Little known Oriental Bird: White-eyed River-Martin

by Joe Tobias, from OBC Bulletin 31, June 2000.

In January 1968, during the course of ringing activities at a wetland site in south-central Thailand, fieldworkers discovered a strange swallow amongst large numbers of migrant hirundines. It proved to be a new species and was christened the White-eyed River-Martin Pseudochelidon sirintarae by Kitti Thonglongya who dedicated this spectacular and beautiful bird to Princess Sirindhorn Thepratanasuda. Over the next three years several more specimens were collected at the same site, but apart from these, and a fleeting observation in 1978, this remarkable bird has effectively vanished. An avian enigma, it has come to epitomise the mythical allure of rarity to the birdwatcher, and for three decades it has symbolised the Asian mystery of the ornithological world. As such it has appeared in logo form in the pages of this journal as the archetypal little-known bird. The time has come to compile our knowledge of the species and to present it afresh in the hope that it might lead to a dramatic rediscovery.

To begin with, we need to retrace the events of January and February 1968 and glean what we can from the available facts. The site of discovery is first misleadingly given as a big marsh on the Chao Praya River (1). The type-locality is then specified as Bung (= Nong = Lake) Boraphet, Amphoe Muang, Nakhon Sawan Province, central Thailand (1), and from its subsequent description as a shallow, marshy, reed-filled lake of 25,000 hectares it seems clear that this is the big marsh originally mentioned (a point confirmed by Thonglongya (2)). Rediscovery efforts in 1980-1981 were apparently concentrated on an island where all of Kitti's river martins had been captured (3), suggesting that, at one time, confidence was high that a very precise origin was known.

White-eyed River-Martin  © H.E. McClure

White-eyed River-Martin © H.E. McClure

This no longer appears to be the case. The first White-eyed River-Martins were reportedly caught while night-trapping roosting swallows (Hirundo rustica, H. daurica, Riparia riparia), wagtails and warblers by casting a fishing net over a reedbed (1) a method repeated by subsequent authors (3,4,5). However, according to a local technician who worked with the original field team, the birds were neither seen in the field nor trapped by any of the team members, but rather were brought in to the teams hotel in nearby Nakhon Sawan by villagers following a broadcast appeal for live wild birds for ringing purposes (6). It seems likely, therefore, that the precise site of collection is impossible to determine, but that it is certainly in the region of Bung Boraphet, and most likely at the lake itself.

Whatever their exact origins, nine specimens were initially collected: one each on 28 and 29 January (although the label on specimen 53-1218 actually states 27 January 1968 (6)) and seven on 10 February 1968 (1). From analysis of the resultant skins its closest ally was deemed to be the African River-Martin Pseudochelidon eurystomina (1). Initially described as congeneric (1), the African species and the Asian species differ markedly in the size of their bills and eyes, suggesting that they have very different feeding ecologies, sirintarae probably being able to take much larger prey and perhaps in different microhabitats (7). The gape of sirintarae is swollen and hardened, unlike the softer, fleshier, much less prominent gape of eurystomina(1,19).

The feet and claws of sirintarae are unusually large and robust for an aerial feeder (1) and the two species also have different toe proportions, which might suggest dissimilar nesting habits (19). These differences are sufficiently pronounced in the view of some taxonomists to permit the allocation of its own genus, Eurochelidon (7), although other authors support the retention of both species in Pseudochelidon, arguing that they mirror patterns in other congeneric hirundines (8). Whether treated as one genus, or two, the syringeal structures of the two river-martins are divergent enough from those of the Hirundininae to confirm subfamily distinction from the true swallows, and apparently enough to suggest that they might belong in a separate family (1,9).

Shortly after these first specimens, a tenth bird was caught in November 1968 (2) and brought alive to Bangkok where it was photographed in December 1968 (3). Furthermore, at least two birds (one pair) reached but soon afterwards died in Dusit Zoo in Bangkok in early 1971 (3). The only widely reported field observation was of six individuals flying low over Bung Boraphet towards dusk on 3 February 1978 (10). In addition, four probable immature White-eyed River-Martins were reportedly observed perched in trees on Temple Island in Bung Boraphet in January 1980 (3,5), and one was reputedly trapped by local people in 1986 (11). Both these records remain unconfirmed. Several subsequent searches have tried to locate the species around the site. For example, eleven amateur birding groups surveyed the lake unsuccessfully during 1979 (3). Investigations were carried out between December 1980 to March 1981 by a team from the Association for the Conservation of Wildlife but, despite netting many roosting Barn Swallows in reedbeds, they failed to reveal any river-martins (12). In 1988 another concerted effort to relocate the species was undertaken at Bung Boraphet, ending with failure as the swallow roosts were highly disturbed and mobile (13).

