| Seed dispersal is a key process in the successful recruitment of plants. For many plant species, ants play an important role in this mobile phase of plant life history. Due to their ubiquitous occurrence in terrestrial ecosystems, seed dispersal by ants (myrmecochory) has therefore been described in different habitats. Myrmecochory is considered to be a mutualism whereby ants benefit by gaining nutrition from lipid-rich elaiosomes attached to seeds and plants benefit from having their seeds dispersed away from parent plants. Myrmecochory plays a key role in seed germination and seedling establishment, with cascading effects on regeneration and spatial distribution pattern of plant populations. This interaction between ants and plants may be modified by external drivers, both biotic (e.g., plant traits, ant traits, the presence of alternative food resources) and abiotic (e.g., temperature, light). Field experiments were conducted with Corydalis giraldii Fedde (Papaveraceae), a typical myrmecochore, to explore the effects of the type, composition and richness of the alternative food resources and habitat on seed dispersal, contributing to the understanding of the fate of diaspores adapted for ant dispersal, and provide scientific theoretical basis for biodiversity conservation and ecosystem balance. The results are mainly as follows:1. The indirect effects of alternative resources type on C. giraldii seed dispersal depended on ant identity and its recruitment mode. In a field experiment, the pitfall trapping method was used to examine ant feeding preference among5items of diets (i.e. meat, honey, apple, bread and seed of C. giraldii). Additionally, two artificial diaspore depots were offered near the ant nests, one with seeds and the other with either meat, honey, bread, apple fragments, or no food (control), to determine the effect of alternative food resources on seed removal. While8ant taxa (ca.837individuals) were captured in the sampling plots, only two species (Lasius alienus and Formica fusca) were dominant and effective dispersers, with mass recruitment and simple cooperative recruitment respectively. In fifteen40minute trials, the mean number of ant visit of the seeds was38.73(±4.57) times, and the mean rate of seed removal was33.87(±4.22) in L. alienus. No significant difference in either ant visit frequency or seed removal rate was found in L. alienus following the addition of alternative food types (P>0.05). Both ant visit frequency and the rate of seed removal in F. fusca were reduced significantly by the addition of alternative food (P<0.05). In particular, ant visit frequency decreased from a mean of30.8(±2.87) individuals (n=15,40min) in the control to15.6(±3.61),9.07(±1.4) and7.67(±1.58) individuals, respectively, with the addition of either bread, apple fragments or honey. The rate of seed removal decreased from16.27(±3.35) individuals (n=15,40min) in the control to3.47(±1.17) and2.87(±0.9) in the addition of apple fragments or honey.2. The indirect effects of alternative resources composition and richness on C. giraldii seed dispersal depended on ant identity and its recruitment mode. In a field experiment, we selected four component foods (meat, honey, apple and bread) and two artificial diaspore depots were offered near the dominant and effective dispersers ant (L. alienus and F. fusca) nests, one with seeds and the other with each of15diets combination treatments or no food (control), to determine the effect of alternative food resources composition and richness on C. giraldii seed removal. The ant visit frequency and the rate of seed removal in L. alienus increased from a mean of38.73(±4.57) and33.87(±4.22) individuals (n=15,40min) in the control to55.47(±4.67) and49.07(±4.1) individuals, with the addition of bread and meat (SBM)(P<0.05), the rate of seed removal decreased to23.13(±3.05) in the addition of honey, apple, bread and meat (SHBMA)(P<0.05); no significant difference in either ant visit frequency or seed removal rate was found in L. alienus following the addition of each level of alternative food richness (P>0.05). The ant visit frequency and the rate of seed removal in F. fusca were30.8(±2.87) and16.27(±3.35) individuals (n=15,40min) in the control, ant visit frequency were reduced significantly by the addition of alternative food, besides SM, the rate of seed removal were reduced significantly by the addition of alternative food, besides SM and SB; The ant visit frequency and the rate of seed removal were reduced significantly by the addition of each level of alternative food richness, and positively correlated with the level of alternative food richness.3. Habitat significantly affected the removal rate and dispersal distance of C. giraldii seeds. In a field experiment, sticky traps were used to measure primary dispersal of seeds up to0.5m from mother plants. Pitfall trapping method was used to estimated ant population density. While9ant taxa (ca.289individuals) were captured in deatritus, only three species (L. alienus, Myrmica sp. and F. fusca) were dominant and effective dispersers, which accounted for about36.58%,14.53%and7.96%of the total ant number, respectively; While9ant taxa (ca.1139individuals) were captured in old-field, only four species (F. polyctena, F. fusca, Aphaenogaster smythiesii and L. alienus) were dominant and effective dispersers, which accounted for about45.48%,21.24%,15.19%and4.83%of the total ant number, respectively. Our results indicated significant difference between the two habitats in disperser abundance (χ2=62.798, P<0.0001). Primary seed dispersal is positively skewed,98%of seeds fall within20cm of the mother plant and primary dispersal distance often being less than the height of the mother plant. The removal number of seeds per hour was17.13(±8.35) and52.8(±17.46) in deatritus and old-field respectively. The average dispersal distance was82.63(±2.17) cm and108.07(±3.07) cm in deatritus and old-field respectively. The removal number of seeds per hour and the average dispersal distance differed significantly between the two habitats (P<0.05).Our results indicated that habitat and the availability of alternative food resources other than the diaspore itself have an important influence on ant-mediated C. giraldii seed dispersal. L. alienus and F. fusca were dominant and effective dispersers, with mass recruitment and simple cooperative recruitment respectively. No significant difference in either ant visit frequency or seed removal rate was found in L. alienus following the addition of alternative food types. Both ant visit frequency and the rate of seed removal in F. fusca were reduced significantly by the addition of honey or apple. Both ant visit frequency and the rate of seed removal in two ants were affected significantly by the addition of alternative food composition. No significant difference in either ant visit frequency or seed removal rate was found in L. alienus following the addition of each level of alternative food richness, while significant reduction in both were found in F. fusca, and positively correlated with the level of alternative food richness. We concluded that the indirect effects of alternative resources on seed dispersal depended on ant identity and its recruitment mode. Ant community composition and the relative abundance of seed dispersers were the major factors to the significant difference in the removal number of seeds per hour and the average dispersal distance between habitats. The study can assist in the understanding of the fate of diaspores adapted for ant dispersal and provide reference to further study on the spatial distribution pattern of myrmecochore. |
PDF Full Text Request