Family PSITTACIDAE

Platycercus (Violania) adscitus eximius

Museums

Museum list

Regional Maps

Family PSITTACIDAE

Compiler and date details

R. Schodde & I.J. Mason, CSIRO Australian National Wildlife Collection, Canberra, Australia

Introduction

Psittacidae (parrots and lories) comprise about 310-335 species in 70-87 genera; 41 species in 21 genera occur in Australia and its territories; one endemic species (Psephotus pulcherrimus Gould, 1845) is now extinct. The Australian fossil record comprises a wide range of unidentified and extant genera and species in north-east NSW, south-east QLD, Norfolk Is., and from southern VIC to southwest WA, including Kangaroo Island. Although with centres in the Old World and New World tropics, where most species occur, parrots are most diverse in Australasia where at least nine of the ten subfamilies or tribes recognised by Smith (1975) are found; six of these groups are present on mainland Australia (cf. Smith loc. cit., Homberger 1980).

Commonly gregarious, parrots and lories feed on the ground or in trees on seeds, fruit, nectar and pollen gathered by their bills; seeds are husked by the cutting action of the lower mandible on nuts held by the tongue on a dental pad on the maxilla, and nectar and pollen are won by tongue-brushing. Parrots and lories may or may not clump-roost; they drink by lapping, ladling or suck-pumping; they bathe variably; and their voices are characteristically weazy chatters and screeches, moderated as well by variable whistles and warbles in Platycercinae. Nests, defended in undefended territories, are usually of unconstructed beds in hollows or holes, or exceptionally are constructed in tree branches or on the ground; eggs are spheroidal, plain-white and slightly glossy, and usually incubated only by female. Young are altricial, nidicolous, blind at hatching, and covered with sparse to long dark grey or whitish down; they are fed by pump-regurgitation, the parent holding the chick's bill in its own.

Family-group Systematics

Although Australian members have been split variously into several families by Salvadori (1891), Mathews (1913, 1927), RAOU Checklist Committee (1926), Verheyen (1956), Brereton (1963), Wolters (1975-1982), Condon (1975) and Forshaw (1978), they are combined here in one, following Finsch (1867), Fürbringer (1888), Stresemann (1927-1934), Peters (1937), Mayr & Amadon (1951), Boetticher (1959), Wetmore (1960), Smith (1975), Homberger (1980) and Christidis et al. (1991). Of the six infra-familial groups in Australia and its territories, the composition and relationships of a number are still not clear, particularly for Cyclopsittini (fig parrots), Psittaculini (as represented by Eclectus Wagler, 1832 and Geoffroyus Bonaparte, 1850), and Polytelini (long-tailed parrots), see Salvadori (loc. cit.), Beddard (1898), Smith (loc. cit.), Homberger (loc. cit.; 1991) and Christidis et al. (loc. cit.). These, therefore, are conservatively given no more than tribal status here; other more clearly defined and better established groups are treated as subfamilies. The groups are as follows, after Smith (loc. cit.) as modified by Homberger (1980) and Christidis et al. (loc. cit.).

Nestorinae Bonaparte, 1849 (kakas)-large, stout, short-tailed parrots with spine-tipped rectrices, bronzed plumage, multi-striped wings, unfused cartilaginous first bronchial rings and cockatoo-like gait and flight. There is one genus, of three species, centred in New Zealand and of obscure affinity: Nestor Lesson, 1830. For additional details of circumscription, see Salvadori (1891) and Holyoak (1973).

Loriinae Selby, 1836 (lorikeets and lories)-small, brightly coloured, wholly arboreal flower-feeding parrots with usually camouflaging green dorsa, pointed wings and tails, dashing flight, brush-tipped tongues with eversible papillae, and short intestinal tract with simple crop and gizzard. Eleven to 13 genera are recognised (Peters 1937; Wolters 1975-1982; Forshaw 1978; Sibley & Monroe 1990), with much the greatest diversity in New Guinea where 8-10 genera occur, including three endemic; three genera are recognised in Australia. For additional details of circumscription, see Salvadori (1891) and Holyoak (1973).

