POWER OF FLIGHT. 417 



which supply unvitiated air to the pulmonary passages during the 

 expiratory act (see below). The regulating mechanism cannot be 

 properly treated here, indeed it is not fully understood, but the 

 protection against loss of heat by radiation from the external 

 surface by the feathers is one obvious factor and the loss by eva- 

 poration from the internal surface of the air-sacs must be another, 

 birds being without the cutaneous sweat-glands which are so 

 characteristic of mammals. Though during prolonged frost and 

 snow numbers of birds perish, it is rather from the want of 

 food than the inability to bear cold, and the habit of migra- 

 tion, which among birds is carried to such an extreme, is un- 

 doubtedly prompted rather by the desire to obtain food than 

 to avoid cold. This is obvious from the fact that in the most 

 northern regions the migratory movement southward begins 

 before the full summer warmth is there felt. This movement dis- 

 penses with the necessity of passing into a torpid condition 

 which is so common among many of the mammals that winter 

 in northern countries, and some of the land-birds which remain 

 to brave a temperature that might otherwise endanger life 

 are endowed with additional feather-clothing (Lagopus, Linota, 

 some owls, etc.). 



The most essential peculiarity of birds is their power of flight. 

 Their whole organisation, both internal and external, is modified 

 in correlation with this peculiarity. In this connection we may 

 call attention to the great uniformity of structure presented by 

 the class, and the sharp definition of its characters. Between 

 the extremes of avine organisation there is less difference than 

 in a single order of mammals, and there are no forms transitional 

 between birds and other classes of vertebrata. It is true that 

 they are not the only vertebrates which have achieved the aerial 

 habit. The pterodactyls amongst reptiles and the bats amongst 

 mammals have also developed the power of flight. But in these 

 animals the power depends upon quite other modifications than 

 in birds, and although it is incontestable that reptiles are the 

 nearest allies of birds, there is no single family of reptiles from 

 which they can be derived, least of all from the pterodactyls. 

 Moreover these groups are comparatively small and unimpor- 

 tant, whereas birds are a dominant group at the present time 

 and exceed all other vertebrate classes in number of species 

 though not in variety of organisation. The origin of birds is a 

 z.-u. E E 



