360 
Birds of Celebes: Hirundinidae. 
Others from East Celebes and elsewhere do not differ in size; in two the wing 
is 110, Ml inm, hut about 103 seems to be tbo average. 
Egg. India: closely resembles that, of II. rustica, but is decidedly smaller; moderately broad 
oval, slightly compressed towards one end, ground-colour i^ink- white, very finely 
speckled and spotted, thinly at the small, more densely at the large end, with 
different shades of dull pur 2 i]e and brownish red: 2 to 5 to a sitting; 16.3 — 19.6 
X 12.2 — 14.5 mm (Hume 26). “Examples in my collection from Celebes and Borneo 
resemble those of our rustica, ai’e, however, smaller and measure 17.5 X 13 mm” 
(Kehrkorn MS.). So, also, two specimens in the Sarasin Coll. (8. Sept.). 
Nest. A deep half-saucer. Of mud containing many fresh-water snail-shells, well lined Avith 
fine black bair-like vegetable fibres, a few feathers, a small piece of cotton (Kenia, 
N. Celebes, specimen collected by Dr.s. Sarasin). In India various writers, {26) 
mention a thick bed of feathers only as forming the lining of the nest; Mr. Theobald 
{26) speaks also of vegetable dowm, and Col. Legge {11) of “feathers, threads, small 
pieces of rag”, and other things it may chance to pick up. Its habits of nidification, 
therefore, appear to differ somewhat between India and Celebes, as is not surprising 
in a case where migration and intermixture of individuals do not ajApear to be 
the rule. 
Breeding season. North Celebes — July (Sarasin MS.); South India — 2 broods in 
succession from February to April (Davison 26); Ajmil to June (Wait 26); Ceylon 
— April to June (Legge 11); Tenasserun — lays in the second week of April 
(Theobald 26). 
Distribution. South India and Ceylon (Legge 11); Andamans (Hume c 2); Tenasserim 
(Davison 8, Theobald 28); Malay Peninsula (Hume JO, etc.); Sumatra (Wallace J7, 
Hartert 21, Modigl. 30); Billiton and Mendanau (Vorderraan 27, 28); Borneo 
(Wallace, Mottlcy, etc. 20); Palawan (Everett 20, Whitehead 24); PhilijApines 
— Mindoro, Samar, Negros, Basilan, Leyte, Bohol, Dinagat, Cebu, hlindanao 
(Steere, Platen, etc. 2.9); Sooloo (Guillem. 16); Great Sangi (Meyer 14, Platen 19); 
Lenibeh, Manado tua and Banka Is. (Nat. Coll, in Dresd. Mus.); Celebes — North 
Peninsula (Wallace 3, Guillem. 16, Sarasin); E. Peninsula (N.O.); South Peninsula 
(Wallace 5, Meyer 9, Weber 32); Togian Islands (Meyer 9); Sula Islands 
(Allen 3, 17); Natuna Is. (Hose 36); Java, Lombok, Timor (Salvad. 12, Sliarj^e IT); 
Timorlaut (Riedel, Mus. Dresd.); Ternate, Morty, Batchian, Bum, Amboina, 
Matabello (rect. Watubella), Kei, Salawatti, Ncav Guinea, Join, Aru, Yule Id., Duke 
of York Id. (Salvad. 12); Islands of Torres Straits (Sharpe J7); Cape York, Australia 
(Ramsay 18). 
The Java Swallow seems to be a stationary species. Davison (J3) speaks 
of it as a resident and very common about the Nilghiris, Bourdillon (7) as 
“a resident travelling but little” in the Travancore Hills. Hume believes it to be 
a migrant to the Andamans, but it is known in literature as a breeding sirecies 
in South India, Ceylon, Tenasserim, Billiton, Borneo, .lava, Celebes, Duke of 
York, and doubtless elsewhere. Sharpe (17) says that “specimens from Tra- 
vancore are much duller beneath than any other birds yet examined, and those 
from South-eastern New Guinea and Torres Straits are palest below . . . but 
light-bellied examples are found in Borneo and other localities”. These may 
be incipient subspecific distinctions, but the bird appears to be darker in 
