HBW 4 - Family text: Steatornithidae (Oldbird)

Family text: 

Class AVES
Order CAPRIMULGIFORMES
Suborder STEATORNITHES
Family STEATORNITHIDAE (OILBIRD)

by Betsy Trent Thomas
Independent Researcher, Virginia, USA

  • Large nocturnal fruit-eating birds, with hooked bill, wide gape, long wings and tail, and short legs.
  • 40-49 cm.
  • NW South America to Trinidad.
  • Roosts and nests gregariously in caves, locally from sea-level up to 3400 m; feeds in subtropical forests.
  • 1 genus, 1 species, 1 taxon.
  • No species threatened; none extinct since 1600.

Systematics
Alexander von Humboldt was the first scientist to see the Oilbird (Steatornis caripensis) in its native habitat in 1799, but he did not publish a detailed report of this encounter until 1817, and it was only much later in the nineteenth century that European anatomists received enough specimens for them to study and attempt to classify. The Oilbird’s affinities to the typical owls (Strigidae) and the nightjars (Caprimulgidae) seemed clear from the beginning and were never seriously questioned, largely because of their similar plumage and nocturnal foraging behaviour. The problem was the degree of relationship to these two groups, and even, by extension, of the Caprimulgiformes to the swifts (Apodidae). A study of the pterylosis of the Oilbird suggested a close relationship with the owls, yet clearly the owl foot, used in the capture and manipulation of food, is quite different from the small leg and weak foot of the Oilbird. The Oilbird’s skeleton was found to be most similar to that of the potoos (Nyctibiidae). Other anatomists have linked the Oilbird to the rollers (Coraciidae), trogons (Trogonidae) and cuckoos (Cuculidae), and Humboldt even suggested a relationship with a passerine species, the Alpine Chough (Pyrrhocorax graculus).

The idea that the Oilbird belongs somewhere within the Caprimulgiformes is nowadays generally recognized, and this has been supported by studies using egg-white protein. Most systematists have tended to place it next to the owls, at the beginning of the Caprimulgiformes, and it is usually believed to be an early offshoot of the order. Its closest affinities may be with the potoos and nightjars, although neither of these appears to be a particularly near relative, so the Oilbird is almost always isolated in its own family, Steatornithidae.

The fossil history of ancestral steatornithids suggests that they may have been widely distributed in a very early radiation of birds over much of the Northern Hemisphere, perhaps as early as the Paleocene. Two nearly complete skeletons from the early Eocene, about 50 million years ago, have been recovered in the Fossil Butte of the Green River Formation in Wyoming, USA. Although these fossils are smaller than the extant Oilbird, they conform closely with it in having an unusually short and stout humerus, a shortened tarsometatarsus with a weak foot and claws, a large hooked bill with a very wide gape, and a large head. These fossils, belonging to a species that has been named Prefica nivea, share some features with early rollers; also, they have a more primitive sternum and pelvis than does Steatornis and less specialized structure of the humerus, toes and tail. This suggests that the fossil form was not adapted either for life in caves or for hovering flight, which are both distinctive forms of behaviour of the present-day Oilbird.

Most interesting is that some of the same families and even genera of angiosperm plants in the contemporary Oilbird’s diet were found as fossils in the same subtropical Green River Formation. It has been postulated that the early steatornithid Prefica nivea may similarly have eaten the fruit and then disseminated the large seeds of these same trees (see Food and Feeding).

Another fossil related to the Oilbird was found in south-western France in the Phosphorites du Quercy, dating from the Upper Eocene to the Lower Oligocene. The French and Wyoming fossil sites of early steatornithids are remote from the present entirely Neotropical distribution, which suggests that the extant Oilbird is a relict species.

Even though it is thought that the family may have diverged from other caprimulgiforms very long ago, its relationship to them remains unequivocal. Nearly all caprimulgiforms have very short tarsi, triangular hooked bills with long rictal bristles, and wide to very wide gapes. All are also nocturnal. The Oilbird differs mainly by its exclusively frugivorous diet, since all other caprimulgiforms studied to date are essentially insectivorous.

A reasonable hypothesis may explain the fossil distribution in Wyoming and France, compared with the family’s present distribution in north-west South America, stretching north to Panama and east to Trinidad and Guyana. North American climatic changes in the Pleistocene could have forced ancestral steatornithids to migrate south following the retreat of their food trees, and on into South America after the Central American landbridge was formed. There they would have lived during the fluctuating climate of the late glacial periods. However, sequences of more favourable (wetter) and less favourable (drier) climatic conditions that prevailed in South America at that time affected the forest habitat, including the fruiting trees on which the Oilbird is totally dependent. Possibly during times of less favourable conditions some population segments became isolated. This would have reduced normal gene flow, and could have created the genetic bottleneck which researchers have identified in the present-day Oilbird. Alternatively, the homogeneity of genes could be a result of the Oilbird’s capacity for far-ranging dispersal (see Movements).

