HBW 12 - Family text: Maluridae (Fairy-wrens)

Texto de familia: 

Class AVES
Order PASSERIFORMES
Suborder OSCINES
Family MALURIDAE (FAIRY-WRENS)

  • Small to very small insectivores with long legs and long cocked tail; plumage strongly sexually dimorphic, breeding males with streaked brown or brilliant, largely blue and black, iridescent plumage, females duller.
  • 10–22 cm.
  • Australia and New Guinea, including some islands.
  • Variety of habitats, ranging from tropical rainforest and Eucalyptus forest to savanna, grassland and shrubland, including arid and semi-arid areas.
  • 5 genera, 27 species, 65 taxa.
  • 2 species threatened; none extinct since 1600.

Systematics

In early classifications, many Australian taxa were placed close to groups of Eurasian or African taxa having similar, convergent morphology or ecological roles, and they were labelled as “ flycatchers”, “thrushes” or “warblers”. The fairy-wrens and their allies were given the name of “ warblers” or “wren-warblers”. From the time when they were first described, their affinities remained uncertain for some 200 years until 1985, when the molecular-genetic work of C. G. Sibley and J. E. Ahlquist revealed the existence of an independent radiation of passerine families in the continental mass that gave rise to Australia. Subsequent work by P. G. P. Ericson, F. K. Barker and others has refined the view of passerine radiation, and suggested that the ancestors of the oscine passerines became isolated in the Australian landmass after the break-up of the southern supercontinent of Gondwana. The oscines underwent a significant radiation in the Australo-Papuan region, and it is thought that it is from here that the ancestors of the vast radiation of the “ parvorder” Passerida spread.

No fossil of any malurid is known before the Quaternary, but Australian fossil material identifiable as passerine in origin dates back to the Eocene, 55 million years ago. Fossil material from the Miocene, 15–25 million years old, represents several modern passerine families and supports the DNA evidence of a radiation of old endemic Australian families. The DNA studies indicate that Maluridae is an ancient lineage which has been isolated for 20 million years or longer from its nearest relatives, the honeyeaters (Meliphagidae), the pardalotes (Pardalotidae) and the thornbills (Acanthizidae).

The first malurid described, Malurus cyaneus, was originally named as Motacilla cyanea, meaning “ blue wagtail”, because of its long cocked tail. J. Gould, in his Handbook to the Birds of Australia, published in 1865, did not assign Malurus to any known family, although later taxonomists of the nineteenth and twentieth centuries included them variously with the Old World warblers (Sylviidae) and the Old World flycatchers (Muscicapidae). E. Mayr and D. Amadon recognized a subfamily Malurinae within the Muscicapidae, which included other small insectivorous species from Australia such as the Acanthiza thornbills and the pardalotes. In 1969, C. J. O. Harrison recognized the mutual affinities of Malurus, Stipiturus, Amytornis and the New Guinea genera Sipodotus and Clytomyias, on the basis of a shared interscapular gap (apterium) in the dorsal feather tracts. He regarded them as a subfamily of the Timaliidae, the Eurasian babblers, which at that time included also the Australasian babblers (Pomatostomidae), because these genera shared with the timaliids the behaviour of co-operative breeding. After the early work by Sibley on egg-white proteins had raised doubts about the relationships of Australian passerine species with the families in which they were at that time placed, the 1975 “Interim List” of the Royal Australian Ornithologists Union (RAOU) recognized the fairy-wrens, grasswrens and emuwrens as constituting a family in its own right, the Maluridae. This treatment was maintained in the taxonomic list adopted by the RAOU in 1994, which is based on the taxonomy of Sibley and B. L. Monroe, with some modifications, and was further refined in the recent systematic treatment by R. Schodde and I. J. Mason in their “Directory” of Australian passerines.

In his 1982 monograph of the Maluridae, Schodde listed as distinguishing characters of the family the interscapular gap, the possession of a tail with no more than ten rectrices, the outer pair much reduced or absent, and, at the base of the skull, enlarged tympanic chambers, or “bullae”, formed from exoccipital and parasphenoid bones. The function of these bullae is not clear, but is probably related to sound reception. Further distinguishing osteological features of the palate and the head of the humerus are noted by Schodde and Mason. The maxillo-palatine processes of the palate are long and strap-like in form, and the vomer over which they extend is bifidly horned and resembles that of the meliphagids. On the other hand, the head of the humerus has only two fossae, and these are furrow-like, shallow and unpneumatized, resembling the condition found in the acanthizids.

The Maluridae are divided into two main groups, easily distinguishable by plumage, and regarded as subfamilies by Sibley and Monroe. One of these, the Malurinae, consists of four genera, two of which, Sipodotus and Clytomyias, are monotypic and are restricted to New Guinea, the other two being Malurus, containing the twelve species of fairy-wren, and Stipiturus, with the three tiny emuwrens. The fairy-wrens have brilliantly coloured males, generally iridescent blue, whereas the emuwrens have streaked brown plumage and very long filamentous tail feathers. The second subfamily, the Amytornithinae, contains only one genus, Amytornis, with ten species of grasswren. These are larger than fairy-wrens, and cryptically streaked with black and white on reddish-brown. While fairy-wrens are found in Australia and New Guinea, emuwrens and grasswrens are present only in Australia.

Although past controversy concerning the definitions of species and subspecies was considerable, recent morphological and molecular studies have helped to clarify species relationships in Australia. Nevertheless, further molecular studies are required in order to assess whether more or fewer subspecies are warranted. In New Guinea, variation within species is not well known, museum specimens are few and scattered, and little recent work has been done on the family. Of the two endemic New Guinea genera, the molecular affinities of Wallace’s Wren (Sipodotus wallacii) within the Malurinae have never been determined; the Orange-crowned Wren (Clytomyias insignis) is more closely related to Malurus than to Stipiturus, although it has no blue colours in its plumage. Neither genus has been intensively studied.

The biochemical studies carried out by L. Christidis and Schodde indicate that the twelve species currently making up the genus Malurus include five lineages. Two of these five are single-species groups, one containing the Broad-billed Fairy-wren (Malurus grayi) and the other the Emperor Fairy-wren (Malurus cyanocephalus); both are blue and both are confined to New Guinea. The third, the “chestnut-shouldered lamberti group”, consists of the Lovely (Malurus amabilis), Variegated (Malurus lamberti), Blue-breasted (Malurus pulcherrimus) and Red-winged Fairy-wrens (Malurus elegans) of Australia, while the fourth, the “Australian blue group”, contains the Superb (Malurus cyaneus), Splendid (Malurus splendens) and Purple-crowned Fairy-wrens (Malurus coronatus). The fifth and final of the five lineages identified by Christidis and Schodde is the “bicoloured group” of Australia and New Guinea, incorporating the Red-backed (Melanurus melanocephalus), White-winged (Malurus leucopterus) and White-shouldered Fairy-wrens (Malurus alboscapulatus). Within the two Australian lineages and the one with members in both Australia and New Guinea, species do not generally overlap in distribution, although representatives of all three lineages may occur together.

The two blue New Guinea species, the Broad-billed and Emperor Fairy-wrens, were originally assumed to be close to the blue Australian forms the Superb, Splendid and Purple-crowned Fairy-wrens, but biochemical evidence indicates that they are separate lineages, long distinct. The Broad-billed Fairy-wren is the least known of all the New Guinea malurids. It comprises two non-overlapping subspecies which were initially regarded as two separate species.

Taxonomically, the most difficult of the Malurus species is the bicoloured White-shouldered Fairy-wren of New Guinea, for which as many as eleven or twelve subspecies have been named, although only six are currently recognized. Variations in size and female plumage with habitat and altitude make this widespread species particularly obscure, and no relevant biochemical data are yet available. Schodde suggested that populations of this lowland species along different river systems have diverged in isolation from each other. The Australian species in this “bicoloured group” have more or less non-overlapping distributions, the Red-backed Fairy-wren occurring in the north and east, with two subspecies, and the White-winged Fairy-wren in central and western drier habitats, with two island subspecies and one mainland one. Although White-winged Fairy-wrens exhibit no distinct morphological differences among populations over the whole of inland arid and semi-arid Australia, recent work by A. C. Driskell and others suggests that there are significant genetic differences between eastern and western populations. Ancestral bicoloured fairy-wrens were clearly widespread across the north of the Australian landmass, including the lowlands of New Guinea, where they became isolated during a period when high sea levels separated northern Australia from New Guinea. Schodde has suggested that the other New Guinea species represent early-diverging branches of the malurid line which became isolated in New Guinea rainforests.

Within the Australian “chestnut-shouldered complex”, the position of the northern forms, the females of which have varying amounts of blue in the plumage, has been uncertain in the past. The north-eastern form amabilis, from Cape York Peninsula, in north Queensland, has been treated by various authors as a separate species or as a subspecies of the Variegated Fairy-wren, or been allied with other blue forms of the latter as the subspecies dulcis. Biochemical information, combined with other differences in plumage, has established amabilis as a sister to the M. lamberti group, the two forming a superspecies. The Variegated Fairy-wren includes two northern subspecies, dulcis and rogersi, which are similar to M. amabilis, the Lovely Fairy-wren, in having blue-plumaged females. The contiguous distribution of the Red-winged, Blue-breasted and Variegated Fairy-wrens in south-western Australia is interpreted by Schodde as the result of successive cycles of aridity, each species representing an episode of expansion to the south-west of an ancestral northern Malurus form in a wetter period, followed by isolation of the population in the south-west with increasing aridity. The Red-winged Fairy-wren, in the extreme south-west, exhibits the greater differences from the Variegated Fairy-wren. Within the latter species, one subspecies, assimilis, is widespread over the arid and semi-arid area of inland Australia; each of the other four has a more restricted distribution at the coastal fringes.

The Purple-crowned Fairy-wren is so distinct in plumage that its affinities with the “Australian blue group” were unrecognized until demonstrated by the biochemical studies of Christidis and Schodde. This species does, however, share the relatively larger size and strong complex calls of the Superb and Splendid Fairy-wrens and, biogeographically, its northern distribution complements theirs in the south. The Superb Fairy-wren occupies the south-east corner of Australia, with different subspecies in Tasmania and on the islands in Bass Strait, between Tasmania and mainland Australia. It forms a superspecies with the Splendid Fairy-wren, which is widely distributed across the drier inland areas south of 20° S, with four subspecies differing markedly in the tone of their deep blue overall colour and in the extent of black bands, especially on the back and the pectoral region. These races intergrade where they abut geographically.

Of the three species in the genus Stipiturus, the Mallee Emuwren (Stipiturus mallee) has at different times been included in either the Southern Emuwren (Stipiturus malachurus) or the Rufous-crowned Emuwren (Stipiturus ruficeps). Biochemical data have suggested that it is, in fact, a distinct species, aligned with ruficeps, and having a very restricted distribution in inland south-east Australia, lying between those of its two congeners. The coastal species, the Southern Emuwren, shows considerable variation over a wide geographical range, among populations on islands or isolated in islands of its preferred swampy heathland habitat; as many as twelve subspecies have been described, eight of which are currently recognized. The other inland species, the Rufous-crowned Emuwren, is restricted to the spinifex, or hummock grasslands, of arid central and western regions of Australia; some differentiation occurs across its wide range, but this has been insufficiently studied.

The grasswrens in the genus Amytornis have proved particularly difficult to study. This is a result of both their secretive nature and the fact that many of them occur as small, isolated populations in remote, inaccessible habitats in northern and inland Australia. The majority of the species live where the understorey vegetation is hummock grassland known as spinifex, consisting of grasses of the genera Triodia and Plectrachne (see Habitat). Much confusion stemmed from misidentification of some early specimens from central Australia. Recent researchers have recognized ten species, with a total of 15 subspecies. Schodde and Mason recognized some superspecies groups, namely the Short-tailed (Amytornis merrotsyi) and Striated Grasswrens (Amytornis striatus), the White-throated (Amytornis woodwardi) and Carpentarian Grasswrens (Amytornis dorotheae), and probably the Dusky (Amytornis purnelli) and Kalkadoon Grasswrens (Amytornis ballarae). Relationships among the other species are not yet clear, although the distinctive Grey Grasswren (Amytornis barbatus) certainly appears to be separate from the rest.

