Annual pathogenicity surveys of Puccinina graminis f. sp. tritici (Pgt), initiated at the University of Sydney in 1919, have continued without interruption to the present day. The population structure of Pgt over the past 85 years has been strongly influenced by exotic introductions in 1925 (race 126), 1954 (race 21), and 1969 (races 194 and 326), subsequent random mutations to virulence, and selection of genotypes with virulence matching resistance genes in cultivars. Pathotypes detected in Australia over the past 10 years trace back to either races 21, 194, or 326. Based on varietal resistance and pathogenic variability, previous workers identified 3 periods between 1919 and 1970: from 1919 to 1938, cultivars lacked effective resistance genes; from 1938 to 1964, cultivars released with single genes for resistance (Sr6, Sr11, Sr9b, Sr36, Sr17), and new pathotypes with corresponding virulences were detected; from 1965 to 1970, and beyond, cultivars with multiple resistance genes were deployed in many regions, significantly reducing yield losses. During this third phase, and until now, cultivars were protected by resistance genes Sr2, Sr9g, Sr12, Sr13, Sr17, Sr22, Sr24, Sr26, Sr30, Sr36, and Sr38, singly or more commonly in combinations. Overall inoculum levels and pathotype diversity in Pgt have declined in all wheat-growing regions since the mid 1970s, likely as a consequence of the release of cultivars with gene combinations. Despite the low levels of stem rust in Australia over the past 30 years, resistance is still a top priority in many breeding programs. The development of virulence for Sr38 in WA in 2001 was a timely reminder of the need for continued vigilance if the sustained genetic control of the past 30 years is to continue.