Platycercinae Selby, 1836 (broad-tailed parrots)-small to medium-sized ground- and tree-feeding flap-gliding parrots with usually long, pale blue-, white- or yellow-tipped tails, and commonly red or yellow under-tail coverts; sexual dimorphism is slight, there is a single narrow under-wing stripe in females and buff-billed juveniles, the furcula is commonly vestigial and cartilaginous, and extrinsic syringeal muscles appear to be absent. Comprising about 14 genera with 35-38 species, Platycercinae are centred in Australia and its territories where 11 genera (10 endemic) and 25 species (24 endemic) occur, cf. Peters (1937), Wolters (1975-1982), Forshaw (1978) and Sibley & Monroe (1990). Fijian Prosopeia Bonaparte, 1854, Australian Lathamus Lesson, 1830, and montane New Guinean Psittacella Schlegel, 1871 are referred to the group, following Homberger (1980) and Christidis et al. (1991). The only other genera found outside Australia are Eunymphicus Peters, 1837 and Cyanoramphus Bonaparte, 1854, both limited to islands in the southwest Pacific. According to Christidis et al. (loc. cit.), Australian Platycercinae comprise a core of six closely related genera (Platycercus Vigors, 1825, Barnardius Bonaparte, 1854, Purpureicephalus Bonaparte, 1854, Lathamus Lesson, 1830, Northiella Mathews, 1912 and Psephotus Gould, 1845), which are rather distant from other more diversely related genera in the group: Cyanoramphus Bonaparte, 1854, Melopsittacus Gould, 1840, Neophema Salvadori, 1891, Neopsephotus Mathews, 1912 and Pezoporus Illiger, 1811 (cf. also Condon 1941; Homberger 1991).

Psittaculini Vigors, 1825 (Palaeotropic parrots)-medium-sized, prevailingly green, arboreal parrots with heavy bills, short or racket-shaped or attenuate tails, variable sexual dimorphism, and no under-wing stripe; both tongue and dental pad are ridged or grooved. Composition is uncertain; the most circumspect definition (Wolters 1975-1982) admits 11 genera centred from India to Papuasia when Psittrichas Lesson, 1831 is excluded (Homberger 1980); there are two genera in Australia. The circumscription of this group and its relationships to other groups of parrots is far from settled, see Christidis et al. (1991) and Homberger (1991).

Cyclopsittacini Salvadori, 1891 (fig parrots)-small, green, wholly arboreal, fruit-eating parrots with stout ridged bills, wedge-shaped tails, marked sexual dimorphism, permanent double under-wing stripes, and completely ossified orbital ring free from well-developed zygomatic processes. There are five species in one to three genera centred in New Guinea, see Peters (1937), Forshaw (1978), and Sibley & Monroe (1990); one genus occurs in Australia. Relationships with Philippine Bolbopsittacus Salvadori, 1891 and other psittaculine parrots are in need of resolution. For further details of circumscription, see Salvadori (1891), Holyoak (1970, 1973) and Wolters (1975-1982).

Polytelini Mathews, 1916 (long-tailed parrots)-slender, medium-sized, swift-flying, seed-eating parrots with prevailingly green plumage and attenuate, pink-tipped tails, marked sexual dimorphism, no under-wing stripe, psittaculine tongue and soft palate, and extrinsic syringeal muscles evidently attached to lung membrane. Their three genera with eight species are centred in Australia where all three genera occur, see Peters (1937), Wolters (1975-1982), Forshaw (1978) and Sibley & Monroe (1990). The Fijian Prosopeia Bonaparte, 1854 is excluded here, following Homberger (1980). The relationships of the Polytelini are in need of clarification: most revisers (e.g. Salvadori 1891; Smith 1975; Homberger 1980) align the group with the Psittaculini but Verheyen (1956) and Christidis et al. (1991) record links with Platycercinae (broad-tailed parrots), cf. Homberger (1991).

The genera are arranged in their infra-familial groups, as defined here, after the familial sequence in Condon (1975), but commencing with Nestorinae Bonaparte, 1849 (not treated by Condon loc. cit.) following Peters (1937) and concluding with tribes incertae sedis.

Genus-group Systematics

Psitteuteles Bonaparte, 1854-Protein screening by Christidis et al. (1991) indicated that generic groupings in the Loriinae are as difficult to distinguish at molecular level as from external morphology, see the varying arrangements in Peters (1937), Wolters (1975-1982) and Sibley & Monroe (1990). Pending a review of character homology and geography for the entire group, it is premature to alter the conventional limits among Australian lorikeets established by Condon (1975). So Psitteuteles Bonaparte, 1854 is retained for the Varied Lorikeet.

Trichoglossus Stephens, 1826-The Scaly-breasted Lorikeet (Psittacus chlorolepidotus Kuhl, 1820) is kept in this genus, for reasons see Psitteuteles Bonaparte, 1854, q.v..

