Search for a command to run...
In hermaphrodites, the of sexuality can favor the spread of parthenogenesis in two ways. First, it can promote higher female fecundity in parthenogens that have reduced to the male function. Second, if parthenogens have retained a fertile male function, they can spread genes for parthenogenesis into a coexisting sexual population. We present evidence for both effects in a natural population of the flatworm Schmidtea polychroa. Parthenogens, which have a reduced male function, had 42% higher female fecundity than coexisting sexuals. New, presumably parthenogenetic, triploids arose out of the diploid sexual population at a frequency of 1.3%, probably as a result of gene flow from parthenogens to sexuals. However, we could also identify a strong compensating fitness benefit for sexuals: they had substantially higher female fertility than coexisting parthenogens, both in terms of cocoon fertility (93% and 74% respectively) and offspring per fertile cocoon (3.6 and 2.8 respectively). Additional key words: evolution of sex, sex allocation, Platyhelminthes, Dugesia, Schmidtea Evolutionary theory predicts that strong counteracting selective forces act on sexual and parthenogenetic reproduction. On the one hand, it is widely accepted that sex pays the high cost of male allocation (Maynard Smith 1978) or cost of meiosis (Williams 1975). On the other hand, benefits of sexual reproduction have been postulated such as advantages in hostparasite coevolution, reduced accumulation of mutations, reduced competition among sexuals, and the opportunity to repair physical DNA damage (Bell 1982; Vrijenhoek 1984; Stearns 1987; Kondrashov 1988; Michod & Levin 1988; Weeks 1993). Some predictions of these theories have been the focus of experimental studies (Lively 1987; Mogie & Ford 1988; Michaels & Bazzaz 1989; Hamilton et al. 1990; Lively et al. 1990; Moritz et al. 1991; Vrijenhoek 1994), but relatively little effort has been put into quantifying the costs and benefits of sex. The only available field estimate of the benefit of sex is the 1.43-fold higher fitness of sexually derived offspring that was found in an experimental study of the grass Anthoxanthum odoratum (Kelley et al. 1988). Estimating the of sex in hermaphrodites is not as easy as in outcrossing gonochoristic species, and has been the subject of several theoretical studies (Charlesworth 1980; Lloyd 1988; Joshi & Moody a Author for correspondence. Present address: Institut fiir Spezielle Zoologie, University of Muenster, Huefferstrasse 1, D-48149 Muenster, Germany. E-mail: michiels @ uni-muenster.de 1995). Briefly summarized, it depends on the sex of the coexisting sexual and parthenogenetic hermaphrodites, as well as on the ability of the male function of the parthenogens to generate parthenogenetic offspring in sexual partners. If the latter is not possible, the male function of a parthenogen is effectively for the parthenogenetic sub-population, as it makes no contribution to the next generation of parthenogens. Selfing rate and inbreeding depression are also important, but not relevant for our model system and are therefore not mentioned here. Let us first consider a newly arisen parthenogenetic lineage with the same sex as its sexual ancestor, but with a sterile male function (e.g., parthenogens are polyploid and so are their sperm). Such individuals will produce the same number of maternal offspring and make the same genetic contribution to the next generation as sexuals. Hence, the of sex is zero and additional changes are needed to offer an advantage to parthenogenetic hermaphrodites (see also Mogie 1996). On the one hand, parthenogens may reduce male rparth and increase female (1-rparth). Only when rse=0.5 in the sexual ancestor, and by complete elimination of the male function in parthenogens, can the latter obtain a 2-fold advantage. In all other cases (rsex O), the of sex [1/(1-rsex)] will be less than 2-fold. On the other hand, parthenogenetic hermaphrodites may generate new parthenogenetic lineages by fatherThis content downloaded from 157.55.39.45 on Thu, 01 Sep 2016 04:44:16 UTC All use subject to http://about.jstor.org/terms Weinzierl, Schmidt, & Michiels ing eggs of sexual partners. Assuming a simple genetic system, and all else being equal between sexuals and parthenogens, this will result in a (maximum) 3/2 advantage for parthenogens (Charlesworth 1980). This advantage will be lower when some partners of parthenogens are not sexual, but parthenogenetic as in sperm-dependent parthenogens (Beukeboom & Vrijenhoek 1998). Loss of eggs and sperm from sexuals to parthenogenetic partners should also be included in the of sex. Hence, in order to reveal the of sex in a natural population of hermaphrodites one has to consider (1) reallocation of resources to the female function in parthenogens and (2) the production of parthenogenetic offspring through gene flow from parthenogens to sexuals. Many examples of reduced male function in parthenogenetic hermaphrodites are known (Christensen et al. 1978; Nogler 1984; 0 Foighil & Eernisse 1988; 0 Foighil & Thiriot-Quievereux 1991; Weinzierl et al. 1998), but the reduction has never been shown to lead to a substantial increase in female reproductive success. This may be explained by the fact that male is low in some sexual hermaphrodites (Bell 1984a,b). Yet, it is clearly high in others (e.g., Rameau & Gouyon 1991). Gene flow from parthenogens to sexuals is possible in a number of hermaphrodites (Suomalainen et al. 1987; Asker & Jerling 1992) and can result in new parthenogenetic lineages (Menken et al. 1995). The hermaphroditic freshwater planarian Schmidtea polychroa BALL (=Dugesia polychroa (SCHMIDT); Tricladida, Paludicola) consists of a diploid sexual and several polyploid parthenogenetic biotypes (Benazzi 1957). Individuals of both modes produce cocoons that contain 1-10 eggs embedded in a common yolk mass. Parthenogenetic reproduction is pseudogamous, i.e., egg development is stimulated by a sperm cell that does not contribute genetically (Benazzi 1950; Benazzi & Benazzi Lentati 1976; Beukeboom & Vrijenhoek 1998). Despite being polyploid, parthenogenetic animals have functional, haploid sperm (Benazzi Lentati 1970), but, like sexuals, always need allosperm from a sexual or parthenogenetic partner in order to produce maternal offspring (Benazzi & Benazzi Lentati 1992). Parthenogens mate frequently, with each other and with sexuals (Peters et al. 1996; M. Storhas, pers. comm.). In S. polychroa, as in other planarians (Hyman 1951), testes in sexual individuals are numerous and occupy significant portions of the body. Parthenogens of S. polychroa, in contrast, have very few testes (Weinzierl et al. 1998). Parthenogenetic worms also appear to have a lower mating rate than sexuals (Weinzierl et al. 1998), which again saves resources that can be reallocated to the female function. We tested whether parthenogens of S. polychroa have higher female fecundity than their coexisting sexual conspecifics. Flexible reallocation is likely in planarians, because they are known to adjust cocoon production in response to fluctuating resource availability (Reynoldson 1964, 1977; Reynoldson & Young 1965; Boddington & Mettrick 1977; Calow & Woolhead 1977). By fertilizing sexual eggs, parthenogens of S. polychroa can inject genes for parthenogenesis into the coexisting sexual population, and this can generate new parthenogenetic lineages in laboratory crosses (Benazzi Lentati 1966). We investigated whether sexual individuals taken directly from our study population produced parthenogens among their offspring. One balancing selective force favoring sexual reproduction is female fertility, which is known to be reduced in a number of parthenogenetic insects (Lamb & Willey 1979). We compared fertility of cocoons produced by sexual and parthenogenetic worms.