Morphological Aspects

The small size, the cocked tail, the preference for dense cover and the grey-brown plumage probably led to the use of the term “wren” in the vernacular names of members of this family. The smallest species are the emuwrens in the genus Stipiturus and Wallace’s Wren in Sipodotus, with a body mass of less than 10 g and, in the case of the Mallee and Rufous-crowned Emuwrens, only 5–6 g. The Malurus fairy-wrens and the Orange-crowned Wren weigh 7–15 g, with most in the region of 9–10 g, while the Amytornis grasswrens are larger, ranging from 17 g to approximately 35 g, these extremes represented by the Eyrean Grasswren (Amytornis goyderi) and the White-throated Grasswren.

Most members of the Maluridae forage mainly on the ground or in low vegetation close to it (see Food and Feeding), and they possess long tarsi and short wings, the morphological features generally associated with this mode of foraging. The wings have ten primaries and are short and rounded, with primaries P1 to P7 similar in length. The length of the tarsus of most species is about 20–25 mm, which is long for passerines in the 10–20 g weight range; only the largest, the White-throated Grasswren, and the two smallest, the Mallee and Rufous-crowned Emuwrens, lie outside this range, having tarsus lengths of, respectively, 28 mm and 15 mm. Purple-crowned Fairy-wrens, which often forage among the spiny fronds of the prickly palm Pandanus aquaticus, have relatively robust feet and tarsi, as also have the Amytornis grasswrens, the most terrestrial members of the family.

The most characteristic morphological feature common to all malurid species is the long tail. With the exception of the three emuwrens, this has ten rectrices, with the outer pair generally reduced. The innermost rectrices are the longest, and, on most fairy-wrens and grasswrens, all of the rest, except the outer pair, are similar in length. On the fairy-wrens, the outermost pair is about one-fifth the length of the other eight, and the outermost rectrix of the grasswrens is about half the length of the rest. Wallace’s Wren exhibits more gradation in length of the inner eight tail feathers, with the outer pair short, as in Malurus. In the tail of the Orange-crowned Wren only four graduated pairs of rectrices are readily apparent, the outer pair being greatly reduced. The emuwrens have what is relatively the longest tail of all the malurids, but with only three pairs of highly modified rectrices. The shafts of the rectrices are strong, but the barbs are sparse and lack the hooks that mesh the barbules together, giving the feathers the filamentous appearance typical of Emus (Dromaius novaehollandiae). The Southern Emuwren has the longest tail with the sparsest barbs. The tail is generally carried cocked nearly at right angles to the body when the bird is perched or hopping, but in flight it trails straight out behind the body, drooping slightly.

Within the fairy-wren genus Malurus, the four members of the “chestnut-shouldered group” (see Systematics) have a relatively long tail for their general size, while the three in the “bicoloured group”, the Red-backed, White-winged and White-shouldered Fairy-wrens, possess a tail that is, by comparison, relatively short. Male fairy-wrens in breeding plumage have a shorter tail than those of females and immature males. Among the Amytornis grasswrens, tail length is generally closely related to body size, except in the case of the Grey Grasswren, which has a particularly long tail. Many Malurus species have a white-tipped tail, but the extent of the white at the tips varies with age, sex and feather wear. These white tips appear to emphasize the tail when the bird is foraging and in general signalling, and they are particularly obvious on the Lovely Fairy-wren.

The other morphological features unique to Maluridae are not immediately obvious. All members of the family have a gap in the mid-dorsal spinal tract of feathers, this being known as the interscapular gap or apterium. This gap is concealed by feathers from the nape and, from each side, by the scapulars, which are long, profuse and often erectile. Unique features of the skull, involving the palate and the auditory region, and the humerus were discussed in detail by Schodde and Mason (see Systematics).

Bill shape varies considerably among the species, reflecting differences in diet and in foraging methods. I. Rowley and E. Russell distinguished three main shapes. A relatively long, narrow, pointed bill, only slightly wider at the base, and with a width greater than the depth, is typical of emuwrens, Australian fairy-wrens and the White-shouldered Fairy-wren of New Guinea. These species are almost entirely insectivorous, and forage mostly on the ground or in low vegetation by gleaning over a surface, probing, and making an occasional aerial sally from a perch. Their rictal bristles are poorly developed. The second type of bill is long, broad at the base, much wider than it is deep along its length, and with the rictal bristles well developed. Such a bill is possessed by the four other New Guinea species, namely the Broad-billed and Emperor Fairy-wrens, Wallace’s Wren and the Orange-crowned Wren; the smallest of the four, Wallace’s Wren, has the largest bill in relation to its body size. All four are insectivorous arboreal foragers, gleaning and striking upwards at prey when in flight. The third bill type, relatively shorter, narrower and deeper, is typical of most grasswrens; the depth of the bill is equal to or greater than its width, and the rictal bristles are well developed. The grasswrens’ diet frequently includes some seeds, and the most extreme bill development is exhibited by the Eyrean Grasswren and the Thick-billed Grasswren (Amytornis textilis), which possess a bill reminiscent of that of a finch (Fringillidae).

Two very different plumage styles are displayed by the family. In the subfamily Malurinae, many male fairy-wrens in the breeding season are brilliantly patterned in iridescent blues and black, these being known as “Bright” males, whereas some other males (“Dull” males), along with females and immatures, are generally brown. Many descriptions of plumages refer to “breeding males”, “ immatures” and “non-breeding males”. With malurids, two brown individuals, one male and the other female, may attend a nest and raise young; they appear to be a breeding pair, and the brown-plumaged male is therefore called a “breeding male”. Since extra-pair copulations are of widespread occurrence among fairy-wrens (see Breeding), however, true paternity is frequently unknown, and it is better therefore to refer to such brown-plumaged males as “Dull” males, in contrast to full-plumaged “Bright” males. The iridescence of the male plumage often makes it difficult to be sure of colours, in particular whether the breast is black, as that of the Variegated Fairy-wren, or navy-blue, as that of the Blue-breasted Fairy-wren, in areas where these two species occur close together, or whether the colour of the crown and ear-tufts is cobalt-blue or violet or somewhere in between. Among the Australian fairy-wrens, most males assume this Bright plumage only in the breeding season, but males of the three New Guinea species exhibit the same plumage coloration and pattern throughout the year. Plumages of male fairy-wrens are predominantly of blue, violet and black, with some tinges of red and white, superimposed on a general body colour of blue, black, or blue with buff or white. The crown, ear-tufts, scapulars and upper back are all significant areas which may be contrastingly coloured against a plainer background, and all are important in displays (see Breeding).

Possession of a blue crown, black face and nape, and erectile ear-tufts is common to the males of most species of all lineages of Malurus except the “bicoloured group” (see Systematics). In the blue and chestnut-shouldered lineages, the crown, upper back and elongated ear-tufts are iridescent blue of various shades, from silvery blue to violet; the crown and ear-tufts, in particular, are significant in threat and breeding displays. In the “chestnut-shouldered group”, rich chestnut scapulars fall over the shoulders of the wing. Although the Purple-crowned Fairy-wren is genetically associated with the Australian blue lineage, it is only the male’s crown that is brilliantly coloured, being deep purple surrounded by a black band through the eye and nape; the rest of the plumage is similar to that of the female. In the bicoloured species, the dark blue or black of the head and body contrasts with the scapulars and upper back, which are red on the Red-backed Fairy-wren and white on the White-winged and White-shouldered Fairy-wrens. These contrasting red or white feathers are erectile and feature in displays (see Breeding). Microstructural analysis of the feathers from the blue, mainland subspecies and the black, island race of the White-winged Fairy-wren suggests that increasing melanin density may mask the blue-producing microstructure of the feathers in the black island race. This difference appears to have a simple genetic basis. In the two endemic New Guinea Malurus lineages, the striking all-blue Emperor Fairy-wren lacks contrasting ear-tufts, whereas the Broad-billed Fairy-wren has the crown blue and black, with the black concentrated in the centre, the ear-tufts blue and elongate, and the overall body colour a paler blue.

Females of the Australian Malurus species are generally greyish-brown above and buff ventrally. They exhibit small differences among species in the colour of the back, lores, ear-tufts and bill, ranging from the Lovely Fairy-wren, with a smoky-blue back, turquoise ear-tufts, white lores and a black bill, to the very plain females of the Australian bicoloured fairy-wrens, which are light reddish-brown above and buff below, with an orange bill and only faint markings around the face and eye. Females of the New Guinea Malurus are more colourful, the female Emperor Fairy-wren having the same head coloration as the male’s, with a red-brown back and a white belly; the immature of this species resembles the female. The female Broad-billed Fairy-wren is similar in pattern to the male, but without blue in the crown and with a white belly. The remarkable variation in the colour of females of the New Guinea bicoloured form the White-shouldered Fairy-wren, discussed in detail by Schodde, is not well understood, although it has been the basis of the description of more than a dozen subspecies at different times. Females of this species may be plain brown as those of Australian bicoloured fairy-wrens, or they can be black and white as the male of the species, although less iridescent and with brown wings, or they can be pied, with a black back and white scapular tufts, brown wings, and whitish breast and flanks blotched with black. Brown females occur in southern New Guinea, in drier regions of eucalypt (Eucalyptus) savanna woodlands similar to those of northern Australia.

In the genus Stipiturus, comprising the three emuwrens, the blue colour is confined to the throat, upper breast and ear-coverts of males, and this becomes visible on five-day-old nestlings. The upper surface is streaked red-brown, similar to that of the Amytornis grasswrens, and the ear-coverts are streaked black, while the ventral surface is unstreaked and varies from reddish-buff to white. Males retain their blue colour throughout the year, and there is no change to a Dull plumage in the non-breeding season.

Also within the Malurinae, the Orange-crowned Wren of New Guinea resembles a rufous version of a female fairy-wren, but with an orange crown. The sexes are indistinguishable in size or plumage. Likewise in the other monotypic New Guinea genus, Sipodotus, the male and the female of Wallace’s Wren are very similar to each other, except for the yellower throat and breast of the female. At first sight they are unlike any other member of the family, but the pattern of a blue-looking crown, albeit black flecked with blue, a black face, a reddish back and a white belly is similar to that exhibited by the female Emperor Fairy-wren, and Wallace’s Wrens have the typical malurine ear-tufts, white and elongated.

The ten grasswrens comprising the subfamily Amytornithinae are more cryptically patterned in reds, browns and buff, disruptively streaked with black and white. The individual feathers can have very complex markings, such as red-brown webs with a paler buff fringe and a white streak edged with black along the central shaft. The dorsal surface is generally more streaked than is the ventral one, and subtle patterns are achieved when one colour, black, white or brown, is emphasized in the streaking of a particular region, such as the ear-coverts, head or breast. Black facial markings are found in some species, the Eyrean, White-throated and Carpentarian Grasswrens sporting a dark malar stripe and the Grey Grasswren exhibiting a complex pattern which includes a black eyestripe. The sexes are generally very similar to each other, except that females have more rufous flanks and sometimes a more rufous belly. Males retain the same appearance through the year, and there is no change to a Dull plumage in the non-breeding season.

During the breeding season, male fairy-wrens develop a sperm-storage structure and testes that, in relation to body size, are among the largest found within the passerines. The sperm-storage structure is a swelling around the vent, and is called the cloacal protuberance. In the breeding season, that of the Superb Fairy-wren has average dimensions of 7•9 mm long by 6•3 mm wide by 5•3 mm high, with a prominent pointed tip at the anterior end. The protuberance changes seasonally, and in winter it is generally not obvious. That of Superb Fairy-wrens in Canberra, in south-east Australia, begins to enlarge, on average, about 3–4 weeks before the first eggs are laid, well after males begin their pre-breeding body moult into Bright plumage; it declines after the breeding season, when the birds start their post-breeding moult. The large cloacal protuberances and testes of fairy-wrens provide large stored reserves of sperm, presumably for extra-pair copulations, which occur with a very high frequency among fairy-wrens (see Breeding). The daily rates of sperm production, sperm reserves and sperm densities are, per gram of body mass, among the largest reported for any of the world’s birds or mammals. Cloacal protuberances are present also in grasswrens, but J. Karubian found that the cloacal protuberance of the Striated Grasswren was relatively much smaller than those of fairy-wrens, and that the number of sperm produced was much lower.