Barnardius Bonaparte, 1854-Barnardius Bonaparte, 1854 is kept generically separate from Platycercus Vigors, 1825 because no documented case for its inclusion has been published since Condon's (1941) revision (cf. Peters 1937). Condon (loc. cit.), followed by Cain (1955) and Holyoak (1973), distinguished Barnardius because of the condition of its auditory meatus and colour pattern; see also Baird (1992). Although Ovenden et al. (1987) also found Barnardius divergent, Christidis et al. (1991) demonstrated faults in their analysis and recorded much closer links with Northiella (Mathews 1912) and Platycercus. These relationships, nevertheless, need better resolution before the status of Barnardius is altered.

Cyanoramphus Bonaparte, 1854-Although Holyoak (1973) combined the crested New Caledonian Eunymphicus Peters, 1937 with Cyanoramphus Bonaparte, 1854, his case based on the structure of the auditory region needs corroboration and extension before the conventional generic arrangements of Peters (1937), Wolters (1975-1982) and Sibley & Monroe (1990) are modified. The splitting-off of Phippsittacus subgenerically by McAllan & Bruce (1989), based on 'larger, darker green form with a different bill morphology', is too vague a justification.

Neophema Salvadori, 1891-Neophema Salvadori, 1891, is limited here to the small, green, yellow-rectriced Australian grass parrots with zig-zag flight, after Boetticher (1959), Immelmann (1966), Homberger (1980) and Sibley & Monroe (1990). For separation of Bourke's Parrot under Neopsephotus Mathews, 1912, see that genus. The species of Neophema fall into two groups recognised subgenerically here after Cain (1955), Immelmann (loc. cit.) and Lendon (1973): subgenus Neophema with rich blue over the head, azure wings, and sexual dimorphism with red in the plumage of males; and subgenus Neonanodes Mathews, 1912 with yellow-green faces, cyanic superciliaries, dark blue remiges and sometimes shoulders, and almost no sexual dimorphism.

Neopsephotus Mathews, 1912-Despite the arguments made by Cain (1955), Ford (1969) and Forshaw (1981) to include Bourke's Parrot (Euphema bourkii Gould, 1841) in Neophema Salvadori, 1891, compelling morphological and behavioural reasons for separating it in a monotypic genus, Neopsephotus Mathews, 1912, have been put forward by Boetticher (1959) and Homberger (1980: 165-166). Data in Christidis et al. (1991) indicate that it is widely divergent in proteins as well. Accordingly, Neopsephotus is recognised generically here, after the recommendation of Lendon (1973: 253).

Northiella Mathews, 1912-Argument over the status of the Bluebonnet (Platycercus haematogaster Gould, 1838) is polarized on one side by Peters (1937), Condon (1941, 1975) and Sibley & Monroe (1990) who refer it to monotypic Northiella Mathews, 1912, and on the other by Cain (1955), Immelmann (1966), Lendon (1973), Wolters (1975-1982) and Forshaw (1981) who include it in Psephotus Gould, 1845. Although Ovenden et al. (1987) recorded it clustering with Psephotus, Christidis et al. (1991) found it linked more closely to the Platycercus Vigors, 1825-Barnardius Bonaparte, 1854 lineage in proteins. Condon's (1941) morphological analysis also concluded that its closest affinities are with Barnardius, contra Cain (loc. cit.). Until these phyletic discrepancies are resolved, Northiella Mathews, 1912 should be retained.

Pezoporus Illiger, 1811-Apart from Forshaw (1978, 1981), the only reviews to have analysed the generic positions of the Ground Parrot (Psittacus wallicus Kerr, 1792) and Night Parrot (Geopsittacus occidentalis Gould, 1861) since Condon (1975) are those of Ford (1969), Schodde & Mason (1981: 35-36) and Leeton et al. (1994). All concluded, on morphological, ecological, biogeographical and molecular grounds, that the two species are congeneric. This assessment is accepted here.

Platycercus Vigors, 1825-For generic status with respect to Barnardius Bonaparte, 1854, see that genus below. Two long-known infra-generic groups of species (Immelmann 1966) are recognised formally here as subgenera: the large, blue-cheeked, broad-winged group and the smaller, white- or yellow-cheeked and more narrowly winged group. Their members are allopatric within and sympatric between groups. Western P. icterotis (Kuhl, 1820) has sometimes been separated from the white-cheeked group as a third infra-generic element, based on its more marked sexual dimorphism and anecdotal assessments of its flight, vocalizations and supposed links with Psephotus Gould, 1845 (Lendon 1973; Forshaw 1981). The case needs better phylogenetic documentation before it can be accepted.

