157 



of this latter hue, which is also considerably developed on the abdominal feathers, and almost wholly 

 supersedes the rufous on the flanks. Another specimen from Tenasserim has but a very slight trace of 

 rufous left towards the vent and bordering the under tail-coverts, being elsewhere wholly cyaneous, excepting 

 the dusky black of the large wing- and tail-feathers, and the tips of some of the wing-coverts, which are 

 whitish. Finally, the Chyebassa (Central India) specimen is totally devoid of any rufous trace whatever, 

 but has most of its clothing-feathers slightly dusky tipped, with minute pallid extreme tips, in which condition 

 of plumage it accords with P. pandoo, Sykes." Mr. Blyth further proceeds to make remarks on the female ; 

 but we cannot enter into considerations of this sex, believing as we do that its history is not clearly worked out, 

 and must be made the subject of a series of careful observations in the field. About the specimen froni Luzon 

 there can be no question ; for, as Mr. Blyth justly observes, it is the true Turdus manillensis , Gm., i. e. P. 

 solitarius of the present article. His Tenasserim specimens are the same bird, the first described being in a 

 transitional state from the blue-and-red dress to the fully blue garb ; the second being still further advanced, 

 and retaining but slight traces of the rufous plumage on the abdomen. In sending to the same journal a little 

 later (xii. p. 182) some critical remarks on his paper published in the previous volume (/. c), Mr. Blyth makes 

 some further remarks on these species : — " Petrocincla manillensis, auct., and P. pandoo aut maal of Sykes. 

 The birds referred to under these denominations are most puzzling ; and I am now inclined to suspect that 

 three, if not four, closely allied species will eventually prove to inhabit South-eastern Asia and its islands. In 

 loc. cit. I have described a male from Lufonia, which is unquestionably the Turdus manillensis, Gmelin, while 

 there is every reason to presume that T. eremita, Gmelin, refers to its female, as Petrocincla maal of Sykes is 

 the female of his P. pandoo." Mr. Blyth then gives a careful description of two specimens from Macao, com- 

 paring the male with his Luzon example ; and he sums up as follows : — " Upon full consideration I consider 

 the Chinese and Philippine-Island specimens to be of the same species, or Petrocincla manillensis vera." This 

 conclusion is undoubtedly correct. Then he proceeds (and here his remarks are most important) : — " A second 

 species appears to exist in the specimens from the Tenasserim provinces ; and to this I refer a fine male from 

 Darjeeling, where the collector lately employed by the Society never obtained more than this one example. 

 Judging from the Darjeeling specimen (for those from Tenasserim have the tail imperfect) it would appear 

 readily distinguishable from P. manillensis by the shape of the tail, which (instead of being squared) has its 

 outermost feathers nearly half an inch shorter than the middle ones. The mottlings of the upper parts are 

 nearly obsolete, and those of the lower parts but little more developed ; and there would appear to be generally 

 some trace of ferruginous, more or less ; in the Darjeeling specimen this is confined to the lateral margins of 

 two or three of the lower tail-coverts ; and successively more developed in two from Tenasserim, as formerly 

 described by me. I shall designate this presumed species P. affinis. The third form is P. pandoo from 

 Hindoostan, which would appear to have never any rufous whatever, and has the tail intermediate in shape 

 to those of the two preceding." The Darjeeling bird mentioned by Mr. Blyth in the above paragraph is 

 another P. solitarius in nearly full blue dress. In a subsequent article (J. A. S. B. xvi. p. 150) Mr. Blyth 

 describes a new species from N.-W. India as P. longirostris. " This species," he says, " I only know from a 

 female, presented to the Society by Captain Boys, who procured it on the march from Scinde to Ferozopore." 

 In a footnote he adds : — " Can this be P. cyanea of Europe ? Lord A. Hay has procured a species in 

 Cashmere, which he thinks is the European one ; and various other European birds occur' there, as Corvus 

 monedula and Coracias garrula, which (as his Lordship informs me) abound in the valley of Kashmir." 

 Subsequent specimens have confirmed the fact that the bird from these parts is P. cyanus ; and Mr. Blyth 

 himself has long ago acquiesced in this decision. In this same paper he gives the range of the eastern 

 Petrocossyphi as follows: — 1. P. affinis, Blyth : rare at Darjeeling; but common along the eastern side of 

 the Bay of Bengal, from Tipperah and Aracan to the Tenasserim provinces. 2. P. pandoo, Sykes : inhabits 

 Central, Western, and Southern India. 3. P. manillensis (Gm:) : inhabits the Philippines and China. 

 Latterly Mr. Blyth's faith in the species P. affinis somewhat wavers ; and the last stage of the controversy has 

 been well set forth by Dr. Jerdon in the ' Birds of India' (i. p. 511) : — " Colonel Sykes separated the Indian 



4P 



