Phylogeny, biogeography, and taxonomy of Australasian teals

Auk, The, Jan 2000 by Kennedy, Martyn, Spencer, Hamish G

ABSTRACT.-The taxonomy of the Australasian teals has been particularly unstable. Australasian Grey Teal (Anas gracilis) and Chestnut Teal (A. castanea) are widely viewed as specifically distinct, but the taxonomy of the New Zealand teals remains unsettled. Because conservation status is affected by taxonomic rank, it is important to resolve the status of the rare subantarctic teals. To estimate phylogenetic relationships of teals, we sequenced three mitochondrial DNA genes (12S, and ATPase 6 and 8). The resultant phylogeny unequivocally groups the Chestnut Teal with the Grey Teal, rather than with the New Zealand teals as has traditionally been held (Fleming 1953). A greater level of sequence divergence occurred within the New Zealand teals than between the Grey and Chestnut teals. This diversity, together with morphological and behavioral differences, implies that the New Zealand teals should be accorded specific status as A. aucklandica, A. nesiotis, and A. chlorotis. Although it is most likely that the teal that colonized the Auckland Islands and Campbell Islands originated in New Zealand, our data do not allow us to determine whether the ancestors of the Campbell Island Teal came from mainland New Zealand or the Auckland Islands. This uncertainty arises because, as our data show, the colonization events were separated by a short period of time. Received 25 November 1998, accepted 16 June 1999.

FIVE EXTANT TEALS inhabit the Australasian region and provide an interesting example of speciation in insular Southern Hemisphere waterfowl (see Livezey 1990). Grey Teal (Anas gracilis), which are small and drab, are widespread throughout Australia. Last century the species was rare and localized within New Zealand, but it is now widely distributed (Marchant and Higgins 1990). Grey Teal are broadly sympatric with the Chestnut Teal (A. castanea) in southeastern and southwestern Australia and with the Brown Teal (A. chlorotis) in mainland New Zealand (Marchant and Higgins 1990); vagrant Chestnut teal have been reported in New Zealand on several occasions (Guest 1992). Chestnut Teal are strongly sexually dichromatic; the male has a green head, chestnut breast, and dark upperparts, and the female closely resembles a Grey Teal. Brown Teal are less obviously dichromatic, with the female a more uniform brown than the male, which has a deep chestnut breast and a green iridescence on the crown and nape. Brown Teal were formerly widespread throughout North, South and Stewart Islands but are now rare and restricted in their distribution, with the main populations in the north of North Island and on Great Barrier Island (Marchant and Higgins 1990). The brown-plumaged teals (i.e. all teals except the Grey Teal) all exist in allopatry. The Auckland Island Teal (A. aucklandica) and Campbell Island Teal (A. nesiotis) occur only on their respective subantarctic island groups (see Fig. 1). They are smaller and darker than the Brown Teal, and unlike this species, they are flightless (Marchant and Higgins 1990).

The taxonomy of the Australasian teals has been the subject of much debate (e.g. Dumbell 1986, Williams and Robertson 1996). Anas gracilis, for example, was previously treated as a subspecies of A. gibberifrons of Indonesia (Livezey 1991). Although now universally accepted as separate species (e.g. Marchant and Higgins 1990, Turbott 1990, Livezey 1991, Christidis and Boles 1994), occasional debate has questioned whether Grey and Chestnut teals warrant separate specific status (Frith 1967) despite their widespread sympatry.

The taxonomic history of the brown-plumaged teals remains unresolved (see Dumbell 1986). Fleming (1953) considered the Brown, Auckland Island, and Campbell Island teals (the endemic New Zealand teals) to be subspecies of the Chestnut Teal, a position accepted by Frith (1967). More recently, however, the New Zealand teals have been accepted as specifically separate from the Chestnut Teal, although several different taxonomic arrangements have been proposed. One arrangement considers them to be three allopatric subspecies (Kinsky 1970, Dumbell 1986, Turbott 1990), another accords all of them separate specific status (Marchant and Higgins 1990), and yet another accords specific status to the Brown Teal and Auckland Island Teal but places the Campbell Island Teal as a subspecies of A. aucklandica (Livezey 1990).

Using morphological characters, Livezey (1991) generated a phylogeny for dabbling ducks that included most of the Australasian teals (Fig. 2). This phylogeny suggests a pattern of speciation in which the monochromatic Grey Teal separated from the common ancestor of the dichromatic brown-plumaged teals within Australia. Within the brown-plumaged teals, the Chestnut Teal later diverged from the progenitor of the New Zealand teals, the latter presumably invading New Zealand from Australia. The Auckland Island Teal then diverged from the Brown Teal after colonizing from the New Zealand mainland. In an earlier paper, Livezey (1990) presented a tree of the same topology that included only the teal and included the Campbell Island Teal as sister taxon to the Auckland Island Teal, consistent with the traditional hypothesis that the Campbell Islands were colonized from the Auckland Islands (Dumbell 1986). Alternatively, if the Auckland and Campbell Island teals are not sister taxa, the Auckland and Campbell Island colonizations represent separate events from the mainland, with Brown Teal stock giving rise to both species (Turbott 1968, Dumbell 1986). Livezey's (1990, 1991) phylogeny implies that the strong sexual dichromatism of the Chestnut Teal represents the ancestral state and is thus reduced in the insular forms, a pattern that traditionally has been interpreted as the loss of an isolating mechanism (i.e. dichromatism isolating the Grey and Chestnut teals).