Psephotus Gould, 1845-The two subgenera diagnosed behaviourally by Lendon (1973: 212-213)-Eyrean, green-backed and pallid-crissumed subgenus Psephotus Gould, 1845 and Torresian, brown-backed, red-crissumed Psephotellus Mathews, 1913-are accepted here (also Wolters 1975-1982). That the Red-rumped Parrot (P. haematonotus (Gould, 1838)) stands apart from its consubgener P. varius Clarke, 1910 and other members of Psephotus (Immelmann 1966) needs further supporting documentation before infra-generic arrangements are modified.

Purpureicephalus Bonaparte, 1854-Protein data (Christidis et al. 1991) corroborate the morphological, behavioural and biogeographical evidence in Serventy (1938), Condon (1941) and Forshaw (1964) that Purpureicephalus Bonaparte, 1854 is an integral member of the Platycercus Vigors, 1825-Barnardius Bonaparte, 1854 lineage, pace Cain (1955). Its pecularities in structure of skull and bill and in behaviour (Lendon 1973: 145) may represent autapomorphic adaptations to its foraging niche (Robinson 1960; Forshaw 1981: 168); but they need phylogenetic documentation before any change to its monogeneric status is considered.

Cyclopsitta Reichenbach, 1850-Generic limits among the fig parrots (Cyclopsittacini) are far from settled: Holyoak (1970) combined them in one (Psittaculirostris J.E. & G.R. Gray, 1859) and Wolters (1975-1982) split them into three. For the present, I have followed the conservative arrangements of Peters (1937), Mayr (1941) and Sibley & Monroe (1990) which recognise two genera: Cyclopsitta and Psittaculirostris, the latter diagnosed, inter alia, by its feathered cere, spiked cheek feathering and bulk.

Alisterus Mathews, 1911-Although treated as a subgenus of Aprosmictus Gould, 1842 by the RAOU Checklist Committee (1926) and Lendon (1973), the Australo-Papuan species of king parrots (Alisterus Mathews, 1911) stand apart in their slenderer build, longer and more graduated tail, extensively red and usually sexually dimorphic ventrum, direct flight and closed forest niche. Current convention separates them generically, see Peters (1937), Condon (1975), Wolters (1975-1982) and Sibley & Monroe (1990).

Polytelis Wagler, 1832-Neither of the conflicting subgeneric arrangements by Mathews (1913, 1916, 1917, 1927) and Wolters (1975-1982) is accepted pending resolution of phylogenetic relationships in this close-knit genus.

Species-group Systematics

Nestor productus (Gould, 1836)-This species forms a superspecies with the New Zealand Kaka (Nestor meridionalis (Gmelin, 1788)), see Sibley & Monroe (1990), and is sometimes considered a subspecies of it. Salvadori's (1891) treatment of Norfolk and Philip Island populations as separate species is based in part on a malformed bill in the only description of material from Norfolk Island available to him; the distinction is no longer accepted (Peters 1937).

Glossopsitta concinna (Shaw, 1791)-Tasmanian populations largely lack the distinct bluish wash to the crown found in adults from mainland Australia and are recognised subspecifically after White (1938). Evidence for trans-Bass Strait migration is conflicting, cf. Blakers et al. (1984) and Green (1989).

Glossopsitta porphyrocephala (Deitrichsen, 1837)-Although geographic differentiation appears to be insignificant, it has yet to be demonstrated comprehensively.

Glossopsitta pusilla (Shaw, 1790)-Geographic variation also appears to be insignificant in this nomadic, nectar-seeking species, but it has yet to be demonstrated analytically.

Trichoglossus chlorolepidotus (Kuhl, 1820)-That northern populations are sub-specifically distinct from southern ones has been assumed at times from their supposedly smaller size, but the differences appear to be slight and clinal, cf. Hall (1974). Acceptance of subspecies is contingent on demonstration-so far lacking-that any gradient in size is stepped and secondary.

Trichoglossus rubritorquis Vigors & Horsfield, 1827-Whether to recognise the Red-collared Lorikeet of NW Australia as a valid species or to combine it with eastern Australian and southwest Pacific green-collared forms in one widespread polytypic species, Trichoglossus haematodus (Linnaeus, 1771), is contentious. Lendon (1973), Condon (1975) and Sibley & Monroe (1990) took the former view, and Peters (1937), Cain (1955), Wolters (1975-1982), Forshaw (1978) and White & Bruce (1986), the latter. Neither alternative has been documented recently; thus a review here is needed to reach a decision.