Morphological Aspects
The Oilbird is a large chestnut-brown bird of 40-49 cm, weighing 350-485 g, and with a wingspan of more than a metre. The female is slightly smaller then the male. The male is a greyer, slightly darker brown, whereas the female is paler and more rufous. Some birds have a greyish flush or sheen on the upperparts, while all are paler brown below. The underparts are speckled with white diamond-shaped spots, finely edged black, these spots being smaller on the throat and becoming larger toward the vent; in addition, there are small diamond-shaped spots on the head and, in some birds, on the uppertail-coverts. The Oilbird has ten primaries, twelve secondaries and ten rectrices; these feathers are also chestnut, but some are darker brown and, to a varying degree, most are banded with blackish lines broadly edged with a black speckled area.

Most outstanding are rows of large white spots which occur on the outer two primaries, the outer two secondaries, and the outer two tail feathers on each side. The carpal edge of the wing is white, and a highly conspicuous line of large white spots is found on the upperwing-coverts. All of the spots are further accented with blackish edging. In faint nocturnal light these white spots would give each bird highly visible warning lines, especially on the long wings. The Oilbird’s wing bones are reported to be quite thin for the size of the bird, so the business of avoiding collisions with other conspecifics could be particularly important. Visually outstanding warning lines might be all the more advantageous to a large bird that forages silently in close groups at clumped fruit in trees at night (see Food and Feeding). Other social nocturnal-flying birds, such as migrating warblers, do not need to fly so close together and generally fly roughly parallel to each other.

The Oilbird’s feathers are soft, as in most night-foraging birds, but not so silky and soft-edged as in either the Strigidae or Caprimulgidae, families of birds that depend on silent flight to catch living prey. Oilbirds can fly very slowly, hover, and turn and twist with agility, all of which contribute to their ability to manoeuvre within the narrow passages of caves. In flight at night in the open, the silhouette has been described as falcon-like, and birds have been reported sometimes to dive down like falcons with their wings half closed.

Distance flight speed has been reported to be 20-25 km/h, not very fast. The main importance of flight speed for the Oilbird is that it would limit how far a bird could fly in a single night, still making enough round-trips from fruiting trees to feed nestlings in caves.

Like vultures, Oilbirds have deep wingtip slotting to reduce stalling speed, and the wing is deeply cambered to give it lift at low speed. The birds have low wing-loading (body weight to wing area), and a low aspect-ratio (wingspan to wing width). These features are aerodynamic advantages for flying slowly, and for carrying heavy loads such as a stomachful of large-seeded fruits. The plumage structure and its particular modifications to allow slow hovering at fruiting trees have been suggested as pre-adaptions to ledge-nesting and ledge-roosting, where hovering is also important. Echolocation, for which the birds are famous, is used only in dark caves, and may be a more recent adaptation (see Voice).

The tail is long and graduated, with unstiffened feathers. An unusual tail arrangement, which involves its being held in an inverted V with the outer rectrices lower, appears to be another adaptation for hovering at fruiting trees, and also probably when approaching nests and roosting ledges. Oilbirds watched feeding in Peru were described as rhythmically elevating and depressing the tail, as some hummingbirds do when feeding: the downward movement of the inverted V-shaped tail generates lift, while the upward movement meets with relatively little air resistance.

The very short unscaled tarsus has a few bristle-like filoplumes. The legs are set far forward, and when the bird is perched on its nest or on an adjacent shelf in a cave, it appears to tip awkwardly forward. The anisodactyl foot has three toes pointing forwards with the hallux at a right angle to the middle toe; in the past, the Oilbird’s foot was erroneously reported to be pamprodactyl, like that of swifts (see page 370), with all four toes pointing forwards. The Oilbird’s toe arrangement and flexible tail do not permit it to perch scansorially like woodpeckers, or to cling firmly to vertical surfaces like swifts, although it will occasionally cling briefly to cave walls. It does, however, aid nestlings, that fall or get pushed over the edge of a nest, to claw their way back to safety using the feet, bill and wings. On nests and ledges, the birds can shuffle about, but a single foot seems too weak to support the bird when it scratches its head without the support of one or both wings against the substrate.

The bill, although wide and triangular at the base, is thin and hawk-like toward the tip. On each side of the bill are 10-12 long (3-5 cm) rictal bristles, or vibrissae, directed forward and probably aiding the birds tactilely in both fruit selection and nest attendance, all aspects of which are commonly performed in the dark. The gape is very wide and can accommodate fruits of more than 6 cm in diameter. The upper mandible has a single notch, presumably as an aid in grasping and pulling off large fruits from trees. The Oilbird has a double bronchial syrinx, as in both owls and nightjars.

