Abstract:
Mycorrhizae are important for plant growth, particularly in nutritionally poor soils. Vesicular-arbuscular mycorrhizae (VAM) are the primary form of mycorrhizae found in tropical plants, and their persistence in the soil as colonizing fungal propagules following deforestation cannot be directly
measured. Utilizing the "most probable number" (MPN) method for estimating infective propagules, values of mycorrhizal inoculum potential (MIP) were obtained for two Costa Rican soils with different vegetation histories following primary forest removal in 1 950's-60's. The soils (Oxic Dystropepts) were collected at La Selva Biological Research Station from sites covered by secondary forest, pasture, and land bare of vegetation for four and for six years. We hypothesized that MIP would be greater in the pasture and forest soils than in the soils bare of vegetation. Two VA-mycotrophic plants, Psidium guajava L. and Alilum cepa L. were used as bioassays in
greenhouse studies to obtain MPN values. Both bioassay studies gave estimates of 0.6 propagules/gr. dry soil from the pasture soil. For the other three soils both bioassays gave significantly lower estimates (p<O.0000l) in the range of 0.002-0.104 propagules/gr. dry soil. Growth responses for
plants grown in the pasture soil were similarly greater than those plants grown in the other three soils. The MPN values were correlated with spore counts made from the
same soils using the wet-sieving and decanting technique. Correlations between spore counts and MPN values were not significant.