The real number of White-eyed River-Martins trapped in the 1960s and 1970s may have been much higher than these figures suggest. In the wave of public and media interest following the sensational discovery of the species, trappers are rumoured to have caught around 120 individuals and sold them to the director of the Nakhon Sawan Fisheries Station (3,5). Moreover, local markets were reported to have had several other specimens in January-February of succeeding years (10). Having been found on Thai soil and decorated with the name of Thai royalty, there was a significant local demand for specimens or caged examples of the species, for zoos, presentation to dignitaries or as curios for the affluent.

What has become of the White-eyed River-Martin? Did this harvest of hirundines extinguish it entirely? Were these last known individuals merely the doomed remnants of a population displaced by disturbance from a specialised breeding habitat? (5) Perhaps. It is quite conceivably extinct, and if it still survives its population seems likely to be tiny. The original series of specimens taken in early 1968 were outnumbered by hordes of trapped Barn Swallows by a ratio of 9:6,000 (1). In spite of this exceptional rarity, it was thought that the species might be regular at Bung Boraphet since the local bird-catchers had a name for it, nok ta phong, the swollen-eyed bird (1).

Unfortunately, there has been a drastic decline in the Bung Boraphet swallow population from hundreds of thousands reported around 1970 to maximum counts of 8,000 made in the winter of 1980-1981, although it is not certain if this represents a real decline or a shift in site in response to persecution (3). However, an estimated 100,000 swallows were present at a roost near Chotiravi, near Bung Boraphet, in August 1986 (11) and there were 30,000 at Bung Boraphet in May 1988 (11). Nevertheless, a dealer working the large Chotiravi roost claimed never to have encountered the species (11). The general feeling is that an absence of sightings since early 1980, despite numerous observational efforts, cast ominous doubts over the survival of the White-eyed River-Martin (3).

Unfortunately, the habits of swallows around the lake appear to have altered recently, with very few birds roosting in the reedbeds until late winter (13). Much of the population now roosts in sugar cane plantations, moving back to the reedbeds after the cane has been harvested (13). The roosts also form well after dark, whereas they once gathered before dusk (13). These changes are probably the result of prolonged disturbance by trappers (11). In any case, the swallow roosts are more mobile and difficult to locate, factors that have further obstructed the rediscovery of the White-eyed River-Martin.

The reduction in Barn Swallow populations in the Bung Boraphet area is difficult to explain but intensive trapping activities for the purpose of selling birds as food in local markets must have played a major role, as must the annual destruction of roosting sites to make way for lotus cultivation (3). Huge areas of reedbed in areas frequented by roosting swallows were being burnt in February 1986 (11). The hunting of hirundines without a licence has been illegal since 1972, although this legislation is rarely enforced (3). Relations between conservationists and bird trappers at Bung Boraphet are occasionally fraught, to the extent that a reserve ranger was killed when trying to apprehend poachers at roosts in 1987 (13).

It seems that any rediscovery efforts should now be targeted away from Bung Boraphet, and indeed perhaps away from Thailand. How might we judge where best to look? What secrets have hitherto been disclosed that might help direct our search? Unhelpfully, the ecology of this bird remains almost totally unknown, and thus ornithologists have looked to its presumed relative, the African River-Martin, to provide clues. Since P. eurystomina feeds largely over both forest and open grassy country, nesting colonially in tunnels dug in sandbars of large rivers (14) it has been inferred that E. sirintarae possibly does or once did the same (4). However, the differently shaped toes might suggest otherwise (19). At least one of the initial specimens had mud or sand adhering to its claws, and while this perhaps suggests a terrestrial perching habit (6), most swallows occasionally do the same, especially when collecting nest material. Another clue: in holding cages used during the swallow ringing programme, the birds stood quietly in the corner of the cage in strong contrast to other swallows which move rapidly from perch to perch calling repeatedly (1).