As soon as an Oilbird flies out of its cave it uses its eyes because the sonar, or echolocation clicks, stop abruptly. In addition to its special adaptations to the dark, the Oilbird may have good night vision. The eyes are brown, not blue as reported by Humboldt, an error that has lingered in the literature, but when spotlighted they appear bright red, as can be seen in many flash photographic pictures. Both above and below the eyes are rows of stiff bristles.

The Oilbird has an enlarged and heavily innervated olfactory organ, with very thick mucous membranes, and a nasal chamber conformation particularly adapted to carrying an airstream to these membranes. This highly developed olfactory morphology is assumed to give it a keen sense of smell, a rare feature in the class Aves that is found, for example, in kiwis (Volume 1, page 104), some petrels (Volume 1, page 218), and cathartid vultures (Volume 2, page 27), all of which also have olfactory adaptations to particular environmental circumstances. Many of the fruits that the Oilbird depends on have a highly aromatic odour (see Food and Feeding). Also, the slightly musty odour of the birds themselves has been suggested as a possible aid in individual recognition. The Oilbird has a bare oil-gland that is relatively large in relation to the bird’s mass, as compared with those of both owls and nightjars.

Moult in the Oilbird has been little studied due to the difficulty in tracking individual birds, and the subsequent disturbance that handling causes to breeding birds. Nevertheless, D. Snow found that Oilbirds, like most birds, moult following breeding. Primary moult is exceedingly slow, and frequently lasts into the next breeding season; indeed, there is even some evidence that birds may begin a second primary moult before the first is completed. The moult is generally outwards from the innermost primary, but several birds were found moulting primaries from two different centres. Some moult takes place in every month of the year. 
Habitat
Caves used for breeding and roosting in are considered to be the main population-limiting resource. However, it is also becoming increasingly clear that the forests used by the birds for their foraging may prove to be the more seriously threatened asset.

The Oilbird is found only in the Neotropics, from sea-level up to 3400 metres. In Trinidad and Venezuela a few sea caves have roosting and breeding populations, but these caves are not far from rich hill forests, the source of food. The combination of acceptable fruit trees and suitable caves within nightly commuting distance is not common. Apparently lacking enough truly deep, dark caves in some places, Oilbirds will roost and breed in narrow, semi-dark gorges and shallow grottos that have suitable rock shelves. One such well-known site is at Spring Hill near Arima, Trinidad, which is a water-worn tunnel with ledges only 3-4 m above a small stream. Similar places have been found at remote Venezuelan, Colombian and Ecuadorian sites with small Oilbird populations. An early suggestion that Oilbirds were associated with water is probably because most caves are found in karst, where water has hollowed out the rock, and usually there is some, or much, water still flowing through these caves.

Oilbird caves have been found in Colombia, Ecuador, Peru and Bolivia, as well as Venezuela, Trinidad and Guyana. However, individual Oilbirds have also been recorded from Brazil, Panama and the islands of Aruba and Tobago, and even as far north as central Costa Rica. Two things may account for this: first, Oilbirds fly far in a single night in search of fruiting trees; and, second, the national borders between Panama and Colombia, and those separating most South American countries are poorly explored, often rough mountainous areas, which may still harbour unknown caves.

In the most extensive caves, with large populations of thousands of birds, such as the cave at Caripe, Venezuela, the floor is covered with mounds 1-3 m high of seeds regurgitated by generations of Oilbirds. Many of these seeds germinate in the moist atmosphere, are blanched pale yellow and white, and die due to the lack of sunlight. These cave floors are alive with fauna such as spiders, cockroaches, whip scorpions, and cave crickets. Many caves, especially large ones, are also shared with thousands of bats.

Within caves, Oilbirds nest on the highest suitable ledges, often in remote sections, and sometimes as far as 770 m from an exterior opening. Since nestling Oilbirds have been harvested for centuries by indigenous peoples throughout their range, it is probable that the birds have retreated over time to the least accessible nesting and roosting areas (see Relationship with Man). Some caves, generally smaller ones, are used only for roosting and not for breeding.

General Habits
Extreme gregariousness influences all aspects of Oilbird life. The birds breed and roost closely together in gatherings of up to many thousands in suitably large caves, but also in groups of less than ten in small or marginally appropriate caves. There is great synchrony in Oilbirds’ crepuscular departure from caves to nocturnal feeding grounds, and this is no doubt aided by vocalizations and perhaps by their unique echolocation clicks. At dusk there is often an hour of restlessness inside the cave when birds fly about calling and clicking, before large numbers leave the cave. Pairs, and associations of adjacently nesting pairs, return at the same time to feed nestlings several times nightly. Most other nocturnal birds feed independently on mammals and insects, but Oilbird associations are thought to be advantageous in finding and tracking ripe fruit of their preferred trees that are scattered over vast mountainous areas.

