Southern Australia's annual- based agricultural system has come at a large cost to the landscape. Dryland salinity is a rapidly expanding environmental problem that is reducing the amount of land available for agriculture, and causing a significant ecological cost to remnant and riparian vegetation. There is an urgent need to increase the area of the landscape that is sown to deep- rooted herbaceous perennials to reduce the increase in dryland salinity, and for their successful adoption by landowners it is recognised that these perennials must be economically viable. Australian perennials are unlikely to provide such options in the short term and therefore there is a need to search for species overseas. Many agricultural weeds have arisen as a direct result of deliberately introduced species escaping cultivation and naturalising in the Australian environment. They cause a huge cost to agriculture in terms of both lost production and control. There is also a cost to natural ecosystems as a result of lost biodiversity and weed management. A conflict of interest thus arises.This paper follows on from a workshop held between the CRC for Plant- based Management of Dryland Salinity and the CRC for Australian Weed Management. It discusses 4 key areas where potential conflict exists between the maintenance of biodiversity in natural ecosystems and the development and introduction of new herbaceous perennials. Each of the issues within pre- entry weed risk assessment, post- entry weed risk assessment, weed risk of translocating native species and field assessments of new species is discussed in detail and suggestions are given on the means to resolve the conflicts. Actions to address the recommendations are urgently required if we are going to resolve the current conflicts of interest between the need for managing present and future environmental weeds and for mitigating dryland salinity.