1 



176 



Degeneration in the Ostrich 



cock No, 182, was mated with hen No. 110, as well as with hen No. 179, 

 and both laid eggs and assisted in the nesting, sitting side by side. The 

 chicks are therefore from two sisters mated with the same brother, but 

 the scutellation is slightly different in the two hens. The series is 

 presented as showing practically all the intermediate phases in the 

 production of the two breaks, mostly with marked accentuation on the 

 parental conditions. 



Fi Breeders : 



Cross cock, No. 182 

 Cross hen, No. 180 

 Cross hen, No. 179 



F2, Chicks : 



One— No. 6 

 Two— N08. 2, 4 ... 

 Two— Nos. 5, 8 ... 



One— No. 1 

 Two— Nos. 3, 7 ... 

 Two— Nos. 10, 11... 

 One— No. 9 



TABLE XVIfl. 



No break ; narrowing at first joint, none at second 

 No break ; narrowing at first joint, none at second 

 Break complete on one toe, incomplete on other 



No break ; slight narrowing at first joint, none at second 

 No break ; medium narrowing at first joint, none at second 

 No break ; very narrow at first joint, fragmentation beginning 



at second 

 Complete "first break ; no fragmentation at second 

 Complete first break ; scales fragmentary at second 

 Complete first break ; scales strongly fragmented at second 

 Complete first break ; complete second break 



As arranged above the eleven F^ chicks present an almost complete 

 series of scutellar changes, comprising all the intermediate stages from 

 a slight narrowing at the first joint to one where the first and second 

 breaks are fully completed (cf Fig. 8). The scales over the first joint 

 simply become smaller and smaller and ultimately disappear ; before 

 this occurs those over the second joint begin to break up and the frag- 

 ments become less and less until no trace of them remains. Like those 

 in Table XVII they demonstrate that the individual losses may repre- 

 sent a considerable advance on the maximum displayed either by the 

 parents or the grandparents. It is difficult to resist the conviction that 

 we are here witnessing the loss of scales in actual progress, and that by 

 means of selection we are able to accentuate retrogressive evolution. 



The two breaks constitute a most decided example of a similar 

 somatic character being produced by two wholly different factorial 

 methods. They justify the remark of Morgan that "experience has 

 •shown that it is very unsafe to judge as to the nature of the mutation 

 from the appearance of the character alone," quite as conclusively as 

 any of the instances he himself adduces. 



Comparison may be made with the retrogressive changes in the 

 plumes. In these a long process of simple reduction in size is followed 

 by a breaking-up and ultimate loss of the constituent structural parts, 

 while in the case of the scales over the second joint the first evidence 



I 