Oilbirds are strictly nocturnal, and spend the daylight hours hidden away in the depths of their dark caves. Due to the darkness, and also to the inaccessibility of most nesting and roosting sites, the birds’ behaviour at the roost is rather poorly known, and these difficulties are accentuated by the fact that any human intrusion to one of the caves is almost certain to disturb the birds. As if these difficulties were not sufficient, the caves are often unhealthy sites for humans to spend much time in (see Relationship with Man).

There is an interesting case of parallel or convergent evolution between Oilbirds and some Aerodramus swiftlets, such as the famous edible-nest species. The two groups share no recent ancestry, and the swiftlets are confined to the Old World, occurring mainly in the Oriental Region. Both the Oilbird and the Edible-nest Swiftlet (Aerodramus fuciphagus), for example, nest and roost on ledges in caves, where they use echolocation for navigation (see page 378). Both species have other loud calls and screams, perhaps because there is natural loud noise in these environments such as running water or breaking ocean waves. In addition, Oilbirds and swiftlets hold their nests together with saliva, have protracted incubation and fledging periods (90-125 days for Oilbirds, up to 75-90 days for some swiftlets). Finally, the birds are highly social, and both have suffered long from persecution by man. Needless to say, much of this convergence may be ultimately attributable to their occupation of dark caves.

Voice
Any disturbance in an Oilbird cave causes a cacophony of nearly deafening shrieks, demonic screams and snarls, amplified by reverberations inside the cave, as the birds circle overhead. Humboldt called this noise "ear-splitting". These loud and intimidating vocalizations, representing a form of mobbing, are described by others as loud, harsh calls with a guttural quality and raucous screeches. A fast series of staccato sounds of many different frequencies and long sustained calls may make it easier for other birds to detect the position of calling individuals in total darkness.

The famous clicks of echolocation are used when Oilbirds fly in pitch darkness, to avoid collisions with other birds and with the walls of the cave itself. Echolocation clicks are actually sonar, because the bird judges from the echo how close it is to an object. The clicks sound like castanets, and are of a low frequency of about 7000 cycles/second, thus being audible to the human ear. Bats use echolocation to find flying food, but it is of much higher frequencies not audible to humans.The duration of Oilbird clicks is about 1 millisecond, and they are given in short bursts of 2-6 clicks repeated an average of 2·6 milliseconds apart. Experiments using adult Oilbirds flying in a totally dark room showed that they easily avoided hitting the walls and ceiling using echolocation, or when the room was lit and they could use their eyes. But in the dark, with their ears plugged they could not avoid collisions until their ears were unplugged again. Other experiments have shown that an object 20 cm in diameter is about the smallest that an Oilbird can detect by echolocation.

Outside their caves at night Oilbirds make shorter, less harsh "karr, karr" or "kuk, kuk" vocalizations, thought to be contact calls between individuals. D. Snow watched Oilbirds feeding at night, when two or three birds circled around each other with low, clucking calls and a long, harsh "karrrr". He believed this group, and other evenly spaced trios of circling birds, might be involved in pair formation or courtship behaviour.

Chicks, while still in the egg, make high-pitched cheeps, and after hatching produce a variety of sounds, which become increasingly louder when they are begging. By the time they are well feathered, at about day 70, they utter harsh screams when alarmed, in the same way as the adults.

Food and Feeding
The Oilbird is an obligate frugivore. At night, when the birds emerge from caves they fly directly in search of fruiting trees. There, while hovering briefly, they tear fruits off the trees and swallow them whole. Nearly all of these fruits have single, usually large seeds. The muscular action of the Oilbird’s stomach then strips the thin outer flesh (pericarp) from the seed, which is regurgitated intact, except in the case of figs (Ficus), an atypical food whose small seeds can be found in faeces.

The food of Oilbirds has been studied in Trinidad, Venezuela and Ecuador, and a total of 147,500 seeds from more than 80 species of tree have been identified. For these studies, the seeds, mostly 1-6 cm in diameter, were typically collected in wire baskets fastened underneath breeding ledges, or in Ecuador, by sampling the first 30 cm of the 2·5 m piles below nest ledges. This has afforded unusually good opportunities to identify the fruits fed to nestlings and also eaten by adult birds over the breeding season and, in Trinidad, throughout the year.

In Trinidad, D. Snow identified more than 36 species of fruit, and he found that Oilbirds fed mainly on three families: Arecaceae 52%, Burseraceae 25%, and Lauraceae 23%. In Venezuela, with a somewhat different climate, 40 species of fruit were found in the Oilbird’s breeding-season diet, with proportions of Lauraceae 51%, Arecaceae 31%, and Burseraceae 7·6%. In undisturbed forests it is difficult to find flowering trees, which are often necessary for the identification of some difficult genera, so a few seeds in all studies remain unidentified.

