442 NATURAL SCIENCE. Dec. 



sideration as probably having no real significance, came back upon 

 him with new force, this time bringing conviction with it. 



The remiges or quill feathers are briefly referred to. Of the 

 primaries (metacarpo-digitals), Professor Dames tells us that they 

 were attached, not only to the metacarpus, but extended down to the 

 claw of the longest finger — digit II. The shafts of these primaries, 

 he considers, having regard to their function as " Schwungfeder "' 

 (quill feathers), to be somewhat thin and slender. Here, again, it 

 will be seen that my restoration does not quite agree with Dames' 

 description. Turning to PI. II., Fig. I., it will be noticed that the 

 quills are supposed not to have extended beyond digit III., and 

 I think there is every reason to believe that this was actually the 

 case. The shafts of these quills are certainly slender, but not 

 relatively more so than is the case with very many birds of active 

 flight to-day. In the large plate accompanying Dames' paper, 

 all the remiges are represented as having a proximal overlap, an 

 arrangement exactly the reverse of that which I have supposed to 

 obtain when discussing this matter in Part I. 



Our fourth restoration is that given by the late Professor 

 Romanes 3 (17), Fig. 5. It is difficult to know what to say of this, 

 for it agrees with neither the London nor the Berlin fossil, and looks 

 rather like the outcome of a process of mental evolution than the 

 result of a careful study of the actual specimens. In the figure it 

 will be seen that though the relative length of the digits roughly 

 corresponds to that of the Berlin fossil, the hand, as a whole, is too 

 short. All three fingers are represented as perfectly distinct, freely 

 moveable independently of one another, and apparently were not 

 supposed to bear any part in the support of the remiges. It is just 

 possible, however, that the third digit may have been credited with 

 supporting, say, two or three primaries. Not even with a great 

 stretch of imagination can the outline of the wing be likened to that 

 of the Berlin fossil. I have taken it for granted that this restora- 

 tion is intended to represent the Berlin example, and is not a com- 

 bination of this and the London fossil, since in Owen's restoration, 

 there were four digits, of which I. and II. were separate and apparently 

 freely moveable, while III. and IV. (resembling digits II. and III. 

 of modern birds) were supposed to have supported the primaries. 

 Had Owen's restoration been present before the mind of the restorer, 

 it is possible that these digits III. and IV., with their primaries, 

 would have been made to supplement what were evidently meant 

 for the three digits of the Berlin specimen. The animal is thus 

 described (p. 172) : " The extinct Archaeopteryx . . . presents 

 throughout its whole organisation a most interesting assemblage of 

 ' generalised characters.' . . . For example, ... its still 



' This, it will be remarked, is stated to be "after Flower." Sir W. Flower 

 tells me, however, that he cannot recollect either having published such a figure or 

 having used it in lectures. 



