W. B. Scott — Variations and Mutations. 369 



also is observable in the phyletic series. Upon such corre- 

 spondences, however, little stress can be laid. (3) Digital 

 variation may be similar and simultaneous on the two sides, 

 but usually differs in degree. In the phyletic succession, on 

 the contrary, there is no such difference of degree. Indi- 

 viduals may and do vary in this way ; the normal does not, but 

 always keeps perfect symmetry and balance. In the changes 

 of manus and pes phylogenetic change agrees with variation 

 in showing a certain independence in the two extremities. As 

 a rule, the pes changes first, while the manus is more conserva- 

 tive and lags behind, though sometimes this relation is reversed. 

 While there can be little doubt that in the earliest mammals, 

 or in their reptilian ancestors, the hand and foot were com- 

 posed of the same number of parts arranged in the same way, 

 yet in all the lines of descent there has been a constant diver- 

 gence from this condition, though the correspondence, even in 

 the most highly differentiated forms, is still remarkable. It 

 may not be " necessary to suppose an independent evolution 

 for the legs of the horse, since in the light of the facts of 

 variation it is as easy for all to take on the new characters as 

 for one," (Bateson, p. 27), and so far as the two hind-legs or 

 the two fore-legs are concerned, this agrees with the facts of 

 the evolutionary history, but, on the other hand, the fore- and 

 hind-legs, not only of the horse, but of nearly all other mam- 

 mals, pursue very independent ways of their own. Thus, the 

 Eocene Hyracotherium had four digits and a rudiment in the 

 manus while the pes had already been reduced to three. This 

 tridactyl condition of the pes persisted till the end of the 

 Miocene by which time the manus had attained the same 

 degree of reduction, and in the next change, which consisted 

 in the diminution of digits II and IT to splint bones, the two 

 extremities keep equal pace. It is hardly necessary to observe 

 that such independence of the manus and pes is limited and 

 conditioned by the unity of the organism and its needs as a 

 whole. 



(4) and (5) When a digit is added, the adjacent ones may be 

 modified according to their new place in the series. There is 

 no sharp distinction between the duplicity of a given digit 

 and other modes of addition to the series. These results 

 apply only to variations ; in the phylogenetic history of mam- 

 mals new digits never are added, by doubling or otherwise. 



(6) Digital variation is often markedly discontinuous. In 

 the history of the terrestrial mammalian lines the facts lead to 

 exactly the opposite conclusion. One of the most remarkable 

 and striking features in the reduction of digits is the perfect 

 continuity of it ; it is often very slow, but gradual and steady, 

 and its completion is often delayed for incredibly long periods 



