276 NATURAL SCIENCE. Oct., 



that the remainder are very nearly so, would not be sufficient. I 

 must beg the reader to lay upon the plate the straight edge of a piece 

 of paper, or a rule, and to see for himself that this is true, and that the 

 concavity of the curvature of every primary quill is turned towards 

 the middle one, i.e., the fourth. The hinder ones are curved — but 

 only very slightly — in such way that the concavity of the curve is 

 turned forwards. If the line of the rhachis of each individual primary 

 quill be traced onwards, a pretty accurate idea may be obtained as to 

 where these primary quills were inserted. 



In the Steinmann-Doderlein figure, which is spreading like a plague 

 in modern books, those hinder primary quills which the photograph 

 shows to be curved slightly forward (i.e., with the concavity forwards), 

 are represented as curving strongly backzvards — curving, that is, through 

 an angle of more than forty degrees ! The distinction between 

 primary and secondary quills is abolished, and all but the first two or 

 three are represented as attached to the ulna ! The second (or ? first) 

 quill is represented as the longest, and the fourth, which is seen in the 

 photograph to be the longest, is drawn much shorter than this, or 

 rather as falling short of it at the tip. The continuation of the quills 

 in a wrong direction as far as the ulna involves, in the case of the 

 third one, the representation of the feather of nearly double its true 

 length. 



Nor is it difficult to guess how the originators of the figure came 

 to draw it thus falsely. Inadvertent error on the part of the engraver 

 is out of the question. The drawing is not to be described simply as 

 erroneous, but as deliberately falsified. The uselessness of strongly- 

 curved feathers for flight may not have occurred to the authors, but 

 the absurdity of supposing (as many do !) that the three long, slender, 

 and especially weak-jointed fingers could bear the torsional stress to 

 which they would be exposed during flight if they supported those 

 large quills, appears to have occurred to them and to have led them 

 to avoid the absurdity of this, the everyday view, by the falsifications 

 to which I have referred. 



If the dissected wing of a common bird, such as a pigeon — the left 

 wing— be laid on the table and compared with the plate of Archaopteryx, 

 the conclusion that those two wings are essentially alike will be inevi- 

 table. It will be impossible to avoid the conclusion that the two digits 

 which support the quills of the ordinary bird existed also in Archceopteryx, 

 and their position will be seen to be indicated faintly in the photo- 

 graph by a shadow which runs parallel with and behind the slender 

 digits. The carpal angle of the wing will be seen in front of the 

 carpal ends of the slender fingers, and from this point the outline of 

 the anterior margin of the wing can be traced to the tip. This margin 

 lies under those fingers. Not only did these, as their form and 

 structure show, not support the quills, but they did not even contribute 

 to the support of them. These fingers lie not in the wing at all, but 

 upun its feather-clad surface. 



