The incidence of viruses in wild Brassica rapa ssp. sylvestris in southern England

Using enzyme‐linked immunosorbent assays, the frequency of occurrence of six viruses was determined in Brassica rapa ssp. sylvestris collected from two Thameside sites (Abingdon and Culham) in Oxfordshire and one near the Avon (Claverton) in Bath & North East Somerset. During 2000–2001, the viruses detected were: Beet western yellows virus (genus Polerovirus) (BWYV), Cauliflower mosaic virus (genus Caulimovirus) (CaMV), Turnip crinkle virus (genus Carmovirus) (TCV), Turnip rosette virus (genus Sobemovirus) (TRoV), and Turnip yellow mosaic virus (genus Tymovirus) (TYMV). BWYV and TYMV were the most frequently detected viruses at the Oxford shire sites, both as single infections (20/1743 and 66/1743 respectively) and as dual infections (7/1743). Turnip mosaic virus (genus Potyvirus) (TuMV) was not detected in the field‐grown plants assayed from any of the sites. There was a highly significant (x²_1]=30.07, P<0.001) difference in the proportion of plants at each Oxfordshire site in which one or more viruses were detected, and essentially the same pattern of virus infection was observed in tests on B. rapa from the site near Claverton. At least one representative isolate of each detected virus was tested for its morphological and serological effects on glasshouse‐grown individuals from different half‐sib families of B. rapa from both Oxfordshire sites. Except for TRoV, where there was a large difference in the frequency of successful infection in B. rapa from the two locations (1/15 vs 11/15), no clear evidence of resistance or immunity to challenge was observed, although tolerance (virus invasion without symptoms) was frequent. Fewer of the plants from Abingdon were infected than those from Culham, when mechanically challenged with TRoV, but the two B. rapa populations were not otherwise consistently different, either in their infectibility by this virus or in their responses to challenge. However, with TCV, viral antigen concentration was closely linked to the severity of disease and the B. rapa from both Oxfordshire sites segregated into two classes: those with symptoms and most viral antigen, and those without symptoms and least viral antigen. These results suggest that generic risk assessments cannot be made due to differences in the way distinct B. rapa populations react to virus challenge.