THE GENETIC STRUCTURE OF A CYCLICAL PARTHENOGEN

Cyclical parthenogenesis is a mode of reproduction largely restricted to parasitic trematodes ofthe order Digenea, rotifers, cladocerans, aphids, gall wasps (Cynipidae), and some gall midges (Cecidomyiidae). In these organisms, unisexual propagation via eggs is periodically interrupted by a bisexual phase. In most cases parthenogenesis is not truly cyclical in that the length of the unisexual phase is a function of environmental circumstances and not of an inherent predetermined program. Nonetheless, such organisms should be considered distinct from facultative parthenogens. The cytogenetic events promoting unisexual and bisexual propagation in cyclical parthenogens are fundamentally different; males are parthenogenetically produced prior to the bisexual phase, and the mode of reproduction is not simply a function of the availability of males. Much has been made of the evolutionary value ofheterogonic life cycles (White, 1973; Williams, 1975; Maynard Smith, 1978). By altering unisexual and bisexual modes of reproduction individuals are thought to combine the short-term advantage of a rapid replication of their genome with the long-term advantage of recombination. Theoretically, parthenogenesis should allow rapid and efficient selection for highly fit linkage groups, and a periodic phase ofrecombination should facilitate the creation of coadapted gene complexes. This does not mean that populations of cyclical parthenogens will rapidly approach genetic homogeneity. Both segregation of heterotic linkage groups and shifting selection pressures in temporally and spatially variable environments will result in the maintenance ofgenetic variability, as will the periodic release of hidden genetic variance that must accumulate in a clonal population subject to mutation and selection (Lynch and Gabriel, 1983). Nonetheless, the above arguments imply that the cyclically parthenogenetic mode of reproduction should provide an exceptionally effective mechanism for coping with environmental uncertainty. Why then is cyclical parthenogenesis so rare? White's (1973) suggestion that the evolution of a reliable mechanism to support two radically different reproductive modes would require a cytogenetic tour-de-force certainly warrants consideration, particularly when one considers how common obligate parthenogenesis is relative to the cyclical mode. However, before resorting to an explanation based on cytological limitations, it is worth considering whether the supposed ecological and genetic advantages ofcyclical parthenogenesis are indeed fulfilled in natural populations. Ecological genetic analyses of cyclical parthenogens are rare. Although valuable genetic observations have been obtained with rotifers (King, 1977, 1980), aphids (Wool et al., 1978; Suomalainen et al., 1980; Tomiuk and Wohrmann, 1980) and cladocerans(Hebert, 1978, 1980; Young, 1979a, 1979b), for the most part these studies are limited in the amount of ecological information that they convey and! or are short-term in duration. Validation of the assumptions in my second paragraph must involve long-term investigations of the genetic dynamics of populations and their association with the selecti ve process. Here I report on the first five years of a continuous genetic