1866.] VISCOUNT WALDEN ON BIRDS FROM TENASSERIM. 553 



Bengal, and agree in having bright yellow irides ; while those from 

 Southern India (P. mahrattensis, Sykes, P. Z. S. 1832), otherwise 

 nearly similar in plumage, are distinguishable by having the irides 

 white or greyish white. 



From the Andamans. 

 1. Halcyon capensis, (Linn.). 



Alcedo capensis, Linn. S. N. ed. 12. 1/66, i. p. 180. no. 9. 



Ispida capitis bonce spei, Briss. Orn. iv. p. 488. no. 8. 



No. 3. Port Blair, Andamans. 



If we resolve to regard all the local races of this Kingfisher, or any 

 of them, as constituting so many distinct species, we must dissever 

 the A. capensis, Linn., from the A. leucocephalus, Gm. Of the Lin- 

 naean species we possess a detailed description in Brisson's ' Ornitho- 

 logia ;' and with it this Andaman specimen very nearly agrees. The 

 upper part of the head is " cinereous inclining to fulvous," and forms 

 a distinct cap. Dr. Jerdon observes, when writing on the Indian 

 form, that specimens from the east, and especially from Tenasserim, 

 have the cap " albescent or dirty brown." Its blues are not pure, 

 but greenish as in Brissou's type. A. leucocephalus, Gm., founded 

 on Buffon's 757th plate, represents the form in which the occipital 

 plumage is uniform with that of the neck, nape, and under surface, 

 the cap being therefore absent, and in which the blues are pure, rich, 

 and brilliant. A Sarawak specimen in my collection fully answers 

 to Buffon's account and plate, even to the brown striae of the occiput, 

 this appearance being the result of each feather having a dark brown 

 centre at its base, and being but narrowly edged with fulvous. When 

 these feathers normally overlap one another, the occiput appears of 

 a colour uniform with the nape, the brown centre of each feather 

 being invisible ; but if any of them are disarranged the occiput 

 puts on the appearance shown in Buffon's plate. Thus A. capensis, 

 Linn., forms the type of all the races with a coronal patch, and A. leu- 

 cocephalus, Gm., of those in which the cap is wanting. To the first 

 belong the Hindostan and Ceylon birds, the Bengal race having been 

 accurately described by Captain Pearson (J. A. S. B. 1841, p. 633) 

 under the name of H. gurial ; and if all the Indian specimens are, 

 as Ur. Jerdon states, uniformly brown on the head, Pearson's name 

 must be applied to them. A Flores specimen in my collection has 

 the head and crest dark brown, as in Ceylon specimens of mine. 

 But while in the Ceylon bird, which seems to be identical with H. 

 gurial, Pears., the scapulars and wing-coverts are dingy greenish 

 blue, in the Flores bird the blue is intense and perfectly pure, as in 

 the Sarawak specimen. The variations of this widely extended form 

 have yet to be studied. Dr. Cabanis observes (Museum Heiueanum, 

 i. p. 156) that every degree of variation is to be found in a series of 

 specimens of this species ; but he does not mention whether he was 

 certain of the localities from which each specimen came. My expe- 

 rience of the form is the same as that of the learned doctor, but with 

 this addition, that the varieties are coincident with changes of loca- 



Proc. Zool. Soc— 1866, No. XXXVI. 



