The long-term persistence and suppressiveness of Pasteuria penetrans against Meloidogyne arenaria race 1 were investigated in a formerly root-knot nematode suppressive site following 9 years of continuous cultivation of three treatments and 4 years of continuous peanut. The three treatments were two M. arenaria race 1 nonhost crops, bahiagrass (Paspalum notatum cv. Pensacola var. Tifton 9), rhizomal peanut (Arachis glabrata cv. Florigraze), and weed fallow. Two root-knot nematode susceptible weeds commonly observed in weed fallow plots were hairy indigo (Indigofera hirsuta) and alyce clover (Alysicarpus vaginalis). The percentage of J2 with endospores attached reached the highest level of 87% in 2000 in weed fallow, and 63% and 53% in 2002 in bahiagrass and rhizomal peanut, respectively. The percentage of endospore-filled females extracted from peanut roots grown in weed fallow plots increased from nondetectable in 1999 to 56% in 2002, whereas the percentages in bahiagrass and rhizomal peanut plots were 41% and 16%, respectively. Over 4 years, however, there was no strong evidence that endospores densities reached suppressive levels because peanut roots, pods, and pegs were heavily galled, and yields were suppressed. This might be attributed to the discovery of M. javanica infecting peanut in this field in early autumn 2001. A laboratory test confirmed that although the P. penetrans isolate specific to M. arenaria attached to M. javanica J2, no development occurred. In summary, P. penetrans increased on M. arenaria over a 4-year period, but apparently because of infection of M. javanica on peanut at the field site root-knot disease was not suppressed. This was confirmed by a suppressive soil test that showed a higher level of soil suppressiveness than occurred in the field (P