Why do more plant species use ants for dispersal on infertile compared with fertile soils?

Mark Westoby*, Kristine French, Lesley Hughes, Barbara Rice, Louise Rodgerson

*Corresponding author for this work

    Research output: Contribution to journalArticlepeer-review

    54 Citations (Scopus)


    Myrmecochores are plant species that bear seeds with food‐bodies adapting them for dispersal by ants. Within Australia and South Africa, myrmecochores account for a large percentage of the flora in sclerophyll vegetation on very infertile soils. On fertile soils, there are few myrmecochores and more species with fleshy fruits adapted for dispersal by vertebrates. This effect of very infertile soils may also account for the high incidence of myrmecochores in the floras of Australia and South Africa compared with other continents. The taxonomic distribution of myrmecochory shows it has evolved many times within the Australian flora. Evidence has been collected in relation to eight hypotheses that might account for the prevalence of myrmecochory on very infertile soils: (1) Myrmecochore seeds do not appear to be relocated to nutrient‐enriched microsites. Seed‐removing ant species relocate their nest entrances frequently; active ant nests are not nutrient‐enriched; and seedlings emerging after fire are not located in nutrient‐enriched soil. (2) Traffic of seed‐removing ants is not greater on infertile than on fertile soils. (3) Burial to avoid predation by small mammals is not a significant factor for seeds in Australian vegetation. (4) Experiments in which diaspore traits are manipulated support the idea that a larger food body increases the likelihood that a diaspore will be taken by a seed‐disperser ant species that will not eat the embryo, rather than by a seed‐predator ant. However, evidence so far does not support the hypothesis that, on very infertile soils, seed‐disperser ants are consistently more available than seed‐predator ants. (5) Removal rates of fruits adapted for dispersal by vertebrates were, on average, two and a half times greater in vegetation on fertile soils. Birds were the main removers. Our current opinion is that the fruit‐removing bird assemblage is more likely to be a consequence of the high incidence of species with vertebrate‐adapted fruits, rather than the reverse. (6) Fleshy fruits are more expensive in potassium than food bodies for ants, but not in nitrogen or phosphorus. This might be a reason why fleshy fruits are uneconomic on very infertile soils. (7) Species with smaller embryo‐plus‐endosperm weight are relatively more likely to be adapted for dispersal by ants rather than vertebrates. However, differences in the frequency distribution of embryo‐plus‐endosperm weights between infertile and fertile soils are moderate in the Sydney area. An indirect association between soil fertility and dispersal mode via embryo‐plus‐endosperm weight cannot account for much of the difference in dispersal spectrum. (8) Species growing to less than 2 m tall are relatively more likely to be adapted for dispersal by ants rather than vertebrates, and also account for more of the flora on infertile than on fertile soils. This indirect correlation is capable of accounting for as much as 80% of the higher incidence of myrmecochory on infertile soils.

    Original languageEnglish
    Pages (from-to)445-455
    Number of pages11
    JournalAustralian Journal of Ecology
    Issue number4
    Publication statusPublished - Dec 1991


    Dive into the research topics of 'Why do more plant species use ants for dispersal on infertile compared with fertile soils?'. Together they form a unique fingerprint.

    Cite this