Australian malurids have a complete moult, which starts after the breeding season, often in January, and is complete by April or May in southern Australia. For the lesser-known northern species, the timing appears to be similar. In southern Australia, the mass of contour plumage, and therefore plumage insulation, of Superb Fairy-wrens is greater in winter than in summer. Although the tail feathers are moulted at this time, they may be lost and replaced at any time of the year; the primary moult is descendant. A pre-breeding partial moult of most body feathers occurs in late winter or early spring. In the case of males, the Dull plumage of the non-breeding season is replaced completely or partly by the Bright plumage of the breeding season. The onset of this moult, which lasts for about a month, varies according to age, social status and condition, younger males beginning in August or September and older, more dominant ones starting as early as April in southern Australia. Some adults acquire little or no non-breeding Dull body plumage, appearing instead to moult directly from one Bright body plumage to another. In studies of Superb Fairy-wrens carried out in Canberra, R. A. Mulder and M. J. L. Magrath found that only five of 426 males, fewer than 1%, did this; all five were more than four years old, but not all males older than four years moulted in this way each year. This direct moult from an old Bright plumage to a new Bright one has been recorded also for the Splendid and Red-winged Fairy-wrens.

Females, too, appear to undergo a pre-breeding body moult with no obvious change in appearance. Juveniles may have a complete moult 2–3 months after fledging, but late-hatched individuals often moult only some flight-feathers. Young males in their first year may undergo a partial or complete moult of the body feathers in spring. In the case of the Superb and Splendid Fairy-wrens, first-year males acquire almost complete adult male Bright plumage. In contrast, first-year males of the Red-winged and Blue-breasted Fairy-wrens rarely attain full Bright plumage but, instead, take on a spotty appearance. Those male White-winged Fairy-wrens which do not achieve Bright plumage until they are 2–3 years old may forgo the pre-breeding moult.

Emuwrens, having no alternate Bright plumage in the breeding season, undergo a complete moult after breeding. Although the primary moult is completed in late summer to early autumn, the moult of the body feathers can be spread over a longer time period. The extent and timing of the juvenile moult are not known. Grasswrens probably perform a single complete moult after breeding. The juveniles of these species undergo a partial moult of the body feathers soon after fledging, before the tail is fully grown.

In New Guinea, moult is less seasonal, and can be protracted over a longer period of time. With no alternation between Bright and Dull plumages, there is only a single, complete annual moult.

Habitat

In global terms, Australia is an arid country and its interior is only sparsely vegetated, but, compared with other areas with similar low rainfall, its deserts are well vegetated, with a variable cover of grasses and shrubs. Except for the higher-rainfall areas near the coast, most of Australia is arid or semi-arid, 75% of the landmass having an average rainfall of less than 250 mm, with extended periods during which little or no rain falls. Even the species of tropical northern Australia live in areas with a protracted dry season. Eucalypt forests and woodlands are restricted to higher-rainfall areas of the periphery. Acacias (Acacia) form the dominant component of much of the arid and semi-arid shrubland. The endemic hummock grasses of the genera Triodia and Plectrachne form a prominent understorey in open woodlands and shrublands over about a quarter of the land surface of Australia, and they are the principal habitat of several species of malurid. The hummocks are made up of a mass of repeatedly branching stems with long, spine-like leaves which form a mound up to 1–2 m high and 2 m wide. Hummock grasses are commonly called “spinifex”, because of their resemblance to the coastal grass Spinifex. In the rocky ranges of arid areas, spinifex may be almost the only vegetative cover.

Throughout Australia and New Guinea, the Maluridae are found in habitats ranging from arid shrublands to tropical forests, but most often in habitats with dense vegetation growing to about 2 m in height. The two genera restricted to New Guinea, Sipodotus and Clytomyias, containing Wallace’s Wren and the Orange-crowned Wren, are more often to be found in the subcanopy at up to 40 m in rainforest, while Malurus, Amytornis and Stipiturus species rarely venture more than 2 m above the ground. The habitats occupied by these three genera include shrubby coastal heathland, shrubby or grassy understorey of forest or of savanna woodland, arid shrubland and hummock grassland. In addition, the Purple-crowned Fairy-wren occurs in dense tropical riverine vegetation, the Red-backed and White-shouldered Fairy-wrens in tall tropical grassland, and the Broad-billed Fairy-wren even in the understorey of tall primary rainforest. Two other species of the wet tropics prefer forest edge and other dense habitats, the Emperor Fairy-wren in New Guinea and the Lovely Fairy-wren in north-east Australia.

Habitat requirements of some fairy-wrens are very broad. One such example is provided by the Superb Fairy-wren, a species which is widespread and survives even in some degraded agricultural and suburban developments. Other malurids, especially the grasswrens and emuwrens, are very specific in their requirements, and habitat modifications have led to large decreases in the ranges of some of these species. The grasswrens typically live in arid and semi-arid habitats, often in hilly regions with extensive rocky slopes covered in spinifex. These rocky habitats are a good year-round source of insects, and they also provide plentiful shelter sites that are good thermal buffers on hot summer days and cold winter nights. Emuwrens prefer habitats with dense vegetation up to 1 m tall, especially in the coastal heathlands. These latter habitats are under increasing pressure from urban development.

General Habits

Apart from the brilliant plumage of male fairy-wrens, one of the striking features of the family is that its members are often seen in groups larger than a pair. These spend most of their time, throughout the year, together, moving, feeding and roosting as a group. In the breeding season, the additional individuals assist in the feeding of nestlings and in tending to the needs of fledglings (see Breeding). For this reason, the co-operatively breeding fairy-wrens are the most intensively studied of all Australian passerines.

Some of the early records referred to Superb Fairy-wrens as “mormon wrens”, since many groups appeared to consist of a single conspicuous full-plumaged male and several brown individuals, assumed to be his “harem”. More careful study revealed that many of these brown birds were, in fact, males in Dull plumage (see Morphological Aspects); fully group-members, they had not yet achieved full breeding plumage. The clarification of the plumages of immatures and those of the two sexes enabled the identity of group-members to be understood, and showed that a group was composed basically of a pair and its progeny from previous years. For all species of fairy-wren so far studied, populations have more males than females. This bias arises because dispersal is undertaken mostly by females, which in consequence suffer higher mortality. In the “blue” and “ chestnut-shouldered” fairy-wrens (see Systematics), groups may include several Bright males. Throughout the year, group-members forage together all over their territory, and react together in defence against trespasses by conspecifics or in alarm when predators approach.

In a study of the Splendid Fairy-wren, it was found that groups consisted of two to eight adults, usually three or four, with both male and female helpers. Although most young birds helped for only one year before disappearing, or before becoming established as a breeder in their own group or elsewhere, a few stayed for longer, one for ten years. A few female helpers even attempted to nest, in addition to the primary breeding female. The length of time over which an individual remained as a helper depended on whether an opportunity arose for it to inherit its natal territory, to move to a vacancy elsewhere, or to establish a new territory if a partner was available. Another of the “blue” fairy-wrens, the Superb Fairy-wren, has been studied at several sites in eastern Australia. With this species, population turnover is relatively high and most females can find a breeding opportunity at one year of age, as a consequence of which the groups, of three to five adults, only rarely included female helpers. Moreover, in one study in western New South Wales, no helpers at all were detected during three dry years when very few pairs bred successfully.

Russell and Rowley studied the Red-winged Fairy-wren, which lives in the understorey of tall eucalypt forest in well-watered habitat in south-western Australia. They found that the mortality of breeding individuals is low and that vacancies for new breeders are uncommon, with the result that the groups are larger and fewer pairs have no helpers. In one year, a group included four full-plumaged males of different ages, as well as females and immatures. Many young birds stayed as helpers for longer than one year, female helpers were common, and the age at first breeding was 2–3 years, rather than 1–2 years as with most other fairy-wrens studied. Similar group organizations have been revealed through detailed studies of marked individuals in populations of Purple-crowned, Variegated, Blue-breasted and Red-backed Fairy-wrens.

Among the White-winged Fairy-wrens of the “bicoloured group” (see Systematics), blue males are rarer, and males seldom become fully blue until they are three years old. Other males in the group, subordinate to the dominant blue male, do not become blue unless the blue male disappears or they become dominant in another group. White-winged Fairy-wrens have a more complex group structure, which is still not fully understood. They appear to live in groups containing several breeding females, each with a male partner in brown Dull plumage, and a single Bright male in full blue plumage. The territory area defended by the dominant Bright male and the rest of the group includes the nesting areas of all of the breeding females in the group. The blue male has his own female partner, and she and other females may also have non-breeding helpers, generally brown males. In studies of the black race of the White-winged Fairy-wren on Dirk Hartog Island, off Western Australia, it was found that helpers were few, and that pairs were largely socially monogamous, perhaps owing to their observed lower reproductive success.

For most of the remaining species in the family, evidence of co-operative breeding is weaker, but, for grasswrens, emuwrens and the five malurids in New Guinea, the occurrence of groups at all times of the year suggests that the young of these species probably delay dispersal (see Breeding).

Since most malurid territories are occupied continuously throughout the year, they tend to persist spatially, and populations tend to fluctuate within that framework according to the productivity achieved in successive seasons, and the level of survival. Most fairy-wrens have small clutches, but they may produce more than one brood in the course of a long breeding season (see Breeding). After a good year, the number of immatures reaching the next breeding season will be reflected in the size of the group. Usually, young females stay in the group only briefly, unless they replace their mother. Males, on the other hand, can remain for many years, until they inherit the lead role or, in some cases, move to an adjacent territory to occupy a vacancy there. After a poor breeding season with low productivity and survival, the incidence of pairs may be high. After a good season, there may be few pairs and groups of 5–8 individuals may occur. Most malurids that have been studied are sedentary, resident within the same territories through the entire year and from one year to the next. When all the suitable habitat is occupied, and because male group leadership generally passes to another group-member, territories persist in much the same place for many years.

In continuous habitat, territories form a mosaic, with each territory bounded by those of several neighbours. Fairy-wren territories are commonly 1–2 ha in size, but the areas occupied by the larger, more complex clan groups of the White-winged Fairy-wren are considerably larger, covering up to 6 ha. For the largest grasswren, the White-throated Grasswren, which is also the largest member of the family, territory size has been estimated at 10 ha. The Purple-crowned Fairy-wren, which lives in the vegetation fringing tropical rivers, maintains linear territories, these extending for 200–300 m along one or both sides of a watercourse. In the non-breeding season, the territory boundaries of Superb Fairy-wrens may be relaxed to some extent, resulting in considerable overlap in the areas utilized by different groups. In the highly seasonal and ephemeral tropical grassland habitats of the Red-backed Fairy-wren, the birds may abandon their territories and congregate in thick vegetation along rivers during the dry season.

Most of the behaviour relating to territory defence takes the form of song battles, especially early in the morning and in the evening. Song given near a boundary is vigorously answered. All members of a group, including females, sing in defence of the territory, and, if trespass does occur, the intruders are vigorously threatened and chased. Intruding females are chased by the breeding female of the territory invaded. A threatening male holds his body feathers and ear-coverts partly erected, his wings away from the body, and his head and tail lowered. R. Hutton saw a similar threat display given by a captive Striated Grasswren, in this case followed by a short “attack flight”.

All members of the family are predominantly insectivorous (see Food and Feeding). In the winter months, when insects are less active, the group spends most of each day in foraging, moving throughout the territory. In summer, when insects are abundant and more active, and the daytime is longer, there is time for periods of rest, especially in the middle of the day. The group seeks shade, and the individuals perch side by side, each in body contact with its neighbour, while preening themselves and each other, jostling for a position in the middle of the line. Many of the grasswrens live in arid, often rocky habitats, and they are forced to seek shade under rocks or in crevices, and have been recorded as using even rabbit (Oryctolagus) burrows for this purpose.

At the end of the day, after a period of intensive foraging, actively interspersed with territorial song, the group moves to a preferred dense shrub. Here, its members will spend the night, perched in physical contact with one another along a horizontal branch.

Voice

The vocal repertoire of Australian fairy-wrens and emuwrens is well known, but the species from New Guinea are not so well known, with few descriptions or recordings. Some vocalizations have been recorded for all of the Australian grasswren species, but the full range and context of these is not always clear. Rowley and Russell provided illustrated sonagrams of the songs of most species in the family. Fairy-wrens and emuwrens have a clearly recognizable song which is similar in basic structure for all species. The song of the grasswrens is quite different and much more varied in structure. All members of the family have a range of contact and alarm calls, in addition to song.