The Trichoglossus haematodus complex comprises (a) monotypic Trichoglossus ornatus (Linnaeus, 1771) from Sulawesi which has been treated consistently as allospecifically distinct since Peters (1937); (b) small, short-tailed and ventrally chlorotic T. weberi Büttikofer, 1894) from Flores which, although commonly treated as a race of haematodus (Peters 1937; Forshaw 1978; White & Bruce 1986), is of disputed affinity (references in White & Bruce loc. cit.); and (c) five principal regional groups of haematodus-like forms with bluish heads, coloured collars, broad red/yellow breast bands, and yellow bands and washes through the under-remiges and retrices respectively. The five regional groups are:

(i) forsteni Bonaparte, 1850-of three to four subspecies in the Lesser Sundas: small and short-tailed, with blackish blue head, broad pale green collar, dark green mantle under-spotted obscurely red, scarlet underwing coverts, plain deep scarlet breast, bluish black upper belly and deep green lower belly mottled diffusely yellow;

(ii) capistratus Bechstein, 1811-of three subspecies in the eastern Lesser Sundas: large and long-tailed, with greenish or dusky-blue face grading to deep blue crown, broad yellowish-green collar, mid green mantle under-spotted yellow, yellow under-wing coverts, almost plain golden-yellow breast, dark or bluish-green upper belly and light yellowish-green lower belly mottled dark green;

(iii) haematodus Linnaeus, 1771-of 10-11 subspecies from the Moluccas and islands in the Arafura Sea through New Guinea and Melanesian archipelagos to Vanuatu and New Caledonia: medium in size and tail, with dull blue face grading to purplish dusky crown, narrow pale green collar, dark green mantle under-spotted red, scarlet breast barred dusky-blue, dark green upper belly and yellow lower belly coarsely spotted dark green;

(iv) moluccanus Gmelin, 1788-of three subspecies in eastern Australia: medium to large in size and tail, with royal blue head, narrow pale green collar, mid green mantle under-spotted red, yellowish scarlet under-wing coverts, unbarred but shaded yellow and scarlet breast, deep royal blue upper and lower belly, with fine green spotting over yellow restricted to crissum and flanks;

(v) rubritorquis Vigors & Horsfield, 1827-apparently monotypic in northwestern Australia and several islands in the eastern Lesser Sundas (Visar, Romang): rather large and long-tailed, with royal blue head, broad orange-red collar, deep blue mantle under-spotted red (grading to mid green on back), orange-red under-wing coverts, plain orange-red breast, bluish-black upper and lower belly, with yellow and green spotting restricted as in moluccanus Gmelin, 1788.

The first three groups, which are widespread in Wallacea, have already been recognised by White & Bruce (1986). With one exception, all five merit separation as allospecies according to the criteria for taxonomic definition given in the Introduction (see Taxonomic Scope), as do also weberi Büttikofer, 1894 and ornatus Linnaeus, 1758. The exception is moluccanus Gmelin, 1788 which is linked to haematodus Linnaeus, 1771 by a morphologically intermediate population, caeruleiceps D'Albertiz & Salvadori, 1879, in the Trans-Fly region of southern New Guinea. Members of none of the others are intermediate or known to intergrade where they approach one another geographically, for example, moluccanus and rubritorquis around the head of the Gulf of Carpentaria. Hartert (1904) reported possible intergradation between rubritorquis and the capistratus group on Romang, but his evidence is based on a single questionable record and requires substantiation.

Accordingly, the arrangement of Condon (1975) and Sibley & Monroe (1990) is followed here for Australian species in the haematodus complex. The form on northern Torres Strait islands adjacent to the south of New Guinea is presumed here to be T. h. caeruleiceps d'Albertis & Salvadori, 1870 (see Mees, G.F. (1982). Birds from the lowlands of southern New Guinea (Merauke and Koembe). Zool. Verh. 191: 1-188 4 pls).