White-eyed River-Martin  © H.E. McClure

White-eyed River-Martin © H.E. McClure

In the unconfirmed report of 1980, individuals were flying after insects with some Barn Swallows and sometimes perching on the tops of trees (20). During the 1978 sighting they were apparently skimming the water surface, possibly to drink (10). While these accounts describe behaviour characteristic of most swallows, the only direct dietary evidence is the fragment of a large beetle found in the stomach of a specimen (1). This fact, along with the mandibular morphology of the species, implies that it consumes sizeable prey.

What about breeding season, distribution and migratory behaviour? Five of the nine specimens collected in late January and early February 1968 were immature (1); they were later termed juvenile, and some of the other material as subadult (2) (although this is not mentioned in the original description). A breeding site within Thailand was initially considered plausible on the grounds that so many of the type series were young (2). It has also been speculated that if nesting occurs in Thailand it is most likely to do so between March and April, as this coincides with the local nesting season for the majority of insectivorous birds, while the monsoon rains from May onwards presumably raise water levels above the riverine sandflats postulated to be the favoured nesting habitat of the species (5,6,10). It is unlikely, however, that juvenile plumage would be retained for eight months, and thus these two facts are difficult to reconcile. The White-eyed River-Martin has otherwise been thought a non-breeding visitor to south-central Thailand (20) and clearly migratory (4), but these assumptions should also be treated with care. Although it has only been found between December and February, and despite the above disparity, there is insufficient information to rule out breeding in the Nakhon Sawan area (6,11). In conclusion, it is unclear whether the species is, or was, a migrant at all.

As recent searches around Bung Boraphet have been unsuccessful, let us assume it is a migrant. If it travelled across Thailand, where did it come from? The riverine nesting grounds might possibly lie along one of the four major watercourses (the Ping, Wang, Yom and Nan) which drain northern Thailand, either in the immediate vicinity of Nakhon Sawan or to the north (5,6). If it came from further afield, perhaps these putative breeding grounds lie on one of the other major river systems of South-East Asia, such as the Mekong in China, Laos or Cambodia, or the Salween and Irrawaddy in Myanmar (5,6). Evidence that the species breeds, or has ever occurred, in China is scant, although a painting by a Chinese artist held in the Sun Fung Art House of Hong Kong appears to depict the species (15). This tentative clue has failed to lead to any further information, and in any case the subject of the painting is more likely to be an Oriental Pratincole Glareola maldivarum (16).

A survey of the Nan, Yom and Wang rivers in northern Thailand was carried out in May 1969, but was not comprehensive and relied chiefly on interviewing villagers, none of whom seemed to know the bird (2). Rivers near the Chinese border of Laos were searched in April 1996, local people being shown illustrations of the species, but without success (W. Robichaud verbally 1997). Very few other surveys have looked for it outside Thailand and there is still scope for research in remote regions where a population might survive.

Throughout its possible range there is a catalogue of pressures potentially imposed on the species (5,6). Man has drastically altered the lowlands of central and northern Thailand: huge areas are now deforested, agriculture has intensified, pesticide use is ubiquitous and urban environments have spread extensively (5,6). In addition, all major lowland rivers and their banks suffer a high level of disturbance by fishermen, hunters, vegetable growers and sand-dredgers (5,6). Whole communities of nesting riverine birds have vanished from large segments of their ranges in South-East Asia owing to habitat destruction, human persecution and intense disturbance of most navigable waterways (5,17,18). Local people routinely trap or shoot birds for food and for sale in local markets (5,6). Even at Bung Boraphet Non-Hunting Area (established in 19793) the trapping of birds has continued, at some level, up to the present (5,6). If the species preferentially forages over forest, its numbers could already have declined to a perilously low level at the time of its discovery because of deforestation and the intensification of agriculture in river valleys (5,6).