Fairy-wren and emuwren song is usually spoken of as a reel. It starts with a few short, high-pitched introductory syllables, descending into a trilled section in which various syllable types are rapidly repeated in strings. The whole song may last for 2–5 seconds, or even longer in the case of some species, and it has a mechanical character, rather than a melodious, musical one. The rate of repetition of the individual syllable types is so rapid that, in most species’ songs, they cannot be distinguished, and the human ear hears a continuous sound that rises and falls in pitch. Humans are unable to register differences between the songs of individual birds of the same species, but analysis of sound spectrograms shows that the basic syllables or elements of the songs differ from one individual to another. In studies of the vocalizations of Splendid, Superb and Red-winged Fairy-wrens, playback of recorded song from familiar conspecifics and strange ones has demonstrated that the birds can distinguish strangers from group-members on the basis of song. In the song of the Purple-crowned Fairy-wren, in which the elements are repeated more slowly, it is possible to hear individual differences in rhythm, such as “cheepa-cheepa-cheepa” and “ ricketty-ricketty-ricketty”.

In the case of fairy-wrens, all adult group-members, including females, sing, and most song is heard early in the morning, just before and during the breeding season. A song from one territory is generally answered by another from a neighbouring territory. Song is probably an important way of defending the territory and of advertising the fact that it is occupied. In the complex promiscuous mating system of fairy-wrens, in which females seek extra-pair matings by visiting males in nearby territories (see Breeding), song probably also allows females to identify and locate individual males of their choice. With most species, individuals sing alone; even when more than one individual responds to playback of a recording, this does not give the impression of an organized chorus. The territorial song of the Purple-crowned Fairy-wren, however, is more co-ordinated, and the male and female fly to a conspicuous perch and sing there together in a loud reply to a neighbour’s song or to playback of a recorded song.

Among Australian fairy-wrens, the loudest songs are those of the “blue group”, the Superb, Splendid and Purple-crowned Fairy-wrens. The individual syllables have a complex structure which gives the overall song a harsh, strident quality; it is also lower in pitch than are the songs of other species, and it can be heard by most people. The songs of emuwrens are weak and high-pitched, and therefore inaudible to many human listeners. In between these two extremes, the songs of all four members of the “chestnut-shouldered group” of fairy-wrens (see Systematics) are similar to one another in quality, and have simpler elements, a less harsh character and a higher pitch than do songs of the “blue group”. The songs of the “bicoloured group” have a simple syllable structure, and that of the White-winged Fairy-wren is typical; this consists of a very characteristic undulating trill, which has been described as musical, but has also been likened to the sound made by a winding fishing reel or a squeaking cart wheel. All the songs may be delivered at what seems to be maximum volume in some contexts. In others they are given very softly, more or less in the manner of a contact call, apparently intended for other group-members which are nearby.

Fairy-wrens feed and move through their territories as a group, and their contact calls are an obvious sign of their presence to anyone trying to locate a group somewhere in 1-2 ha of thick cover. The calls are generally soft and brief, given singly or in a series, and can vary from the strong “chet” of the Purple-crowned Fairy-wren to the very soft “tsi” of the Blue-breasted Fairy-wren, one of the quietest species in the family. Alarm calls are a loud and noisy “chit” or “ zit”, uttered singly or, if a threat continues, repeated, sometimes in a harsh chatter. Various other calls have been described, in association with such other aspects as copulation, brooding, and the feeding of young, but there is little specific information on the nature of these calls and their contexts.

The songs of grasswrens are very different from those of fairy-wrens and emuwrens. They are generally more complex, varied and melodious. The exception is the Grey Grasswren, the only song of which appears to be a series of ringing, high-pitched metallic notes, transcribed as “tsit tsit tsit”. Other grasswrens produce a much more complex song, incorporating whistles, trills, buzzes and melodious phrases in a long and varied sequence. Despite their generally secretive nature, the songs of grasswrens, given at dusk or in the early morning, are often delivered from an exposed perch or rock. Contact calls of grasswrens consist of a variety of very high “seeet” notes and soft twittering trills, too high to be heard by many people, and with the source very difficult to pinpoint. Other, more noisy alarm calls are also described for some of the ten species, but the full range and context of the vocalizations of grasswrens are yet to be studied.

As indicated at the start of this section, the songs of the New Guinea malurid species have not been studied, and few recordings are available. The White-shouldered and Broad-billed Fairy-wrens have songs similar to those of their Australian congeners in Malurus. The song of the Emperor Fairy-wren is more complex and variable, and incorporates some more melodic syllables likened to those in the song of a European Common Blackbird (Turdus merula), in addition to the typical Malurus reel. Wallace’s Wren and the Rufous-crowned Wren are poorly known vocally, apart from the fact that they emit a variety of high-pitched sibilant “see” contact calls as they move rapidly overhead during foraging.

Food and Feeding

The Australian fairy-wrens, emuwrens and grasswrens spend most of their life within one metre or so of the ground. They are mainly insectivorous, but for some, especially the grasswrens, seeds and fruits can be a significant component of the diet at some times of the year. Emuwrens are almost entirely insectivorous.

Dietary studies have revealed that malurids consume a great variety of insects and spiders (Araneae). R. D. Barker and W. J. M. Vestjens, in their survey of all the studies of food eaten by Superb Fairy-wrens, listed items from 40 insect families. Ants (Hymenoptera) are a very abundant component of the Australian insect fauna and form a very significant item in the diet of fairy-wrens, particularly as an important “fall-back” or reserve component in winter, during drought and after fire. The present authors watched as Splendid Fairy-wrens, after an intense fire had destroyed their habitat, foraged along a trail of worker ants, systematically collecting the ants as they ferried their loads back to the colony. When they are feeding nestlings or fledglings, adult fairy-wrens eat very small items themselves, and carry larger ones, such as grasshoppers and crickets (Orthoptera), cicadas (Cicadidae), larger moths and caterpillars (Lepidoptera), back to the nest. The small fairy-wrens carry only single food items, but the larger grasswrens can carry several items at once.

It is often suggested that fairy-wrens ingest seeds only by accident, but in some specimens so many seeds have been found that the ingestion of them must surely have been deliberate. Some seeds are probably present in the stomach as a result of the individual having eaten the small fleshy fruits produced by many shrubs of arid and semi-arid regions.

Grasswrens consume significantly more seeds than do fairy-wrens, seeds accounting for perhaps up to half of their diet. Seeds are one dietary component that persists in an arid environment long after the good conditions that produced them have passed. The small seeds of the various species of Triodia, the hummock grass, or spinifex, that is characteristic of the habitat of all except three species of grasswren (see Habitat), are one of the important components in their diet, as also are acacia seeds. The relatively fine bill possessed by most grasswrens is well suited for the small grass seeds of Triodia. In the case of the Eyrean Grasswren, the large hard seeds of the dune grass Zygochloa paradoxa are an important food, dealt with by this species’ stout finch-like bill. The habitat occupied by the Thick-billed Grasswren is arid shrubland, rather than spinifex grassland, and this species’ bill, heavier than that of the spinifex-dwelling grasswrens, enables it to exploit the larger seeds and fruits of a diverse range of shrubs, including those of the genera Chenopodium, Atriplex and Rhagodia.

For all malurids, the most common feeding movement on the ground is the “hop-search”. This involves a hop with the two feet together, then a pause in order to probe litter or to glean from a leaf or stem, and then another hop or two and a repeated search. When feeding above ground in a shrub, these birds use a similar pattern, hopping from branch to branch and gleaning from bark or leaves.

Most Australian malurids forage on or near the ground, generally in or below dense vegetation such as shrubs and grasses, either in the understorey of forest or woodland, or in heathland where trees are few. Dense clumps of introduced plants such as blackberry (Rubus fruticosus) are a favourite feeding place of some fairy-wrens and emuwrens in less arid habitats, and fallen trees are popular as foraging sites among forest and woodland species. The birds rarely venture far from cover or from other members of their group. The Purple-crowned Fairy-wren forages slightly higher up, in the prickly Pandanus palm fringing tropical rivers, probing among the detritus that collects in the leaf axils and gleaning insects from the fronds and bark. The least terrestrial of the other Australian malurids is the Lovely Fairy-wren, which seeks its food at up to about 5 m above the ground in the thick shrubs and vines of tropical rainforest margins, in regrowth vegetation and even at the edges of mangroves. Most malurids forage as a group, keeping in vocal contact with one another as they move. They spend little time in the open, moving rapidly through open vegetation, and emerging from cover only briefly in order to forage at the edges of shrubs. They sometimes feed temporarily with roving mixed flocks of small insectivores, joining such a flock as it moves through their territory and leaving it at the boundary.

Grasswrens are even more terrestrial than are fairy-wrens and emuwrens. Indeed, most observations of foraging grasswrens refer to these birds as feeding on the ground in the litter that accumulates under shrubs or clumps of Triodia. In arid areas, they run across the open ground between the clumps of vegetation, moving so fast that the human observer sees only a blur.

The New Guinea species are less well known. One of them, the White-shouldered Fairy-wren, is similar in morphology to the Australian fairy-wrens and it forages in a comparable manner, by gleaning from low vegetation and from the ground. H. L. Bell studied the foraging behaviour of rainforest birds in New Guinea, and recorded that the Emperor Fairy-wren spent about one-third of its foraging time on the ground or among litter and two-thirds of the time in gleaning on leaves and branches up to 2 m above the ground. The other New Guinea species, the Broad-billed Fairy-wren, Wallace’s Wren and the Orange-crowned Wren, display increased adaptation to arboreal foraging, having shorter tarsi and a shorter tail (see Morphological Aspects). They forage at 2–10 m above the ground, where they move through the canopy of tall undergrowth below tall forest with its tangle of vines and rattans, while gleaning from leaves, probing in cracks, and flying to snatch prey from a surface or to hawk flying insects.

Breeding

Although the nest-sites and eggs have been described for all members of the Maluridae except the Orange-crowned Wren, little more is known about the breeding biology of most species. All those for which the relevant information is available build a domed nest and lay relatively small clutches of 2–4 eggs. A few species of Malurus have been the subject of long-term studies, and their breeding habits are probably better known than are those of any other Australian passerines. What little is known of the more secretive emuwrens and grasswrens suggests that many aspects of their breeding habits are similar to those of fairy-wrens.

All species in the family appear to be group-living, and probably all of them also engage in co-operative breeding. Co-operative breeding is characterized by the presence of breeding and non-breeding mature individuals in a social unit, or group; the non-breeders are usually referred to as “helpers”, and they contribute to the reproductive effort of the breeding female by feeding nestlings and by caring for fledglings after they leave the nest. Helpers also assist in the defence of the territory and in attacking or distracting predators or parasitic cuckoos (Cuculidae).

Detailed studies of colour-ringed individuals have identified co-operative breeding by ten species, and, for almost all others, group-living and reports of helpers at nests indicate that they, too, are probably co-operative breeders. All of the Australian fairy-wrens breed co-operatively. Although, for most of the remaining species in the family, the evidence for such behaviour is weaker, the grasswrens, the emuwrens and the five malurids in New Guinea do occur in groups at all times of the year, which suggests that the young of these species probably delay dispersal. In the case of the White-shouldered Fairy-wren and Wallace’s Wren in New Guinea, more than two individuals have been seen to feed chicks at a nest. In Australia, helpers have been recorded at nests of the Kalkadoon, Black (Amytornis housei), Grey and White-throated Grasswrens, and breeding pairs of Thick-billed and Striated Grasswrens occasionally have a third bird with them, suggestive of co-operative breeding. A study of colour-ringed Southern Emuwrens has clearly revealed the existence of this behaviour, some male offspring delaying breeding and, instead, helping to raise younger siblings.

For many years, groups of fairy-wrens were regarded as consisting of typical monogamous pairs with their non-breeding offspring. Rowley’s studies of the Superb Fairy-wren in Canberra, in south-east Australia, and of the Splendid Fairy-wren in Perth, in Western Australia, provided the first hints that a very different breeding system was operating. Rowley found that the “ Petal-carrying” display recorded in many older reports was performed most often in the presence of a female by a male intruding into another’s territory. He initially suggested that, because the intruder was rarely chased by the resident male, the display had an appeasement function. Further observation of Splendid Fairy-wrens in Western Australia and of Superb Fairy-wrens in Canberra, however, established that in the winter months, before nesting begins, the females are frequently visited by displaying males from other territories, and this phenomenon continues throughout the breeding season. Although this suggested the possibility that females might mate with these intruding males, copulations are observed so rarely that the only way in which to investigate the significance of these courting intruders was to determine the parentage of the young produced by the females. In Western Australia, more than half of all the Splendid Fairy-wren young produced were fathered by males from outside the female’s social group. In another population of the Splendid Fairy-wren, in South Australia, the use of DNA “fingerprinting” to investigate parentage has confirmed a high level of extra-pair paternity, ranging from 24% to 52%, with an average of 42%. This population was at a lower density that that in the Western Australia study. Most offspring were sired by dominant breeding males, but helper males sired about 25% of all extra-pair young. This was most likely to occur when the helper was unrelated to the breeding female, suggesting that incest avoidance constrained mating between helper males and their mothers. A long-running study of the Superb Fairy-wren population in the Australian National Botanic Gardens, in Canberra, undertaken by A. Cockburn and a succession of students and colleagues, has also used DNA-fingerprinting methods to show that, despite the observed monogamous social system, Superb Fairy-wrens have a promiscuous mating system, with extreme infidelity between male and female partners. Not only were 76% of all offspring sired by males outside the female’s social group, but most of these young were sired by only a small number of preferred males; a few were sired by subordinate males in those groups with preferred males. The females’ social partners were, therefore, feeding young to which they were not related.

