Search

Seed dispersal is qualitatively effective when it increases recruitment probability. A poorly studied factor likely affecting recruitment is the spatial distribution of dispersed seeds. Seed-caching animals are thought to disperse seeds in a way that reduces clumping and density to impede cache pilfering. Furthermore, dispersal might differ depending on whether the seed is immediately consumed versus cached for later consumption, and might differ depending on the ecological context. The main objectives of this study were to determine: 1) the spatial pattern of seed dispersal by rodents in a heterogeneous environment; 2) whether the patterns differ among years and among acorn competitor exclosure treatments, and 3) whether rodents create different spatial patterns of dispersal for acorns that are cached versus consumed immediately following dispersal. We studied the degree of spatial aggregation of acorn dispersal by rodents using two different estimators derived from the Ripley K and the Diggle G functions. We also analyzed various metrics of dispersal distances. For both analyses we used observed acorn dispersal patterns in two years differing in crop size and inside versus outside exclosures restricting access to acorn-consuming ungulates. During 2003, a year with a larger crop size, maximum seed dispersal distances were less, and the pattern of dispersed seeds was more clumped, than in 2004, a year with a smaller crop size. Median dispersal distances did not differ between years. In the presence of ungulates, seed dispersal was marginally sparser than in their absence. Cached acorns were dispersed more sparsely than acorns eaten immediately. These results have important implications for the quality of seed dispersal for oak recruitment that are likely relevant to other systems as well. [ABSTRACT FROM AUTHOR]

Seed-hoarding vertebrates may survive yearly periods of food scarcity by storing seeds during the preceding fruiting season. It is poorly understood why rodents creating long-term reserves, especially those in the tropics, incorporate seeds from plant species that germinate rapidly and hence seem unsuitable for long-term storage. We carried out a series of experiments to understand why red acouchies ( Myoprocta exilis) scatter-hoard Carapa procera (Meliaceae) seeds, which lack dormancy and germinate rapidly. Hypotheses tested were: (1) even germinated seeds are still significant long-term energy reserves for acouchies, (2) acouchies use the seeds only as short-term supplies, and (3) acouchies manipulate the seeds to slow down germination. The first two hypotheses were not supported; seed reserves were rapidly depleted during experimental planting, and acouchies did use seeds for long-term storage. We did find support for the third hypothesis. Acouchies intervened in germination by removing the protruding radicle and epicotyl after which they re-cached seeds. Pruning stopped further seedling development, yet the pruned seeds did not decay. The cotyledons apparently stayed alive and physiologically active as “zombie seeds” that only formed undifferentiated calli. Pruned seeds were suitable for long-term storage, with negligible loss of endosperm over time. Pruning was most effective after sprouting of the epicotyl, and germinating seeds were most susceptible to pruning during two weeks upon emergence. Acouchies actively managed their food supplies and must continuously survey for germinating seeds to timely intervene in seed reserve depletion, within the brief period of seedling emergence. We suggest that the trees use the rodents to achieve seed dispersal, and gain from imperfect intervention in germination by the rodents. Because scatter-hoarding rodents and large-seeded plant species with similar germination co-occur in tropical forests world-wide, it is plausible that the phenomenon of seed pruning to preserve seeds is more general than currently appreciated. [ABSTRACT FROM AUTHOR]

In addition to acting as seed predators, some terrestrial mammals bury seeds via scatter hoarding. This study system used two permanent plots in examining the interaction between small rodents and the seeds of the palm Astrocaryum mexicanum . We tested how experimental burial, and fruiting status of the parent, distance to the parent, seed size, and microsite characteristics affect the survival of these seeds. Up to 34% of the buried seeds that were exposed only to ignorant rodent foragers (individuals not responsible for burial) survived. In comparison, less than 1% of seeds buried by scatter hoarding rodents survived in previous studies, a percentage that is comparable to the low survival of unburied seeds in this study (<2%). Although unburied seeds had very low survival, increasing distance and/or seed density positively affected survival of unburied seeds. Distance to parent had no effect on buried seed survival. Buried seed survival was most strongly and significantly determined by the fruiting status of the trees under which they occurred. Seeds experienced significantly greater predation if buried under “parent” trees that fruited during the experiment. Buried seed survival was also negatively affected by germination, as germination may signal the presence of a seed to foraging rodents. There was some indication of a positive effect of tree density on seed survival between the two plots, whereas differences in rodent abundance appear to have no effect on seed survival. Seed size and microsite characteristics had no significant effect on buried seed survival, likely due to the greater proportional effects of other factors and the longevity of A. mexicanum seeds. The results of this study were used to generate a hypothetical causal network showing how comparatively low recovery of buried seeds by ignorant foragers – combined with processes determining the removal of scatter hoarding foragers from their scattered seed caches – may affect seedling recruitment in A. mexicanum . [ABSTRACT FROM AUTHOR]

Seed size may vary greatly among individuals within plant species. What effects the extremes of this variation have for seeds taken by small mammals are poorly understood. Not all seeds removed by small mammals are necessarily eaten. Small rodents are common seed predators, but they may disperse a significant proportion of seeds by scatter hoarding them via burial. Size-dependent predation and dispersal of seeds has not been directly tested within a plant species for tropical rodents. This study tested whether or not large and small nuts of Astrocaryum mexicanum (Palmae) differed in their fates due to handling by the spiny pocket mouse Heteromys desmarestianus (Heteromyidae). Exclosures were used to give small rodents exclusive access to A. mexicanum nuts. H. desmarestianus preferentially consumed large over small A. mexicanum nuts, but cached (in burrows and by scatter hoarding) similar proportions of these nuts by size. Small nuts tended to be buried farther away from exclosures than large nuts. Although sample sizes of buried nuts were small, the rodents retrieved all buried large nuts, but 30% of the small nuts remained buried long enough to germinate. I also examined predispersal predation by insects and found that insects appear to have no size preference for A. mexicanum nuts, but insect predation appears to hinder nut development. Thus, nuts attacked by insects develop to be significantly smaller, with a low proportion of undamaged endosperm, than uninfested nuts. It is hypothesized that the preferential predation of large A. mexicanum nuts by H. desmarestianus is a response by these rodents to insect predation. [ABSTRACT FROM AUTHOR]