254 BRITISH BIRDS. 



The characters of this genus consist in the smallness of the first pri- 

 mary, which is less than the primary- coverts, and sometimes apparently 

 ahsent altogether. 



It contains about twenty species, all confined in the breeding-season to 

 the Palsearctic Region, except one species, which is circumpolar, and two 

 nearly allied species, one of which is Nearctic and the other Neotropical. 

 Six species are included in the British list, but only two of them breed in 

 our islands. Four other species are European. 



The Larks chiefly frequent sandy open plains, and are also found on 

 cultivated land, but a few species are more arboreal. They do not perch 

 much on trees, but walk and run along the ground with great ease and 

 quickness. The Larks furnish a most interesting instance of protective 

 colouring, and their plumage harmonizes very closely with surrounding 

 objects. Most of the species are fond of dusting themselves. They sepa- 

 rate into pairs in spring, but are more or less gregarious in autumn and 

 winter. Many species are migratory. They are moderately good 

 songsters, and generally sing whilst fluttering in the air, often ascending 

 to an immense height. Their flight is powerful, slightly undulatory, and 

 performed with rapid beatings of the wings with occasional cessations. 

 Their food consists of insects, worms, and small seeds in summer ; but in 

 winter they are almost exclusively granivorous. All the species breed on 

 the ground, and their nests are slightly made of dry grass, lined with roots 



and of the complicated synonymy of each species whicli are so characteristic of the author 

 of the principal volumes of the ' Catalogue of Birds in the British Museum.' This valuahle 

 and important contrihution to our knowledge of the ornithology of the Ethiopian Region 

 is prefaced by a key to the genera of the subfamily Alaudinss, which is elevated to the 

 rank of a family, and is subdivided into no fewer than nineteen genera. The so-called 

 structural characters on which these genera are based are no more ridiculous than are 

 most of the characters which modern ornithologists select in their mania for genus- 

 making ; and I only point to this article as a fair illustration of the absurdity of the system. 

 Many of the so-called generic characters are not even of specific value, and vary in individuals 

 of the same species ; others, though true of typical species, do not hold good of the aberrant 

 members of the genus ; whilst most of them are of such a trivial nature that it is impossible 

 to read them without a smile. The whole key to the genera reads like an ornithological 

 jeu d'esprit. It is scarcely conceivable that generic distinctions founded on the relative 

 length of the culmen and the middle toe, or of the crest-feathers and the tarsus, can 

 be meant to be taken seriously. Of the seventy or more species of Larks, many are so 

 nearly related that they can only be regarded as local races, whilst others have be- 

 come specifically distinct ; but the whole group is of such recent origin that there has 

 not been time for a sufficient number of species to have become extinct to make gaps 

 wide enough to be recognized as well-defined natural genera. We may perhaps sepa- 

 rate the fiat-winged Palsearctic Larks from then- Preglacial round-winged confreres in 

 the Ethiopian and Oriental Kegions ; but the two groups are coimected together by the 

 intermediate groups of the Larks which were driven to the confines of these regions 

 during the Glacial period. 



