Department of Zoology, University of Queensland, Brisbane, Queensland 4072, Australia; and 2Department of Ecology, Massey University, Palmerston North, New Zealand In many cooperatively breeding species of birds, adult males are thought to outnumber adult females (e.g. Red-cockaded Woodpecker [Picoides borealis], Gowaty and Lennartz 1985; Splendid Fairy-Wren [Malurus splendens], Rowley and Russell 1990; Pied Kingfisher [Ceryle rudis], Reyer 1990; see Emlen 1984, Brown 1987). The occurrence of male-biased sex ra- tios in some populations of species with helpers-at- the-nest has led to the hypothesis (the differential mortality model of Emlen et al. [1986]) that a short- age of females could explain--at least in part--de- layed dispersal, helping behavior, and mate-sharing by males (Rowley 1965, Maynard Smith and Ridpath E-mail: agoldizen@zoology.uq.edu. au 1972, Emlen 1984, Curry and Grant 1989, Reyer 1990, Davies 1992). A shortage of females might result from a higher rate of mortality compared with males, perhaps associated with female-biased dis- persal. An experimental test on Superb Fairy-Wrens (Malurus cyaneus) provided support for this model (Pruett-Jones and Lewis 1990). A classic example in the debate on the link be- tween sex ratios and cooperative breeding is the Tas- martian Native Hen (Gallinula mortierii). Ridpath (1972b) reported male-biased sex ratios among both adults (1.5 males per female) and chicks (2.8 males per female) in his study population, and an overall sex ratio of 1.22 males per female among 489 indi- viduals collected near his study area and sexed by dissection. Maynard Smith and Ridpath (1972) used this evidence to propose that a male-biased sex ratio sets the stage for the occurrence of "wife-sharing" in this species by causing a shortage of available breed- ing females. This case has been cited in reviews as an example where male-biased sex ratios may lead to mate-sharing by males (e.g. Oring 1986, Emlen et al. 1986, Emlen 1991). Here, we present data on sex ra- tios and mating patterns from two populations of Tasmanian Native Hens to challenge the assumption that cooperative breeding in this species occurs as a result of a male-biased sex ratio. Tasmanian Native Hens are flightless rails endem- ic to Tasmania that live in areas with a combination of open pasture, dense vegetative cover, and water. Tasmanian Native Hens live in groups of 2 to 17 in- dividuals that defend territories year-round (Rid- path 1972b). Their mating system includes frequent mate-sharing by both sexes, at least in some popu- lations (Goldizen et al. 1998). Tasmanian Native Hens exhibit monogamy (a single male mated to a single female), cooperative polyandry (two or more males mated to a single female), polygyny (two or more females mated to a single male), and polygyn- andry (two or more males and two or more females all breeding together). In groups with multiple breeding females, the females lay eggs in communal clutches. In groups with multiple breeding males, all such males copulate with the female(s) and then as- sist with all aspects of parental care (Ridpath 1972a, b, Gibbs et al. 1994). Mate-sharing in Tasmanian Na- tive Hens is more common among males than among females and usually involves closely related males (Ridpath 1972b). Helping behavior also is exhibited by some nonbreeding one- and two-year-olds of both sexes that remain in their natal groups (Ridpath 1972a, b, Gibbs et al. 1994), but such helping behav- ior usually is minor relative to that provided by breeding adults (Goldizen and Goldizen unpubl. data). We studied our first Tasmanian Native Hen pop- ulation in cattle pastures near Geeveston, in southern Tasmania, Australia (4310'S, 146 55'E), from De- cember 1988 to December 1989 (Goldizen et al. 1993). Our second study population inhabited an area of cleared pasture surrounded by eucalypt woodland at the northern end of Maria Island off the eastern coast of Tasmania (4235'S, 148 04'E). We studied this population through the consecutive breeding seasons of September to December, 1990 to 1996. We observed the population continuously