CHAEACTEES OF HIRUNl^INID^ 365 



toes. The digits possess the normal number of phalanges ; 

 the basal phalanx of the middle digit is commonly coherent 

 with one or both lateral toes ; the hallux is ordinary, and not 

 reversible. The digits are commonly naked and scutellate, 

 rarely feathered to the claws. The claws are comparatively 

 strong, compressed, well-curved, and acute, apt for clinging. 

 The plumage is soft, smooth, and l)lended, most frequently 

 glossy or even iridescent, but sometimes lustreless. Head 

 short, broad, and depressed ; neck short. Mouth capacious, 

 its greatest width equalling that of the head. The tongue is 

 short and not extensible. The pharynx and oesophagus arc 

 large, the latter having no crop. The syrinx or lower larynx 

 is perfectly Oscine, and said to possess four pairs of muscles. 

 The voice is sharp and rather thin, though melodious and sus- 

 ceptible of rapid and various modulation. The stomach is 

 elliptical or roundish, moderately muscular, and lined with 

 thick rugose epithelium ; the coeca are very small. 



Such characters distinguish the Eirundinidce as a perfectly 

 natural family of Oscines, which may, indeed, be recognized 

 on sight by the combination of fissured bill, lengthened wings, 

 and weakened feet. The group scarcely inosculates with any 

 other, so perfectly is the Swallow type circumscribed. Its re- 

 lationships appear to be, on the one hand, with the Old World 

 Muscicapidce, and on the other, through Progne, with the Ajii- 

 pelidcB. Within the family, the extremes of modification are 

 seen in the genus Hirundo, of which the Barn Swallow is 

 typical, and in Progne, which includes the Purple Martin and 

 its allies. The minor characters have been made by some 

 ornithologists the basis for separating the species into twenty 

 or thirty different genera, while other writers retain them all 

 under the single genus Hirundo. Between these extremes 

 there seems to be a more judicious middle course, following 

 which the hundred or more Swallows which have been de- 

 scribed may be thrown into a few generic or subgeneric groups, 

 founded on certain ulterior modifications of structure. 



As pertinent to this portion of the subject, I here introduce 

 Sundevall's method of arranging the Swallows, which will 

 give a good idea of the leading modifications of structure 

 throughout the family. The schedule is abridged from the 

 famous Methodi Natiiralis Avium Disponendarum Tentamen 

 (8vo, Holmise, 1872, pp. 51-53). 



