Seed ecology and growth comparisons of native (Virgilia divaricata) and invasive alien (Acacia mearnsii and A. melanoxylon) plants: implications for conservation

The abundance of invasive alien plants, with their numerous associated negative effects, has resulted in challenging and costly management requirements in South Africa. Management of invasives is further complicated when species have rapid inherent growth rates, atmospheric nitrogen fixing capabilities and large, persistent seed banks. Management strategies other than the usual mechanical, chemical, and biological (and combinations thereof) methods may be more effective. An alternative strategy is ’habitat management’, whereby management may be aimed at favouring native species. If the native species could compete favourably with, and possibly exclude, the invasive species, which is more likely when the species are ecologically similar, management may be more effective. This study investigated the competitive performance of two invasive species, Acacia mearnsii De Wild. and Acacia melanoxylon R.Br., and an indigenous ecological analogue, Virgilia divaricata Adamson, on the southern Cape Coast of South Africa. Both acacias are included among the five worst invasive species in the native forest and fynbos shrubland ecosystems. Performance was compared among the three species in terms of germination success (responses to five stimuli, namely chipping, hot water exposure, dry heat pulse, smoke, and control, incubated at 12 or 20°C), soil-stored seed bank densities and viability at different soil depths, and individual growth performance in seedling and sapling stages (and the effect of inter-specific competition on the latter). Virgilia divaricata performed comparably to the acacias in most aspects. Responses to germination stimuli at both temperature regimes were comparable among species, with chipping and hot water exposure resulting in the highest germination success. Acacia melanoxylon displayed poorer germination response under the cool temperature regime than the warm regime. Of the three species, A. mearnsii (7 596 seeds m-2) had the highest seed bank density, followed by V. divaricata (938 seeds m-2) and A. melanoxylon (274 seeds m-2). Seed viability was high (87-91%) in all three species. Neither seed density nor seed viability differed significantly between soil depths, and there were no clear differences in this aspect between species. In terms of growth during the seedling stage, V. divaricata outperformed A. mearnsii. In the sapling stage, growth was comparable between the three species and was not affected by the extent of competition from neighbouring saplings. Overall, the findings of this study suggest that the success of A. mearnsii as an invader is primarily attributable to its large seed banks. However, the good performance of V. divaricata seedlings and no apparent negative effect of competition on sapling growth from A. mearnsii and A. melanoxylon suggest that it may be used in the management of the acacias along with other management strategies. Due to the similarity observed between the three species, any management operation (such as fire) that would benefit V. divaricata recruitment would benefit the acacias equally (except for potentially reduced germination of A. melanoxylon in cooler seasons). Management should thus attempt to prevent the germination of Acacia seeds. Alternatively, dense localised Acacia seed banks may be reduced by repeat burning. On a small scale, areas where V. divaricata would occur naturally (or which are completely compromised ecologically) can be seeded with pre-treated (chipped or hot water exposure) V. divaricata seeds. If V. divaricata is given enough time to establish dense stands, subsequent germination and growth of the acacias should be suppressed and simultaneously promote forest re-establishment. To further decrease Acacia success, biocontrol (seed-boring weevils, flower-galling midges, and rust fungus) should be widely disseminated