Barnardius zonarius (Shaw, 1805)-The four principal forms of ringnecks (barnardi Vigors & Horsfield, 1827, macgillivrayi North, 1900, semitorquatus Quoy & Gaimard, 1830, and zonarius Shaw, 1805) are treated as subspecies within one species, contrary to much current convention but consistent with the polytypic species and intergradation criteria adopted here for the definition of species. All four forms are allopatric/parapatric and intergrade through wide zones where they meet. B. z. semitorquatus intergrades with nominotypical zonarius through south-western Australia (Condon 1941, 1975; Hall 1974; Serventy & Whittell 1976; Ford 1987). Nominotypical zonarius intergrades with B. z. barnardi through the Flinders Range, SA east towards NSW (Cain 1955; Fisher 1970; Hall loc. cit.). Because of this, nominal infra-specific taxa based on intergrading material from that region by Peters (1937), Condon (ll.cc.), Cain (loc. cit.) and Forshaw (1981) are rejected here as indeterminable. B. z. barnardi, in turn, intergrades with B. z. macgillivrayi through the central-upper Diamantina River drainage, Qld (material in ANWC). Differences of opinion over the affinities of macgillivrayi-whether with barnardi, based on head and cheek pattern and discolorous back and rump (Condon 1941; Forshaw loc. cit.), or with zonarius, based on green frons and broad yellow ventral band (Ford & Parker 1974)-serve to stress the specific coherence of the group. Within nominotypical zonarius, northern and centralian populations have been separated subspecifically by Condon (loc. cit.), Cain (loc. cit.) and Forshaw (loc. cit.). Nevertheless, their supposed pallidness is slight, apparently clinal and often due to wear (also Hall loc. cit.). A comprehensive analysis of geographic differentiation and regional introgression in this species has yet to be published.

Cyanoramphus novaezelandiae (Sparrman, 1787)-The case made by McAllan & Bruce (1989) to separate the large form on Norfolk Island specifically lacks sufficient comparative data to support a departure from conventional polytypic arrangements of this species (Peters 1937).

Neophema splendida (Gould, 1841)-Although no regional differentiation has been recognised by recent revisers (e.g. Cain 1955), geographic variation has never been documented comprehensively.

Neophema chrysogaster (Latham, 1790)-Separation of Tasmanian from Australian mainland populations subspecifically by Mathews (1912) and his subsequent revisions, and by Peters (1937) did not appreciate that their mainland records were probably of non-breeding migrants from Tasmania.

Neophema elegans (Gould, 1837)-Peters (1937), Cain (1955), and Condon (1975) agreed that the two western and south-central Australian isolates are subspecifically differentiated. It is an interpretation followed tentatively here because south-central populations appear to be more extensively and intensely yellow ventrally, sex for sex, than the more citrine-washed form in the west, and to have a brighter blue frons band, cf. Hall (1974). Nevertheless, because Lendon (1973) and Forshaw (1981) disagree, a comprehensive analysis of geographic variation is needed.

Neophema petrophila (Gould, 1841)-Circumstances and status as for N. elegans (Gould, 1837).

Neopsephotus bourkii (Gould, 1841)-Although no subspecies are recognised, there has been no comprehensive assessment of geographic variation in confirmation (cf. Cain 1955; Ford 1969). Because of the semi-isolation of western, centralian and Lake Eyre basin populations, a modern survey is needed.

Northiella haematogaster (Gould, 1938)-The arrangement and geographic limits of infra-specific taxa in the Lake Eyre and Murray-Darling basins are tentative here, pending clarification of their zones of intergradation. Gene flow, judged by variation in plumage, appears to be wide and deep.

Pezoporus wallicus (Kerr, 1792)-Infra-specific differentiation in eastern Australian populations is disputed, Lendon (1973), Condon (1975) and Forshaw (1981) accepting none, and Peters (1937), Condon (1942) and Ford (1969) separating the Tasmanian isolate. The latter interpretation is followed here pending comparison of fresh unfaded south-eastern Australian and Tasmanian material; Tasmanian populations appear to be deeper green and more heavily marked with black than those on the mainland, particularly their females.

Platycercus caledonicus (Gmelin, 1788)-Contrary to convention, King Island populations, which were isolated from Tasmania well before those on the other Bass Strait islands (Schodde 1982: 47), are separated subspecifically here, being duller in tone with a green cast to the ventrum, orange cast to the brow, and citrine cast to the coppery rump (material in QVM). Such a situation highlights the need for a comprehensive analysis of differentiation between Tasmanian and all Bass Strait island populations. Cain (1955) has argued a case, supported by data in Ovenden et al. (1987: figs 3-5), for treating this group as specifically discrete from P. elegans (Gmelin, 1788), cf. Peters (1937).

Platycercus elegans (Gmelin, 1788)-Specific and infra-specific systematics is subject to continued argument. Here the principal forms (crimson elegans Gmelin, 1788, orange 'adelaidae` Gould, 1841, and yellow flaveolus Gould, 1837) are treated as members of one polytypic species for three reasons:

(1) they are allopatric/parapatric yet intergrade wherever they meet (Boehm 1959; Joseph & Hope 1984; Ford 1987);

(2) they form a closely related cluster relative to other rosellas, judged by both morphology and their mitochondrial genomes (Ovenden et al. 1987: figs 3-5); and

(3) most recent revisers have combined them in this way, see Condon (1941, 1975), Cain (1955), Immelmann (1966) and Wolters (1975-1982).

