90 



onward motion, with wings raised high above the back hke a shuttlecock, 

 drop down into the darkness. This habit of theirs of nesting in chim" 

 neys, it may as well be admitted, seems at first to detract much from 

 their claim to an etherial nature as dwellers in the aii, but it should be 

 borne in mind that the swift never makes his habitation amongst soot 

 and smoke, for he is always careful to choose a chimney that is not in 

 present use. Moreover, in the days when the human lord of this con- 

 tinent was living in a wigwam filled with soot and smoke, the home 

 of the swift was the shaft of a tall and hollow tree. 



The Cypselidce or Swifts are a family of swallow-like birds of medium 

 size and generally of dull plumage. In the classification of the older 

 ornithologists, on account of many superficial points of resemblance' 

 they were closely associated with the true swallows ; and as popular 

 language even in the present day applies the name "swallow" indis- 

 criminately to all those birds of graceful flight which live on insects 

 caught upon the wing, it may be well to consider for a moment the 

 reasons that have led to the modern classification ; for now while the 

 swallows are closely linked with the finches, tanagers and other singing 

 birds of the Passerine order, so unlike them externally, the swifts on 

 the other hand are placed in a distinct order and as intimately coupled 

 with a family of entirely different appearance, the humming birds. For 

 a vindication of what seems at first an unnatural classification it 

 would be hard to find anything more satisfactory or conclusive than the 

 words of Prof. Garrod as quoted in Cassell's Natural History. At the 

 same time they will give us a glimpse of the internal structure of the 

 swifts which may serve to explain some of their curious habits. I give 

 them in abridged and somewhat modified form : 



' Most of us know that unlike the hair upon a quadruped the 

 feathers of a bird are not distributed evenly over the body, but grow 

 in linear clusters, called tracts, with narrow naked spaces between. 

 A similarity of the arrangement of these feather tracts in different 

 species has been found to be closely associated with that general simi- 

 larity of the important organs of the body which leads to the grouping 

 of species together under one order, while the different orders frequently 

 show different patterns in this respect. Now the arrangement of the 

 feather tracts on the swift is found to be almost identical with that of 





4 

 J 



