BIRDS OF ETHIOPIA AND KENYA COLONY 217 



Swainson-^ described the West African birds as Turtur erythro- 

 fhrys at about the same time as Riippell described serrvitorquata from 

 Ethiopia, but the hitter's work was pul)lished first and his name 

 therefore stands. However, erytkrophrys has been usually considered 

 a valid race, based on more rufescent color of the undei'parts. This 

 character is extremely variable, both extremes beino; found in all 

 parts of the range of the species, and can not therefore be used as 

 a subspecific criterion. Some idea of the similarity between erythro- 

 phrys and semitorquata may be gained from the fact that Gylden- 

 stolpe -- considers South African birds the same as West African, 

 although other workers consider them the same as East African 

 examples. Gyldenstolpe, however, writes that, " * * * there are 

 no differences whatever — at least no constant ones — between birds 

 from West Africa and those from eastern Belgian Congo, Uganda, 

 Central Kenya Colony, and Tanganyika Territory," and unites them 

 with intermedia^ using Swainson's name on account of priority. 

 However, inasmuch as intermedia is not distinct from semitorquata 

 it follows that erythrophry^ is likewise not recognizable. 



The third proposed race is Granvik's form elgonensis -^ from 

 Mount Elgon, said to differ from semitorquata in having the crown, 

 belly, lower rump, and under tail coverts dark gray ; the neck, throat, 

 and breast dark wine color, tinged with rusty, and with a pale, some- 

 times scarcely discernible, shade of gray on only a few of the wing 

 coverts. Although he comments on the great variation in semJ,- 

 torquata he bases this form on a single skin and another which was 

 shot but not preserved. The important thing from our standpoint 

 however, is his note that in the Berlin Museum he found a specimen 

 from Bukoba which has tlie belly, lower rump, and under tail 

 coverts dark gray, " * * * but the neck * * * wine colored 

 * * *." In the Museum of Comparative Zoology is a male from 

 Chantwara, Bukoba, that has the crown as in Granvik's type speci- 

 men of elgonensis but is otherwise like semitorquata. Apparently 

 then, two birds from Bukoba exhibit some of the characters of 

 elgonensh^ but different ones in each case — an indication at least, 

 that elgonensis is probably based on some such individual variant. 

 More material from Mount Elgon is needed before this form can be 

 allowed recognition. 



In studying the systematics of this bird, I have assembled a series 

 of 63 specimens. The wing length of adults are as follows: Ethi- 

 opia; males: 179.5, 185, 185.5, 187, 187, 188, 190, 191.5, 192.5, 194; fe- 

 males: 175.5, 179, 179.5, 179.5, 179.5, 180, 181.5, 184, 184, 186, 186.5 

 195.5. Kenya Colony, males : 180, 181, 185.5, 186, 191 ; females : 168, 



^ Bds. West Afr., vol. 2, p. 207, pi. 22, 18.37. 

 " KuiiKl. Sv. Vet. Akad. Handlgr., 1924, p. 310. 

 =2Jourii. f. Oniith., 1923, Sonderheft, p. 'A. 