Although extra-pair copulations were clearly frequent, they were hardly ever observed. In the Canberra study, M. Double and colleagues attached tiny radio transmitters to female Superb Fairy-wrens when they were building a new nest. These females were tracked from before sunrise; they were found to leave their territory and travel directly to a nearby territory, pause briefly, unseen in the dim light, and then to return home just before sunrise. Females made these excursions only in the few days before egg-laying, when they were at their most fertile. DNA analysis indicated that the young that hatched from the eggs laid after these excursions had been fathered by the male of the territory visited by the wandering female.

As most male fairy-wrens look equally beautiful in their blue Bright plumage (see Morphological Aspects) in the breeding season, the Canberra workers looked for what it was that made a few particular males so attractive to females. They discovered that the preferred males shared one distinguishing characteristic. They moulted from the brown Dull plumage into the blue plumage well before the start of the breeding season, and throughout the winter these few Bright males were visiting nearby territories, and there displaying to females. Blue males had elevated levels of the male sex hormone testosterone, which is known to have the side effect of making males more subject to diseases and parasites. In studies by A. Peters, however, it was found that those males that were blue in winter showed no sign of the adverse effects of high testosterone levels, suggesting that they were in particularly good condition. Females that chose to mate with these males were choosing the best possible fathers for their offspring.

A similar high level of extra-pair mating has been established from DNA-fingerprinting studies of the Red-backed Fairy-wren in northern Australia and the White-winged Fairy-wren in Western Australia and South Australia. The size of the cloacal protuberance (see Morphological Aspects) of other Australian fairy-wrens and the recorded occurrence of Petal-carrying displays by most species suggests that they, too, have a similarly promiscuous mating system. The mating systems of emuwrens, grasswrens and the New Guinea malurids have not been investigated. Karubian, however, found that male Striated Grasswrens had a relatively small cloacal protuberance and produced lower numbers of sperm, suggesting, for this species, at least, that high levels of extra-pair matings do not occur and that a monogamous mating system is more likely.

Although extra-pair copulations are not uncommon among birds in general, the high frequency of such events found in fairy-wrens is rare. Many reasons why females engage in extra-pair copulations have been put forward, but so far no explanation has received convincing support. One suggestion is that a female mates with a male of higher “quality” than her social partner. Others are that she benefits by increasing the genetic diversity of her brood, or by increasing the chances of genetic compatibility with at least one of her partners. There is some support for the view that extra-pair mating is a means of incest avoidance for species that do not disperse far, and some evidence exists that group-living males that have mated with the group’s female are likely to contribute more to the feeding of her brood, so that the female gains a direct benefit from the extra-pair copulations.

The timing of breeding among Australian malurids varies considerably, being influenced mainly by the latitude and by seasonal distribution of rainfall for a particular population. Fairy-wrens in southern Australia breed in spring and early summer; few clutches are laid before September or after the end of December. In more northern, subtropical populations with summer rainfall, breeding may start as early as July or August. With winter rainfall in south-western Australia, Red-winged Fairy-wrens and southern populations of Blue-breasted Fairy-wrens rarely lay before October, and hot dry summers generally render conditions unsuitable for breeding after December. A population of the Blue-breasted Fairy-wren only 150 km to the north-east began to breed about one month earlier, in August. In arid regions, breeding is influenced by rainfall. If conditions allow, seasonal spring breeding occurs, from July onwards depending on latitude, but breeding can take place at any time following adequate rainfall. Rowley and Russell found that the White-winged Fairy-wren breeds in spring in coastal habitat in south-western Australia, which has regular winter rainfall, but in more arid regions this species has been observed to nest in all months except June. After exceptional rainfall, it may continue to breed for as long as conditions allow.

Even those species living predominantly in tropical summer-rainfall areas, the Red-backed, Purple-crowned and Lovely Fairy-wrens, exhibit a peak of breeding activity in spring and early summer, although there is some breeding in most months, especially in autumn after summer rain. In studies of the Purple-crowned Fairy-wren, Rowley and Russell found that a good summer wet season was followed by breeding in the autumn, from March to mid-May, and the spring, from July to September, with little, if any, nesting from mid-May to mid-July. In a year with a very poor wet season, however, no breeding was attempted in the subsequent autumn or spring. Although not well studied, the Lovely Fairy-wren of tropical north-eastern Australia appears to breed in most months, including the wet season of December and January. Even in arid areas with regular summer or winter rainfall, the length of the breeding season may be protracted or curtailed, depending on the amount and timing of rainfall.

The few available data on the reproductive behaviour of emuwrens and grasswrens suggest that these two genera have similar breeding seasons to those of fairy-wrens occurring in the same habitat. Thus, the Southern and Mallee Emuwrens breed in spring and early summer, and the Rufous-crowned Emuwren breeds at similar times to the White-winged and Variegated Fairy-wrens living in the same arid areas. Even less is known about breeding seasons of grasswrens, the few records suggesting that the northern species nest during the wet season, generally from November to March or April, and possibly for as long as conditions are suitable. Records for grasswrens in arid inland regions of Australia indicate that breeding takes place in spring if conditions allow, or at other times following good rain. B. Brooker’s study of the Thick-billed Grasswren at Shark Bay, at 25° S in Western Australia, where the average May–August rainfall is 200 mm, revealed that eggs were laid in July–September. In the Striated Grasswren population studied by Karubian in inland South Australia, where rain falls predominantly in winter, laying took place from August to December.

Little is known of the breeding habits of most of the malurids in New Guinea. Here, some rain may fall in all months, but the wettest period is from November to April. Information for other insectivorous birds suggests that most breeding occurs before or after the peak of the wet season, although Bell’s study of a lowland rainforest community near Port Moresby recorded some breeding in all months of the year.

For the majority of fairy-wrens, the breeding season is long enough for pairs to produce two broods, or sometimes even three, or to rear the young of a parasitic cuckoo and still have time for two broods of their own. The shorter reproductive season of the Red-winged Fairy-wren in south-western Australia allows fewer females to raise more than one brood; females which failed with the first nest had sufficient time to nest again, but only those females which began nesting on an earlier date than the average had time for a second brood after a first succeeded. A female may be able to produce more broods within the limits of the breeding season if helpers are present in the group, as these can take over the care of some of the fledglings. In the study by Rowley and others of the Splendid Fairy-wren in south-western Australia, it was found that, in groups with two or more helpers, females laid their next clutch almost two weeks after the previous brood fledged or even sooner, whereas females with no helpers looked after fledglings for about one month before laying their next clutch.

The displays performed by intruding males before and during the breeding season exploit the bright colours of their iridescent plumage, including the erectile contrasting feathers of the scapulars, mantle and nape. Those males which possess elongated erectile ear-tufts incorporate these, too, in displays. A common element in many displays is the “Face-fan”, when the elongated iridescent ear-tufts are erected, projecting on each side of the head. The malurid displays which have been most closely observed are those of the Superb and Splendid Fairy-wrens, and they have many features in common, as summarized by Rowley and Russell for the Splendid Fairy-wren and by Mulder for the Superb Fairy-wren. Only very rarely are these displays given by a male to a female in his own territory. Displays are given only by males in Bright plumage (see Morphological Aspects). In a typical sequence, a male Superb Fairy-wren approaches a female while performing the Face-fan, and holding the tail lowered, the crown feathers flattened and the feathers of the nape and mantle fluffed out; he twists from side to side, displaying the contrasting colours of his plumage in the “Blue and Black” display, the term used by E. Bradley and J. Bradley in 1958. R. Hutton named a similar display that she saw performed by the Splendid Fairy-wren in aviaries as the “Lizard” display, because the erectile feathers around the neck were reminiscent of the frill of a frill-necked lizard (Chlamydosaurus kingii). When departing from a territory after displaying to a female, a male often continues his display as he flies off, flying with his body held erect and the tail lowered. Rowley, having initially referred to this as “Impeded flight”, later termed it “Sea Horse flight”, because of the vertical body posture like that of the marine fish the sea-horse (Hippocampus).

Those male fairy-wrens which have the scapulars and mantle feathers of a striking colour, and sometimes elongated, make full use of these plumage parts in displays. In the case of the Red-backed Fairy-wren, which has no contrasting ear-tufts, the red feathers of the mantle and scapulars cover the lower back and uppertail-coverts, and are erectile, so that, when raised in display, they make the male look like a red-and-black powder-puff, especially striking in the Sea Horse flight. In another display, male Red-backed Fairy-wrens hold their primaries close together and depress the wings slightly, thereby exposing the striking red colour of the back, while at the same time erecting the black crown feathers. Displays of White-winged Fairy-wrens emphasize the contrast of the overall dark blue plumage with the brilliant white of the scapulars, secondary wing-coverts and innermost flight-feathers, both in display-flights and in perched displays, when the back is exposed.

In some of their displays to females, intruding males augment the exhibiting of the plumage with the striking addition of a coloured petal. Eight of the nine Australian fairy-wrens have been observed to perform this display, in which the male plucks and carries a flower petal or fruit of a colour that contrasts with its own plumage. Most such records refer to the Superb Fairy-wren, which carries yellow petals, and the Splendid Fairy-wren, which carries pink, blue or purple ones, with an occasional yellow or white petal. Other species observed with petals are the Red-winged Fairy-wren, with yellow and white ones, the Lovely and Variegated Fairy-wrens, both with yellow ones, the Blue-breasted Fairy-wren, carrying orange petals, and the Red-backed and White-winged Fairy-wrens; of these last two, the former has been seen to carry orange, red and yellow petals and red berries, and the White-winged to carry purple, blue, yellow and white petals. Displays are seen mainly during the breeding season and the two preceding months. Mulder scored petal-carrying in about 20% of displays to females by male Superb Fairy-wrens. The resident male was present in about one-third of visits by displaying intruders; he sometimes chased the intruder out of the territory, but in many cases made no attempt to do so. Females generally show little interest in the displaying visitor, and avoid him if he pursues. Displays by intruders do not generally lead to copulation. Studies of the Red-backed Fairy-wren found that most Petal-carrying displays are directed at a female, suggesting that they have a courtship function.

Copulation by malurids is rarely witnessed by human observers, but the few accounts, relating to a number of species of fairy-wren, indicate that it is brief, with little or no preceding display. The female soliciting copulation crouches, and flutters her wings. If the male is not close by, she sings before soliciting, at which the male immediately approaches and mounts her. Coition takes only a few seconds, and it is easy to understand how a female seeking extra-pair copulation can visit a male in another territory, copulate and return home very quickly; in the dim early-morning light, when such forays occur, females can presumably identify and locate individual males by their song. A similar wing-fluttering display is also given in other contexts, by a juvenile begging food from an adult, by a helper giving way to an older individual when approaching a nest with food, or by a young male to an older male.

Few displays by emuwrens and grasswrens have been recorded. Hutton described a captive male Striated Grasswren as feeding the female frequently during the week before mating and egg-laying. When the female is ready to accept mating, she wing-flutters; the male hops up and down briefly in front of her, and then mounts. Courtship and mating by the Southern Emuwren follow a similar course. In addition, feeding of the female by the male Thick-billed Grasswren has been documented.

