TERPSIPHONE PAKADISI. 405 



Wlim fully fledged the back, wings, and tail are chestnut-red, the inner webs of the quills dusky, the head, crest, and 

 hind neck glossy black, and the chin and throat dark iron-grey, almost black on the chin, and blending into the paler 

 grey of the chest and breast, which changes into greyish white on the lower parts. The female has the throat 

 paler than the male. 



Change of plumage. At a certain age, and at a season of the year varying in Ceylon from November until May, the 

 male birds ehauge by an alteration in the colour of the feathers from chestnut-red to white. The red colour on 

 the quills, scapulars, and rectrices changes or fades into white, and the shaft-streaks of black simultaneously appear. 

 The scapulars and primaries usually change first, and then the tail-feathers ; and of the body-feathers, I have 

 generally noticed that the upper tail-coverts are the first to fade. While this is going on (and, in fact, from what 

 I have been able to gather from all the specimens that I have examined, almost before any of the upper-surface 

 feathers have changed) the grey breast just beneath the black throat turns white. I have a short-tailed example 

 with a pure white chest and only one white feather (in the scapulars) on the upper surface. There are a series 

 of long-tailed chestnut examples in the national collection with various white feathers among the primaries, 

 secondaries, scapulars, and rectrices, and all with chests pure white. 



Obs. Mr. Sharpe remarks, in his Catalogue, that South-Indian and Ceylonese red birds have the chest greyish, and 

 those from Northern India white. I observe, however, that these white-chested examples are all in a state of 

 change on the upper surface to the white plumage ; and at this period the chest is always white in Ceylon birds, 

 inasmuch as it seems to be the first part to change. 



The nearest ally to this species is T. affinis, which inhabits Burmah, Malacca, and portions of the Malayan archipelago. 

 It is distinguished from T. parodist by having the feathers of the crest all of the same length, giving it a broad 

 and " bushy " appearance, and also by the white bird having the feathers of the hind neck and back with black 

 shafts, and the adjacent edges of the web grey, which imparts a streaky appearance to the upper surface. Eed 

 birds which I have examined in the national collection have nearly all the under surface dark iron-grey, the 

 abdomen and under tail-coverts only being white. Toung birds have the back yellowish chestnut, changing into 

 a darker hue on the rump, and the under surface yellowish or fulvous grey. The measurements of a red bird 

 from Burmah are — wing3'5 inches, tail 13-5 ; a white example (Sumatra) — wing 3-5, tail 105 ; another (Flores) — 

 wing 3 - 55, tail 134. 



Distribution. — The Paradise Flycatcher is a partial migrant to Ceylon, and its movements are perhaps the 

 most singular and the most difficult to study of any Ceylonese bird. The adults, in red and white plumage, 

 arrive in the island about the last week in October, spreading over the whole country, and not finally leaving 

 again until the latter end of May. In the clamp districts on the western side lying between Negombo and 

 Tangalla it remains no longer than March. x\n inland movement then takes place north and east, many birds, 

 however, at the same time (according to my observations) quitting the island entirely. Others remain in the 

 last-named quarters to breed, and do not leave until the end of May, or even the first week in June. By this 

 time the whole of the white birds have disappeared, and I believe also the adult red ones. I have never seen 

 a long-tailed red bird between the months of May and October, nor can I find any one who is certain of the 

 contrary. Should I be correct, therefore, in this hypothesis, the fact of a total migration of the adults is 

 established. The young birds remain in the island, inhabiting the northern half and the eastern side as far as 

 Hambantota ; and on the arrival of the adults in the following season many of these yearlings follow them 

 into the west. It appears, however, probable that with the general inflow in October many yearling birds 

 arrive from abroad, as the numbers to be met with in all parts of the low country preclude the possibility of 

 their all being recruited from inland-bred birds. Here, then, we have the extraordinary fact of the disappear- 

 ance of all old birds in the island, whereas their progeny are left behind to await their return in the following 

 season, and likewise the arrival, with these latter, of many more young from the mainland, who partake in the 

 general stream of migration throughout the country. As regards the mountainous districts of the island, 

 Dumbara and other parts of corresponding altitude in the Kandyan Province, and also portions of the southern 

 ranges, are visited for the same period as the west coast, the birds quitting the hills in March. I am not 

 aware of its occurrence anywhere above an altitude of 2000 feet, at which Mr. Bligh has seen it in the 

 Kandy district. It inhabits the northern and eastern portions of the island in greater numbers than the west 

 coast, there appearing to be an appreciable diminution of the species south of the Maha-Oya, north of which. 



