Dormancy and endozoochorous dispersal ofOpuntia rastreraseeds in the southern Chihuahuan Desert

Seed dormancy and endozoochorous dispersal have important effects on population dynamics, i.e. seed bank formation, spatial and temporal bet-hedging, avoidance of crowding and decreased sib-competition. Seeds having primary dormancy do not germinate, even if adequate environmental and dispersal conditions are met, until an after-ripening period is completed. This study examined the effects on germination rates of seed dispersal by vertebrates and primary dormancy (i.e. how germination rates changed in time) in the prickly pearOpuntia rastrerafrom the southern Chihuahuan Desert. Ripe fruits, scats of coyote, mule deer, woodrat, common pig and pellets of the northern raven were collected in 1992 during the ripe fruit peak (August). A factorial design was used to assess the effect of disperser and seed age on germination rates using these seeds collected in 1992. In 1992, 1993 and 1994 (fresh, 1- and 2-year-old seeds, respectively) germination success after 3 months was determined for 100 seeds of each disperser, plus a control of uneaten seeds. We found differences in germination rates both between dispersers and years. All but one disperser (deer) decreased germination rates with respect to the control. Germination steadily increased with ageing of seeds, implying the presence of primary dormancy (embryo immaturity). Our results show that despite the large disperser coterie associated withOpuntia rastrera, seeds have an obligate primary dormancy which seems to result in seed bank formation. Seed germination and seedling establishment will depend on selection pressures caused mainly by environmental variability in deserts (unpredictable rainfall, extreme temperatures, safe-site availability). Nevertheless, endozoochorous dispersal provides some additional advantages (e.g. habitat colonization, genetic variability, avoidance of long-term local extinction) that needs further research