.--The Sey- cheiles Islands in the Indian Ocean have had two endemic species of white-eyes, Zosterops modesta on the main island, Mahe, and the now extinct Z. semifiava on the small island of Marianne. Newton (1867) found Z. modesta to be "tolerably plentiful" in groves of clove trees Eugenia caryophyllata. He noted that they did not sing, but from dissection he thought that the specimens he shot in late January would soon have bred. Subsequent workers in the Seychelles (NicoI1 1906, Vesey- Fitzgerald 1940, Lousteau-Lalanne 1962, Gaymer et al. 1969) did not add to our knowledge of this species, and Watson et al. (1963) said that "nothing is known of its biology." The following observations on the Seychelles White-eye were made between January and August 1972, and November 1972 and May 1973. The Seychelles have two seasons: the wet northwest monsoon from November to March, and the drier southeast trades, from April to October, so that these observations were mainly during the rainy season. Like many white-eyes, Z. modesto is relatively tame, and on many occasions I watched them from a distance of about 2 m. The head, back and tail are olive- gray, the rump is paler with a yellowish tinge. The breast is paler, less olivaceous, and the belly is tinged yellow. The throat is white, and in poor light the fore- head also appeared white, but this was less noticeable in sunlight. The flanks are rusty, this being especially noticeable during wing-flicking, when the crown feathers are also raised. The white eye-ring is complete but for a small gap in front of the eye. Four males and one female, collected on 16 February 1867 by E. Newton and now in the University of Cambridge Museum of Zoology, showed the rusty flanks to variable extents, being barely visible in one of the males. These specimens also showed a variable amount of rust on the primaries. In one male, the pri- maries were almost entirely rusty, while in another they were brown with rust restricted to spots around the rachis. In the other three specimens this feature was hardly noticeable. Thus Z. modesta retains a limited amount of yellow pigment that is not men- tioned in Newton's (1867) original description or in subsequent reports (Moreau 1957, Lousteau-Lalanne 1962). Mees (1957) however, stated that the mantle "still has a definite greenish hue." Nevertheless like many insular forms Z. modesta has lost much of the carotenoid pigment of continental white-eyes. It appears to fall between the tvo Mascarene white-eyes, where Z. olivacea retains more green coloration and has a broad eye-ring, while Z. borbonicc has lost both yellow pig- ment and the eye-ring (Gill 1971). Z. modesta is known only from the island of Mahe. Apart from one pair of xvhite-eyes at Sans Souci (ca. 170 m), all white-eyes seen were above 300 m. They were heard infrequently at La Misere, but appeared to be most abundant in the Mission-Casse Dents area, where most of these observations were made. In all areas they inhabited mixed secondary forest, especially where tall trees Alblzzia )alcataria, Swietenia macrophylla, and Pterocarpus indicus had an underlying, or were adjacent to, a scrub layer of young Albizzia and Pterocarpus, together with Cinnamonurn zeylanicum, Chrysobalanus icaco, Phyllanthus casticum, and Dillenia )erruglnea. Most feeding was done in this scrub layer, usually below 3 m. Between November and January most of the birds seemed to be in nomadic feeding parties of either four or five birds, no larger or smaller parties being seen. These parties were con- stantly moving through the undergrowth and calling. While feeding the call was a low and intermittent "cher chew but when the birds moved from one bush to another the calls became louder and more frequent, as described by Lousteau- Lalanne (1962). Two other calls, both probably alarm calls, were heard: a loud and hard chewick when a mynah (Acrldotheres tristis) flew overhead, and a chat- tering trill when a Seychelles Bulbul (Hypsipetes crassirostris) flew close to a feed- ing party. The birds appeared to be largely insectivorous, feeding especially on a mealy bug Coccus sp. that was common on young Albizzla. In addition, they took a wide variety of caterpillars in quantity. They may also have eaten ants, these being abundant in flowers of tea (Camellia thea) in which they had been feeding. The only vegetable matter seen to be eaten were berries of Lantana camara, but Lousteau-Lalanne (1962) thought they may take nectar. This insect and vegetable diet is typical of most members of the family (Mees 1957). White-eyes rarely sang in November-January when most of them were in feeding parties, but they sang regularly from February to April. The song is varied and melodic, each phrase consisting of several notes that are not organized into a cadence. Singing occurs throughout the day but is most persistent in the morning and evening. All notes on breeding were made in 1973, and the first indications of courtship were seen in two feeding parties, each of four birds, on 25 and 31 January. On each occasion one bird (presumably a female) moved close to another, and lowered and flattened its half-extended wings. The second bird then gave it a caterpillar and during this period neither bird called. Shortly afterwards the second presumed female begged to her mate who gave her a caterpillar. This courtship feeding lasted only a few seconds, after which normal feeding, with the cher call, was resumed. In February song became more frequent and white-eyes were less commonly seen in feeding parties, until by mid-March all birds seemed to be in pairs, each pair having one or two favored song perches, usually an elevated position in a bush, tree, or on telegraph wires. They may have been territorial, but I saw nc, con- flicts between birds. From mid-March onward I concentrated my observations on one pair that I could nearly always find within a small area. On 22 March copulation was seen. What eventually proved to be the female flew into a young Pterocarpus tree, giving the feeding call. She did not feed, but moved along the branches with highly exaggerated wing-flicking movements, giving intermittent low phrases of