HBW 9 - Family text: Pipridae (Manakins)

Family text: 

Class AVES
Suborder TYRANNI

  • Small arboreal birds of compact build, with short, rather wide bill, short wings and, except for males of a few species, short tail; feed mainly on small fruits, plucked in flight, and insects.
  • 7-16·5 cm.
  • Central and South America, and some offlying islands.
  • Mainly tropical forest, a few species in scrubby woodland and thickets.
  • 18 genera, 58 species, 146 taxa.
  • 4 species threatened; none extinct since 1600.

The classification of the Pipridae has long been controversial, and has still not been finally resolved. The long-standing and continuing difficulty has been that the "typical" manakins, consisting of 13 genera, form a well-defined group, while several other more or less manakin-like genera have characters that link them, variously, to the cotingas (Cotingidae) or the tyrant-flycatchers (Tyrannidae). The latter families are the two other main groups which, with the manakins, comprise the "tyrannoids", formerly combined in the superfamily Tyrannoidea. Although this taxon is now rarely employed, the three families are usually still kept together in a single suborder, the Tyranni, a useful grouping in this context.

P. L. Sclater, in 1888, separated the manakins from the two other related families on the basis of their "syndactylous" foot structure, in which the second phalanx of the outer toe is united with that of the middle toe, and on details of tarsal scutellation. This treatment was followed by R. Ridgway, in 1907. Later, however, it became apparent that too rigid an application of these criteria led to the probable misplacement of some genera. The syndactyly, for example, is shared with the cotingid genus Rupicola, and may originally have been an adaptation for perching on vertical stems. More recently, R. O. Prum has made a thorough reappraisal of manakin classification, based on extensive research into the structure of the syrinx and other anatomical, as well as behavioural, characters of the manakins and other tyrannoid genera. While it is clear that a number of changes very definitely need to be made to the traditional classification, there are several suggested ones for which the evidence is still somewhat uncertain, or even conflicting. Furthermore, it is likely that, within the near future, further research will throw new light on these and other controversial points. The classification currently adopted is therefore conservative, but incorporates the following generic changes, involving Pipra, the need for which seems incontestable.

Until recently, the genus Pipra was much the largest in the family, containing as many as 17 species. Such an arrangement has now been recognized as rendering the genus polyphyletic, and it has consequently been divided into three: Pipra and Lepidothrix, each with eight species, and Dixiphia, with a single species. In its syringeal anatomy Lepidothrix is much closer to the other manakin genera than to Pipra, from which it differs considerably in behaviour; its allocation to Pipra was based on the superficial similarity of the male plumage to that of others in the genus. Dixiphia is closer to Pipra but is not its nearest relative, and it, too, differs greatly in its behaviour. It is worth noting that Teleonema, a monotypic genus erected for the Wire-tailed Manakin (Pipra filicauda), has for some time been merged with Pipra, as this species, notwithstanding its extraordinary wire-like tail-feather projections (see Morphological Aspects), is a very close relative and allospecies of two typical Pipra species, the Crimson-hooded (Pipra aureola) and Band-tailed Manakins (Pipra fasciicauda).

The five genera placed at the end of the family are all taxonomically controversial. They have traditionally been treated as piprid, following the criteria used by Sclater and Ridgway, but almost certainly they are only distantly related to the typical manakins. Since there is no overwhelming evidence for placing them elsewhere, however, they are retained here. Schiffornis is especially puzzling, as an analysis by Prum of 19 characters, mostly anatomical, suggested that it is a member of a monophyletic assemblage of six superficially very diverse genera currently divided among the Pipridae, the Cotingidae and the Tyrannidae. Furthermore, recent investigations by U. S. Johansson and colleagues, who analysed sequence data from two nuclear genes and one mitochondrial gene, resulted in Schiffornis grouping with Pachyramphus and Tityra in a tityrine clade, whereas Pipra, Manacus and Chiroxiphia were found to be monophyletic.

The genera Piprites, Tyranneutes and Neopelma also share with the cotingids and tyrannids some apparently derived character states, in other words, characters which have evolved later, rather than being ancestral to the tyrannoid stock. Tyranneutes and Neopelma may be closely related to each other. The monotypic genus Sapayoa contains the little-known Broad-billed Sapayoa (Sapayoa aenigma). It seems to be close to the tyrannids, and this likelihood has been strengthened by the recent discovery of its nest, which is pendent and pear-shaped, quite unlike a typical manakin's nest, and contained nestlings which were fed by both parents, indicating a social organization unlike that of a typical manakin.

At the species level, several taxonomic uncertainties remain. Further research may well indicate that, of the 58 piprid species currently recognized by most authorities, six should be split. These are the Blue-crowned (Lepidothrix coronata) and Blue-rumped Manakins (Lepidothrix isidorei), the White-crowned Manakin (Dixiphia pipra), the White-ruffed Manakin (Corapipo leucorrhoa), the Green Manakin (Chloropipo holochlora) and the Thrush-like Mourner (Schiffornis turdina). In most of these cases, the issue in question is that of whether lowland populations east and west of the Andes, presently treated as different subspecies, have diverged sufficiently to warrant the status of full species. Critical studies of voice, ecology and other aspects of their biology have not yet been undertaken, and any decision must necessarily be arbitrary. This is, of course, not a peculiarity of the manakin family; throughout the world, the same problem affects numerous closely related life forms with disjunct populations that have diverged to varying degrees.

A different and less usual situation complicates the taxonomy of the species of Manacus. Three of the four species in this genus come into contact with one another in Panama and Colombia, where they have produced narrow zones of hybridization. They could perhaps be considered to constitute a single species, as they differ mainly in the colour of the adult male plumage and are extremely similar in their courtship displays (see Breeding), as well as in their voice and ecology. On the other hand, the narrowness of the hybrid zones, and evidence that, at a lek within one such zone, males of the two hybridizing populations kept to separate ends, with hybrid males in the middle, indicate a degree of reproductive isolation warranting their treatment as separate species. This case is further complicated by recent evidence, based on protein analysis, which suggests that the yellow throat colour of the male Golden-collared Manakin (Manacus vitellinus) has spread into adjacent white-throated populations in north-west South America, and that white-throated White-bearded Manakin (Manacus manacus) populations west of the Andes are more closely related to the Golden-collared Manakin than they are to populations of the White-bearded Manakin living east of the Andes.

Morphological Aspects
Manakins are small birds, mostly in the weight range 8-30 g. They are arboreal, typically possessing a stout body and short tail, and a short bill which is broad at the gape and without rictal bristles. The outer toe is partly united with the middle toe.

Most of the typical manakins (see Systematics) exhibit a high degree of sexual dimorphism. Adult males of the majority of the species are largely black, with patches of brilliant colour, either pigment-based red, orange or yellow, or structural blue, opalescent or white, mainly on the crown or other parts of the upperside. Females and young birds are olive-green or green above and paler below, and in some cases they are so similar to other, co-existing species that they present problems of identification in the field. Adult males not only have patches of brilliant colour but, in many instances, also have highly modified wing or tail feathers, most or, probably, all of which are used in courtship display (see Breeding).

In the genus Manacus, the four outer primaries are very narrow, stiff and slightly curved, the outer webs being extremely narrow. Whenever the bird flies, and not just during display, these produce a grasshopper-like whirring sound. The secondaries are more highly modified, their shafts being unusually thick and the outer webs very stiff. As described later (see Voice, Breeding), these are involved in the production of the loud snaps that accompany courtship display. In several other species, in the genera Pipra and Ilicura, the secondaries are modified to a much lesser degree, the shafts being thicker and the vanes somewhat stiffer than in the unmodified condition. In these species, snapping sounds, less loud than those made by Manacus species, are a feature of the courtship displays, but the mechanism of sound production is quite different from that of Manacus (see Voice). Two of the eight Pipra species, the Golden-headed (Pipra erythrocephala) and Red-headed Manakins (Pipra rubrocapilla), lack the modification and make no snaps when displaying, although other elements of their display are quite similar to those of their most closely related congeners.

In males of Machaeropterus, the secondaries are extraordinarily modified in one species, the Club-winged Manakin (Machaeropterus deliciosus), and similarly so but to a far lesser extent in the other three species. The Club-winged Manakin's secondaries become progressively more curved from the outer end of the row inwards, the shafts becoming thickened distally, culminating at S6 and S7, which have the end of the shaft modified into a hollow, curved and pointed club. In display, male Club-winged Manakins raise the wings vertically above the back and vibrate their modified secondaries to produce a unique ringing sound.