Fruiting in many of these tropical trees is strongly seasonal, although in some species it varies from year to year, and at least one tree species used was found to be a biannual fruiting species. Generally laurel and incense trees produce abundant seasonal fruit crops, while the palms have smaller amounts of fruit but this tends to be available all year. Analysis of the most frequently selected fruits showed that the pericarp which is non-succulent is high in proteins and fat. Comparison of seeds dropped in different Trinidad caves showed that the birds in one cave apparently exploited traditional feeding areas different from those used by the birds of other caves. Furthermore, individual birds may have special preferences. Seed collections from below active nests revealed that more tree species had ripe fruit during the months of maximum nestling feeding.

Except for palm fruits, Snow found that nearly all of the fruits that Oilbirds ate or fed to nestlings are spicy or aromatic when ripe, and a similar conclusion was also drawn by B. Tannenbaum and P. H. Wrege in their Venezuelan study. Interestingly, Snow found a species of odourless laurel fruit that was available at the same time as two other frequently eaten laurels. Analysis showed it was almost equally nutritious, but it was virtually absent from his seed collections. Thus, there is a strong case for the hypothesis that the birds find much ripe fruit by odour, which would tie in with the particularly well-developed olfactory organs (see Morphological Aspects).

The birds forage on canopy fruits from trees and palms that are 5-20 m high, though rarely they will also feed from the understorey. Groups of up to 20 or more birds circle or hover silently near a fruiting tree. Then with rapid, shallow wingbeats they fly into the foliage briefly, pluck and swallow fruits, and fly a short distance away, often returning subsequently to a different spot on the same tree. E. de Bellard noted that feeding Oilbirds settled in trees when clouds covered a bright moon, but began feeding again when the clouds had passed. Snow suggested that Oilbirds may also locate fruiting palms by sight because of the distinctive palm shape; their fruits are without noticeable odour.

At the large Caripe cave in Venezuela, R. L. Roca used radio-telemetry to follow ten adults, including birds of both sexes. Individual marked birds left the cave at the same time each night, along with hundreds of others. Although the colony as a whole appeared to forage at random each night close to the cave, single birds constantly chose to forage in the same directions and repeatedly in the same patches. The Oilbirds exploited discrete areas of primary forest and avoided disturbed forest. Most foraging was within a radius of 40 km of the cave but some birds flew as far as 150 km away in a single night.

Over time, one radio-tagged bird used a range of 85 km², while another used 96 km². One male was apparently a floater which visited the colony briefly several times then was absent from the radio signal area for periods of several weeks to two months. During the breeding season, April-July, the members of the Caripe colony together used about 300 km², but later their home range expanded to 1350 km², as fruit availability decreased. This wider area included the large area of primary forest at Mata de Mango to the east, and still further away the extensive floodplain of the Orinoco Delta (see Status and Conservation).

Although the Trinidad birds used their caves all year, roosting in them after breeding, there was a marked post-breeding exodus from the Caripe cave. This same behaviour was postulated by B. K. Snow for the caves she investigated in Ecuador, where numbers of birds in the post-breeding period were dramatically lower than earlier in the season. These differences suggest that ripe fruit near some caves was not sufficient to support the birds following breeding, especially when fledglings increased the size of the local population.

All three of the principal Oilbird food families, and even two genera of laurels, plus three other families of fruit eaten occasionally by the extant Oilbird species, were found at the Green River Wyoming fossil site. S. L. Olson suggested that because early steatornithids "would have been effective disseminators of large angiosperm seeds, rapid co-evolution of both seed plants and their avian dispersal agents at the beginning of the Tertiary is likely".

Breeding
Everywhere that the Oilbird has been studied, it has been found to initiate breeding at the end of the dry season, although there is some variability from year to year probably due to differences in the timing and volume of rains. A typical breeding effort takes from four to more than five months, a considerable time, and this does not include post-fledging care, about which almost nothing is known.

Oilbirds appear to be monogamous on a long-term basis. Partners regularly roost together, and during breeding the off-duty bird typically perches close beside the nest, while courtship allopreening at the time of egg-laying may strengthen the pair-bond. All nesting responsibilities are shared, including nest building, incubation, brooding and the feeding of nestlings. This apparent long-term monogamy may be linked with a shortage of suitable nesting and roosting sites that appears to be the norm in most caves. Certainly some pairs repeatedly use the same nest over several years, and one ringed adult always returned to the same nest for at least 12 years.