through each of these breeding seasons. Birds were trapped, banded with numbered metal bands and unique combinations of color bands, weighed, measured, and aged; blood samples were collected as described by Gibbs et al. (1994). A total of 83 native hens was banded in the Geeveston pop- ulation. At Maria Island, 135 of 146 birds present in the population were trapped and banded between December 1989 and August 1990. After that, all young were banded between six weeks and three months of age; immigrants were banded as soon as possible after their establishment in the study pop- ulation. In totaI, 371 birds were banded at Maria Is- land. We found no evidence for a sex bias in capture probability. Of the 11 birds that we failed to catch at Maria Island in 1990, 8 that we captured subsequent- ly included four males and four females. The sex of Tasmanian Native Hens cannot be dis- tinguished by external characteristics. Ridpath (1972b) determined the sex of individuals either by behavior, which is reliable only for some individuals, or by dissection. We determined the sex of 74 indi- viduals from the Geeveston population by laparos- copy or dissection. At Maria Island, we assigned the sex of 231 of the 371 captured birds by laparoscopy, 118 by genetic means, and 12 using other criteria (e.g. copulation positions). Sex could not be deter- mined for 10 of the banded birds. Sex determination by laparoscopy involved the di- rect examination of gonads using the procedures de- scribed in Gibbs et al. (1994). The surgical procedure had no observable negative effects on the birds; all 231 individuals that underwent surgery at Maria Is- land were alive and well at least one week after sur- gery. We did not resight all of the birds that under- went surgery at Geeveston, but we believe that this was due to a high level of social instability in that population (Goldizen et al. 1993). Sex determination by laparoscopy also was extremely accurate; no bird sexed by this method was observed behaving in a manner contradictory to its assigned sex. We used two different methods for the genetic determination of sex. The first employed techniques similar to those given in Millar et al. (1996), with the following mod- ification: Because only a single female-specific DNA fragment was detected in Tasmanian Native Hens, a control gene, beta-actin, subsequently was hybrid- ized to ensure the presence of sufficient high-quality DNA in those individuals assigned as males. The techniques used in the second method are described by Griffiths (unpubl. data). Copulation position was used to determine sex only if the bird was seen cop- ulating more than once. We monitored the compositions of most of the Ma- ria Island groups daily through each breeding sea- son. Data on the frequencies of the different mating patterns (monogamy, polyandry, polygyny, polygyn- andry) refer to the mating pattern of each group each year (group-year), either at the time when the group laid its first clutch of eggs or at the beginning of No- vember (the middle of the breeding season), which- ever came first. The criteria used to determine which birds were breeders are presented in Goldizen et al. (1998). We compiled data on adult sex ratios from the information on the memberships of all groups pres- ent in this population each year at the beginning of November. Adults included all individuals that hatched in the previous breeding season or earlier TABLE 1. Juvenile (i.e. hatched during current breeding season) and adult (i.e. hatched during a previous breeding season) sex ratios in the Maria Island population of Tasmanian Native Hens. 1989 1990 1991 1992 1993 1994 1995 1996 All Juveniles No. of males 31 29 6 21 6 8 15 19 135 No. of females 29 19 11 20 5 8 13 18 123 No. unknown 0 0 1 6 0 0 1 0 8 Ratio M:F 1.07 1.53 0.55 1.05 1.20 1.00 1.15 1.06 1.10 Gad  0.066 2.077 1.450 0.024 0.087 0.000 0.140 0.027 0.557 P greater than 0.70 0.10 0.20 0.80 0.70 0.99 0.70 0.80 0.30 Adults No. of males -- 72 88 86 83 75 71 75 -- No. of females -- 65 72 63 73 61 64 66 -- No. unknown -- 5 4 4 8 7 7 13 -- Ratio M:F -- 1.11 1.22 1.37 1.14 1.23 1.11 