Ibis zyxwvutsrqp (1999) zyxwvutsrq BA 141, 644-651 Sexual dimorphism in the North Island Kaka zy aZ FE Nestor meridionalis septentrionalis: selection for enhanced male provisioning ability? RON J. MOORHOUSE'*, MlCK J. SIBLEY2, BRIAN D. LLOYD3 & TERRY C. GREENE4 'School of Biological Sciences, Victoria University of Wellington, PO Box 600, Wellington, New Zealand 2Auckland Zoological Park, Motions Rd, Western Springs, Auckland, New Zealand 3Department of Conservation, PO Box 10-420, Wellington, New Zealand 4Department of Conservation, Private Bag 68-908, Auckland 1, New Zealand zyx DC The North IslandKaka zyxwvu HGFED Nestor rneridionalis septentrionalis, an endemic New Zealand parrot, displays a disproportionate degree of sexual dimorphism in bill-size, males being 13.6% and 12.4%) larger than females in culmen-length and -depth respectively, but only 2.0Yo and 4.2'yo larger in tarsus-length and cube root of body-mass. Culmen-length provides a reliable means of sexing Kakas if the age-class of each bird is known; all adult and subadult males had exposed culmens 2 47 mm long, while those of females were less than this value. Juveniles with culmen-lengths in excess of 44 mm were male. Similar sexual dimorphism in the Kea zyxwvu FEDCBA N. notabilis, the sole extant congener of the Kaka, suggests a phylogenetic basis for this condition. The monogamous mating system of the Kaka and Kea, together with the prolonged provisioning of females and young by males in both species, suggests that selection for enhanced male provisioning ability, rather than sexual selection, could be maintaining sexual bill dimorphism in these species. Sexual size dimorphism is a common phenomenon in birds, males being larger than females in most species (Amadon 1959, Lack 1968, Jehl & Murray 1986). The vast majority of parrot species (Psittaciformes) are, however, monomorphic in body size (Smith 1975, Forshaw 1989). As sexual size dimorphism is usually associated with intermale competition for females or breeding territories (Murray 1984, Payne 1984, Oakes 1992), its rarity in parrots suggests reduced inter-male competition compared with other avian taxa. This is consistent with the virtually universal monogamy and non-territoriality of parrots as a group (Forshaw 1989). Sexual size dimorphism has been recorded in only seven of the 81 parrot genera (op. zyxwvut A cit.). Four of these, Nestor, Probosciger, Barnardius and Platycercus (nomenclature follows Forshaw 1989) are further unusual in that sexual size differences are predomi- nantly manifested in bill-size, rather than in overall body size (op. cit.). In the most extreme example, a *Correspondingauthor. Present address: Department of Conservation, Private Bag 5, Nelson, New Zealand. Email: rmoorhouse@doc.govt.nz subspecies of the Palm Cockatoo Prubosciger aterrimus stenolophus, males exceed females by an average of 27.4 mm (24.8%) in length of the exposed culm but are only 2.1 mm (5.8%) larger in tarsus-length (up. cit.). Such disproportionate sexual dimorphism in bill- size suggests that the latter may be an adaptation, rather than simply an allometric consequence o overall size difference between the sexes (Selander 1966, 1972). The evolutionary significance of sexual dimorphi in feeding morphology is the subject of conside debate (Selander 1972, Payne 1984, Jehl & Murray 1986, 1989, Mueller 1989, Shine 1989, Jonsson z L X Alerstam 1990). Excluding the possibility that such differences are merely epiphenomena, or 'spandrels' with no adaptive significance (Could & Lewontin 1979), possible explanations for sexual differences in bill-size are: Sexual selection: intersexual differences in bill-size evolved as a consequence of selection on males fo enhanced courtship or combat ability (Darwin 1 Huxley 1938, Trivers 1972). @ 1999 British Ornithologists' Union