232 BULLETIN 15 3, UNITED STATES NATIONAL MUSEUM 



Palestine, and sokotrae from Socotra. Hartert *^ considered Er- 

 langer's form the same as the typical race and at the same time- 

 separated the birds of the region south of the Atlas Mountains in 

 Algeria. Tunisia, and Morocco under the name phoenicophUa. 



In 1920 Sclater and Praed *" once more delved into the systematics 

 of this pigeon and not only recognized aequatoriaUs but also de- 

 scribed another race from the Anglo-Egyptian Sudan which they 

 named sudanenMS, characterized by its distinctly pale coloration, 

 the breast below the collar being pale vinous pink of a more bluish, 

 less reddish tinge than in its nearest ally, S. s. aequatoriaUs, and also 

 by its small size (wing 128 to 133 millimeters). Gyldenstolpe ^"^ 

 and Van Someren ^^ found that individual variation seemed to ac- 

 count for all the so-called racial characters of aeqvatorialis and con- 

 sequently synonymized it with senegalensis. The latter author, how- 

 ever, more recently " has reversed his decision and grants Erlanger's 

 form racial standing. Granvik ^^ likewise recognizes aequatorialis. 



The series studied in this connection, including in all some 44 

 specimens, of which 36 are aequatorialis, indicates that individual 

 variation is unquestionably very great. Unfortunately no typical 

 senegalensis are available for comparison, but the specimens collected 

 by Mearns fall into two distinct groups — a light group (from Dire 

 Daoua and Sadi Malka, Ethiopia, and northern Kenya Colony) and 

 a darker, redder series from southern Ethiopia (Tertale, Wobok, and 

 Gato River near Gardula). At first glance it would seeiii that two 

 races are represented, but in this case geography fails to conform 

 with color variation, as farther south (East Africa generally) the 

 birds are light like those from Dire Daoua. Furthermore, the dark 

 birds seem hardly a recognizable aggregate, as two of the Dire 

 Daoua specimens are just about as dark as those from Wobok. The 

 presence of darker individuals in the range of aequatorialis is prob- 

 ably the reason why many workers have concluded that that form 

 was not valid. However, although the difference between dark and 

 light aequatotmlis is very marked, the darkest specimens are not 

 as dark or as red as the colored figure of senegalensis given by 

 Erlanger.-'' In that plate the coloration of senegalensis and that of 

 aegyptiaca are alike. I have compared the darkest specimens of 

 aeqiMtormlis Avith some of aegyptiaca and find a constant difference 

 in color. The typical form and the Egyptian one are redder below 



^8 Nov. Zool., 1916, pp. 82-83. 



«ldpm, 191G, pp. 832-834. 



MRungl. Sv. Vet. Akad. Handlgr., 1924. p. 311. 



=1 Nov. Zool., vol. 29, 1922, p. 36. 



^■^ Journ. East Afr. and Uganda Nat. Hist. Soc, No. 30, July, 1927, p. 93. 



"Journ. f. Ornith., 1923. Sondcrlieft, p. 49. 



"Idem, 1905, pi. 5, fig. 1. 