1.14 -- Gad  -- 0.357 1.598 3.553 0.639 1.438 0.362 0.573 -- P greater than -- 0.50 0.20 0.05 0.30 0.20 0.50 0.30 -- G-test with Williams' correction testing for departure from sex ratio of 1:1. Immature sex ratios for the Maria Island population were calculated from the total of all male and female young that hatched and survived to the age of first banding (between six weeks and three months) each season. For the Geeveston population, because of smaller sample sizes and an inability to accurately determine the age of many individuals, we calculat- ed a sex ratio based on all individuals of all ages trapped during 1988. Results.--At Geeveston, 74 immatures and adults (an estimated three-quarters of the total population) were successfully sexed by laparoscopy or dissec- tion; 9 others could not be sexed. The sex ratio of this sample was not significantly different from parity (35 males, 39 females; sex ratio = 0.90 males per fe- male; G-test with Williams' correction, G = 0.215, df = 1, P > 0.5). The sex ratio of 266 immature birds that hatched at Maria Island during the 1989 through 1996 breeding seasons also was not significantly dif- 0.40  o 30 0.00 by males by females 1.10 1.15 1.20 1.25 1 30 1.35 1.40 annual adult sex ratio (M:F) FIG. 1. Relationship between annual adult sex ra- tio and frequency of mate sharing by males and fe- males in the Maria Island population of Tasmanian Native Hens. ferent from 1:1 (1.10 males per female; Table 1). The sex ratios of the young that hatched during individ- ual breeding seasons varied from 0.55 males per fe- male in 1991 to 1.53 males per female in 1990. There was no significant sex-ratio bias for any year's cohort of young (Table 1). However, with such small yearly sample sizes, the chances of sex ratios being signifi- cantly biased are small (Koenig and Dickinson 1996). The sex ratio of all adults in the Maria Island pop- ulation ranged from a low of 1.11 males per female in 1990 and 1995, to a high of 1.37 males per female in 1992 (Table 1). We have no information on adult sex ratio for the 1989 season because we began band- ing in December 1989. No annual adult sex ratio was significantly different from 1:1 (Table 1). These sep- arate annual sex ratios are not statistically indepen- dent because many individuals were present during successive years. Of the 258 group-years for which the mating pat- tern was known with certainty, 56% were monoga- mous, 23% were polyandrous, 13% were polygy- nous, and 8% were polygynandrous. Thus, 31% of group-years exhibited mate-sharing by males, whereas females shared mates in 21% of group- years. The annual proportion of groups exhibiting mate-sharing by males varied from 0.23 (1996) to 0.37 (1992), whereas the proportion of groups exhib- iting mate-sharing by females ranged from 0.13 (1991) to 0.26 (1992). There was no relationship be- tween annual variation in sex ratios and mate-shar- ing frequencies for either males or females (rank re- gressions, n = 7; for males, r 2 = 0.241, P = 0.263; for females, r 2 = 0.001, P = 0.938; Fig. 1). Each year there were many nonbreeding but re- productively mature birds present in their natal groups at Maria Island (Table 2). Tasmanian Native Hens are capable of breeding at one year of age; in- TABLE 2. Number of nonbreeding adults in natal groups in the middle of each breeding season (i.e. 1 November) in the Maria Island population of Tasmanian Native Hens. 1990 1991 1992 1993 1994 1995 1996 No. of males 22 28 14 13 13 8 9 No. of females 20 23 7 12 8 2 10 deed, most yearlings bred in Ridpath's study popu- lation (Ridpath 1972b). Discussion.