of propagules, and grow rapidly (Menge, 1985). Arbuscular mycorrhizal fungal species that adapt to a wide Survival of arbuscular mycorrhizal (AM) fungi in soil may be afrange of hosts and edaphic contingencies presumably fected by the presence or absence of crops and by the crop being grown. Field studies were conducted in central Iowa during three have a better chance of long-term survival. One means growing seasons with cropping to continuous corn (Zea mays L.) by which AM fungi survive is the production of spores (two cultivars), continuous soybean [Glycine max (L.) Merr.] (two although the length of survival of spores in field soils cultivars), or fallow in three soils to determine AM selection and under varying conditions is not well documented. Presurvival. The initial numbers of spores (all following soybean) in May sumably, when conditions are favorable, these spores 1996 averaged 0.9 g Ϫ1 soil in Clarion (well drained), 1.1 in Nicollet germinate, grow, and produce other fungal structures. (somewhat poorly drained), and 3.6 in Webster (poorly drained) soils. Many cornfields, especially in low-lying areas, exhib-In May 1998, the highest spore count average was 11.2 g Ϫ1 soil in Webster under corn and 3.0 spores g Ϫ1 under soybean. Nicollet soil ited stunting and purple coloration characteristic of seaveraged 6.8 spores g Ϫ1 for corn and 0.9 for soybean in May 1998, vere P deficiency in the fall following the central U.S. whereas Clarion soil had 6.3 for corn and 2.0 for soybean. The fallow floods of summer 1993 (Fixen et al., 1994). Plants treatments consistently had low spore counts, ranging from 0.7 to 1.0 seemed not to be absorbing sufficient P for fast early spores g Ϫ1 for all three soils. After 3 yr under the same cropping growth even though soil tests indicated adequate P. regime, spore numbers in soil were corn Ͼ soybean Ͼ fallow; no Farmers were advised to use 35 to 40 kg P ha Ϫ1 as starter significant differences were found between cultivars of the same crop. fertilizer in areas that had been flooded the previous Most probable number counts were correlated with spore counts and averaged 11% of spore counts, suggesting that only a portion of the year. Ellis (1998) conducted a survey to determine the spores were viable (or culturable in our determination). By the end status of AM fungal populations in areas that were of the study, Glomus albidum and G. etunicatum dominated under flooded in 1993 and reported a decline in AM fungal corn, whereas G. constrictum dominated under soybean.