338 University of California Publications in Zoology [Vol. 11 



if deprived of their barbules. A closer examination shows that 

 all but the most proximal rami of both vanes, at their junction 

 with the shaft, are continued to the ventral surface of it as more 

 or less distinct ridges, suggesting that the stiff, broad, type of 

 barb may be the more primitive, and that the form of the outer 

 vane beyond the point of incision may be merely a reversion to 

 an old type, rather than a new structure. As these barbs are 

 serviceable without barbules, this condition might be thought 

 of as a possible intermediate condition between a reptile's scale 

 and a normal feather, at a time when its evolution was taking 

 place and barbules had not yet reached a perfected state of 

 development. 



From the point of view of ontogenetic development, it is easy 

 to think of the deep type of ramus as more primitive than the 

 shallow one. At a time M'hen shaft and barbs are mere ridges 

 on the inner surface of the feather germ, the shaft ridge is 

 probably to be considered merely as a number of barb ridges 

 combined, as described and figured by Strong (1902, p. 160, and 

 pi. 1, fig. 2). This being the case, the more primitive condition 

 would be that in which the barbs and shaft are of equal depth, 

 as in the case of the comparatively primitive feathers of casso- 

 waries and penguins. In the latter the very broad, flattened 

 shafts, while higher in the middle than the barbs, slope off on 

 the sides, and there is no disparagement between depth of barb 

 and shaft at their junction. In this case, however, the similar 

 depth is due not to the greater depth of the barbs, but to the 

 shallowness of the shaft. In the majority of ordinary feathers, 

 where the shaft is enlarged and deepened, the barbs have not 

 kept pace with this increase, and are therefore not so deep as 

 the shaft. In the terminal portion of the primaries of Circus 

 and other .strong flyers, as shown above, there is a re-establish- 

 ment of the equivalence in depth of barb and shaft, but whether 

 this is to be considered as a reversion, or as a new specialization, 

 depends upon the point of view. 



The inner vane is very much broader and better developed 

 than the outer vane, its surface constituting over eighty-five per 

 cent of the entire surface of the feather. There are no downy 

 barbs whatever, except a few which spring from the lips of the 



