[Truncated abstract] Dryland salinity is a major environmental challenge facing agriculture in Australia. One option to manage dryland salinity is the use of perennial forages that increase water use of agricultural systems. However, the current array of perennial forages is limited. Forage species that satisfy the range of climatic and edaphic environments, and production systems, in southern Australia are needed (Chapter 1). In particular, low rainfall regions lack options other than lucerne (Medicago sativa L.) (Chapter 1). The Dorycnium genus (canary clovers) contains perennial species that might be useful forage plants for southern Australia. Dorycnium are sub-shrubs and their plant form differs from current perennial forages (Chapter 1). The aim of this project was to investigate some of the agronomic traits of several species of the genus Dorycnium to explore where they might be used in Australia and how they might be integrated into agricultural systems for management of dryland salinity. First, two desktop investigations assessed the potential adaptation and role of Dorycnium species in southern Australia: a review of the current literature on the agronomic characteristics of Dorycnium (Chapter 2) and an eco-geographical analysis to explore the ecology of Dorycnium species (Chapter 3). The agronomy of Dorycnium has been previously researched mainly in New Zealand, and although this provides some indications on where and how Dorycnium might be best used in Australia, this still requires testing in Australia. In particular, the aluminium tolerance of Dorycnium species indicates that they may be more suitable for acid soils than lucerne. Little ecological data was obtained for germplasm and herbarium collection sites of Dorycnium species. Climate comparisons between the native distribution of Dorycnium species in the Mediterranean basin and Australia, using spatial aridity data and CLIMEX climate match modelling, revealed that D. hirsutum and D. rectum might be suitably adapted to the temperate pasture regions of southern Australia. Suitable germplasm of D. pentaphyllum may also exist, but subsequent investigations in this project focussed on D. hirsutum and D. rectum.