With the single exception of the Orange-crowned Wren in New Guinea, the nests of all members of the family have been described. The majority of malurids build a domed nest, taller than it is wide, with a slight hood over the entrance, which is about two-thirds of the way up on one side. Among fairy-wrens, the largest nests are those of the Purple-crowned, which are, on average, 15•2 cm high and 10 cm wide, and are placed about 90 cm above ground; these bulky nests look like a pile of flood debris in Pandanus palms growing at the edges of tropical rivers. Unsurprisingly, the smaller species build smaller structures, such as the neat, tightly woven nests of Variegated and White-winged Fairy-wrens, which measure about 10 cm by 6•5 cm, and are sited about 20 cm above ground in low shrubs. The nests of emuwrens are similar in size and structure to those of the smaller species among the fairy-wrens. Some grasswren nests are fully domed, as those of the Striated, White-throated, Carpentarian and Kalkadoon Grasswrens. Others are more variable, as is the case with the Thick-billed Grasswren, some nests of which are almost an open cup, others having a partial hood and still others a complete hood with a side entrance. The biggest malurid nests are those of the larger grasswrens, such as the White-throated Grasswren. The nest of this species has dimensions of about 16–18 cm by 12–14 cm.

Few nests of any malurid species are built more than 1 m above ground, where they are placed up to 0•5 m from the outer surface of a shrub or bush. Their openings often face in a direction away from the afternoon sun or from prevailing winds or rain. Nests are frequently placed higher above ground later in the breeding season, when the weather is usually hotter. In the case of the Blue-breasted Fairy-wren, for example, those in September–October were close to the ground, about 25 cm up, in piles of dead brush, whereas nests built after mid-November were about 62 cm above ground within the canopy of live green bushes. At any one site, the majority of any malurid’s nests are built in one or two favoured species of shrub. Superb Fairy-wrens inhabiting agricultural and urban areas utilize a variety of exotic shrubs and weeds such as blackberry, and hawthorn (Crataegus). Emuwrens and grasswrens living in arid habitats nest within mature spinifex clumps, which afford great protection, while the coastal emuwrens often use clumps of sedges and tussock grasses.

Nest-building is undertaken solely by females, which at this stage are often accompanied by their male partner, and are visited by intruding males. The construction is begun with a framework of grass and strips of bark, held in place with spider web. Walls and a roof of finer grass and fibre are added, and the nest is lined with fur, feathers and vegetable down. Some of the species inhabiting high-rainfall tropical regions sometimes incorporate moss into the roof, which may help to keep the nest dry.

The eggs of all malurids are very similar to each other in appearance. They have a tapered oval shape, and in colour they are dull white or pinkish-white, spotted with red-brown, the spots especially evident at the blunter end. The most densely spotted eggs are found in clutches of the Grey, Thick-billed and Dusky Grasswrens. In contrast, those of the Black Grasswren are the least patterned. Egg size increases with the body size of the female, and ranges from an average of 15 × 11 mm, as with the smallest emuwrens, to 22 × 16 mm, as the eggs of the Black and White-throated Grasswrens. An even smaller egg, 13•5 × 9•5 mm, has been described for Wallace’s Wren, but so far only one measurement of an egg known for certain to have been laid by this species has been recorded. The eggs of the small emuwrens are large in relation to body size, corresponding to about 20% of the female body mass; by comparison, those of the largest grasswren, the White-throated, are only about 9% of the female body mass.

Clutch sizes are small and show little variation. Most Australian fairy-wrens and emuwrens lay clutches of two to four eggs, with three the most frequent number. No four-egg clutches has been recorded for the Purple-crowned, Blue-breasted and Red-winged Fairy-wrens, and, among Red-winged Fairy-wrens, clutches of two eggs are slightly more frequent than are clutches of three. In the case of White-winged Fairy-wrens, clutches of three and four were equally frequent in coastal south-western Australia, and studies in other areas have reported a mean clutch size of slightly greater than three. Among the emuwrens, the larger, coastal Southern Emuwren lays two to four eggs, with three the most common clutch size, whereas the tiny Rufous-crowned Emuwren of the arid interior most frequently lays two eggs but occasionally three. Few of the grasswrens have been studied in detail, but information from egg collections suggests that, in this genus, clutches consist of two or three eggs, with two the more common number. For the Thick-billed Grasswren at Shark Bay, in Western Australia, Brooker found that the mean clutch size was 2•7, with three eggs the most frequent. Among the little-known New Guinea malurids, clutches of two eggs are usual.

The clutch size of fairy-wrens so far studied varies little during the course of the breeding season, and exhibits no sign of variation with latitude. Blue-breasted Fairy-wrens, however, usually lay a first clutch of three eggs, but clutches laid later in the season or in years when rainfall during the breeding period is low may consist of only two eggs.

Fairy-wrens and emuwrens lay at intervals of one day. Although it has been assumed that the same applies to grasswrens, the Thick-billed Grasswren has been reported as laying at two-day intervals.

Among the fairy-wrens, only the female incubates, starting when she lays the last egg of the clutch. She sits facing the entrance of the domed nest, and, in the confined space, the tips of her long tail feathers develop an obvious bend to one side or the other. Bouts of incubation last for about an hour. The female may leave the nest in response to a song by the male; alternatively, she may leave, and he will appear in response to her song. During recesses from incubation, she forages very urgently, without pause, close to the male, while he remains vigilant and forages little. The female moves away from the nest as she forages, and her return to it is made in a long, low, direct flight. The incubation period of fairy-wrens is typically 13–15 days. In the Splendid Fairy-wren study carried out by Rowley and others, the duration of incubation ranged from 12–13 days to 16–17 days, most clutches hatching on the fifteenth day of incubation. For most species, incubation in the summer months of December and January is about one day shorter. In captivity, the male has been seen to bring food to the female on the nest, but this has not been recorded in the field.

Aviary observations of emuwrens and grasswrens suggest that the males occasionally incubate for a short period. Field observations confirm this.

Nestling fairy-wrens hatch within 24 hours of each other. They are initially naked and blind, and red in colour, darkening within one day and becoming blue-grey as the feathers develop beneath the skin. Chick growth is rapid, and on the third day the primary feathers begin to protrude through the skin. The eyes are usually open on the sixth day, when the primaries start to burst from their sheaths and the rectrices begin to emerge; by the seventh day the body feathers, too, are starting to emerge. By the time when they leave the nest, at 11–12 days of age, the young are well feathered but the tail is still very short, less than one-quarter of its final length, which is not attained until 40–45 days after hatching; the wings too, are not fully grown, and they continue to grow for a further ten days. The young, on leaving the nest, can fly only weakly, and for 7–10 days they remain well hidden in thick vegetation until the wings have finished growing, the tail is about half-grown, and they are able to fly better. No sexual differences are apparent at this stage.

Grasswrens are covered with charcoal-grey down on hatching. Other aspects of their growth and development have been little studied, but reports suggest that the chicks leave the nest 10–15 days after hatching and remain cryptically concealed for about one to two weeks thereafter. It is not clear whether male and female grasswrens are distinguishable from one another at this early stage of their life. Emuwrens hatch with tufts of blackish down on the head, shoulders and tail. In the case of the Southern Emuwren, differences between the sexes are apparent in the nest; the male’s throat is pale blue-grey and the female’s is reddish-brown.

Once malurid nestlings hatch, the male and female, together with any helpers associated with the pair, begin to carry food to the nest. At first, the breeding female broods the nestlings almost continuously, and she takes food brought to the nest, eats some of it and passes some on to the nestlings. Later, when more of the chicks’ feathers have emerged, she leaves the nest for increasingly long periods, and attendants deliver the food directly to the young and remove their faecal sacs. The contribution of individual group-members is very variable. The breeding female usually makes the greatest contribution; if she has only one male to help her, however, he may contribute nearly as much as she does. Helpers can reduce the amount of work done by the breeding female, and when female helpers are present, as is so in the case of Splendid and Red-winged Fairy-wrens, they often provide nearly as much food for the nestlings as does the breeding female. When a single male helper is present, the senior male reduces his contribution, and there is little effect on the work done by the breeding female. In larger groups, those with two or more helpers, the total feeding rate does not increase significantly, but the contribution of each individual is lower. Young helpers, in the first breeding season after the one in which they hatched, are only 9–1 2 months old, and some of them feed nestlings assiduously, whereas others make little contribution; older helpers are more constant in their effort. With species which commonly produce second broods, such as the Splendid and Superb Fairy-wrens, immatures from a first brood may help to feed a later brood. Their visits are usually infrequent, they deliver small items, and they may even beg from adults bringing food to the nest.

Many attempts have been made to explain the presence of helpers at nests of birds, to shed light on the reasons why individual birds capable of breeding choose, instead, to remain with their parents and to feed the young of other adults. Most explanations recognize that there has to be some benefit for the helpers, something that balances the latter’s own forgone potential reproduction. This benefit may be indirect, if the helpers are assisting their parents to rear nestlings with whom they share genes, a so-called “kin benefit”; or it may be direct, with some positive benefit to the helper, and also to the parents for allowing the helper to stay. With malurids, there is little evidence of indirect benefits, since the helpers have no great effect on the production of fledglings, and extra-pair copulations further reduce any kin benefit. On the other hand, several direct benefits for helpers and breeders have been identified among malurids. Helpers reduce the amount of work done by the breeding female and her partner in feeding nestlings; studies of fairy-wrens in Canberra have shown that this frees the dominant male from parental duties and allows him to spend more time in visiting females in other territories and seeking extra-pair copulations. Another possible direct benefit is an improved chance of gaining a breeding territory in habitats in which all good territories are occupied throughout the year. Long-term studies have revealed that male Splendid Fairy-wrens that stay at home are likely to fill a vacancy for a breeder either in their natal territory or in an adjoining one, and their best chance of surviving to do this is to remain in their familiar natal territory.

Having left the nest, young malurids are fed by their parents and other adults in the group for about one month. After they become independent of parental feeding, they remain with the group, joining in group activities. Some young females may disperse early in the autumn, whereas others remain with the group through the winter, and disperse in late winter and early spring. Both males and females are sexually mature and capable of breeding at one year, although not all at that age have become established in a territory with a partner. Those that are not paired may remain as helpers, generally in the natal territory, and among many malurid species a significant proportion of males and some females do not breed until they reach two years or even more. Because of the promiscuous mating system, young males paired for the first time are more likely to be cuckolded. In the case of the Red-winged Fairy-wren, with high survival of breeding adults and few new vacancies for breeders, the median age at first breeding was three years for both males and females.

For fairy-wrens, the major causes of nest failure are predation and brood parasitism, the latter by cuckoos. Desertion is a relatively minor cause of failure, rarely affecting more than 10% of nests, and can be due to rain, excessive cold or heat, or predation of the female. Considering all species in the family Maluridae, parasitism by cuckoos has ranged from 0% to 36%, with 10–20% the usual range. The majority of parasitism of most species is by Horsfield’s Bronze-cuckoo (Chrysococcyx basalis), and in tropical Australia the Purple-crowned, Variegated and Red-backed Fairy-wrens are hosts of the Brush Cuckoo (Cacomantis variolosus); several other cuckoos have been recorded occasionally as nest parasites of fairy-wrens, grasswrens and emuwrens. In addition to causing the complete failure of host nests, cuckoos are significant nest predators of fairy-wrens. Although Horsfield’s Bronze-cuckoo is able to lay eggs in the nests of Superb Fairy-wrens, and these eggs are then incubated to hatching, one study in Canberra has shown that as many as 40% of nests containing a lone Horsfield’s Bronze-cuckoo chick were deserted by Superb Fairy-wrens, a species which rarely deserts its own nestlings.

Despite the supposed protection offered by the domed nest structure, nest predation is a significant cause of failure, ranging from 9% in one population of the White-winged Fairy-wren to 55% in another, and accounting for a failure rate of 55% recorded for the Superb Fairy-wren in Canberra. Predators include snakes and lizards, rats (Rattus), feral cats and foxes (Vulpes), and a variety of birds, such as the Pied Currawong (Strepera graculina) and the Grey Shrike-thrush (Colluricincla harmonica). The success of nests varies both within and among species, and year-to-year variation can be considerable. Over a 17-year study of the Splendid Fairy-wren near Perth, in south-west Australia, the proportion of clutches from which young fledged ranged from 30% to 86%, with an average of 57%, and the average rate of cuckoo parasitism was 20%. In populations of other species with low levels of cuckoo parasitism, the success rate of nests is higher, 68% in the Red-winged Fairy-wren and 88% in the coastal White-winged Fairy-wren. In two other studies of White-winged Fairy-wrens, both in drier areas and with a greater degree of parasitism by cuckoos, only 28% and 36%, respectively, of nests were successful.