Another extraordinary and unique modification, in this case of the tail feathers, is exhibited by the Wire-tailed Manakin, which for that reason was for long placed in a monotypic genus Teleonema. The shafts are prolonged into fine wire-like filaments curving inwards and downwards, those of the outermost feather pair being longest and the others progressively shorter. The filaments of the outer pair, if straightened, may be up to 50 mm in length, about half as long as the head and body combined. As described later (see Breeding), these tail filaments, which are rather inconspicuous, are used in display not for visual effect, but as tactile organs; they represent one of the very few such organs known in the avian world. It is of particular interest that two very closely related parapatric species, the Crimson-hooded Manakin and the Band-tailed Manakin, have unmodified tail feathers. The group may have differentiated in the Quaternary period, at a time when the Amazonian forest was reduced to a number of isolated refuges; if so, this is a striking example of the rapid evolution of a secondary sexual character.

Male manakins in the genus Chiroxiphia have the three outermost primaries pointed, the shafts being thickened and the barbs reduced. It is not clear in what way, if any, these feathers are used in display, but they may be aerodynamically important in the fluttering jump and hover which is the most significant element in the uniquely complex courtship display of this genus. In three of the five species the central tail feathers are elongated to different extents, most extremely so in the Long-tailed Manakin (Chiroxiphia linearis). This, too, appears to be a case of rapid evolution, as four of the species, at least, are certainly very closely related. It is curious that only in the Long-tailed Manakin does the elongation seem great enough to enhance the visual effect of the courtship displays, which are very similar among all five species.

Turning to other aspects related to morphology, the relative sizes of the sexes, based on body mass, or weight, vary according to the genus involved. In most genera, males are slightly heavier than females, except during the latter's egg-laying period. Conversely, in Pipra, Lepidothrix, Dixiphia, Corapipo and, probably, Masius, it is the females that are the heavier, most markedly so in the case of the White-crowned Manakin, of which a large sample of both sexes in French Guiana showed the average weight of males to be 86% of that of females. In these five genera, displaying males execute rapid and complex flight manoeuvres. Owing to the modifications of the flight-feathers of the adult males of some piprid genera, the standard wing measurement is not a reliable measure of overall size.

In common with most other passerines, adult manakins undergo a complete moult immediately after breeding. Prior to that, the young birds first acquire a female-like juvenile plumage. Their subsequent plumage changes are not well known, and only those of the White-bearded Manakin and the Golden-headed Manakin have been studied in some detail. In these two cases, the juvenile plumage is replaced after a few months in an apparently prolonged moult of the head, body and wing-coverts, resulting in a plumage still of female type in both sexes, except that young males often acquire a few new feathers of adult male colour. At the next moult, which is complete and begins when the birds are a little over a year old, males acquire fully adult plumage.

The time taken by young males to gain the full adult plumage, with the concomitant structural feather modifications, varies according to species. As mentioned above, young male White-bearded and Golden-headed Manakins moult into adult plumage at the beginning of their second year of life, the moult commencing a little earlier in the season than that of the breeding adults. At the other extreme, males of the Long-tailed Manakin pass through three distinct, age-specific subadult plumages and do not acquire fully adult plumage until their fourth year. This delayed plumage maturation is related to the complex dominance system obtaining at Chiroxiphia leks (see Breeding), each subadult plumage stage apparently serving as a "badge of status" within a group of males of different ages.

For many manakin species, details of plumage sequences are not well known. Indeed, only for a small number of species is significant information available.

The manakins are almost exclusively birds of tropical forest, living mainly in lowlands below about 1000 m. A few species are found at subtropical levels in the Andes, a notable example being the Golden-winged Manakin (Masius chrysopterus); none ascends as high as temperate levels, where several of their relatives, the cotingas, and many species of tyrant-flycatcher occur. Piprids live almost entirely in the lower and middle strata of the forest, rarely moving up into the high canopy.

Whereas some manakins, such as the Lepidothrix species, are rather strictly confined to humid primary forest or mature second-growth forest, others are tolerant of, or even prefer, more disturbed or more open woodland. Thus, the members of the genus Manacus prefer forest-edge habitats, quite recent second-growth woodland, and similar kinds of vegetation where there is an abundance of more or less closely spaced saplings with vertical stems, suitable for their leks (see Breeding). Many species, living mainly in the forest interior, come regularly to feed at fruiting trees at the forest edge. This is, in fact, much the easiest place for the casual observer to see them, as some of their most important sources of small fruits, especially small trees in the families Melastomataceae and Rubiaceae (see Food and Feeding), tend to be in greatest abundance along forest edges.

A general requirement for all typical manakins is the availability of a year-round supply of small fruits. Most humid forests in the Neotropics meet this requirement, and well-grown secondary forest, which, like forest edge, tends to be rich in suitable berry-bearing trees, can hold especially high populations of Pipra and Manacus species.

Within the forest, the display sites of the different manakin species, at and around which the males spend most of their time (see General Habits), are far from randomly distributed. Much of what is known about this aspect of habitat selection by manakins comes from studies carried out by M. Théry in French Guiana. Here, the leks of White-bearded Manakins and the display-logs of White-throated Manakins (Corapipo gutturalis) were in sites, such as east-facing slopes, which received sunlight at the two times of day when display activity is at its height, namely, the early morning and the afternoon. Similarly, the much higher display-perches of Golden-headed Manakins, 5-15 m above ground, were sited so as to be illuminated through gaps in the canopy to the east and west. The display sites of White-fronted Manakins (Lepidothrix serena), near the ground in the densest patches of forest, were nevertheless situated in places where there was most light, the males regularly alternating between a morning site and an afternoon site, each best illuminated at those times of day.

Recent studies being undertaken in southern Venezuela have corroborated these findings. Observations on four piprid species living in forest bordering the upper River Orinoco, in Amazonas, confirmed that the males, for their leks, selected positions along the vertical gradient where ambient light increased chromatic and/or achromatic contrast of their colour signals against the background or within their coloration patterns.

As mentioned above, the Manacus species prefer forest edge and secondary woodland to the interior of primary forest. Some other piprids, most notably those in the genus Chiroxiphia, occur, at least locally, in more open and drier types of woodland; the Lance-tailed Manakin (Chiroxiphia lanceolata), for instance, lives mainly in thick scrub and the lower growth of rather dry woodland. At the other extreme, the closely related Crimson-hooded, Band-tailed and Wire-tailed Manakins are found mainly in seasonally flooded várzea forest or in drier forest along watercourses. The three Heterocercus manakins also show a strong preference for seasonally flooded forest.

Two manakins occur in habitats which are atypical for the family. The Helmeted Manakin (Antilophia galeata) inhabits gallery woodland in the cerrado region of central Brazil, and the Black Manakin (Xenopipo atronitens) lives in patches of scrubby woodland and gallery forest in mainly unforested areas of the Amazon Basin. Both of these species are "true" manakins (see Systematics), and one of them, the Helmeted Manakin, is apparently quite closely related to Chiroxiphia. It is noteworthy, therefore, that they have not developed, or perhaps have lost, the elaborate courtship displays of their forest-living relatives and appear, instead, to have a more conventional social organization, although this aspect of their life is not well known.

General Habits
Typical manakins share a combination of habits that marks them out from other Neotropical forest birds. Almost all of their movements, facilitated by the small size of the species, are quick and abrupt. When foraging, they do not hop about among branches or foliage, but perch motionless and make sudden aerial sallies. When displaying, they perform very rapid, often complex flights. Their actions when perched consist either of rapid "about-faces", so sudden that the foot movements are hard to see, or apparent "slides" along the perch, either sideways or backwards, by means of very fast, short steps, again too rapid to be followed by the unaided eye. Adult males spend a great part of their lives in more or less elaborate advertising display at fixed sites, either singly or in groups, and females undertake single-handedly all nesting and young-rearing duties (see Breeding).

The members of this family are generally unsociable when foraging and feeding, but some piprids do at times accompany mixed-species foraging flocks. Not uncommonly, however, several individuals may be found together at a fruiting tree, independently attracted by the same food source.

Manakins bathe communally in forest streams in the late afternoon, mainly between 16:00 and 17:30 hours. In his detailed study in French Guiana, Théry found that four species, the Golden-headed, White-crowned, White-fronted and White-throated Manakins, regularly bathed together, whereas the White-bearded Manakin bathed separately. Individuals took turns to bathe, the others perching a few metres away. Evidently, it is important to keep a keen lookout, as predators such as forest-falcons (Micrastur) and snakes were seen near bathing sites, and a bathing tanager (Thraupidae) was observed to be taken by a snake.

