COLLOCALIA FRANCICA. 327 



manager of which informed us that he had manured 100 acres of coffee with it during that season. Besides 

 this colony, there are two other smaller offshoots on the adjoining estate, in one of which, Mr. Bligh tells me, 

 the birds have to pass through a cloud of spray in order to gain access to their nests. 



Concerning the large breeding-station on the Bentota river, Mr. Layard writes, in the ' Annals and 

 Magazine of Natural History' for 1853, xii. p. 168, as follows : — " Having fully described my acquaintance with 

 these birds in a letter to my friend Mr. Blyth, I cannot do better than copy what I then wrote : — 'The cave is 

 situated at a place called Havissay, about thirty-five miles from the sea and twenty from the river, and about 

 500 feet up a fine wood-clad hill called Diagallagoolawa or Hoonoomooloocota. Its dimensions are as 

 follows — length between 50 and 60 feet, about 25 broad and 20 high. It is a mass of limestone rock, which 

 has cracked off the hill-side and slipped down on to some boulders below its original position, forming a 

 hollow triangle. There are three entrances to the cave, one at each end, and one very small in the centre. 

 The floor consists of large boulders, covered, to the depth of 2 or 3 inches, with the droppings of the birds, old and 

 young, and the bits of grass they bring in to fabricate their nests. The only light which penetrates the cavern 

 from the entrances above mentioned is very dim ; when my eyes, however, got accustomed to the light, I could 

 see many hundreds of nests glued to the side of the fallen rock, but none to the other side, or hill itself. 

 This I attribute to the fact of the face of the main rock being evidently subject to the influence of the weather, 

 and perhaps even to the heavy dews off the trees ; but for this the side in question would have been far more 

 convenient for the birds to have built on, as it sloped gently outward, whereas the other was much overhung 

 and caused the birds to build their nests of an awkward shape, besides taking up more substance. I was 

 at the spot a few days before Christmas, and fancy that must be about the time to see the nests in perfection. 

 This is corroborated by the fact of my finding young birds in all the nests taken by me, and by what the old 

 Chinaman said, that the ' take ' came on in October. I find that they have three different qualities of nests, 

 and send two for your inspection ; the best is very clean, white as snow and thin, and is also very expensive. 

 The most inferior are composed of dry grasses, hair, &c. ; but I could not detect any thing like the bloody 

 secretion as described (though only under peculiar circumstances of exhaustion) by Mr. Barbe, even in a fresh 

 nest. I was in the cave late (after 5 p.m.) in the evening of a day which threatened rain ; but the old birds 

 were still flying round the summit of the mountain at a vast altitude, occasionally dashing down into the cave 

 with food for their nestlings.'" 



Genus DENDROCHELIDON. 

 Bill much as in Collocalia, but smaller, if any thing, and much hooked at the tip. Wings 

 with the 1st quill the longest. Tail very long and deeply forked, the lateral feathers much 

 attenuated. Tarsus very short, much less than the middle toe ; three anterior toes subequal ; 

 hallux long, directed backwards, and not reversible. 