Nests are situated on cave ledges and shelves close to the cave ceilings, sometimes 20-30 m high, though rarely as low as 4 m above the ground, and they are occasionally very deep inside caves. As pairs re-use the same nests each year, only a minor amount of reconditioning of the top nesting cup is required. These traditional nests are hard, truncated cones 20-45 cm high and 25-38 cm wide at the top. They are made of concentric layers of regurgitated seeds, some excreta, and fruit pulp, all held together with the birds’ saliva. Because the same nests are used again and again they gradually increase in height. At the top, a thin vertical rim of 2-3 cm surrounds the shallow nesting cup. In one case, F. Ortiz-Crespo described six nests in a very shallow cavity, set in a circle with the bases touching.

Some eggs are laid as early as March and as late as July in Trinidad, with the main period of egg-laying falling in April and early May; there, the clutch is of 2-4 eggs. In contrast, the main egg-laying months in Venezuela are May-June, and the clutch consists of 1-3 eggs. Eggs tend to be somewhat spherical, like some owl eggs. Ten eggs averaged 41·7 x 32·7 mm and the mean weight of ten Trinidad eggs was 20·2 g (17-22·5). Although the eggs are white they quickly become stained in the nest.

Egg-laying intervals are variable but long, at 2-9 days, and incubation begins with the laying of the first egg, so hatching is highly asynchronous, each egg hatching after 32-35 days. Some females incubate slightly more than males. As mentioned above, the non-incubating bird, when not foraging at night, remains perched alertly beside or even on the nest. Experiments showed that Oilbirds have a poor sense of egg retrieval if an egg is accidentally moved.

There is great synchrony of egg-laying within subsets of each colony, which is thought to facilitate the social foraging of off-duty birds. Pairs of birds with young return together to feed them, and at the same time neighbouring nesters also return to feed their young. During the 12-hour night very small young are fed 5-6 times, but larger young only 3-4 times.

As the diet of nestlings is entirely frugivorous, this results in slow and protracted development. Hatchlings weigh 12-16·5 g and, except for some sparse pale grey down on the underparts, they are entirely naked. At about three weeks old, they are covered in dense dark down. Also at about this time, nestlings begin to thermoregulate and are no longer brooded, and they also begin self-preening. A second generation of darker grey down pushes out the first down feathers, except for some of those on the underparts. Adult-style feathers begin erupting from sheaths on the tail, secondary coverts and scapulars by day 35, and at around ten weeks old the nestlings look similar to the adults, though they are about 50% heavier than the average adult weight of about 410 g. There is no juvenile plumage and nestlings never develop other juvenile visual signs even on the head or the bill, probably because they are raised in darkness.

Nestlings are fed entirely by regurgitation. Young birds squeal and call when an adult returns to the nest. Then the parent leans down with open bill and the chick places its bill inside. The two birds are often linked in violent struggles while fruit is transferred. Nestlings as young as 12 days old are fed whole undigested fruits and are able to regurgitate the seeds. Apart from these feeding bouts, the young are very lethargic in the nest.

Once a nestling is 1-2 months old, it receives nightly feedings equal to a quarter or a third of its body weight. Most fruit-eating birds eat large volumes, and the fruit is generally high in water and passes through the bird quickly, but the fruit fed to nestling Oilbirds is processed very slowly: plastic markers that researchers fed to nestlings averaged more than six hours in the birds’ guts.

There is great variability in the development of nestlings of the same age, but at about 70-80 days they begin to lose their excess fat, while their plumage matures. At this same time the feeding rate decreases. Whether this slowing in the feeding rate is deliberate on the part of the adults, or a result of seasonal decrease in available fruits near the caves is not known; it may well turn out to vary from place to place.

The nestlings are brooded constantly for their first 25 days or so, at which point they are able to thermoregulate. D. Snow suggested that the extraordinary nestling fat aids the birds in thermoregulation when they are no longer brooded in dark and often wet caves, although the fact that the fat is concentrated on the chicks’ ventral surface is perhaps something of an obstacle to this view.

Because of the asynchronous hatching, the growth of younger nestlings always lags behind that of their older siblings, and Tannenbaum and Wrege observed heavy brood reduction at the Caripe cave in a year of presumed fruit shortage. Oilbirds typically fledge at 100-115 (88-125) days old, after their weight has dropped close to the average adult weight and their wing feathers have attained adult length. A few ringed fledglings returned to their nests by day over a short time, but most were not found again in their natal caves.

Nearly 60% of the young that hatched in the Trinidad study reached the flying stage, so an average of a little more than 2 young were reared annually by each pair; at the Caripe cave 1·7 of the hatched young fledged. Some losses occur when eggs or young fall out of nests. In Trinidad, the crab Pseudothelphusia garmani was thought to be the cause of the death of some young nestlings. Venezuelan studies found that the spiny rat Proechimys urichi stole eggs and ate young, particularly those that fell to the cave floor. There, the highest, most inaccessible nest ledges suffered fewer losses. Predation by man at the time of maximum chick weight, for oil, appears to be the most widespread cause of nesting failure (see Relationship with Man).