Roosting takes place on slender perches in the lower levels of the forest. Théry found that, in the breeding season, the roost-sites of adult males were always outside their display areas, and generally near a main feeding area. Two males with neighbouring display sites might sleep less than 5 m apart. Females roosted in different areas, either alone or with a dependent juvenile on the same branch.

It is, however, for their courtship displays that the manakins are famous. These displays involve the most elaborate and complex movements, and interactions between individuals, that are known among passerine birds. A main feature of the displays of many manakin species is that the males perform in groups, or "leks", which persist in the same place from year to year. Moreover, an adult male manakin's entire life, with the exception of the period of the annual moult (see Morphological Habits), is centred on the lek, at which he may, furthermore, be present for up to 90% of the daylight hours (see Movements), with brief absences for foraging. This is typical of, for example, the Golden-headed Manakin. Observations in Trinidad, where this species is numerous, reveal that the males may perch inactively for long periods, frequently emitting brief calls; at times they peck at excrescences on their perches, and, if there are leaves or vine tendrils hanging nearby, they habitually pull at these while hovering, thus keeping the area immediately around the perch clear of obstruction.

A highly complex social organization has been found to exist at these display arenas, where the males compete with one another for the females, which visit the leks to mate. Nevertheless, displays are also performed intermittently when no female is present. As this intriguing behaviour is so closely associated with reproduction, however, it is discussed in greater detail in that context (see Breeding).

Manakins are generally rather silent birds, except at their display grounds. There, in association with their displays (see Breeding), the males utter a variety of mainly simple but often strange or striking calls. In addition, many species produce mechanical sounds, either in flight or when perched; the wing feathers and associated musculature of some piprids are specially modified to this end (see Morphological Aspects). These mechanical sounds are functionally equivalent to vocal sounds. Indeed, because of the rapidity of the bird's movements, it is in some cases by no means obvious to the observer whether a particular sound is vocal or mechanical. Furthermore, even when it is obviously mechanical, it is usually impossible to determine with the naked eye exactly how the sound is made. The following account is not exhaustive, but aims to give some idea of the range of sounds produced by manakins, and the contexts in which they are made.

Males of all of the well-studied piprids with fixed, often traditional, display territories or "courts", namely the typical manakins (see Systematics), have special advertising calls which they utter more or less regularly when at the display site. Away from the display area, both sexes are generally silent. These advertising calls, unlike those of many of the related cotingas, are mostly not very loud or far-carrying; they may be monosyllabic or they may be more complex, for instance a rapid phrase of a few notes, or a trill. Some are curiously unlike any "normal" sound expected from a bird. Thus, the male Golden-winged Manakin utters, at intervals that can be as short as 5-6 seconds but are usually rather longer, a low-pitched frog-like "nurrt", which is audible for only a short distance. This note seems to be designed to make any conspecific which may be in the near vicinity aware that he is in his display territory. Male Pin-tailed Manakins (Ilicura militaris) advertise their presence by emitting a soft, regularly repeated "see-see-see", usually of 5-8 notes, ending faintly and sounding plaintive to the human ear. In both of these species, males occupy display areas well separated from those of their fellows, and they can easily be overlooked by a human observer who is unacquainted with their voices. The lek-displaying Pipra and Manacus species, by contrast, have louder, more incisive advertising calls which are mutually stimulating, so that they tend to be given in bursts interspersed with more silent periods.

The loudest advertising calls are given by the blue-backed Chiroxiphia species, notable for their co-operative courtship displays (see Breeding). Those of the four best-known species are strikingly different from one another. In the case of the Blue-backed Manakin (Chiroxiphia pareola), the dominant male of a group uses a single, incisive "chup", preceded by a rolling "wrrrr", to call up a subordinate male, which perches close beside him. The two then commence a duet of almost perfectly synchronized phrases, usually with 2-4 notes per phrase, "chup-chup, chup-chup-chup" and so on, the dominant male beginning each note about 0·04-0·06 seconds before the subordinate male. The very small time difference between the two birds gives the duet a curious quality, which has been aptly described as "a resounding phrase that rang through the forest like the clicking of billiard balls". In contrast, the duets of the Long-tailed and Lance-tailed Manakins, which replace the Blue-backed in Central America and northern South America, are musical and almost flute-like, and the intervals between the utterances of the two birds are longer. From the resulting medley of pure whistled phrases, these two species have earned the local name of toledo. The fourth of the well-known species in the genus, the Blue Manakin (Chiroxiphia caudata) from eastern South America, is different again; its duetting calls, irregularly overlapping, consist of phrases of sharply ascending and descending pitch, and sound to the human ear like a confused gabbling.

Quite different from the calls of the typical manakins, in both their form and the context in which they are given, are the calls of the three mourners in the genus Schiffornis, included within the Pipridae more for convenience than for any other reason (see Systematics). Far more easily heard than seen, the inconspicuously coloured males, as they move through the forest undergrowth while foraging, give a pure whistled phrase of several notes and characteristic rhythm at irregular, usually rather long intervals. The function of this vocalization is presumably that of advertising possession of a territory to neighbouring males, and the caller's position to interested females.

The closely related Manacus species are outstanding for the loudness of some of the mechanical sounds that accompany their displays. The main sound is a loud snap, produced at the moment when a bird takes off from a vertical stem bordering one side of its court, to land on another on the opposite side. It had been thought that the snap was caused by the modified secondaries brushing very rapidly against one another, but recent research by K. Bostwick and Prum, using high-speed video recording, has shown that the sound is made by the dorsal surfaces of the wings being clapped together above the back in a movement of extreme rapidity, apparently meeting one another at or near the swollen bases of the secondaries. This is made possible by an enlargement and modification of the muscles that raise the wing. A different form of the snap, the "roll-snap", is produced, much less frequently, by a perched individual; as it is made, the wings can be seen momentarily to be raised above the back.

Some other manakin species, when displaying, make a snap or less loud clicks. Only one such species, the Red-capped Manakin (Pipra mentalis), has been studied by means of high-speed video recording, and this has produced some remarkable findings. The display of the male Red-capped Manakin features three kinds of sound of this type, brief broad-frequency sounds with a very sharp beginning, and all three of them are produced in different ways, differing also from the mechanism used in the Manacus snap. One is produced by the wings alone, another by brushing the tips of the primaries very rapidly against the raised and fanned tail, and the third by striking the ventral surface of the wing against the side of the body and the thigh.

The extraordinarily modified secondaries of the adult male Club-winged Manakin are apparently responsible for the second part of a double mechanical sound, "tip-beeuwww", produced by displaying birds. For the "tip" the wings are flicked downwards, and for the "beeuwww", like the sound of a miniature trumpet, they are held up vertically above the back and vibrated, the expanded club-like endings of the middle secondaries presumably acting as resonators and producing harmonics that are apparently unique among avian mechanical sounds. Displaying males of the related Eastern Striped Manakin (Machaeropterus regulus), having less extreme club-like endings of the secondaries, make a resonant grasshopper-like whirring as they cling head downwards on a vertical perch, turning from side to side so rapidly as to become a blur. In the Western Striped Manakin (Machaeropterus striolatus) and the Fiery-capped Manakin (Machaeropterus pyrocephalus), but not the congeneric Club-winged Manakin, the shafts of the tail feathers are thickened and are probably used for mechanical sound production. It is significant that in the genus Machaeropterus, with these extreme specializations for mechanical sound, the syringeal muscles are reduced in comparison with those of other manakins. In particular, the Club-winged Manakin, the most specialized of all, has almost no syringeal musculature, and in its courtship display makes only two simple vocal sounds. The advertising call common to most other manakins has been replaced in this species by the mechanical trumpet sound.

Several members of the Pipridae produce mechanical sounds during a rapid, stereotyped fly-in to the display perch. The three members of the "Pipra aureola group", consisting of the Crimson-hooded, Band-tailed and Wire-tailed Manakins, make a soft "kloop", apparently with the wings, at the lowest point of the flight before swooping up and landing on the perch. Similarly, the White-throated Manakin produces a muffled "pop" at the moment of stalling in mid-air, with a flash of his white wing patches, just before alighting on his display-log. As described later (see Breeding), the Orange-crested Manakin (Heterocercus aurantiivertex), after a spiral ascent high above the forest canopy, makes a loud hissing sound as it plummets vertically downwards at high speed; again, it is thought that this extraordinary sound is made with the wings.

It is likely that further research employing high-speed video recording will reveal equally amazing methods of sound production in manakin genera that have so far been studied only by means of field observation and sound recording. It is probable that modifications of musculature will come to light, and extreme rates of muscle contraction, comparable to those exhibited by hummingbirds (Trochilidae) in flight. This is an area of manakin research that is only just beginning.