Of minor regional forms, the status of crimson populations in central-east Queensland and of orange populations in the Mt Lofty Range, SA, needs further clarification. Although McAllan & Bruce (1989) distinguished the former subspecifically, a larger range of material in ANWC and QM shows it to be no darker or smaller than populations on the coast ranges of the Brisbane River valley at the northern end of a cline in these traits in nominotypical elegans; so the central-east Queensland enclave is not distinguished here pending evaluation of its reportedly red juveniles. Populations through the Mt Lofty Range north to the Bundaleer hills are individually variable and intermediate on a gradient between a rather uniformly orange-red form on Fleurieu Peninsula (fleurieuensis Ashby, 1917) and a similarly uniform but orange-yellow form in the south Flinders Range (subadelaidae Mathews, 1912). Accordingly, the central Mt Lofty Range populations are treated here as intergrades between fleurieuensis and subadelaidae and the name based on them (adelaidae Gould, 1841) as indeterminable, see Incertae sedis. By implication, fleurieuensis and subadelaidae represent ancestral populations; their relationships to one another and to other members of the complex, however, need clarification. For assessments of the status of populations in the Mt Lofty Range, see Ashby in Mathews (1916, introductory chapter), Ashby (1925), Condon (1941, 1954), Serventy (1953), Cain (loc. cit.), and Joseph & Hope (loc. cit.). The crimson isolate on Kangaroo Island is distinct in the obsolete crimson scalloping on its mantle and in its black nuchal patch.

Platycercus eximius (Shaw, 1792)-The principal forms of white-cheeked rosellas-red-headed eximius Shaw, 1792, white-headed adscitus Latham, 1790, and black-headed venustus Kuhl, 1820-may prove members of one polytypic species, given the morphological intermediacy of Cape York Peninsula populations between white-headed populations to the south and venustus in north-western Australia, see Schodde & Tidemann (1986). Nevertheless, only adscitus is combined with eximius here, for two reasons consistent with the criteria for taxonomic definition used in this Catalogue:

(1) adscitus is established as intergrading with eximius in the NSW-QLD border region (Cannon 1984; Ford 1987);

(2) no case has yet been made in the primary scientific literature for combining allopatric venustus as well.

Ovenden et al. (1987) also recorded adscitus as closer to eximius than venustus. Cannon's (loc. cit.) reported segregation between adscitus and eximius in their zone of intergradation cannot be taken at face value because her identifications were based on field observations incapable of determining all intergrading individuals, particularly those approaching parental forms phenotypically. Cape York Penisula populations are kept with the pale-headed group, adscitus Latham, 1790, as is conventional; and only two subspecies are accepted in that group, after Condon (1941), Forshaw (1981) and Ford (1986). Of the additional subspecies recognised by Cain (1955), mackaiensis Cain, 1955 is indistinguishable from southern palliceps Lear, 1832, and elseyi Mathews, 1912 is from the zone of intergradation between Cape York Peninsula and more southerly populations and so unidentifiable, see Incertae sedis.

Platycercus venustus (Kuhl, 1820)-Although Condon (1941), Cain (1955), Immelmann (1966) and Forshaw (1981) rejected subspecies, Kimberley and Arnhem Land populations differ in both the extent of blue edging to cheek patches and of black scalloping on the ventrum (Lendon 1973: 192).

Psephotus dissimilis Collett, 1898-Despite a trend to combine this form with P. chrysopterygius Gould, 1858 (Cain 1955; Wolters 1975-1982; Forshaw 1981), it is kept separate here for two reasons. Firstly, its differences in size, sexual dimorphism and patterns on head and ventrum are considerable and relatively greater, for example, than those between Northern and Cape York Peninsula Eastern Rosellas (Platycercus venustus (Kuhl, 1820) and P. eximius adscitus (Latham, 1790)). Secondly, to merge it with chrysopterygius creates an unbalanced taxonomy with respect to allospecific Psephotus pulcherrimus (Gould, 1845), see Introduction (Taxonomic scope) to this volume. Speciation among all three members of subgenus Psephotellus Mathews, 1913 is in need of clarification.

