Solanum elaeagnifolium Cav. is considered as one of the worst weeds of crop and pasture systems in temperate Australia. Effective long-term control is difficult due to the extensive root system. Field experiments were conducted at two locations in south-eastern Australia between 2006 and 2008 to examine a range of herbicides for control of S. elaeagnifolium on seed production and root regrowth. Herbicide performance was affected by herbicide, weed growth stage and environmental factors. Pyridine herbicides, such as pre-packed mixtures of aminopyralid + fluroxypyr and triclopyr + picloram + aminopyralid were the most effective and consistently reduced within-season aerial growth by 60–90% in both seasons. Overall control using glyphosate-based treatments was generally reduced due to emergence of new stems following herbicide application. Three picloram-based treatments provided the best and most consistent long-term control on root regrowth after two seasons, reducing stem emergence by 45–88%, especially with a late application of herbicides. The efficacy of residual herbicides such as atrazine or imazapic + imazapyr depends on rainfall conditions. Seedset control was best achieved with herbicides applied at the start of flowering stage, with no viable seed produced following treatments of 2,4-D amine + picloram and triclopyr + picloram + aminopyralid. These two treatments also significantly reduced viable seed production when applied at the early berry stage. The results indicate that an application at early flowering followed by a late application in autumn is necessary to effectively control the seedset (seedbank) and the root regrowth (rootbank) of S. elaeagnifolium.