Food and Feeding
Piprids feed mainly on small fruits, supplemented by insects and spiders. They usually pluck fruit in a rapid aerial sally from a nearby perch, landing with the food in the bill and then swallowing it whole, but they will sometimes pluck fruit from a perched position, if it is accessible by this means. Insects, and also small spiders, are likewise typically snatched from the vegetation following a brief flight. When termites (Isoptera) swarm after heavy rains, manakins may hawk for them from treetop perches, as do many other bird species. Some members of the family occasionally accompany mixed-species foraging flocks, and a few regularly do so. Among the few species for which the diets of the sexes have been analysed separately, females have been found to take a higher proportion of animal food than do males.

Sometimes, manakins join other birds in order to feed on insects disturbed by army ants. They usually do so only briefly, however, and they are not very efficient ant-followers, as they hardly ever sally to the ground. Moreover, most manakins use slender horizontal perches, and these are rarely to be found near ground level. E. O. Willis, who has made prolonged studies of ant-following birds in different parts of tropical America, found that in Amazonia the many manakin species rarely follow ants, but some of the same or closely related species attend ant swarms fairly often in Trinidad, Panama and eastern Brazil, where few specialized ant-followers occur. He concluded that the wealth of ant-following antbirds (Thamnophilidae) and woodcreepers (Dendrocolaptidae) which saturate the ant swarms in Amazonia makes ant-following rather unrewarding for manakins.

Manakins have a wide gape, and are able to swallow fruits that are very large in relation to the size of the bird. They often have to "juggle" large fruits in the bill, probably softening them in the process, before they can swallow them. The biggest fruits that the White-bearded Manakin has been recorded as swallowing, those of the rubiaceous plant Coussarea paniculata, were 19 mm long and 16 mm in diameter, and the largest ones recorded as eaten by the smaller Golden-headed Manakin, those of the burseraceous Protium heptaphyllum, measured 15 × 11 mm. The very wide gape is clearly important in enabling manakins to exploit fruits that cannot be eaten by considerably larger but less specialized frugivores, such as many tanagers.

A greater proportion of insect food is fed to the young than is eaten by the adults. For example, of 93 samples of regurgitated or defecated matter collected from below the display perches of White-bearded Manakins, insect remains were found in only four, the remainder containing seeds; by contrast, eight of 15 samples collected below nests with young contained insect remains. Nearly all of the insect remains consisted of hard parts of flies (Diptera) and beetles (Coleoptera). The Thrush-like Mourner, atypical also in other ways (see Systematics), is more insectivorous than the typical manakins and has been recorded as feeding large caterpillars to its nestlings.

The forests in which manakins live are rich in plants that bear small fruits, and the diets of manakin species are correspondingly varied. In a small area in Trinidad, White-bearded Manakins were documented as feeding on the fruits of 105 different plant species belonging to 27 families, and Golden-headed Manakins, less intensively studied, fed on 43 species. For both piprids, the trees and shrubs of species of Melastomataceae were of greatest importance numerically, and those of the Rubiaceae were second in importance. The major role of these two plant families in the diet of the Pipridae has been confirmed in studies elsewhere, especially in those undertaken by Théry in French Guiana.

Théry's research showed that manakins were the main dispersers of the Melastomataceae species and of the rubiaceous genus Psychotria present in his study area. They took an estimated 52-75% of the fruit crop of melastome species, and 61-89% of the fruit crop of Psychotria. In addition, they are very efficient dispersers. This is because their flight sallies enable them to select the ripest fruits, including terminal ones not available to other birds, and also because they feed in short bouts, shorter than the time taken by seeds to pass through the gut, resulting in good dispersal away from the parent plant. Moreover, and significantly, they tend to defecate at forest edges, in small clearings and along the courses of streams, sites that are suitable for the establishment of "pioneer plants", a category which includes most of the Melastomataceae and Rubiaceae. Seeds dropped elsewhere remain dormant for a long time, and in the event of a treefall they may germinate; in fact, a high percentage of such seeds in soil in French Guiana are from fruits eaten by manakins. In all these ways, manakins help to ensure the continuing abundance of their food supply.

As outstanding examples among small specialized frugivores in tropical forest, the adaptations of manakins to a diet of small fruits, which typically have a high water content but low concentrations of nutrients, are of particular interest. An important aspect of these adaptations concerns the fact that the diversity of available fruits may be highly variable, both in regular seasonal succession and unpredictably at times of occasional scarcity. Manakins, therefore, in addition to being able to swallow fruits of a wide range of sizes, must be able to process them rapidly, assimilating the nutrients, mainly carbohydrates, efficiently and getting rid of the seeds as quickly as possible. Consistent with these needs is the fact that manakins' stomachs are small and their intestines relatively short, and they have no crop for storing food. Experimental studies by A. H. Worthington, using temporarily caged Golden-collared and Red-capped Manakins in Panama, revealed that the average passage time of small seeds through the gut was 15 minutes for the former species and 12 minutes, with a minimum of 7 minutes, for the Red-capped Manakin. Seeds with a maximum dimension of more than 5 mm did not pass through the gut but were quickly regurgitated, after an average of 7 minutes and 9 minutes, respectively, for the two species. Efficiency of assimilation of the nutrients in the pulp was variable, as has been found for other frugivores, but it was very high, at 86-98%, for non-structural carbohydrates, the most readily available energy source. These studies also showed that feeding bouts were short, lasting up to 2 minutes, as is the case in the wild, and the birds did not begin a session until the remnants from the previous bout of feeding had been evacuated.

Like the majority of bird species inhabiting tropical forest, manakins nest solitarily. For all the typical piprids that have been studied, only the female is concerned with the nest. Manakins do not form pairs, but females that are ready to mate visit the males at their display grounds, where probably all matings take place. Studies of colour-ringed local populations of Golden-headed Manakins and White-bearded Manakins have demonstrated that most females mate with a single male in any one season, and the few that mate with more than one show a strong preference for one particular male. It is always a small number of dominant males at a lek that perform the majority of all matings; indeed, in the co-operatively displaying Chiroxiphia species, all matings are performed by the dominant male of the group. Since it may take several years to acquire a dominant position, males may not breed until they are several years old; females, as a rule, probably begin to breed at the age of about one year.

If there is one thing for which the family is renowned, it is its courtship displays. These displays, while not set off by such extraordinary feather adornments as those of some of the cotingas, or the birds-of-paradise (Paradisaeidae), are nevertheless remarkable in that they involve the most complex and elaborate movements, as well as the most highly evolved interactions among individuals, that are known among passerine birds. A predominant feature of this aspect of behaviour is that the males of many piprids display in groups at special leks, which are situated in the same places from one year to the next. The males compete with one another for the females, which visit the leks to mate. The spacing of males at a lek varies greatly, males of Manacus species being regularly as little as a metre or less apart, whereas several other species perform at dispersed or "exploded" leks, where males are within hearing distance of, but not in visual contact with, one another. Apart from during the period of the annual moult (see Morphological Aspects), an adult male manakin's whole life is centred on the lek, at which he may be present for up to 90% of the daylight hours (see Movements).

The complexity of manakin displays is two-fold, involving, on the one hand, the display movements and associated sounds, and, on the other, the complexity of social organization. The main features tend to be similar among members of the same genus. Limitation of space prevents a detailed description of all those which have been described, and the following is a necessarily selective summary.

The lek behaviour of the Golden-headed Manakin is a good example, central, so to speak, in terms of both its complexity and the relationship between the individual males, and it has been well studied. The males display under the canopy of lower-storey trees, on more or less horizontal small branches at a height generally of about 6-12 m. Each has a main perch, on which, if he is successful, mating takes place. The perches may be as little as 2-3 m apart and in the same tree, or 10 m or more apart and in adjacent trees. The number of males constituting a lek is variable, depending, at least in part, on the local population density. In Trinidad, where the Golden-headed Manakin is particularly numerous, leks can consist of up to twelve males. For most of the day, apart from short absences to forage, and for most of the year, except for the moult period, males are present at the lek. They often perch, inactively, for long periods, although they frequently utter short calls, and they keep the immediate surroundings of the perch clear of obstruction by continually attempting to remove any hanging leaves or vine tendrils.

