172 Mr. W. L. Sclater on [Ibis, 



This species was compared by Mr. O.-Grant y\hh 

 C. umbrovirens from northern Abyssinia, but the only 

 examples of tliis form in the British INIuseum are now 

 identified with C. u. erythrem, and there are no examples ol: 

 the typical C. umbrovirens from central Abyssinia in the 

 collection. However, the present subspecies from the Yemen 

 seems to be quite distinct, and can be at once recognized by 

 its much paler uuderparts — almost white on the chin and 

 belly, and by its white lores. The upper parts, too, are paler 

 and of a more olivaceous shade of brown as compared with 

 those of C. u. erythre<s. 



The three additional examples from Wasil agree with the 

 types in every way. 



Riparia riparia. 



Hirundo riparia Linnseus, Syst. Nat. 1758, p. 192 : 

 Sweden, 



O.-Grant, 1900, p. 287. 



a. ? . Sanaa, 7600 ft. 11. ix. 13. 



The Egyptian Sand-Martin is considered by Hartert to 

 form a distinct subspecies {R. r. littoralis), but the Yemen 

 bird seems to be on the whole referable to the typical 

 form. 



Eiparia obsoleta, 



Cotr/le obsoleta Cabanis, Mus. Hein. i. 1850, p. 50 : North- 

 east Africa. 



Y'erbury, 86, p. 14, 96, p. 22 ; Barnes, 93, p. 69 ; 

 O.-Grant, 1900 p. 257 ; Lorenz & Hellmayr, 1901, p. 242, 

 1907, p. 115. 



a.?. Menakha, 7500 ft. 27.xii. 12. 



A Crag-Martin from southern Arabia is distinguished by 

 Reichenow (Vog. Afr. iii. 1905, p. 828) as R. arabica, and is 

 compared with R.fuligula of north-eastern Africa. Whether 

 this is a distinct form or not 1 am unable to say, but the 

 bird from Menakha as well as others from Aden and 

 Socotra in the British Museum closely resemble the true 

 R. obsoleta of Egypt and Palestine (if this is the true typical 

 form ; the exact type-locality seems uncertain). There is a 