The Trinidad study found that the birds renested following early loss, and that in a few cases second nests were made by early successful pairs. A few replacement nests were reported in Venezuela, but no second nestings, which perhaps corresponds with the smaller clutch there. It has been suggested that in Venezuela, and perhaps also in the Ecuador study area, the available ripe fruit within energetically economic flight range of breeding caves may limit the breeding season at those sites.

Movements
Although radio-tracking in Venezuela was used to monitor nocturnal behaviour of Oilbirds for seven months, it can not be used to track individual birds over longer time. Apparently, fledglings disperse from their caves of origin, and they are not philopatric in either Trinidad or Venezuela. Adults in Trinidad, however, continue to roost on their nests all year, where one ringed individual there was found to be faithful to the same nest for over 12 years. Nevertheless, at the Caripe cave in Venezuela there is usually a strong post-breeding departure to the Mata de Mango forest area east of the cave. In that area there are numerous smaller caves that are suitable for roosting, but apparently few that are adequate for breeding, and after the breeding season the forest there has more fruit (see Status and Conservation). The difference between these two tendencies may be that in Trinidad, at the time of the studies in the 1960’s, there was a shortage of nesting sites in caves, and no feeding area was too far distant.

The massive post-breeding departure of an estimated 10,000 birds from the Caripe cave, leaving only a few hundred in the post-breeding months, may be a somewhat irregular occurrence. C. Bosque has suggested that such behaviour may be due to the loss of palms locally, as these should form one of the principal sources of food at that time of year. In other Venezuelan caves with breeding populations of over 1000 birds, however, the numbers roosting there in the post-breeding months falls to as low as only two birds. Similar behaviour has been suggested for caves in both Colombia and Ecuador, so post-breeding dispersal or nomadism may be of widespread occurrence in the Oilbird.

The home range of breeding Oilbirds has been estimated to be 85-95 km², with feeding localities as much as 150 km away. The remains of an Oilbird, thought possibly to have been an immature, and presumably killed by collision with a utility wire at 3100 m in Costa Rica, suggests that the species may be resident locally. However, the nearest known Oilbird cave is some 700 km away, by direct flight, in Colombia, as no Costa Rican or Panamanian Oilbird caves have been reported so far. The distance that Oilbirds disperse from known roosting caves remains mostly a matter of conjecture at the present time, and it is worth noting that one marked Oilbird in Venezuela flew 240 km in a single night to another cave.

Relationship with Man
In 1799 Humboldt and his companion, the botanist Amie Bonpland, were led by missionaries to the famous cave near the village of Caripe, Venezuela, to view the local ornithological sight: the impressive emergence at dusk of thousands of Oilbirds from this large cave. It was Humboldt who coined the name Steatornis from the Greek stear meaning "fat" or "tallow", and the English translation, Oilbird, has been used ever since. The village of Caripe is just six kilometres from the cave, and the source of the species name caripensis.

Far earlier, in 1678, a Capuchin monk, Fray Francisco de Tauste, described, in missionary documents, the dense clouds of birds that came out of the Caripe cave at night. He described the tasty oil that was made from squabs, and reported that the local Indians believed the souls of their dead went into this same cave. Oilbirds were then, and still are, called Guácharos in most of their native lands, a word said to some from archaic Spanish or possibly even Catalan meaning "crying, or making a loud lament". Throughout the species’ range, fat nestling Oilbirds are, or have been, collected by indigenous inhabitants for their famous oil. The young are cooked to extract the oil which is reported to be odourless and clear, and to keep well. This oil is used mainly in cooking, but also in some places for lamp oil. There are reports of the use of adult and nestling Oilbirds for their meat, said to be palatable, even though the birds have a somewhat musky odour. Venezuelans have also used the stomach contents of Oilbirds in folk medicine.

Visitors to an Oilbird cave should take suitable precautions against histoplasmosis. This disease is caused by the airborne fungus Histoplasma capsulatum that, although widespread throughout the world, can cause in humans a reaction from mild to occasionally fatal, and can permanently damage the lungs or eyes of some individuals. The fungus, not known from the Oilbirds themselves, lives in highly nitrogen-enriched soils below large bird and bat roosts. When the dry soil is stirred by human activity or winds, the fungus is easily inhaled. Recent Oilbird researchers in Venezuela have all worn surgical masks at all times, as a precaution against this disease because the fungus is known to be especially widespread in Venezuelan caves; it is generally believed that wetter caves are less dangerous.