Active display can occur at almost any time throughout the day, but is most frequent in the hour after sunrise and in mid-afternoon. Although it may occur spontaneously, it is conspicuously stimulated by the arrival of a female at the lek. The posture and movements involved in the display are highly stereotyped and very diverse, consisting of the following five elements. The first is known as "darting back and forth": with the legs fully stretched so that the red and white thigh feathers become conspicuous, and with the body held horizontally, the manakin makes rapid flights to and fro between its main perch and an adjacent one 1-1·5 m away, the wings producing a brisk humming sound in flight between the two perches; on landing, he at once turns about rapidly to face in the direction from which he came. In the second element, the "about-face", the bird makes rapid about-face turns, so quick that the foot movements cannot be clearly seen; these may be repeated while the bird remains in the same place, and each turn may be accompanied by an upward flick of the wings. Next is the "backward slide", in which the bird, with legs stretched, head held low and tail elevated, moves with very short and rapid steps, so that it seems to slide, backwards along the perch for about 10-20 cm; at the end of the slide, the tail is suddenly depressed and fanned and the wings are raised and held vertically above the back. Sometimes, half-way through the slide, the wings are suddenly spread horizontally and then closed again. The fourth display element, the "upright posture", is a stiff, almost statuesque posture, with head and body nearly vertical, which may be held for several seconds, the bird clinging to the side of the perch rather than perching on it; during this, the pupil is greatly contracted and the bill may be slightly opened. The fifth and final element is the "display-flight". This begins with the male leaving his perch and flying to a somewhat higher one some 15-25 m away, where he perches, facing the main perch, utters two or three sharp monosyllables, and then takes off and returns to the main perch in extremely rapid flight, uttering a succession of the same monosyllables which speed up and become sharper as he nears it; he approaches the main perch with a rapid downward swoop followed immediately by an upward swoop, and as he lands he utters a sharp buzzing call. Immediately on landing, he usually executes a backward slide.

As already mentioned, some or all of these displays are intermittently performed when no female is present. On the arrival of a female, however, display by all the males at the lek becomes intense and continuous. A female's behaviour is variable, being probably dependent on her familiarity with the lek and on her breeding condition; she may herself perform some of the display movements, but in less intense form. When ready to mate, she eventually goes to the main perch of one of the males. Occasionally, when the two are perched close together, she pecks very lightly at the male's brilliant orange crown feathers. Copulation may be preceded by any of the above-mentioned displays with the exception of the upright posture, which is used in more aggressive contexts. Mating often ensues immediately after a display-flight, the male landing beside the female and then mounting.

Although the details differ, the courtship displays of the Red-headed and Red-capped Manakins, both closely related to the Golden-headed Manakin, follow the same pattern. Very different is the lek behaviour of the White-bearded Manakin, the males of which display in compact groups on and near the forest floor. Each male clears a small "court", usually roughly circular or oval and about a metre across, removing all fallen leaves and other debris. Each court must have around it at least two, and generally three or four, vertical stems, usually of saplings; one of these is of special importance, as mating takes place on it. Courts are usually compactly grouped, sometimes even immediately adjacent to one another, although more often a metre or two apart. The number of courts at a lek varies greatly, and can number 50 or more in places where the manakins are abundant. At large leks, competition for central courts, where most of the matings take place, can be intense, leading to prolonged and violent fighting between the owner and a would-be usurper.

Like Golden-headed Manakins, male White-bearded Manakins are present at their courts for the greater part of the day and of the year. When not actively displaying, they utter simple, mainly monosyllabic calls. Bouts of display may occur throughout the day, but are concentrated in two periods, the hour after dawn and the early afternoon, and these are the times when females most often visit the leks. The display repertoire of this species consists of three distinct elements. The first of these, and the commonest display, involves jumps between vertical stems. Perched sideways on one of the vertical stems on the court's periphery, with the long throat feathers thrust forward to produce a "beard" that projects beyond the bill tip, the bird leaps across to one of the opposite stems, turning as he lands so as to face the way from which he came; at the moment of take-off, he makes a loud snap by striking the upper surfaces of the wings together above the back (see Voice). He may rapidly repeat the manoeuvre, either to and fro between two stems or irregularly around all the vertical stems surrounding the court. When a closely packed group of males is performing together, the impression is of a confusing collection of jumping black-and-white fire-crackers. The second element of display is less common, being performed only occasionally. After a series of snap-jumps ending low down on the mating perch, a male remains on the perch for a few moments, tense and quivering, with head and body pointing obliquely downwards; then, with a snap, he leaps to the ground, turning in the air, and on landing immediately leaps, making a curious grunting sound, back up to a higher position on the mating perch, and then, taking very rapid short steps, he "slides" down the perch while beating his wings. This sequence, of snap to ground, upward leap with grunt, and slide down perch, is in fact the stereotyped sequence of pre-mating movements. The third element, fanning, is a very distinct but comparatively infrequent display, most often directed to a female on her initial approach to a court. The male crouches, with head retracted between the shoulders, sways from side to side, and beats his wings with a low whirring sound; as the wings are raised, the long neck feathers are pushed forwards and upwards and appear as two flickering white puffs, one at each side.

When a female visits a male's court in order to mate, she usually first joins him in a co-ordinated dance of to-and-fro jumps between the mating perch and an adjacent perch, the two birds crossing one another in mid-air, until finally the female lands on the mating perch and remains there. This is the signal for the male to perform the pre-mating sequence described in the preceding paragraph, culminating in the "slide down the pole" on to the female's back.

The courtship behaviour of the three other Manacus species is extremely similar, and in most details in fact identical, to that of the White-bearded Manakin. This indicates a relatively recent splitting of the four species, which differ mainly in the presence or absence of yellow or orange pigment in the male plumage, and which interbreed to form narrow hybrid zones where their ranges meet (see Systematics).

A new level of complexity is reached in the lek displays of the "Pipra aureola group", comprising the Crimson-hooded, Band-tailed and Wire-tailed Manakins. In this case, additional males, referred to as "beta males", perform co-ordinated displays centred on the mating perch of a dominant male, "the alpha male", which alone mates with the females visiting that perch. The most striking feature in the co-ordinated display consists of precisely alternating display-flights similar to those of the Golden-headed Manakin. In quick succession, each male flies in and lands on the mating perch. As one bird approaches in rapid flight, the other, a moment before the incomer alights, leaves and flies back to the taking-off perch some 20 m away, then at once flies back in and displaces the first, and so it may continue for minutes on end. The beta males are younger birds, which, unlike the alpha males, are not sedentary but move around, competing with one another to take part in the older, established males' displays. The display repertoires of the three species are very similar, but with one extraordinary difference. In the culminating phase, the male, with tail raised, backs towards the female on the mating perch. Whereas two of the species have a tail of normal shape, the male Wire-tailed Manakin has uniquely modified rectrices ending in long wire-like projections, with which, when he has backed close to the female, he vigorously brushes her chin, rapidly pivoting his upturned body from side to side. This treatment is evidently agreeable to the female, as she perches quietly, facing him, with bill raised to present her chin.

The most elaborate display repertoire and social organization in the family, and perhaps among all birds, is that found in the genus Chiroxiphia. The display of the Yungas Manakin (Chiroxiphia boliviana) is unrecorded, but for the other four species the pattern is essentially the same. The Blue-backed Manakin may serve as the example.

Male Blue-backed Manakins display, and mate, on slender, more or less horizontal and typically slightly bowed stems or branches about 1-1·5 m above the ground in forest undergrowth. Each mating perch is "owned" by a dominant male, with which a small number of subordinate males are associated. The courtship display consists of four distinct stages. The first stage is duetting. The dominant male, perched at some height near his mating perch, calls at regular intervals until he is joined by one of his subordinates. The two then perch side by side, facing in the same direction and almost touching each other, and begin a loud synchronized duet in which the subordinate bird utters each note a fraction of a second after the dominant bird, the slight time difference producing a characteristic ringing sound (see Voice). After calling thus for a little time, the two birds fly down to the mating perch for the next stage. This second stage, the jumping display, has as its basic movement a fluttering jump. The bird crouches, and then jumps up with wings fluttering, bill pointing downwards and legs dangling, and hangs momentarily in the air before landing back on the perch; while in the air, he utters a strange vibrant, twanging call with the bill wide open. Typically, the two birds jump alternately, side by side, one beginning his jump as the other lands. The rhythmic twanging calls of the two jumping birds is often the main clue guiding a human observer to a mating perch, which is otherwise often difficult to locate in the tangled woody undergrowth. As the display proceeds, the jumps become lower and more rapid, and the calls become more irregular and confused, and take on a bleating quality. Finally, the last bird to jump emits a louder and more distinct bleating note, followed at once by one or more quite different, high-pitched "zeek" notes, after which both fly off. The third stage is the cartwheel dance. This is a modification of the jumping display and is performed in response to the arrival of a female on the mating perch. The two males turn to face her; the foremost male jumps up, hovers in the air, facing the female, and then moves back, still hovering, while the other male shuffles forward to take its place and in turn jumps up as the first bird lands behind it. Thus, the two males move in the form of a revolving wheel. As in the jumping display of two males by themselves, the cartwheel dance in front of a female, accompanied by the rhythmic twanging, becomes more and more rapid, seemingly frenzied, until it is brought to an end by the sharp "zeek". This is, in fact, uttered by the dominant male, the perch-owner, and is the signal to his dancing partner to leave, the latter's part in the display being completed. The fourth and final stage is the pre-copulatory display. If the female has stood her ground and remained on the mating perch, rather than, as often happens, leaving when the cartwheel dance ends, the dominant male begins a further, quite different and mainly silent display. He flutters around the female with a bouncing, butterfly-like flight, silently crossing and recrossing the display perch, every few seconds alighting momentarily and then flying on. If he alights near the female, he faces her, crouches with head lowered, thus exhibiting the blue patch on his back, and vibrates his wings. As the head is lowered, his red crown is presented squarely to the female, appearing shield-like as the two small lateral "horns" are extended. The female is thus presented with a red shield surmounted by a vibrating patch of shimmering blue. The female turns constantly to face the male, intently watching his performance. Occasionally, the male flies out to a more distant perch, faces the female, and, uttering a low twanging note, takes off with an audible click of the wings and resumes the bouncing flight. Finally, he flies to the mating perch, lands beside the female, and mounts.

