224 PROCEEDINGS OF TEE NATIONAL MUSEUM. vol.55. 



Mr. Ridgway in volume 3 and other early volumes of his Birds of 

 North and Middle America/ arc j^artly wrong, since some of the 

 specimens were measured for total culmen and others for exposed 

 culmen, but all averaged together. In other cases the measurement 

 furnished is called exposed culmen, when it should be total culmen, 

 or the reverse. The measurements hereinafter given for Nannus 

 troglodytes hiemalis and Nannus troglodytes pacificus are taken from 

 Ridgway's work, except for the culmen, which has been remeasured. 

 The names of colors are based on Mr. Ridgway's recently published 

 Color Standards and Color Nomenclature. 



The genus Nannus Billberg^ is by some authors merged with 

 Troglodytes Vieillot, but it is undoubtedly sufficiently well marked 

 to be treated as a separate group. It differs from Troglodytes in its 

 short tail, which is less than three-fourths of the length of the wing; 

 its slender and depressed bill, the culmen being almost straight; 

 and in its narrow and silt-like nares, almost covered by overhanging 

 membranes which are very slightly or not at all thickened. 



The forms of this well circumscribed genus show a remarkable 

 tendency to vary geographically, particularly on islands, and this 

 has given rise to a large number of local races. There is, further- 

 more, a considerable amount of individual variation, although this 

 is much more evident in some than in others. The seasonal difference 

 is also in many cases great, as birds in the summer become much 

 paler and less rufescent. 



Until recently a number of the forms of Nannus were considered 

 distinct species. The writer in 1902 reduced some of these to sub- 

 species,^ but still retained several specific groups. Several years 

 later Doctor Hartert* treated all as races of a single species. This 

 action the present writer, after a careful study of nearly all the forms 

 of the genus, is now prepared to indorse. The North American birds 

 alone have hitherto been referred to three distinct species, but it is 

 evident from the following comparisons that they must all be con- 

 sidered subspecies of the Old World Nannus troglodytes (Linnaeus). 

 Wide individual variation overlaps the difference between Nannus 

 hiemalis of eastern North America and Nannus troglodytes of 

 Europe. Inter gradation of Nannus hiemalis with Nannus hiemalis 

 pacificus is already understood; the latter intergrades individually 

 with Nannus fumlgatus of Japan, and this in the same manner with 

 Nannus nipalensis of northern India. Baird's Nannus alascensis 



iBull. U. S. Nat. Mus., No. 50. 



" Synopsis Faunae Scand., vol. 1, pt. 2, 1828, Table A, anrl p. 57 (type, by monotypy, 

 Motacilla troglodytes Linnaeus). 

 »Auk, vol. 19, 1902, pp. 177-180. 

 * Vogel paiaarkt. Fauna, vol. 1, Heft VI, June, 1910, pp. 77G-784. 



