J. E. DUERDEN 155 



involved in the loss of the cephalic and crural feathers are germinal, 

 though they only gain expression in the late chick. It may be surmised 

 that both will gradually retrogress to an earlier stage in the ontogeny 

 of the bird, the coverings first failing to appear on hatching and 

 afterwards becoming wholly eliminated in the developing chick. A 

 hastening of the time at which structures ontogenetically disappear is 

 suggested in several other directions, particularly in connection with 

 the digits. 



The losses of feathers from the head and legs are obviously examples 

 of the discontinuous variations of Bateson and of the mutations of 

 de Vries, while all the other degenerative changes noticed are suc- 

 cessional variations with a definite trend, illustrations of orthogenesis 

 as understood by Eimer, and of the rectigradations of Osborn. The 

 changes in the one are entire and detached, while in the other they are 

 successional and cumulative, following a definite order. The former 

 are examples of negative, germinal mutations, though expressing them- 

 selves late in ontogeny, and seem altogether fortuitous in their nature, 

 whereas the latter appear to be parts of a definite, deep-seated, ortho- 

 genetic plan. The two types of variation are clear and distinct, and 

 represent the two groups into which students of evolutionary genetics 

 are often classed at the present time, namely, those who maintain that 

 evolution takes place by large variations, mutations or saltation's, and 

 those who conceive it to be effected by minute, successional variations 

 with a definite trend. Here in the same bird we have concrete examples 

 of both types of variation, showing the danger of presuming that bio- 

 logical phenomena can ever be arranged in water-tight compartments. 



Down. 



In most flying birds a covering of down feathers or plumules occurs 

 beneath the ordinary contour feathers of the body and wings, and a 

 number of small, degenerate, hair-like feathers or filo-plumes may also 

 be present. All accounts of the plumage of the ostrich however refer 

 to the absence of down feathers and filo-plumes ^ Yet on close exami- 

 nation every Nigerian and Cape bird is found to display them around 

 the base of the larger feathers on the wing and tail, the remiges and 

 rectrices, the degree of development varying much in individual birds. 



1 Chandler [Univ. Calif. Publ. Zool. Vol. xiii. 1915, p. 284) gives a " total absence of 

 differentiated plumules and filoplumes" as one of the characters of the Eatitae, as well as 

 of the ostrich. 



