198 NATURAL SCIENCE. Sept.. 



are filled up by exceedingly small processes called radii ; of these, 

 there will be found two rows to every ramus, one directed towards 

 the tip of the feather — the distal series — and one looking towards its 

 outer or inner edge, as the case may be, and neither to the tip, nor 

 the base of the feather, as has been described. Now if we press a 

 sharp knife along the edges of one of these barbs, we shall succeed in 

 getting a number of radii of two quite distinct varieties, though ihey 

 have this much in common, both may be likened to lamella? which 

 have undergone the process familiarly known as " warping," in a 

 longitudinal direction. Those of the distal series (Fig. 4) are 

 found to be lamellate for about half their length, when the lamella 

 might be imagined to have been cut up into a number of long, 

 slender, filiform processes, some of which are hooked at the tip. If 

 hooked, they are called hamuli, otherwise they are commonly known 

 as cilia. The hamuli are confined to the ventral edge of the radius, 

 but the cilia may occur on both dorsal and ventral edges. 



The radii of the proximal series will be found, in addition to the 

 warping, to have their upper edges folded somewhat sharply over 

 towards the hollow surface, as in Fig. 3. After rather more than 

 half of its length, this edge gives place to a series of notches, which 

 soon — rapidly diminishing in size — blend with the almost and often 

 quite filiform tip, into which the distal moieties of these radii are 

 produced. It must here be noticed that, in the region of the notched 

 edge, the radii are bent at a slight angle upon themselves, the bent 

 ends being closely approximated, as shown in the figure. 



So much for the general conformation of the radii. What is the 

 reason for this warping? and what is the actual method of inter- 

 locking ? are questions that my readers will here naturally ask. 



Probably a glance at Fig. 2 will explain more than could ever 

 be done by mere description. Here we have a semi-diagrammatic 

 representation of an oblique section across two rami (r), taken parallel 

 with the distal radii, or what amounts to the same thing, parallel 

 with the shaft of the feather (p). As I have hinted, the figure tells its 

 own story, but I should like to draw attention to the following 

 details. 



It will be observed in Fig. 2 that the proximal halves of the distal radii 

 appear as flattened, overlapping laminae, which at about the middle 

 of their length suffer a sudden twist in a vertical direction, then break- 

 ing up into the hooklets and cilia previously described. These 

 booklets, it will be seen, are thrust down between the scroll-like upper 

 margins of the proximal series, between which there is only just room 

 enough to allow the hooklets to pass. The proximal radii are some- 

 what puzzling, perhaps, because seen in section ; this, however, is 

 unavoidable, as a reference to the diagram, Fig. 1, will, I think, make 

 clear. The true relation of this series having been grasped, there is 

 little else to note for the moment, except the folded upper edge, whose 

 function is obviously to afford a secure hold to the hooklets, just 