--ln the two populations of Tasmanian Native Hens that we studied, sex ratios were not sig- nificantly male biased. In contrast to the small de- gree of male bias in our study population, Ridpath (1972b) reported sex ratios of 2.8 males per female among a sample of 68 immatures (four to six months old) and 1.5 males per female among 185 adults in his study population. Although the sex-determina- tion procedures used by Ridpath are not totally re- liable, it is unlikely that sexing errors alone account- ed for the difference in the observed sex ratios be- tween Maria Island and Ridpath's study site. It also is unlikely that Ridpath would have captured males more easily than females, because the patterns of habitat use by the sexes do not differ in any obvious way, and both sexes participate in territorial defense (Ridpath 1972a, b). It is not clear why the sex ratios of both immature and adult birds in Ridpath's pop- ulation were more male biased than were the sex ra- tios in our populations. What is clear, however, is that not all populations of TasmanJan Native Hens are strongly male biased. Despite differences in sex ratios between the Maria Island population and Ridpath's population at Hunt- ing Ground in central Tasmania, the social systems exhibited by these two populations were quite sim- ilar (Ridpath 1972b, Goldizen et al. 1993, Gibbs et al. 1994, Goldizen and Goldizen unpubl. data). For in- stance, the percentage of group-years that displayed mate-sharing was 44% at Maria Island and 51% in Ridpath's population (Ridpath 1972b). The lack of a relationship between adult sex ratios and frequencies of mate-sharing in seven seasons at Maria Island (Fig. 1) provides further evidence that mate-sharing is not related to sex-ratio biases in this species. Additional evidence against the hypothesized link between biased sex ratios and mate-sharing by males is provided by two other observations. First, in each year of our study at Maria Island, both males and fe- males of reproductive age have remained in their na- tal groups without breeding (Table 2). This indicates that there is no shortage of available breeding birds of either sex, and this is the necessary link in May- nard Smith and Ridpath's (1972) argument. Second, mate-sharing also was commonly practiced by fe- males, occurring in approximately 21% of all group- years at Maria Island (Fig. 1) and, in fact, was most frequent in 1992, the year with the most male-biased sex ratio. This last point makes it obvious that we need to examine alternatives to biased sex ratios to determine which factors are associated with mate- sharing in TasmanJan Native Hens. Ridpath (1972b) also observed groups with mul- tiple adult females in his study population at Hunt- ing Ground, but Maynard Smith and Ridpath (1972) did not mention this in their discussion on the evo- lution of mate-sharing. Moreover, Ridpath (1972b) observed that most "roving" birds of both sexes (i.e. first-year birds searching for their first breeding po- sition) eventually left the study area without set- tling. This suggests that potential mates were not in short supply. In conclusion, the mate-sharing frequently exhib- ited by male and female Tasmanian Native Hens does not necessarily occur because of a lack of po- tential mates for either sex. Apparently, a slight male bias in the sex ratio existed in our Maria Island pop- ulation, which may have affected the frequencies of mating patterns to some extent, but this sex bias does not explain the occurrence of mate-sharing in this species. Our long-term study at Maria Island suggests that a shortage of high-quality breeding territories (i.e. territories with water and large amounts of edge be- tween pasture and thick cover) is the limiting factor causing mate-sharing in this population. Multivar- iate models used to investigate relationships be- tween reproductive success and a number of aspects of group compositions and territories found that breeding success was most significantly related to the amount of edge between pasture and thick cover (Goldizen et al. 1998). However, we do not under- stand how this link between territory quality and breeding success promotes mate-sharing. Acknowledgments.