These threats are based on the ecological traits inferred by its suspected taxonomic affinities. It should be borne in mind that riverine nesting habits and preferences for forest are only an assumption, and that it might conceivably utilise some entirely different habitat. Even the name river-martin is perhaps a complete misnomer, as the species has never been seen on a river and is no longer considered congeneric with the African River-Martin (7). Interestingly, the most recent scrutiny of specimens suggested that it was perhaps nocturnal, or at least highly crepuscular, based principally on its unusually large eyes (19). This raises the possibility that it is normally a cave dweller or a hole-rooster in trees or rock, emerging to feed in twilight or darkness, and this opens up new avenues of exploration. There are, for example, limestone caves not far from Bung Boraphet, and many more in Laos and southern China.

While there is only a faint chance that this, one of the most elusive species in the world (15) still survives, it bears the extraordinary distinction of being highly unusual in appearance yet overlooked by naturalists in a well-worked country until the late 1960s. As it is thus either extremely rare or inexplicably cryptic, it is not yet time to give up hope for the swollen-eyed bird. Its possible range should be revisited with a broader outlook. The prize, to any successful searcher, is considerable: solving one of the most puzzling mysteries of Asian ornithology.


  1. Thonglongya, Kitti (1968) A new martin of the genus Pseudochelidon from Thailand. Thai National Scientific Papers, Fauna Series no. 1. Bangkok: Applied Scientific Research Corporation of Thailand.

  2. Thonglongya, Kitti (1969) Report on an expedition in northern Thailand to look for breeding sites of Pseudochelidon sirintarae (21 May to 27 June). Bangkok: Applied Scientific Research Corporation of Thailand

  3. Sophason and Dobias (1984) The fate of the Princess Bird, or White-eyed River Martin (Pseudochelidon sirintarae). Nat. Hist. Bull. Siam Soc. 32(1): 1-10.

  4. Turner and Rose (1989) Swallows and martins of the world. Bromley, UK: Christopher Helm.

  5. Round, P. D. (1990) Bird of the month: White-eyed River-Martin. Bangkok Bird Club Bulletin 7(1): 10-11.

  6. BirdLife International (in press) Threatened birds of Asia.

  7. Brooke, R. K. (1972) Generic limits in old world Apodidae and Hirundinidae. Bull. Brit. Orn. Club. 93: 53-57.

  8. Zusi, R. L. (1978) Remarks on the generic allocation of Pseudochelidon sirintarae. Bull. Brit. Orn. Club. 98(1): 13-15.

  9. Mayr, E. and Amadon, D. (1951) A classification of recent birds. Amer. Mus. Novit. 1496: 1-42.

  10. King and Kanwanich (1978) First wild sighting of the White-eyed River-Martin, Pseudochelidon sirintarae. Biol. Cons. 13: 183-185.

  11. D. Ogle in litt. (1986).

  12. Anon. (1981) A search for the White-eyed River Martin, Pseudochelidon sirintarae, at Bung Boraphet, central Thailand. Bangkok: Association for the conservation of Wildlife of Thailand. Unpublished report.

  13. D. Ogle in litt. (19871988).

  14. Keith. S., Urban, E. K. and Fry, C. H. (1992) The Birds of Africa, volume 4. London: Academic Press.

  15. Dickinson, E. (1986) Does the White-eyed River-Martin Pseudochelidon sirintarae breed in China? Forktail 2: 95-96.

  16. Parkes, K. C. (1987) Letter: was the Chinese White-eyed River-Martin an Oriental Pratincole? Forktail 3: 68-69.

  17. Scott, D. A. (ed.) (1989) A Directory of Asian Wetlands. IUCN, Gland, Switzerland and Cambridge, UK.

  18. Duckworth, J. W., Salter, R. E. and Khounboline, K. (compilers) (1999) Wildlife in Lao PDR: 1999 Status Report. Vientiane: IUCN-The World Conservation Union/Wildlife Conservation Society/Centre for Protected Areas and Watershed Management.

  19. P. M. Rasmussen in litt. (2000).

  20. Ogle, D. (1986) The status and seasonality of birds in Nakhon Sawan Province, Thailand. Nat. Hist. Bull. Siam Soc. 34: 115-143.

This article is based on information to be published in the forthcoming Threatened Birds of Asia, a project sponsored by the Japanese governments Environment Agency and conducted through the BirdLife Asia Council working through the Wild Bird Society of Japan.

See Sales for prices and availability of Bulletin past issues Return to Bulletin index