PEDIGREES AND FAMILY TIES. 221 



These are the parrots, the owls, night-jars, 

 swifts and humming-birds, and the kingfishers. 



All these, it is commonly admitted, are more 

 or less closely related, although they have never 

 before been quite so intimately associated. I 

 have arranged them in this sequence with a 

 purpose. All agree in one particular, which, 

 though an apparently trifling one, is yet pro- 

 bably important — all are gap-winged forms. 

 You will remember that, on p. 67, we pointed 

 out that a gap-winged bird was one in which 

 the fifth pair of great or, as we might say, 

 greatest coverts of the forearm differs from all 

 the other coverts of that series in that they 

 embrace no quill-feather, which all the others do. 



We said all agree. This is not quite true, be- 

 cause humming-birds, some of the kingfishers 

 and swifts curiously enough do not seem to be 

 gap-winged. Kecent researches on this subject, 

 however, have shown that in som£ birds — the 

 pigeons — this pair of covert feathers gradually 

 dwindle in size till they come to look like the 

 coverts of the row next above, and thus till 

 lately not only have escaped notice but have 

 made the wing appear to be non-gap-winged. 

 Now, when we come to study the wings of these 

 exceptional kingfishers and swifts, we shall find 

 they will tell the same story. They will only 

 appear to have no gap, or they may even in 

 some cases have quite lost it. But this will not 

 matter ; they belong all the same to the gap- 

 winged type. 



Having got this little coterie of gap-winged 

 forms cut off from the main body of this large 



