There were two experiments in which three phosphite concentrations were sprayed onto the foliage of either Banksia brownii, B. baxteri or B. coccinea naturally infected with Phytophthora cinnamomi in the Esperance Plains bioregion of Western Australia, or B. grandis artificially infected with P. cinnamomi in Eucalyptus marginata forest of south-western Australia. Plots were either sprayed with one of three concentrations of phosphite (2.5, 5 or 10 g phosphite/L) and penetrant-spreading agent or sprayed only with penetrant-spreading agent. Mortality of the Banksia species, understorey and crown health, plant height and soil inoculum at each site were the dependent variables. There were no significant differences in overstorey phytotoxic necrosis rating of B. brownii, B. baxteri and B. coccinea between phosphite treatments 2.7 and 4 years after the first spray. The greatest phytotoxic effect of phosphite application was the stunting of B. coccinea growth, which did not occur in B. brownii, B. baxteri and B. grandis. For naturally infested Banksia species, greatest mortality occurred in B. brownii and least in B. coccinea, with B. baxteri being intermediate between mortality for B. brownii and B. coccinea. There was a significant reduction in apparent mortality rate and increase in years to 50% mortality following foliar application of phosphite to the four Banksia species. Greatest reduction in mortality of the Banksia species occurred for concentrations between 5–10 g phosphite/L. Challenge inoculation showed that spray treatments of 5 g and 10 g of phosphite/L were effective in B. grandis, 2.2 years after spray. The proportion of positive P. cinnamomi soil samples from the top 3 cm of soil was significantly greater than the proportion positive from 30 cm below the soil surface in infested B. brownii and B. coccinea, but there was no significant difference in the proportion of positive P. cinnamomi soil samples between depth of sampling for B. baxteri. Phosphite had no significant effect on the frequency of isolation of P. cinnamomi from the soil of infested B. baxteri and B. coccinea. Determination of phosphite effectiveness against P. cinnamomi and phytotoxic responses between plant species will assist prescription optimisation for the most effective protection of threatened flora.