BIEDS OF ETHIOPIA AND KENYA COLONY 375 



taneously, all still basally inclosed in sheaths, a condition possible 

 only with the advent of the first jDennaceous feathering. 



Reichenow ^- records a bird from Osi as a hybrid between caudatus 

 and lorti. Plowever, he considers the former to occur north to Shoa 

 (Sekwala and Lake Zwai) , but Neumann ^^ writes that the specimens 

 from these localities are really lorti and that the so-called hybrid is 

 merel}' an intergrade between the two races. It is said by Reichenow 

 to be like lorti with the breast bluish on the sides as in lorti but vio- 

 laceous in the center as in caudatus. Four of the birds from the 

 Gato River (two males and two females) have the breast blue with a 

 large median posterior purplish spot, so that it seems that in the 

 southern half of its range lorti frequently exhibits tendencies toward 

 caudatus coloration. In two other specimens the purplish breast 

 spot is continuous with the same color on the throat, forming a 

 fairly broad median purple stripe. 



Neumann ^^ writes that two females from Lake Zwai and Barssa 

 River, Maleland have a few lilac feathers between the blue ones on 

 the breast, and can not therefore be considered typical lorti. 



Erlanger ^* writes that with increasing age of the birds the color 

 of the back and interscapulars becomes purer reddish brown, while 

 in younger birds it has an olive-green sheen, and that likewise the 

 throat is green at first (young birds) and is bluer in older examples. 



Van Someren ^^ notes that the rump varies from deep to pale 

 blue in birds from the same locality. The present series corroborates 

 this, but it would be more accurate to say the variation is from deep 

 violet to light blue or ultramarine blue. The size variations are as 

 follows : 



Adult males (23): Wing 158-171 (average 163.3), tail 151-209 

 (172.5), middle tail feather 108-126 (117.5), and culmen 31-35.5 

 (33.5 millimeters). 



Adult females (13) : Wing 156-167 (average 161.6). tail 143-177 

 (162.6), middle tail feather 107-121 (115.7), and culmen 32-37 (34 

 millimeters). 



Erlanger °* writes that lorti and abyssinicus have complementary 

 ranges, but this has since been shown to be wrong. However, their 

 ranges are not wholly coincident by any means ; thus, Lort Lovat "® 

 writes that lorti was " * * * never met with in Abyssinia 

 where its place is taken by Coracias abyssinicus.'''' The fact that 

 the ranges of the two do overlap in parts of Ethiopia, together with 



"■■^ VOsr. Afr.. vol. 2, p. 224. 



" Journ. f. Ornith., 1905, p. 184. 



" Idem, pp. 458-459. 



■^^^Nov. ZooL, vol. 29, 1922, p. 73. 



°' Ibis, 1900, pp. 317-318. 



