Background/Question/Methods One of the most important agents for improving soil structure is arbuscular mycorrhizal (AM) fungi. In native and unplowed soils, these microbes bind soil particles into aggregates and thereby improve aeration and infiltration of water. This study examined whether plowed agricultural soils are dominated by AM fungus species that invest little into soil aggregation. Soil disturbance like tillage should select species that have “weedy” traits such as quick reproduction via spores at the expense of their ability to bind soil aggregates.

This hypothesis was tested using a sterile, degraded sand-loam inoculated with soil microbes from an organic farm in Massachusetts. Soil microbes came from three farm treatments: 1) Tilled soil with annual crops; 2) Unplowed soil with perennial crops; and 3) Native maple forest. A sterilized soil served as a control. Each treatment was replicated with three host plant species: corn (Zea mays), the legume Illinois bundleflower (Desmanthus illinoensis) and red maple (Acer rubrum). At harvest, the size structure of soil aggregates was measured through wet sieving.

Results/Conclusions Compared to sterile soil, inoculated treatments had significantly more aggregates in the larger size fractions (1-2 mm, 2-4 mm, and 4-8 mm), indicating that mycorrhizal fungi improved soil structure. However, the effects of inoculum source (tilled, unplowed, or forest) depended on the host plant. Only with Illinois bundleflower did the unplowed inoculum produce significantly larger aggregates than tilled soil. With maple seedlings, the unplowed inoculum generated larger aggregates than forest but did not differ from the tilled treatment. With corn, there were no differences among inoculated treatments. These results suggest that tillage does not leave behind an AM fungus community incapable of binding soil aggregates, but rather the ability of the AM fungus community to stabilize soil aggregates depends on identity of the host plants.