Over 2 growing seasons, the incidences of infection with beet western yellows (BWYV), cauliflower mosaic (CaMV), and turnip mosaic (TuMV) viruses were determined in canola (Brassica napus) crops growing in the agricultural area of south-west Australia. Tissue blot immunoassay was used to detect BWYV and enzyme-linked immunosorbent assay to detect CaMV and TuMV. In 1998, BWYV was detected in 59% of 159 crops surveyed, whereas in 1999 it was found in 66% of 56 crops. Incidences within individual infected crops were 1-65% (1998) and 1-61% (1999). Infection occurred widely in high and medium rainfall zones, but was also readily detected in the low rainfall zone. In addition, BWYV was found in canola samples from 5 sites in New South Wales. Most cultivars tested (9 of 10) in the canola crop survey were infected with BWYV. No clear relationship was found between BWYV infection and any particular type of disease symptom. Overall, the incidence of BWYV at the crop edge was marginally greater than that inside the crop. CaMV was detected in 27% of 143 crops in 1998 but in only 2 of 47 in 1999. Incidences within individual infected crops were 1-17% in 1998 but only 1% in 1999. CaMV infected 6 of 10 cultivars and was present in high, medium, and low rainfall zones. Obvious chlorotic ringspot symptoms were associated with CaMV infection. TuMV was detected in 5% of 139 crops in 1998 but in only 1 of 47 from 1999. Incidences within the individual infected crops were 1-5% in 1998 and 1% in 1999; 3 of 10 cultivars were infected and it was found in high and medium rainfall zones. BWYV, CaMV, and TuMV were all found infecting wild radish (Raphanus raphinistrum). In general, incidences of BWYV were greater in wild radish than in canola. In 1998, BWYV was detected in wild radish at 9 of 12 sites sampled in 5 of 6 districts, with infection incidences up to 48%. In 1999, it was detected at all 10 sites sampled in 7 districts, with incidences up to 96%. Infected samples came from all rainfall zones, and from several different types of sites, some of which were distant from canola crops. Despite the presence of possible viral symptoms in wild radish, none was clearly associated with BWYV infection. In contrast, TuMV caused obvious mottle and 'oak leaf' patterns in wild radish plants. The finding of widespread virus infection in canola crops and a substantial virus reservoir in wild radish weeds is cause for concern to the canola industry.