dif- 

 ferent complexion from what it has in other lights. According to this view, 

 the important evolutionary change at first was in size. It is doubtless of ad- 

 vantage to a horse to be large. For one thing, large animals do not fall so 