The courtship displays of the other well-known Chiroxiphia species differ mainly in the first stage, the duetting calls being quite distinct. The display of one of them, the Blue Manakin, further differs in that three males regularly take part in the cartwheel dance, and occasionally more do so, perching close together and shuffling forward in line as the individual nearest the female jumps up.

By contrast, the two Corapipo species, the White-ruffed and White-throated Manakins, as well as the Golden-winged Manakin, are unusual in that they mate on fallen logs on the forest floor. The social organization of these three has been puzzling. Whereas some accounts describe single males in apparently uncontested ownership of a log, others mention groups of males, up to seven in one case, displaying together on and around a log in a more or less co-ordinated manner, and apparently moving as a group from one log to another. The most recent study of the White-ruffed Manakin, by L. Rosselli and co-workers, which included colour-ringed birds, indicates that each log is the property of a dominant resident male. When females visit for mating, which occurred at 12:45 and 14:47 hours in the two cases observed, that male alone is present, while earlier in the day other males, both adult and immature, are permitted to come and take part in communal, and at times co-ordinated, display among themselves or with the resident male. The function of these communal displays is probably that of establishing and maintaining a hierarchy of males within the local population.

The pre-mating display of the White-ruffed Manakin is one of the most elaborate and impressive. Following a series of flight displays close to the ground, ending with the female perching on the central part of the log, the male flies up from the log in a series of steep spirals until he is above the forest canopy; there he hovers briefly, while uttering one or more very high-pitched calls, before plummeting down to the display-log, landing beside the female and, at the moment of landing, making an explosive snap, almost certainly with the wings, followed instantly by one or two high notes and a sharp buzzing "wah", both probably vocal. Thereupon he immediately jumps over the female, hops towards her, and mounts.

Recently, it has been found that Heterocercus and Tyranneutes species also perform high flight displays. The most spectacular of such performances that is known is that of the Orange-crested Manakin. The male flies steeply upwards, in a series of spirals of decreasing size, to a height of 60-100 m above the forest canopy. From there it plummets down, gaining momentum for the first 15-20 m with rapid wingbeats, and then partially closes its wings and continues its descent in this posture, producing a peculiar hissing sound that becomes louder as the bird's velocity increases. When it is 2-3 m above the canopy it abruptly changes direction by nearly 90 degrees, producing an explosive "pop". Finally, the manakin flies in a wide horizontal arc above the canopy, before entering the trees at great speed and returning to its terrestrial court.

The impossibility of adequately observing such a display in closed forest will be obvious. Detailed observation of the Orange-crested Manakin's display, by J. Álvarez, was made possible by the species' riverside habitat, enabling the whole performance to be seen from a boat. The high aerial displays of the two Tyranneutes species have been watched from a canopy crane and a canopy tower. B. Walther, watching from a canopy crane in Venezuela, has observed the Dwarf Tyrant-manakin (Tyranneutes stolzmanni) to fly straight up with very fast wingbeats from a perch at the top of a very tall tree to a height of 20-30 m and then "dive-bomb", in the manner of an aggressive hummingbird, back to the same perch. In Brazil, K. J. Zimmer, watching from canopy towers, has seen similar displays performed by the Dwarf Tyrant-manakin in Para and by the Tiny Tyrant-manakin (Tyranneutes virescens) near Manaus. How these spectacular displays relate to the species' behaviour at ground level is unknown. Male Tiny Tyrant-manakins call from low perches about 2-6 m above the ground, and make floating flights with rapidly beating wings between these perches.

A common feature of all the manakin lek displays that have been well studied, with individually marked birds, is that, in any one season, most or all of the matings are made by only one or a few of the males, and some males have no mating success at all. In large leks, the successful males are usually those occupying the more central display perches or courts. This at once raises a number of questions, which have been extensively discussed but not all unequivocally answered, concerning the factors that influence females in their choice of a mate. Among males, there is constant competition for central positions in a lek or, in the case of species with joint displays, for dominant status at a mating perch; therefore, a female would choose wisely by simply mating with central or dominant males, because they are likely to be efficient, vigorous individuals. This would not, however, explain the evolution of the males' extravagant displays and adornments, unless the central or dominant males also have the most spectacular performances, and yet the most careful observations and measurements of the few species studied in detail have failed to show any such correlation. However, one must bear in mind that birds certainly recognize many of their neighbours individually, presumably by using very subtle differences in appearance, voice and behaviour that humans cannot detect. In addition, there is some evidence that females are, as it were, "fascinated" by the males' adornments and show this by lightly pecking or touching them with the bill; thus, there would be scope for selection of males on the basis of their individual structural or colour differences. It is difficult otherwise to understand how the extravagant displays and associated plumage have evolved, as Darwin argued long ago. It has been suggested that competition between males, rather than female choice, may be the main selective agency affecting male physical characters. This is presumably the case for animal species in which the males fight for dominance, as with, for example, many mammals, but it seems highly unlikely to apply to manakins and other birds with highly ornate males, as the most elaborate displays are so clearly directed towards, and normally visible only to, the females.

Another intriguing question raised by such displays concerns those in which two or more males engage in some form of co-ordinated display, culminating almost invariably in mating being achieved only by the dominant male. What do the subordinate individuals get out of it? Long-term study of colour-ringed birds, most notably D. B. McDonald's study of the Long-tailed Manakin, has provided what is almost certainly the correct explanation: namely, that the subordinate males, in a ranking order based on age, are awaiting their turn to take over the dominant male's position when he loses it. Meanwhile, by displaying with him, they are gaining experience and maintaining the attractiveness, for females in the area, of the traditional display area and mating perch. This would not be a sensible strategy if annual mortality were high and senility in consequence rare, as is prevalent among many small north-temperate passerine birds. Adult manakins, however, are known to have high annual survival rates, of the order of 90%, and there is evidence that senility or loss of vigour may be quite usual, so that a younger male may have a very good chance of being able to take over the position of a dominant older bird. Even so, it is amazing that a male Long-tailed Manakin may have to wait up to ten years before achieving dominant status and being able to mate.

The evolutionary interpretation of piprid behaviour can be only briefly touched on. There is general agreement that an underlying prerequisite for the kind of social organization so strikingly illustrated by the manakins, characterized by conspicuously ornamented and conspicuously displaying polygamous males, and cryptically coloured females which single-handedly undertake all nesting duties, is the mainly frugivorous diet, which allows adults living in a rich forest habitat to feed themselves in a very small fraction of the daylight hours. Combined with this, there is good reason to suppose that the intense nest predation to which manakins, as other tropical-forest birds, are subject has resulted in their building very small, inconspicuous nests, capable of holding a maximum of two nestlings, which the female can rear alone, the presence of a brightly coloured male at the nest being positively disadvantageous. Under such conditions, intense sexual competition among males has arisen.

Exactly how the elaborate and even bizarre courtship behaviour has evolved in the Pipridae has been much discussed, with outstanding contributions by Prum. What has clearly emerged from Prum's analyses of manakin evolution, based on morphology, behaviour and DNA, is that the extraordinarily varied elements of their displays, including their spatial organization, have a strong phylogenetic component; in other words, they must have a long evolutionary history and, for the most part, are not recent responses to the present-day ecological conditions to which the different species are subject. The most likely evolutionary mechanism seems to be one proposed long ago by R. A. Fisher, who suggested that there has been some kind of linkage between genes producing extreme traits in males and genes determining female preference; put simply, females prefer the most extravagant males. Such a mechanism could, as seems to have occurred with the manakins, lead to the evolution of essentially arbitrary male characters and displays.

