J. E. DUERDBN 187 



Factorial Changes. 



Another striking feature in the retrogressive changes going on in the 

 ostrich is the multiplicity of the methods followed, from which it may 

 be inferred that a corresponding diversity of germinal changes is in- 

 volved. Without the necessary experimental support it may be held 

 that no good purpose will be served by discussion of the latter, 

 especially in view of the warning from Prof. T. H. Morgan as to the 

 danger of interpreting the nature of a germinal change from its ex- 

 pression in the soma. The absence of any likelihood however of carrying 

 out detailed Mendelian experiments with the slow-breeding ostrich 

 emboldens one to attempt to bring the many observations into some sort 

 of line with the principles embodied in the factorial hypothesis, repre- 

 senting as it does the furthest attempt yet made to understand the 

 germinal nature of organisms and the changes which they undergo. 



That degeneration has long been in progress in the history of the 

 ostrich, and is in operation at the present day, appears to admit of no 

 question. Evidence has been adduced that many structural parts have 

 already disappeared, while the imperfect and vestigial conditions of 

 others are deemed to be the various mutational stages leading to still 

 further losses, and in the case of the claw and scales experiments have 

 shown that selection can induce and accelerate the losses. These are 

 but the outward manifestation of internal changes going on in the germ 

 plasm. Hitherto studies in mutation have been largely confined to 

 fortuitous changes, and the genetic behaviour of large and conspicuous 

 characters ; but with more intensive studies and continued application 

 workers like Prof Morgan and his Columbia associates have reached 

 mutations so small as to be scarcely appreciable, and have been able to 

 extend their observations to changes in the actual chromosome itself. 

 In addition, most of the variations yet studied have been discontinuous 

 in their nature and apparently unconnected with others, a notable excep- 

 tion being those of Prof. W. E. Castle on hooded rats ; but in the ostrich 

 we encounter variations of the first kind along with others which are 

 successional and seem part of a determinate plan. The former may well 

 be due to irregularities in germinal mitosis, but the latter certainly call 

 for consistent factorial changes, generation after generation. So deter- 

 minate are the changes that we can with every assurance predict what 

 some of the later ones will be. Thus there can be no reasonable doubt 

 that the fourth toe will disappear in time and that the upper-coverts 