For most malurids, the productivity per nest is 1–2 fledglings, but because the breeding season is long, and because females renest after fledging one brood and re-lay repeatedly after failure, the number of young successfully reared by a pair or group over the whole breeding season is a more significant measure of productivity. Individual pairs or groups of fairy-wrens produce an average of 2–3 fledglings per year, significant numbers of females raising two broods, and with occasional records of three broods. In the population of Splendid Fairy-wrens near Perth, over the 17 years of the study, approximately 20% of females produced two broods and, perhaps more significantly, an average of 85% of groups had at least one successful nest per year; in no year did the latter figure fall below 65%.
At least half of all fledglings leaving malurid nests survive to become independent of their parents after one month, and studies of several species have demonstrated that 25–40% of fledglings live to one year or more. In a study of the Southern Emuwren at Portland, in Victoria, it was found that females, although capable of producing more than one brood per season, rarely did so. This was because of high rates of nest failure and of loss of nesting females, both a result of predation by snakes.

The mortality of adults varies with species and habitat, but is remarkably low for such small, brilliantly coloured birds. In populations of Red-winged Fairy-wrens from the high-rainfall eucalypt forests of south-western Australia, 78% of breeding adults survive from one breeding season to the next, and the maximum recorded longevity is at least 17 years for a male. Among other malurids that have been studied over several years, adult survival has been found to be in excess of 60%.

Movements

For most malurids, few data about movements are available from studies of individually marked birds, but there is no information from any source to suggest that any of the 27 species is anything other than resident. In studies of marked individuals, it has always been found that breeding pairs or groups are resident in a territory throughout the year, some young dispersing from the natal territory in their first year or later. This pattern probably applies to all members of the family.

In the case of the Superb Fairy-wren in Canberra, in south-east Australia, all young females disperse in their first year, in two distinct phases. Some females leave early, mostly 1–2 months after they fledge, and their dispersal is not in response to any obvious aggressive behaviour by the group; these females are most likely to be some of those hatched early in the breeding season. The second phase of dispersal takes place shortly before the next breeding season, and occurs in response to persistent maternal aggression. Early-dispersing females travel much farther than do those dispersing later; the longest movement recorded in Canberra was made by a female found breeding, in the year following her hatch, at a site 8 km from her natal territory. Late-dispersing females did not move so far, and took over breeding vacancies that had become available. While some early-dispersing females found breeding vacancies, others joined another group and remained with it through the winter before dispersing again to find a breeding vacancy. Since no “floating” females occur in the population, it appears that females unable to join a group do not survive. Males generally do not disperse, unless to occupy a breeding vacancy that becomes available in a territory nearby. Many males inherit their natal territory, remaining as helpers (see General Habits, Breeding) until they replace the dominant male. Although most relationships were ended by the death of one of the breeding pair, some females initiated divorce by dispersing to another territory. Occasionally, this was to a better-quality territory with more helpers. In other cases, it occurred when the female became paired to her son after the death of a previous senior male, or when a son was the senior helper; in such circumstances, the dispersal can be interpreted as a means of incest avoidance.

Among other fairy-wrens, dispersal follows a similar course, unless above-average productivity and high survival of juveniles and adults lead to increasing competition for a few breeding vacancies. In the relatively benign forest habitats of south-western Australia, about 80% of breeding adult Red-winged Fairy-wrens survive to the next nesting season, and most females do not disperse until their second year. This leads to larger groups and to the occurrence of helping behaviour by females, something that is not generally found in such species as the Superb Fairy-wren, all females of which disperse in their first year. It is likely that later-dispersing individuals make exploratory forays from the natal territory, remaining in a new territory if they encounter a vacancy or returning home if none is located.

Studies of the Splendid Fairy-wren in south-western Australia revealed that, when a run of good seasons produced a large excess of one-year-old and two-year-old males and females and there were few vacancies for breeders, many females did not disperse until their second or third year. Instead, they remained as helpers in the natal territory, and some even attempted to breed there. After a series of fires significantly reduced productivity and survival, fewer birds were available to fill more breeding vacancies, so that most females dispersed in their first year, and few groups had female helpers.

One of the best-studied examples of dispersal concerns the Blue-breasted Fairy-wren in south-western Australia. Throughout most of this species’ range, clearance for agriculture has removed all but a small proportion of its habitat, and what remains in the Western Australian wheatbelt is fragmented, often degraded, and isolated from other such remnants. Nevertheless, the Blue-breasted Fairy-wren persists in this habitat. In a study of its dispersal, the cleared habitat with fragments of native vegetation made it possible to locate most of the dispersing individuals. Males tended to be philopatric, only about one-third of young males dispersing in their first or later years. Those that dispersed rarely moved farther than the next territory, particularly in larger remnants of habitat. On the other hand, nearly half of the young females dispersed in their first year, and a further 16% did so in subsequent years. The median dispersal distance, measured as a straight line, was 1•2 km, and the farthest recorded straight-line distance was 8•8 km, or 13•8 km via the shortest possible route that followed vegetation “corridors” between the two sites. Since dispersing females are not familiar with the landscape and need to explore many options, however, the total distance traversed is probably much greater. In this study, it was calculated that the fairy-wrens rarely crossed a gap of more than 60 m, so that females dispersing in fragmented habitat might need to travel many times the distance covered by a female in unfragmented habitat, at the same time crossing gaps in vegetation. This implies a greater risk to dispersing females, and female mortality is, in fact, greater than that of males. It also implies that Blue-breasted Fairy-wrens will be reluctant to disperse to fragments isolated by more than 100 m of cleared land unless a vegetation corridor exists. These factors could lead to the decline of the species in habitat with inadequate connections between fragments. The study showed that a once continuous population is now a series of isolated subgroups.

For all malurid species hitherto studied, movement by breeding adults is uncommon, and occurs in one of two contexts. The first is in the rare cases of divorce, when an individual leaves one mate and territory for another, and the second is after a mate has died, and the survivor then moves to a vacancy in another territory. These movements rarely traversed more than one or two territories. There is so far no evidence to indicate that quality of mate or of territory is a actor reponsible for any dispersive movements.

The few studies carried out on emuwrens and grasswrens suggest that these, too, are resident throughout the year in all-purpose territories. Studies of such species as the Grey Grasswren from arid central Australia suggest that in times of prolonged drought, when little vegetation cover remains, territory boundaries are relaxed and surviving individuals from a number of territories congregate in a few refuge areas of persistent vegetation.

Relationship with Man

Fairy-wrens are one of the most widely known and recognized genera of Australian birds, while emuwrens and grasswrens are probably among the least known. The brilliant colour of the male fairy-wrens, combined with the habit of Malurus species of foraging on the ground in relatively open areas, their persistence in many modified habitats so long as sufficiently dense cover remains, and their easily identifiable song, means that they are among the few birds familiar to people beyond the birdwatching community. Fairy-wrens in large suburban gardens and in parks can become very tame and, if food such as cheese or peanut butter is provided, they will come to a bird feeding station, thus adding to their general appeal. “Blue wrens” as a category are, along with kookaburras (Dacelo), Australian magpies (Gymnorhina), magpie-larks (Grallina), parrots (Psittacidae), cockatoos (Cacatuidae), Willie-wagtails (Rhipidura leucophrys), crows (Corvidae) and ducks (Anatidae), one of the general bird groups recognized by most Australian residents. The distinctions among species are not generally made by the lay public, who are familiar with local species and are surprised to find different ones in other parts of Australia. These fairy-wrens have achieved the status of cultural icon, one of the necessary species in any decorative art with a wildlife theme, obligatory in bird calendars, popular wildlife books for children, and the tourist souvenir and postcard trade. This recognition has meant that they are one of the first species enlisted in attempts to gain public support for the preservation of remnants of native vegetation on the fringes of towns and cities.

Several members of the family have been kept in captivity. These include emuwrens and grasswrens, as well as fairy-wrens. The very attractive plumage of some of these species could perhaps explain why people would wish to keep them in captivity, but it is debatable whether they could be considered to make good household pets. On the other hand, observations on aviary-held malurids have provided some useful information on various aspects of the behaviour of these birds.

Status and Conservation

Despite the ability of some malurid species to persist in both modified and man-made habitats, the changes in the Australian landscape wrought since the arrival of Europeans have reduced the range and numbers of all members of the family. The widespread land clearance for agriculture and the large-scale logging of forests for timber production in higher-rainfall areas have greatly reduced the area of land available to malurids, both in Australia and in New Guinea. While some large trees may be left when land is cleared for agriculture or forestry, without an understorey there is no suitable habitat for malurids. Urban settlement in the region clings to the coastal fringe, and the inexorable development of coastal lands for new houses has consumed much of the coastal swamplands and sandy heathlands that were the habitats of the various subspecies of the Southern Emuwren. Some habitat remains in reserves and national parks, but the increasing fragmentation of remnants is a focus of much conservation effort. In agricultural areas, this is of particular significance to malurids, which have relatively weak powers of flight.

Other modifications of habitat have also been of particular significance to malurids, because these are species that forage mainly on the ground or in low vegetation (see Food and Feeding). Much of the uncleared remnants of native vegetation and roadside vegetation in agricultural landscapes has been seriously degraded by grazing, fire, weed-control spraying and increased salinity. In forest habitats, not only is logging activity a problem, but frequent low-intensity burning is also carried out as a means of protecting forests from uncontrollable high-intensity wildfires. These “fuel-reduction” fires have an adverse effect particularly on resident, weak-flying understorey species such as the fairy-wrens; this is significant since, with increased human population density and access to forests, greater numbers of accidental and deliberately lit fires are occurring, chiefly in areas close to major population centres. The overall increase in fire frequency can lead to a decline in both productivity and density of fairy-wren populations.

A further significant interaction is with exotic animals, in particular foxes, cats and rats, brought to Australia by humans. All of these introduced mammals are significant predators of birds and their nests, and they have spread to most parts of Australia, including the arid areas. Domestic cats are significant predators, especially on the fringes of cities and towns. In south-eastern Australia, introduced birds such as the Common Blackbird, the Song Thrush (Turdus philomelos) and the Common Myna (Acridotheres tristis) may be competitors of the Superb Fairy-wren.

For the Australian taxa, the main causes of decline are habitat loss and degradation. In the last 200 years, large areas of native vegetation in Australia have been cleared for agriculture, and much more has been subject to grazing by sheep, cattle and introduced feral goats and camels (Camelus). A problem for species that forage on the ground, or spend their time in the vegetation layer 1–2 m above it, is that many habitat remnants on private land have been degraded at ground level by fire and by the grazing of livestock.

Of the 27 species in the family Maluridae, two are considered by BirdLife International and the International Union for the Conservation of Nature and Natural Resources (IUCN) to be globally threatened. These are the Mallee Emuwren and the White-throated Grasswren, both of which are categorized as Vulnerable. A further species, the Carpentarian Grasswren, has recently been placed on the list of Near-threatened species. No species has become extinct since the time of European settlement in Australia, in 1788.

The tiny Mallee Emuwren has suffered a considerable loss of habitat, since large areas of mallee eucalypt have been cleared and are now used for wheat-farming. Although clearance work has now ceased, the species’ population has been fragmented, and the isolated subpopulations are vulnerable to extensive fires that destroy the understorey of Triodia spinifex, where the emuwrens live. Although the mallee eucalypts survive fires and soon resprout, it takes three or four years for the spinifex understorey to regenerate to a stage that represents good emuwren habitat. In the past, the birds could disperse from unburnt areas into regenerating burnt ones, but increased fragmentation makes it more likely that subpopulations will progressively be eliminated from even quite large patches of habitat. This has already happened in one substantial area of mallee after an extensive fire. Possible options for population recovery are the preservation and protection of the remaining habitat, appropriate management to reduce the extent and frequency of fires, and possible relocation of the species to areas from which it has been eliminated by fire. The global population of the Mallee Emuwren in the first years of the twenty-first century is thought to number no more than approximately 3000 individuals. The great majority of these are based in two core areas in Victoria, one being Murray Sunset National Park and the other Hattah Kulkyne National Park and adjacent land. The conservation status of the species likely soon to be upgraded to that of Endangered.

The other globally threatened malurid species, the White-throated Grasswren, is restricted to the rocky escarpment of north-west Arnhem Land, in the Northern Territory of Australia. The grasswren lives on the bare flat plateaux and rocky hillsides of this highly dissected habitat. The hillsides are covered with spinifex, the nesting habitat of the White-throated Grasswrens, and the birds prefer areas which include boulders or bare rocky pavements. The extent of their habitat is limited, and the changing frequency and timing of fire are leading to significant habitat alteration. Although the rocky nature of the area gives the birds protection from the fire itself, the greatly increased frequency of hot fires at the end of the dry season, instead of cooler fires early in the dry season, is leading to progressive replacement of spinifex by annual sorghum grass (Sorghum), and the gradual decline of the grasswren population is considered likely. The species occurs within the Kakadu National Park, where extensive fire management is practised. It is the population outside conservation reserves that is subject to the uncontrolled late-dry-season fires, which are a threat to habitats and wildlife across the whole of tropical northern Australia.