Status and Conservation
There are no figures on the world population of Oilbirds. The birds are easy enough to count at dusk when they are leaving small caves, but the large numbers that fly out together from such caves as Caripe, estimated by some researchers to be some 10,000 but by others only 5000, can not be counted accurately. Furthermore, counts are complicated by the recently discovered dispersal or nomadic behaviour of some South American populations.

Historically, suitable breeding and roosting caves have been the limiting resource for all Oilbird populations. Recent cave explorations in Venezuela have raised the total to as many as 54 caves used, or recently used, by Oilbirds. The discovery, in the mid-1980’s, of several large Oilbird caves in the Venezuelan state of Bolívar on the border of Brazil may have doubled the estimated number of Oilbirds in Venezuela. In this mountainous tepuí area one cave, Aguapira, was reported to support as many as 10,000 Oilbirds. The combination of caves located in difficult forested terrain, and often on the poorly explored borders of South American countries, suggests that there may be more unknown populations of Oilbirds awaiting discovery.

While in most countries the Oilbird is protected by law, active nestling collecting was still being practised in Peru in the 1960’s, and in Trinidad, Venezuela and Ecuador in the 1970’s; in some remote areas it probably continues up to the present. Over time Oilbirds probably reacted by nesting on the highest and least accessible ledges and shelves in their caves. Even so, in Venezuela seven of 54 colonies, all in small caves, have been deserted, though this is apparently due to loss of foraging habitat close to large human populations; similarly, five of 13 caves in Trinidad have been abandoned by Oilbirds in recent years.

Humboldt recognized the problem of conservation when he learned of the massive raids on nestlings in the outer chambers of the Caripe cave. But he concluded that the annual harvest of thousands of nestlings was apparently sustainable due to reinforcements by birds breeding where the people could not reach the nestlings. There were, he said, Oilbird caves in the area with entrances too restricted for humans to enter, and also he heard Oilbirds in the Caripe cave far beyond the point where the native Indians refused to pass. In those days even the missionaries could not make the Indians go deep into the cave. The Indians believed the birds were ghosts of their ancestors, and to go far into the cave would be to enter the land of the dead. B. Snow, in her investigation of Oilbirds in the Los Tayos caves of Ecuador in 1976, again found human predation on nestlings, but she also believed that it too was sustainable because many of the nests were inaccessible.

In spite of the known and well-documented records of Oilbird predation by humans, it is the relentless attack on subtropical forests by the ever-increasing number of subsistence farmers in all range countries that pose the greatest long-term danger, as they threaten traditional Oilbird food resources. Tannenbaum and Wrege point out one fortuitous situation: many of the Lauraceae trees traditionally saved to shade coffee plantations are also those that are important in the Oilbird’s diet. On the other hand, many of the palms that Oilbirds depend on during non-breeding months are selectively sought out and destroyed by man. An important Euterpe species is the source of the well-known edible Heart-of-palm, which is commercially canned and exported from Venezuela. Much of this product is gathered from palm trees in unregulated forests and the extraction of the heart kills the palm. Thus man’s interests are in direct competition with the Oilbird’s requirements. Few efforts are being made to cultivate the Euterpe palm for its heart, because it takes so long to grow to harvestable maturity.

Oilbirds, however, sow their own seeds. Roca’s study of cast seeds in the Caripe cave suggested that most of the seeds covering the cave floor were regurgitated by nestlings. No seeds were found in collecting trays under nests that were being incubated, so adult birds were regurgitating the seeds, as many as 50 per bird per day, outside the cave before they returned. These he calculated to amount to about 21 t of seeds a month for this colony.

This fact formed the key to a major conservation action taken recently in Venezuela. Radio-telemetry had shown that through the year the birds of the Caripe cave spent much of their time foraging outside the protected area. Following the recommendations of Tannenbaum and Wrege, and also of Roca, the government in 1989 set aside an additional 66,400 ha area, the Mata de Mango, 15 km east of the already protected 15,500 ha Caripe cave area for the Oilbirds.Two other significant factors were involved: the Caripe cave and its Oilbirds are very well known throughout Venezuela, and a large tourist attraction; furthermore, by setting aside the new area as a National Park, it also protected the entire catchment area of the small Río San Juan which is currently important to both local agriculture and nearby commercial maritime petroleum transportation.

In Bolivia, Peru, Trinidad, and Colombia governments have set aside areas near and surrounding Oilbird caves, or are planning to do so. Protecting just the breeding or roosting caves will not succeed in saving the birds, because sizeable tracts of primary forest must also be included. In order for these plans to succeed, specific efforts will have to be made to overcome the usual problems of lack of money for personnel to enforce the national laws, and the lack of official will to oppose the small agricultural enterprises that constantly and permanently damage subtropical forests.