Like those of many birds of tropical forest, manakin nests, although not placed in thick cover, are small and hard to see; some are camouflaged. They are generally sited within a few metres of the ground, and the nests of some species are very low down. The result is that, while the nests of some manakins have never been found, those of a few of the most abundant species have been located in good numbers. Most of what is known of the family's breeding biology is therefore based on just a few species.

Breeding seasons of the few well-known species are long, extending over several months. Most of the available information originates from studies carried out north of the equator, in Central America and Trinidad around latitude 10° N, where the dry season, usually from about January to mid-May, is followed by a wet season lasting, usually with a minor interruption in September or October, until the end of the year. Under this seasonal regime, manakins begin to breed at some time in the dry period and continue into the wet, breeding usually being at its height in April-June. Individual females may make repeated nesting attempts in the course of a single season. Following the breeding season, adults undergo a complete moult which lasts, in the case of the White-bearded Manakin, for about 80 days. Studies of manakins in Trinidad and Panama have indicated that initiation of breeding is associated with, and apparently dependent on, an increase in the availability of fruit, and that the timing of this increase may be annually highly variable. Nevertheless, caution is needed in generalizing from findings in these two areas. This is demonstrated by the fact that, in Costa Rica, the White-ruffed Manakin, an altitudinal migrant (see Movements), was found to leave its breeding area when fruit was abundant and to return to breed at a time of overall scarcity.

A typical manakin nest is a very characteristic structure, easily recognized as such even if the owner is not seen. It is a small cup, constructed from vegetable fibres, leaf midribs, fungal rhizomorphs or other fine material, suspended in a fork between two horizontal diverging twigs, to which it is bound by cobweb or fungal rhizomorphs, or a combination of both. In places where large ferns are abundant, it may be suspended between two adjacent fern frondlets. The nest of some species is covered on the outside with pieces of dead leaves or moss, and may have a "tail" of dead leaves, moss or other material hanging down below it, thus increasing its apparent bulk but obscuring the rounded bottom of the nest cup. Such nests, when seen from the side or below, can be almost indistinguishable from the bunches of dead leaves that are commonly caught up in the forest undergrowth. The only known exceptions to this general form of nest are those of the Thrush-like Mourner, which builds a bulky nest in a quite different kind of site, and the Wing-barred Piprites (Piprites chloris), with a cavity nest. However, Schiffornis and Piprites are among the genera of uncertain affinities that are included in Pipridae only for historical reasons and for want of a clearly preferable alternative (see Systematics).

Perhaps the most extreme instance of adaptations designed to make the nest inconspicuous is that of the Orange-crested Manakin, the nest of which has only recently been discovered. Typically slung in a horizontal fork, the nest was so loosely woven, from vegetable fibres, that the observer could easily see through it from below. The incubating female adopted an exaggeratedly upright posture, with most of her breast well above the nest rim and the entire tail and undertail-coverts clearly in view, angled downwards, the nest itself being practically invisible. Thus, the visual impression was of a perched bird in a state of vigilance. There are clear analogies to this in the nesting strategies of some cotingas in the genera Lipaugus, Xipholena and Procnias.

Manakin eggs are rather large for the size of the bird. For example, those of the Red-capped Manakin measure 21·5 × 15·9 mm and those of the Orange-collared Manakin (Manacus aurantiacus) 20·5 × 14·9 mm. The ground colour is whitish, or pale cream-coloured, buff or grey, variably but often quite heavily marked with brown. So far as is known, the clutch size of all species is almost invariably of two eggs; occasional apparent single-egg clutches are probably the result of egg loss caused by predation. Eggs are usually laid around midday, the second about 48 hours after the first. Bearing in mind the small size of these birds, the incubation period is relatively long: a period of 16-17 days is recorded for the Golden-headed Manakin, 17·5-19 days for the Blue-crowned Manakin, 18-19 days for the White-bearded and Orange-collared Manakins, and 20-21 days for the Thrush-like Mourner. Fledging periods are less well documented, as nestling predation tends to be high, but they are not correspondingly long, being 13-15 days in the case of the Blue-crowned, White-bearded and Golden-collared Manakins.

Incubating female manakins spend long sessions on the nest, with relatively short absences for foraging. Thus, they devote a higher proportion of the day to incubation than do most small birds of other families. This is made possible by their diet of small fruits, which are usually abundant in the near vicinity. They are often very "tame", permitting a very close approach, even to the extent of allowing themselves to be gently touched. Sessions of incubation frequently last for more than an hour, and can be as long as 3-4 hours. The newly hatched young are pink-skinned, with a sparse covering of grey down, an exception being the chicks of Schiffornis, which have a copious down. They are fed with a mixture of small insects and small fruits, which the female carries mainly in her throat, although a few items may be held in the bill. Feeds are infrequent, usually one or two per hour for older nestlings. The female at first removes the chicks' droppings, but older nestlings defecate over the nest rim, so that the ground below is spattered with droppings by the time they are ready to fly. This is a rather surprising habit, as the low nests of, for instance, the White-bearded Manakin can be less than a metre above ground and clearly visible from below. Presumably, potential predators are not attracted in this way.

Be that as it may, nesting success of all piprid species studied is very low. The following figures take into account only nests located before the clutch was complete, as those found later are biased in favour of success. In terms of percentages of nests that yielded at least one fledged young, the success rate for 23 Orange-collared Manakin nests was 13% and that for 227 nests of White-bearded Manakins was 19%. For Pipra species, the higher nests of which are more difficult to find, extremely low success has been recorded; of eight nests of Red-capped Manakins and seven of Golden-headed Manakins, not one produced any fledglings.

With such low breeding success, followed probably by a considerable post-fledging mortality, about which nothing is known, it follows that, for populations to remain stable, adult survival must be high. This has been found to be so for the four species to which long-term studies have been devoted. Thus, for the White-bearded Manakin in Trinidad an annual survival rate of at least 89% has been recorded, with individual males living for a minimum of 14 years and females for 12·5 years. In the same area, male Golden-headed Manakins were recorded as surviving for at least 11·5 years and females for 12 years.

The White-ruffed Manakin in Costa Rica breeds in the mountains, mainly at altitudes of 400-900 m, descending to adjacent foothills and lowlands in the non-breeding season, and there is evidence that similar movements are made by the Red-capped and White-crowned Manakins. In Ecuador, the Club-winged Manakin apparently makes similar elevational displacements. No other manakin is known to undertake seasonal migratory movements of any kind. Indeed, those species that have been investigated in detail have been found to be highly sedentary. For example, in a Trinidad population of White-bearded Manakins studied intensively for 3·5 years, a colour-ringed male was seen while foraging about 800 m from where it had first been trapped, and three colour-ringed females were found nesting at about the same distance from the ringing site; these were the longest movements recorded. Except for the moulting period, most adults, at least, spent the whole time within very restricted areas. During the moult, when the display grounds were largely abandoned, systematic trapping indicated that some individuals wandered farther, since a considerable proportion of adults trapped at that time were never recorded again. Trapping of Golden-headed Manakins in the same area produced similar results.

Established males at display grounds, in particular, are extremely sedentary for much of the year. Male White-bearded Manakins were recorded as spending up to 90% of the daylight hours at their courts, making only brief excursions, nearly all less than 5 minutes in duration, in order to forage at nearby fruit sources, with a slightly longer absence in the late afternoon to bathe in a stream close by. Similarly, a male Golden-headed Manakin, watched throughout the day, was present on his display perch for 88% of the time; he made brief foraging excursions and, in the afternoon, one longer one, of 5 minutes, during which he bathed. A male Pin-tailed Manakin in south-east Brazil was observed throughout the hours of daylight, and was found to be present in his display territory for 87% of the period.

These findings have been confirmed and much amplified by the results of radio-tracking experiments by Théry in French Guiana, involving five manakin species. For four species of typical piprid, the Golden-headed, White-fronted, White-crowned and White-bearded Manakins, the home ranges of adult males during the breeding season were very restricted, from 2·1 ha to 3·4 ha, and included the individual bird's display area and one bathing site. The home ranges of females were more extensive, from 8·8 ha to 14·4 ha, and contained several display areas and bathing sites, and those of immature males were larger still, from 18·7 ha to 24·1 ha. Outside the breeding season, they all used much wider areas; radio-tracking over short periods indicated ranges of 30-32 ha, and these would probably have extended to several hundred hectares if the tracking had continued for long periods.