--We thank J. Cornell, T. Devlin, E. Kelchlin, E. Krebs, T. Male, M. Mather, D. Mc- Cutchen, K. Mcintyre, G. Roberts, K. Robertson, J. Schade, K. Taylor, and numerous Earthwatch volun- teers for help with field work. We also thank D. Rounsevell and past and present rangers at Maria Is- land for their generous logistical support. Financial support was provided by a National Science Foun- dation Environmental Biology Postdoctoral Fellow- ship, the Australian Research Council, the National Geographic Society, Earthwatch, the Frank M. Chap- man Fund of the American Museum of Natural His- tory, and the M. A. Ingram Trust of Victoria, Austra- lia. An otoscope was kindly donated by Thomas Wakefield. Thanks to the New Zealand Foundation for Research Science and Technology for funding some of the genetic component of this research. Fi- nally, we thank K. E. Arnold, I. G. Jamieson, T. Male, K. Martin, I. P. E Owens, ED. Putland, G. Roberts, and K. Robertson for commenting on the manu- script. LITERATURE CITED BROWN, J. R. 1987. Helping and communal breeding in birds. Princeton University Press, Princeton, New Jersey. CURRY, g. L., AND P. g. GRANT. 1989. Demography of the cooperatively breeding Galapagos Mocking- bird, Nesomimus parvulus, in a climatically vari- able environment. Journal of Animal Ecology 58: 441-463. DAVIES, N. B. 1992. Dunnock behaviour and social evolution. Oxford University Press, Oxford. EMLEN, S. t. 1984. Cooperative breeding in birds and mammals. Pages 305-339 in Behavioural ecolo- gy: An evolutionary approach (J. R. Krebs and N. B. Davies, Eds.). Sinauer Associates, Sunder- land, Massachusetts. EMLEN, S. t. 1991. Evolution of cooperative breeding in birds and mammals. Pages 301-337 in Behav- ioural ecology: An evolutionary approach, 3rd ed. (J. R. Krebs and N. B. Davies, Eds.). Blackwell Scientific Publications, Oxford. EMLEN, S. T., J. M. EMLEN, AND S. g. LEVIN. 1986. Sex- ratio selection in species with helpers-at-the- nest. American Naturalist 127:1-8. GIBBS, H. L., A. W. COLDIZEN, C. BULLOUCH, AND A. R. COLDIZEN. 1994. Parentage analysis of multi- male social groups of TasmanJan Native Hens (Tribonyx mortierii): Genetic evidence for monog- amy and polyandry. Behavioral Ecology and So- ciobiology 35:363-371. COLDIZEN, A. W., A. R. COLDIZEN, AND T. DEVLIN. 1993. Unstable social structure associated with a population crash in the Tasmanian Native Hen, Tribonyx mortierii. Animal Behaviour 46:1013- 1016. COLDIZEN, g. W., D. g. PUTLAND, AND g. R. COLDI- ZEN. 1998. Variable mating patterns in Tasma- njan Native Hens (Gallinula mortierii): Correlates of reproductive success. Journal of Animal Ecol- ogy 67: in press. GOWATY, P. A., AND M. g. LENNARTZ. 1985. Sex ratios of nestling and fledgling Red-cockaded Wood- peckers (Picoides borealis) favor males. American Naturalist 126:347-353. KOENIG, W. D., AND J. L. DICKINSON. 1996. Nestling sex-ratio variation in Western Bluebirds. Auk 113:902-910. MAYNARD SMITH, J., AND M. G. RIDPATH. 1972. Wife sharing in the Tasmanian Native Hen, Tribonyx mortierii: A case of kin selection? American Nat- uralist 106:447-452. MILLAR, C. D., D. M. LAMBERT, S. ANDERSON, AND J. L. HALVERSON. 1996. Molecular sexing of the communally breeding Pukeko: An important ecological tool. Molecular Ecology 5:289-293. ORING, L. W. 1986. Avian polyandry. Current Orni- thology 3:309-351. PRUETT-JONES, S.G., AND M. J. LEWIS. 1990. Sex ratio and habitat limitation promote delayed dispers- al in Superb Fairy Wrens. Nature 348:541-542. REYER, H.-U. 1990. Pied Kingfishers: Ecological caus- es and reproductive consequences of coopera- tive breeding. Pages 529-557 in Cooperative breeding in birds: Long-term studies of ecology and behavior (P. B. Stacey and W. D. Koenig, Eds.). Cambridge University Press, Cambridge, United Kingdom. RIDPATH, M. G. 1972a. The TasmanJan Native Hen, Tribonyx mortierii. I. Patterns of behaviour. CSI- RO Wildlife Research 17:1-51. RIDPATH, M. G. 1972b. The Tasmanian Native Hen, Tribonyx mortierii. II. The individual, the group and the population. CSIRO Wildlife Research 17: 53-90. ROWLEY, I. 1965. The life history of the Superb Blue Wren, Malurus cyaneus. Emu 64:251-297. ROWLEY, I., AND E. RUSSELL. 1990. Splendid Fairy- wrens: Demonstrating the importance of lon- gevity. Pages 3-30 in Cooperative breeding in birds: Long-term studies of ecology and behav- ior (P. B. Stacey and W. D. Koenig, Eds.). Cam- bridge University Press, Cambridge, United Kingdom. Received 10 February 1997, accepted 20 October 1997. Associate Editor: K. Martin