song. Shortly afterward the male arrived singing in flight, and settled high in a Pterocarpus tree about 10 m tall. The female flew up to the still singing male and crouched low on the branch. The male moved closer and began pecking at the female's eye-ring before mounting. Mees (1969) suggested that the eye-ring of the Zos- teropidae might have some recognition function, but this observation, of which I have found no parallel in reports of other species, suggests that where an eye- ring is present, it may have an epigamlc function. Copulation lasted a few seconds, after which the two birds preened briefly before separating, the male singing loudly and the female again giving the intermittent low song. This pair was first seen carrying nest material on 27 March, when each bird took a fine root of Chrysobalanus into an overhang. No nest was found here, but the following day they were seen breaking the seed stalks of the grass Panicurn rnaxirnurn and carrying them to a half-completed nest in a mahogany Swietenia rnacrophylla. Both members of the pair took part in building the nest, which was a small cup made of grass and root fibers and supported by petioles about 1 m from the end of a mahogany branch, about 10 m above the ground. The nest was completely inaccessible, but appeared typical of nests of the Zosteropidae (Mees 1969). Subse- quently the birds appeared to remove material from the nest and took it together with more panicurn stalks to a second mahogany. The foliage of this was too dense for me to see what they were doing. On 29 and 30 March both birds visited the nest but were not building, and the nest was no further advanced. After this they were not seen to visit the nest again, and the pair became less vocal and spent less time together. Song was last heard on 12 April, and thereafter the birds were occasionally seen singly until 5 May, after which they were neither seen nor heard. Searches in the other areas where white-eyes had been found earlier in the year also failed to reveal birds, and this followed the pattern of May-August 1972, when I was unable to find them, either as pairs or as feeding parties. In July 1974 Keith (pers. comm.) found a mixed party of sunbirds Nectarinia dussurnieri and up to  dozen xvhite-eyes at Casse Dents, feeding in the tops of trees. On 19 July 1974 I located a mixed party in the same place from the feeding calls of the white-eyes. They were in the tops of tall Alblzzia, again ap- parently feeding. Out of the breeding season Z. rnodesta may therefore be more sociable and more nomadic than during it. The Seychelles White-eye is inconspicuous and difficult to find in the rain forest it inhabits. It also appears to be highly localized, and at present is nowhere abundant. The species may have declined since its discovery by Newton (1867), but in view of the difficulty of finding it and of the paucity of observers in the first half of this century, there is no evidence for the "decline and apparent recovery" of the species reported by Gaymer et al. (1969). The reason for its scarcity and restriction to Mahe is unknown. On other Indian Ocean islands (except Assumption) white-eyes, together with the sunbirds Nec- tarinia sp., are generally among the commonest endemic passerines (Benson 1960, Benson and Penny 1971, Gill 1971, Watson et al. 1963). Sunbirds N. dussurnferi are common and widely distributed in the Seychelles, but it is unlikely that com- petition between sunbirds and white-eyes, which Stoddart et al. (1970) and Ben- son and Penny (1971) thought accounted for the absence of Zosterops on Assumption, would be important on the larger and more diverse granitic islands of Seychelles. The extinction of Z. semifiava on Marianne was undoubtedly due to the clear- ance of the indigenous forest and its replacement with coconuts Cocos nuclfera. Very little indigenous forest remains on Mahe, the existing forest comprising a mixture of endemic, indigenous, and exotic trees, with in many places a scrub layer of Chrysobalanus and Cinnamonurn, which appears to be the habitat the white-eyes preferred. I am grateful to the Natural Environment Research Council for a Research Fellowship to Aberdeen University that enabled me to undertake this work. C.W. Benson kindly showed me the white-eyes in the Cambridge University Museum of Zoology, and commented on an early draft of this paper. LITERATURE CITED BEnSOn, C.W. 1960. The birds of the Comoro Islands. Ibis 103b: 5-106. BEnSOn, C. W., AD M. J. PE3/4. 1971. The land birds of Aldabra. Phil. Tram. Royal Soc. London B 260: 417-527. GA3/4VrER, R., R. A. A. BLAC:A, P. G. DAWSOn, M. PE3/4, An C. M. PE3/4. 1969. The endemic birds of Seychelles. Ibis 111: 157-176. GiL, F. B. 1971. Ecology and evolution of the sympatric mascarene white-eyes Zosterops borbonica and Zosterops olivacea. Auk 88: 35-60. LOrrSTEArr-LAAE, P. 1962. Land birds of the granitic islands of Seychelles. Occ. Publ. Seychelles Soc. No. 1. MEES, G. F. 1957. A systematic review of the Indo-Australian Zosteropidae, part 1. Zool. Verhandelingen 35: 1-204. MEES, G.F. 1969. A systematic review of the Indo-Australian Zosteropidae, part 3. Zool. Verhandelingen 102: 1-390. MOREArT, R. E. 1957. Variation in the western Zosteropidae (Aves). Bull. Brit. Mus. Nat. Hist. Zool. 4: 311-433. NEWTOn, E. 1867. On the land birds of the Seychelles archipelago. Ibis 3, 2nd Ser.: 334-360. NiCOLL, M. J. 1906. On the birds collected and observed during the voyage of the "Valhella" R.Y.S. from November 1905 to May 1906. Ibis 6, 8th ser.: 666-712. SrODDAT, D. R., C. W. BEnSOn, An J. F. PEAXE. 1970. Ecological change and effect of phosphate mining on Assumption Island. Atoll Res. Bull. 136: 121-145. VESE3/4-FITZOEaA;n, D. 1940. On the birds of Seychelles. 1. The endemic birds. Ibis 4, 14th ser.: 480-489. WATSOn, G. E., R. L. Zrrsi, nD R. E. STOaEa. 1963. Preliminary field guide to the birds of the Indian Ocean. Washington, D.C., Smithsonian Institution. CmISTOPHER J. FE^aE, University of Aberdeen, Culterty Field Station, Newburgh, Aberdeenshire AB4 0AA Scotland. Present address: Pest Infestation Control Labora- tory, Tangley Place, Worplesdon, Surrey, England. Accepted 20 Aug. 74.