90 PASSERES. 



In the colour of the upper parts the various species of Wrens com- 

 pletely intergrade. It is impossible to draw a line anywhere between 

 the palest desert forms from Algeria or Turkestan, and the darkest 

 tropical forms from Cashmere and Sikkim. The barring is on an 

 average most conspicuous in the tropical form, but examples showing 

 the extreme amount of barring occur in France, Norway, Mongolia, 

 and other localities. 



In the colour of the underparts it seems possible to draw a line, 

 which may be a natural one. Troglodytes parvulus and its subspecific 

 allies form a pale group, which range across Europe to Algeria in the 

 south and to Russian Turkestan in the east ; whilst Troglodytes 

 fumigatus and its subspecific allies form a dark group, which range 

 from Japan across Asia to the Himalayas and the Altai Mountains. 

 On Bering Island a pale form occurs. Troglodytes fumigatus pallescens, 

 which probably came from Alaska. The underparts are generally 

 much more barred in the dark species than in the pale one, but the 

 amount of individual variation in this respect is very great. The 

 variations in the size of the bill and feet are considerable, but no 

 genetic value can be attached to them. The large bill and feet 

 characteristic of the races of St. Kilda, the Faroe Islands, Bering 

 Island, and the Kuriles have probably each been independently 

 acquired. 



The Kurile Island race of the Common Wren is remarkable for its 

 long bill, the exposed culmen measuring -55 inch (14 millimetres), 

 a length exceeding that of the Faroese Wren and equalling that of 

 the Commander Island Wren. The length of the hind toe, '4 inch 

 (10 millimetres), or with the claw "6 inch (15^ millimetres), agrees 

 with that of the Commander Island Wren and that of the St. Kilda 

 Wren, but exceeds that of the European Wren and that of the 

 Japanese Wren. In colour it agrees with the least rufous of the 

 Japanese Wrens, but is much less rufous than the ordinary type of 

 that race, and much more rufous than the Commander Island Wren. 

 It is more rufous than typical examples from Europe, but scarcelv 

 diflFers in colour from an example collected by General Prjevalski in 

 the Chuan-Che range of mountains in Mongolia. The bars on 

 the upper parts are not quite obsolete on the mantle, and on the 

 breast are well marked, but this is probably only a sign of summer 

 plumage. 



I have two examples collected by Mr. Snow in June on Uschisir, 

 one of the small central islands of the Kurile range, and Dr. Stej- 