Within its relatively small range, in hummock grassland or low open woodland to the south of the Gulf of Carpentaria, the Carpentarian Grasswren appears to be more widespread than was hitherto thought and there is no evidence that it has declined. Nevertheless, the frequent burning of its habitat in order to produce pasture for cattle appears to eliminate the species, which prefers grassland that has remained unburnt for a long time. It is now “officially” listed as Near-threatened. Regular monitoring of Carpentarian Grasswren populations is required, as is fire management that favours the species.

None of the New Guinea members of the family is included in the IUCN list of species at risk, and all of them are insufficiently known for any realistic assessment of their status. The main threat is likely to be the large-scale clearing of lowland and foothills forests for timber and wood-chip production. The Orange-crowned Wren is a denizen of mountain forest, inaccessible habitat not subject to logging, and is unlikely to be seriously threatened. The widespread White-shouldered Fairy-wren has probably benefited from clearance, as it thrives in regrowth, roadside vegetation and overgrown gardens. The other three species live in lowland and foothill forest that is likely to be under threat from logging. Although Wallace’s Wren and the Emperor Fairy-wren are more familiar to ornithologists than is the Broad-billed Fairy-wren, which is known from fewer than 50 specimens, all three are dependent on forest, and the extent of recent clearing, especially in Irian Jaya, is not well documented.

At the end of the twentieth century, a review initiated by the Australian Government Department of the Environment and Heritage (Natural Heritage Trust) was carried out by S. T. Garnett and G. M. Crowley, the results of which were published in 2000 as The Action Plan for Australian Birds. In addition, the Australian Environmental Protection and Biodiversity Conservation Act, referred to as the “EPBC Act”, came into being in 1999. These cover only Australian species; the five species of New Guinea, comprising 15 subspecies, are not included. These two regional assessments investigated the conservation status of taxa down to the subspecies level. On this basis, one subspecies of the Maluridae is rated as “Critically Endangered”, ten as “Vulnerable” and three as “Near-threatened”.

Few fairy-wrens were listed in Garnett and Crowley’s review. Three island subspecies are regarded as “Vulnerable”, because each occurs only on a single, relatively small island. These are the Shark Bay race bernieri of the Variegated Fairy-wren, confined to Bernier Island, and two subspecies of the White-winged Fairy-wren, namely the nominate race on Dirk Hartog Island and, a little farther north, edouardi on Barrow Island; all three of these islands are off the west coast of Western Australia. In each case, there is no possibility of recolonization if the population becomes severely reduced by fire or drought. The only mainland fairy-wren included is the western, nominate race of the Purple-crowned Fairy-wren of north-west Australia, which is “Near-threatened” according to Garnett and Crowley and “Vulnerable” according to the EPBC Act. This subspecies occurs at reduced density over much of its original range and has disappeared from parts of it. It is restricted to the dense riparian vegetation of tropical rivers, and this habitat has been significantly damaged by cattle seeking access to water and, in some areas, by fire. Efforts to exclude cattle from riparian areas in order to reduce the erosion of riverbanks are also of benefit to the Purple-crowned Fairy-wren.

Of all the malurids, it is the tiny emuwrens that are most at risk. Their small size, inconspicuous plumage, weak song and secretive nature render them all but invisible except to the most determined searcher. Most human beings who share the emuwrens’ habitat are quite unaware of the latter’s existence. As mentioned at the start of this section, the monotypic Mallee Emuwren is globally threatened, being placed in the category of Vulnerable. The Rufous-crowned Emuwren is far more widespread, if patchy in distribution, and is locally common. It is not believed to be at any risk, but its near-dependence on mature Triodia clumps could render it vulnerable to excessive, frequent burning.

The Southern Emuwren contains eight subspecies distributed around the coasts of the southern half of Australia and Tasmania. In South Australia, the Fleurieu Peninsula subspecies intermedius is “Critically Endangered” and the Eyre Peninsula race parimeda is “Vulnerable”; likewise, hartogi of Dirk Hartog Island, in Western Australia, is rated as “Vulnerable”. The area originally occupied by intermedius covered only about 400 km² in the Mount Lofty Ranges near Adelaide and south into the Fleurieu Peninsula. This taxon’s range has been greatly reduced by habitat clearance and degradation, and it is estimated that the few remaining subpopulations now occupy no more than about 10 km², with a remnant breeding population totalling fewer than 500 individuals. These prefer swampy land with grasses, sedges and dense low shrubby vegetation along watercourses and in gulleys, but much of this swampy habitat has been drained, cleared, and grazed by stock. The few surviving patches are vulnerable to fire, and they are so isolated from each other that recolonization by dispersing individuals is no longer possible, especially since emuwrens are so small and such weak fliers. The South Australian Department of Environment and many other conservation agencies are involved in attempts to implement a recovery plan, which involves habitat protection, revegetation, population surveys, studies of breeding biology, and a programme to increase community awareness and involvement.

Each of the two “Vulnerable” races of the Southern Emuwren, parimeda and hartogi, occurs in a small area of 100–200 km², and each in a single population of a few thousand individuals. Their numbers are currently stable, but very prone to serious decline in the face of fire or drought. The habitat of parimeda on the Eyre Peninsula has been fragmented by past clearance for agriculture, and the frequency of fire has increased. Although clearance work has now ceased, the remaining patches of habitat are isolated, recolonization after fire is unlikely, and the total area in which this taxon now occurs is so small that a single extensive fire could threaten the entire population. Although the problems of this subspecies have been identified, no programme of fire management and revegetation has been specifically aimed at reducing them. The subspecies hartogi is found only on a long, narrow island of 420 km² off the coast of Western Australia. The island is held as a single grazing lease and, as well as sheep, it carries feral goats, house mice (Mus musculus) and feral cats; it is subject to occasional extensive fires. Because it lacks free water, sheep grazing has been most intensive in the southern half of the island, rendering the habitat unsuitable for emuwrens. The classification of hartogi as “Vulnerable” stems from the fact that it consists of a single, small population; although this was assessed as being stable in Garnett and Crowley’s 2000 review, its continued survival will require careful monitoring and the management of grazing, fire and feral animals. The Thick-billed Grasswren has already disappeared from the island, where it was first recorded in 1916; a dedicated search in 1920, by F. L. Whitlock, failed to find the grasswren, and Whitlock attributed its disappearance to the very large feral cat population.

All five remaining races of the Southern Emuwren are considered not to be at any risk, although all have suffered considerable habitat loss in the past 200 years. These have the misfortune to share the coastal strip of south-western and south-eastern Australia with the majority of the Australian human population, and much of the low-lying coastal heath and swamp habitat has been drained and cleared for urban development. Nevertheless, because these subspecies originally had more extensive ranges than those of the races at risk, most of them still have significant areas of occurrence within conservation areas such as national parks.

The nominate, Bulloo River subspecies of the Grey Grasswren is rated as “Vulnerable”. It occurs in a single location, probably less than 100 km² in extent, around the border between south-west Queensland and north-west New South Wales, and it could be subject to catastrophic decline through the combined effects of drought, fire, cattle grazing and the harvesting of water. Although the population, which probably fluctuates normally, is currently assessed as stable, both it and its habitat require careful monitoring.
According to the EPBC Act, all three subspecies of the Thick-billed Grasswren are “ Vulnerable”. The nominate, western race has suffered a considerable decrease in range. Having once occurred over the arid and semi-arid regions of south-western Australia, from the South Australian end of the Nullabor Plain westwards to Shark Bay, and including Dirk Hartog Island, it is now found only near Shark Bay, within an area of about 1200 km². Its disappearance from most of its former range is attributed to the combined effects of sheep and rabbits on vegetation at ground level and to 1 m above it. The taxon is quite common in the small area to which it is now restricted, especially on the Peron Peninsula, where extensive predator control is enforced over a large national park, and in a buffer zone of sheep-grazing properties. The eastern subspecies, modestus, once widespread in arid and semi-arid regions from the southern Northern Territory south and east to New South Wales, is now restricted to the region of Lake Eyre, in the north-east of South Australia. The Gawler Range race myall, in the south part of South Australia, was never so widespread as the other two subspecies of the Thick-billed Grasswren, but it, likewise, has suffered a significant reduction in range. Again, sheep grazing and rabbits are implicated in the decline of these subspecies. As with all other taxa with restricted distributions, extensive wildfire is a major threat.
One other Amytornis taxon is regarded as “Near-threatened”. This is the nominate race of the Striated Grasswren, which has disappeared from the southern edge of its range and declined in density over the rest of it. While the large extent of its range does not justify the classifying of this taxon as threatened, the processes of habitat degradation and fragmentation, fire and grazing continue. Fire is of particular significance in the spinifex grasslands of arid central Australia, and any management plan must attempt to limit the frequency and severity of fire.

Although other grasswrens, such as the Black, Dusky and Short-tailed Grasswrens, have restricted ranges, their populations and areas of distribution appear not to have diminished to such an extent that they are considered at risk. Most of these malurids live in rugged, rocky habitats where grazing by stock has not had such a significant influence.
For all malurids, whether considered to be at risk or not, habitat loss through land clearance is still a threat, especially in New Guinea but also in some parts of Australia. In addition, the problems of drought, fire, introduced predators, and habitat degradation by grazing stock are significant throughout Australia, and regular monitoring is necessary in order to ensure that population levels of bird species are maintained. The harnessing of the efforts of amateur birdwatchers made possible the large-scale surveys that have produced the two atlases of Australian birds, the first in 1984 and the second in 2003. These provide some indication of population trends, but they are not a substitute for detailed surveys of individual species perceived to be at risk.

General Bibliography

Anon. (2006b), Baker (1995), Barker, F.K. et al. (2001), Barker, R.D. & Vestjens (1990), Barrett et al. (2003), Bell (1980, 1982a, 1982b), Blakers et al. (1984), Bock (1994), Boles (1995, 1997), Brooker, B. (1998a), Brooker, L.C. & Brooker (1995, 2002, 2003), Brooker, M.G. (1988), Brooker, M.G. et al. (1990), Campbell & Lack (1985), Cayley (1949), Christidis & Boles (1994), Christidis & Schodde (1997), Cockburn (2004), Cockburn et al. (2003), Condon (1984), Cooney & Cockburn (1995), Dickinson (2003), Double & Cockburn (2000, 2003), Doucet et al. (2004), Dunn & Cockburn (1996, 1999), Ericson, Christidis et al. (2001), Ericson, Irestedt & Johansson (2003), Garnett & Crowley (2000), Gould (1865), Harrison (1969), Higgins et al. (2001), Hutton (1991), Johnstone & Storr (2004), Jurisevic & Sanderson (1994), Karubian (2001, 2002), Karubian & Alvarado (2003), Koenig & Dickinson (2004), Langmore & Mulder (1992), Langmore et al. (2003), Lill et al. (2006), Maguire & Mulder (2004), Mayr (1986), Mayr & Amadon (1951), Mees (2003), Mulder (1995, 1997), Mulder & Cockburn (1993), Mulder & Magrath (1994), Mulder et al. (1994), Parker (1972, 1982), Payne, Payne & Rowley (1985, 1988), Payne, Payne, Rowley & Russell (1991), Peters (2000), Peters, Astheimer, Boland & Cockburn (2000), Peters, Astheimer & Cockburn (2002), Pizzey et al. (2005), Rathburn & Montgomerie (2003, 2004), Rowley (1965, 1981b, 1991), Rowley & Russell (1993, 1995, 1997, 2002), Rowley et al. (1991), Russell & Rowley (1993a, 1993b, 1998, 2000), Schodde (1975, 1982), Schodde & Mason (1999), Sibley (1970, 1976, 1996), Sibley & Ahlquist (1982, 1985, 1990), Sibley & Monroe (1990, 1993), Simpson et al. (2004), Slater et al. (2003), Tarvin et al. (2005), Tidemann (1983, 2004), Tidemann & Schodde (1989), Tuttle & Pruett-Jones (2004), Tuttle et al. (1996), Webster et al. (2004), Whitlock (1910, 1921), Wilson & Paton (2004).