Psephotus varius Clark, 1910-Males of south-eastern Murray Mallee populations are marginally richer green dorsally and deeper red ventrally than elsewhere, but the differences are slight and clinal, and do not justify the subspecies recognised by Peters (1937), Condon (1941, 1975) and Cain (1955).

Eclectus roratus (P.L.S. Müller, 1776)-It has not been appreciated that members of this species visiting the northern Torres Strait islands of Boigu, Dauan and Saibai adjacent to the southern New Guinea coast are of the New Guinea sub-species, E. r. polychloros Scopoli, 1786 which differs from the Cape York Peninsula form in its brighter grass-green males and smaller size: male wings about 250-270 mm and female wings about 240-260 mm vs. about 285-300 mm and about 275-290 mm respectively on Cape York Peninsula. For a survey of infra-specific groups, see White & Bruce (1986).

Geoffroyus geoffroyi (Bechstein, 1811)-Although usually combined subspecifically with the southern New Guinean form (e.g. Forshaw 1966), Cape York Peninsula populations nevertheless differ in their lighter, pale sky-blue underwing coverts, reduced occipital blue in males, and (?) deeper plum-chestnut heads in mature females. Circumscription of the species is based on Rothschild & Hartert (1901).

Cyclopsitta diophthalma (Hombron & Jacquinot, 1841)-Although Mayr (1947) correctly presumed that Cape York Peninsula populations were the Australian representatives of southern New Guinean C. p. aruensis Schlegel, 1874, Forshaw (1967) subsequently showed them to be distinct subspecifically.

Alisterus scapularis (Lichtenstein, 1816)-Recorded geographic limits of the two recognised Australian subspecies are conflicting, see Condon (1975), cf. Forshaw (1981). Those defined here are based on a review of all material in all Australian museums, including recently gathered specimens from the Clarke Range and Kroombit Tops, central-eastern Queensland, in the ANWC and QM. They indicate that large nominotypical scapularis extends north to the Clarke Range (pace Forshaw loc. cit.) and that small, dark A. s. minor Mathews, 1911 ranges no further south than the Seaview-Paluma Ranges, north Queensland (pace Condon loc. cit.).

Diagnosis

Small to medium-sized, short-necked and brilliantly coloured birds with hooked and cered bills, and plumage coloured by non-carotenoid pigments of as yet unknown composition and back-scattering of light through hollow cortical cylinders (blue-and-green 'dyck texture') in feather barbs; body feathering rather coarse, in diffuse tracts, powder downs obsolete; aftershafts present, downy; uropygial gland developed and tufted (except some Arinae). Feet zygodactylous and specialized for climbing; tarsi short with fine reticulate scaling, toes longer, reticulately scaled, may or may not hold food to bill. Sexes usually similar; markedly dimorphic only in Eclectus and some Neophema. Wings rounded or pointed, with or without coloured stripes or patches: 10 primaries moulting usually regularly in both directions from the 6th, and 10 diastataxic secondaries moulting descendently, at least in Platycercinae; tail long and/or tapered: 12 rectrices (14 in Oreopsittacus) moulting erratically. Nares holorhinal and impervious, nasal septum imperforate; desmognathous palate without vomer, maxillaries articulating with deep, twisted and laterally flattened palatines, maxillae attached to skull by a flexible joint, lachrymals extended in processes that rarely reach squamosals, leaving orbit open; basipterygoid processes absent; cervical vertebrae 13-14; sternum entire or 1-fenestrated on each side, only spina externa present, clavicles ankylosed in furcula or vestigial. Musculus expansor secundariorum and biceps slip absent, M. tensores patagii huge and covering the M. deltoidei, M. t. p. brevis with a wristward slip, and anterior M. deltoideus usually larger than the posterior (except Nestorini); pelvic muscle formula AXY, M. ambiens present or absent (Australian continental taxa); deep plantar tendons Type I. Carotid arteries paired, fused or left superficial. Syrinx bronchial, the first rings upcurved, ossified and fused (except Nestorini). Tongue muscular, tactile, grooved, pocked or with a cluster of eversible papillae enclosed at tip, moved by hyoid apparatus with extensive median foramen in entoglossum and broad basihyal process; crop and zonary proventriculus present; no gall bladder; no caeca. Diploid karyotype of c. 70 chromosomes, with 6-7 pairs of non-telocentric macrochromosomes (examined taxa limited).

Published	As part of group	Action Date	Action Type	Compiler(s)
10-Nov-2020		01-Mar-2012	MODIFIED
		25-Aug-2011	MODIFIED

Australian Biological Resources Study

Australian Faunal Directory