The extent of tuber contamination with Erwinia carotovora in B.C. grown seed potatoes was determined by incubating tubers in a moist, anaerobic atmosphere. Only 2.1% and 11.3% of the tubers indexed in 1972 and 1975, respectively, were found to be latently-infected with 15. carotovora. Recoveries within individual lots ranged from 0 to a high of 27.7%. Both var. carotovora and var. atroseptica were recovered from tubers with equal frequency. No correlation was found between the field incidence of blackleg and the degree of latent tuber infection. Representative isolates of both varietal forms were able to infect faba beans, cucumbers and peppers while none was able to infect cereals, forage legumes or common weeds like black nightshade, horsetail, and lamb's quarters. Approximately 50% of the var. carotovora isolates recovered in the surveys incited blackleg symptoms in stem-prick inoculated Erwinia-tested potato stem cuttings grown at 18° C. Pathogenic carotovora isolates were also found to have a much wider host range than the nonpathogenic carotovora isolates. The additional hosts expressing symptoms were onions, peas, petunias, tobacco, tomato, lady's thumb and Chenopodium quinoa L. The majority of the pathogenic isolates (73%) reacted with an antiserum prepared against a pathogenic var. carotovora isolate. The atroseptica varietal form was found not to survive in artificially infested fallow field soils beyond 8 weeks under winter conditions and 4 weeks under summer conditions. By contrast var. carotovora was recovered from similarly infested soil at the conclusion of the overwintering experiments (139 and 225 days). The carotovora form also survived 29 days longer in soils in the oversummering experiment. The atroseptica form could not be detected around inoculated seed pieces after 4 weeks. However, var. carotovora was still recovered from soil in 2 out of 5 replicates at the end of the experiment (70 days). Both pathogenic and nonpathogenic isolates of the carotovora form were easily recovered from the rhizosphere of horsetail, lady's thumb, lamb's quarters, nightshade and redroot pigweed during the fall but rarely in similar surveys during the spring. The atroseptica form was recovered only once, from the rhizosphere of a horsetail plant during the September 1974 survey. The atroseptica form could still be recovered from artificially infested rhizospheres of lamb's quarters at the end of the experiment (140 days) but was not detected in the rhizospheres of lady's thumb after 42 days or in the rhizospheres of horsetail after 112 days. Parallel experiments with var. carotovora indicated that while this varietal form could survive in soils, weed rhizospheres help to maintain a high inoculum potential. Survival on tubers was found to be the only way in which var. atroseptica could survive winter in numbers which were detectable the following spring. Recovery f rom tubers buried 10 cm deep was greater than from those on the surface after 225 days. Surface-borne tubers inoculated with the atroseptica form tended to decay more rapidly which accounted for the lower recovery rate. The survival of var. carotovora was less dependent upon burial of tubers but the recovery rates were generally higher if tubers were buried. / Land and Food Systems, Faculty of / Graduate