230 ME. p. CHALMERS MITCHELL OK SO-CALLED 



contour row. In the feathering o£ a wing two series of rows 

 are evident, the intersections of the rows forming a diamond 

 pattern, most conspicuous on the surface of a plucked wing. 

 One series is horizontal, or at least more or less parallel to the 

 contour of the ulnar edge. Of these horizontal series, the rows 

 of quills and of major and second and thirdcoverts are most obvious, 

 partly because of the large size of the individual feathers, but 

 these horizontal rows are much more diflScult to trace outside 

 above the larger feathered rows. The other series run rather 

 diagonally to the ulnar edge and are starting upwards and with a 

 forward slope from the quills. They are very beautifully seen 

 in some of Mr. Pycraft's figures (9. plate xxiv.), but are obvious 

 enough in roost wings. These diagonal rows run round the sur- 

 face of the wing posteriorly almost at right angles to its long 

 axis, but anteriorly with an increasingly forward inclination, to 

 which I shall presently refer. They resemble the general dis- 

 position of colour-markings or scales on a cylindrical surface 

 which usually occur as hoops running round it; and it appears 

 to me that the apparentlj^ longitudinal rows are composed of 

 members of the transverse rows at different levels, being in fact 

 simply the enlarged individuals of the transverse rows which 

 come to lie on the ulnar edge. 



In actual development in the wing of the chick, it is true 

 that the two great horizontal rows which are to form the quills 

 and the major coverts appear first as longitudinal rows. It is 

 only when these become obvious, and when one or two other 

 longitudinal rows appear from before backwards, that the 

 diagonal rows begin to be marked. Later on these latter 

 acquire increasing coherence until the adult stage is reached. 

 I am not, however, prepared to attach great importance to 

 this early ontogenetic appearance of the longitudinal rows as 

 such. In the first place, considering what we know of the 

 extraordinary accelerations and retardations that occur during 

 larval development, it would appear to be pushing the recapitu- 

 lation theory to a ridiculous point, to attempt to found a theory 

 of the ancestral nature of the rows from the order of their 

 ontogenetic appearance. Secondly, there is a very obvious 

 reason for the early appearance of the quill-rows. These 

 and the coverts are much larger than the other feathers ; they 

 take longer to grow and must begin first; they are more 



