NOVITATKS ZOOLOtillAE XXIV. 1!)17. 4-57 



Hajeilah, also two taken by Dodson at Lahej near Aden, and I am sorry to say 

 that I cannot recognise their differences from P. d. indicus. 



(I may here add that the boundary between P. d. domesticus, which, according 

 to Zarudny, is still found in Ghilan and Masanderan in northern Persia, and 

 P. d. indicus which inhabits at least the eastern parts of Transcaspia, must be 

 somewhere in the latter province, as some specimens from there have the long 

 wing of P. d. domestictis. while others appear to be intermediate.) 



Emberiza tahapisi arabica (Lorenz and HcDni.). 



Friiigillaria iiiabica Lurenz and Hellmayr. Orn. Momilsb. 1902. |). 5.5 (Ycshbuiii, 8. Arabia) : Sclatcr, 

 Ibis. 1917. p. 148. 



I cannot see the necessity or advisabiUty to separate the so-called Fringillariu 

 from Emberiza. As to E. arabica it is undoubtedly a subspecies of E. lahafisi. 

 and is much nearer to E. t. tahapisi than to the Socotran form, which is con- 

 siderably paler. 



Anthus sordidus arabicus subsp. iiov. 



.inthiii kiKophri/s caytus (nee Haitert!) Sclater, Ibis, 1917. ji. 157. 



The Pipit of South Arabia is very unlike capius, but after due consideration 

 can be regarded as a subspecies of the same species. It has the same elongated 

 bill and general structure, but differs at a glance by having the upper side 

 considerably darker, the wide edges to the wing-coverts browner and darker, 

 sides of the head darker, underside more brownish and much more heavily 

 spotted on the chest. Eliminating a number of apparently wrongly sexed speci- 

 mens I measure: 10 S wings 195-101-5, 10 ^ 89-94 mm. 'i ype : "§" ad. 

 (apparently <J) Menakha 29. i. 1913, No. 331 G. W. Bury Collection. (Tring 

 Museum). (31 .skins examined and compared.) 



This form differs from A. s. hararensis in being much larger and more 

 heavily spotted on the chest, while A. s. longirostris has the edges to the 

 feathers of the upperside and the rump more cinnamon-brownish. 



Hab. : Yemen (Menakha, Wasil. Hajeilah, Souk-al-Khamis) and Amiri 

 district. South Arabia. No doubt breeding in these countries. 



While describing this new form ray attention has been called tu the binl 

 from Sokotra. which is another distinct subspecies. It has very little to do 

 with A. sordidus sordidus from Abyssinia, which is very dark brown on the 

 upperside, with rusty browTi edges to the wing-coverts, and a much more rusty 

 cinnamon undersurface. In coloration of the upperside the Sokotra form is 

 intermediate between arabicus and capius, being dark brown with pale edges 

 to the feathers, but underneath it is even lighter than capius. Unfortunately, 

 tile specimens collected by the Grant-Forbes expedition are all in very worn 

 plumage. Wing shorter than in arabicus. i.e. S ad. 89-90, $ 84-85 mm. 



I call this form, which breeds in Sokotra : 



Anthus sordidus sokotrae subsp. nuv. 



Type : 5 Ahlo Pass, 3,500 ft.. Sokotin i>. ii. 1899, Ogilvie-Grant and H. 0. 

 Forbes coll. (Tring Museum). 



Neumann {I.e. p. 234) has already called attention to these new forms. 



In 1905, when describing the large pale Pipit, which inhabits Persia, Bahi- 