The fifth species tracked, the Thrush-like Mourner, was found to have much larger home ranges than those of the four typical manakins during the breeding season. The figures for two males were 14·3 ha and 16·5 ha, respectively, and the home range of one female was 20·1 ha. Thrush-like Mourners are more insectivorous than the typical manakins; the males have no fixed display area but give their advertising calls while on the move, as they forage through the forest undergrowth.

Relationship with Man
The native Americans of Amazonia and other areas of humid forest could hardly fail to be familiar with many manakin species, especially those with conspicuous displays. In Amazonian Brazil, the brightly coloured Pipra species are associated with a variety of legends, superstitions and witchcraft. Dried skins, as well as whole birds preserved in alcohol, are used as charms, talismans and amulets. No doubt there were many local names for the different species. The word manaquim, from which Manacus and the English vernacular "manakin" are derived, may be a transliteration of an indigenous name used in the upper Amazon.

More recently, the amazing co-operative "dancing" displays of the blue-backed Chiroxiphia manakins became familiar enough to the Spanish and Portuguese settlers to earn the name bailador for the Long-tailed Manakin and dançador for the Blue Manakin. In Trinidad, where the White-bearded Manakin is especially abundant and its leks, with wing-snapping birds leaping to and fro between vertical stems close to the ground, often near forest paths, can hardly be missed, it is known to country people as the "stick-man".

Status and Conservation
In many Neotropical forest areas, manakins are among the most abundant birds of the forest understorey. Thus, in three years of mist-netting in French Guiana, the area's seven species of manakin made up one third of all birds trapped, and in 3·5 years of mist-netting in Trinidad the number of manakins trapped, of two species, far exceeded the combined total for eleven species of tyrant-flycatcher of the forest understorey. The efficiency of manakins as dispersers of the main fruits on which they feed (see Food and Feeding) is probably a main reason for their general abundance. A few species, however, are unaccountably scarce or local.

Of the 58 piprid species currently recognized, four are considered to be globally threatened and a further one could potentially join that category. The species most at risk is the Araripe Manakin (Antilophia bokermanni), the conservation status of which is listed as Critical. This little-known species, the handsome male sporting a plumage of white, black and red, is confined to an extremely tiny range in north-east Brazil. It was first discovered as recently as 1996, in the foothills of the Chapada do Araripe, in south-east Ceará, and has not been found anywhere away from the type locality. Its known range, therefore, covers no more than about 1 km². Although there is hope that it may be present elsewhere in nearby foothills, it has not yet been found there, despite searches having been carried out in the region. The total population, of fewer than 250 individuals, lives in tall secondary forest in the Araripe National Forest, which is encompassed within a large Environmental Protection Area, but legal protection is minimal. Lowland areas adjacent to the species' habitat have already been cleared for commercial plantations, farmsteads and human settlement, and associated developments, including road-building and the construction of recreational facilities, constitute a major threat. Unless effective protection of this manakin's habitat is achieved, its future prospects appear bleak. Even in the event of additional populations of the Araripe Manakin being located in nearby forest, there is no guarantee that these could be safeguarded.

Farther south, in the lowland forests of south-east Brazil, Wied's Tyrant-manakin (Neopelma aurifrons) has similarly suffered from extensive loss of habitat and is now classified as Endangered. It has been found at a number of localities from Bahia south to Rio de Janeiro, but there are very few records since 1990. The species was formerly treated as conspecific with the Serra Tyrant-manakin (Neopelma chrysolophum), which may have obscured its true status. Nevertheless, it appears certainly to have been more widespread in the past in Bahia and Minas Gerais, in both of which states it has not been seen for some years. It was present in Chapada Diamantina National Park, in Bahia, but was last recorded there in 1990. In Espírito Santo, it was uncommon in Sooretama Biological Reserve in 1981 and has been seen there only once since, and it occurs locally within Augusto Ruschi Biological Reserve. Otherwise, Wied's Tyrant-manakin is known from a single locality in the state of Rio de Janeiro, and one target of conservationists is to achieve full and effective protection for this forest, situated north-east of the city of Rio de Janeiro itself.

Two piprids are categorized as Vulnerable. These are the Golden-crowned Manakin (Lepidothrix vilasboasi) and the Black-capped Piprites (Piprites pileata), both of which, as the two other threatened species, are Brazilian, although the latter has been recorded once in north-east Argentina. The Golden-crowned Manakin is so little known that it is difficult to say anything definite about possible threats to its survival. It has been seen only in south-west Pará. The type locality, where the species was discovered by H. Sick in 1957, is on the upper River Cururu-ri, a tributary of the River Teles Pires. It lies within an area of 400,000 ha belonging to the Brazilian air force; here, access is limited and there is no logging. The manakin had not been seen again since its initial discovery, until, in May 2002, it was rediscovered at Consórcio Jamunxim, 200 km north-east of the type locality. This second site, however, is under pressure from an increasing human population and the planned paving of the Cuiabá-Santarém road, which traverses the region. Clearly, fieldwork is needed in order to determine the species' current status and distribution. It seems that the earlier failure of ornithologists to relocate this manakin was probably the result of some confusion over its precise type locality. There are two right-bank tributaries of the Teles Pires bearing the name Cururu. The original specimens were found on the Cururu-ri (often referred to simply as Cururu), which flows north-west, but a better-known river is the Cururu-açu, farther south, which flows south-west to join the main Teles Pires.

The other Vulnerable member of the family, the Black-capped Piprites, has apparently always been rare and local, but it has evidently become more so in recent years, especially in the south of its range, where the Araucaria forest in which it occurred has been extensively destroyed. It is restricted to south-east Brazil, from Minas Gerais and Rio de Janeiro south to Rio Grande do Sul, and there is a single record from extreme north-east Argentina, where an individual was caught in September 1959. Although very sparsely distributed and certainly rare, this species does occur in several protected areas in Brazil, including Itatiaia, Serra da Bocaína and Aparados da Serra National Parks and Campos do Jordão State Park. Rumours of its presence in Iguazú National Park, in Argentina, have never been substantiated. It appears to prefer montane forest, which is not at so great a risk of being cleared as are forests in the lowlands, and its prospects for survival should, therefore, be reasonably secure, at least in the short term.

In addition to these four threatened piprids, one other, the Yellow-headed Manakin (Chloropipo flavicapilla), is currently considered Near-threatened. This species, patchily distributed in montane forest and mature second growth in the Andes of Colombia and Ecuador, is generally uncommon to rare. There are few recent records, although it is an inconspicuous bird and may have been overlooked. Much of the range within which the Yellow-headed Manakin occurs consists of potentially good-quality agricultural land, and a good deal of its habitat has already been lost as a consequence. It is desirable that the precise status of this manakin be established by means of further fieldwork and research, and any possible changes in its circumstances should be closely monitored.

Finally, two members of the family with rather limited ranges, the Opal-crowned Manakin (Lepidothrix iris) of Brazil and the Grey-headed Piprites (Piprites griseiceps) of Central America, were formerly included in the category of Near-threatened. Both had suffered habitat loss as a result of forest clearance, but recent work suggests that there is no longer any reason for immediate concern over their conservation status.

In general, the future conservation of manakins, as that of many other Neotropical birds, will depend on the effective protection of adequate areas of forest. Because of their small size, enabling populations to be maintained within relatively small areas, and because many species occur in mature secondary woodland as well as in primary forest, the prospects for the manakins should be better than those of larger and ecologically more specialized forest birds.

General Bibliography
Álvarez (2000, 2001), Bierregaard (1988), Birdsley (2002), Blake & Loiselle (2002), Bock (1994), Bostwick (1998, 2002a, 2002b), Brumfield & Braun (2001), Bucher & Worthington (1982), Cracraft (1981), Endler & Théry (1996), Foster (1991), Graves (1993), Haffer (1974), Hilty (1994), Irestedt et al. (2001), Johansson et al. (2002), Lanyon, S.M. (1985), Lill (1974, 1975), McDonald (1989a, 1989b, 1993b), McDonald & Potts (1994), Miller (1908), Mobley & Prum (1995), Prum (1990a, 1990b, 1992, 1994a, 1997, 1998, 2000), Prum & Lanyon (1989), Read (1994), Sibley (1996), Sibley & Ahlquist (1990), Sibley & Monroe (1990, 1993), Sibley et al. (1988), Sick (1967), Snow, D.W. (1962a, 1963c, 1975a, 1976a, 1977b, 1985c), Snow, D.W. & Lill (1974), Théry (1990b, 1990c), Traylor (1979), Walther (2004), Whitney et al. (1995), Willis (1984), Worthington (1982a, 1989).