Humid tropical leguminous tree and pasture grass responsiveness to vesicular-arbuscular mycorrhizal infection

Phosphorus is the major nutrient limiting plant growth in a Costa Rican silvopastoral system located on an acid, high P-retaining, volcanic soil. We investigated plant responsiveness to vesicular-arbuscular mycorrhizal (VAM) inoculation using the leguminous tree species Erythrina berteroana Urban, and the two dominant grass species Paspalum conjugatum Berg and Homolepsis aturensis Chase of this silvopastoral system. We grew grass seedlings in the greenhouse for 15 weeks in a methyl bromide-sterilized study soil to which either mixed-species VAM inoculum (Theobroma cacao feeder roots) or autoclave-sterilized cacao roots (non-inoculated control) were added. E. berteroana was grown from both seedlings and vegetative stakes (40 cm long) for 30 and 19 weeks, respectively. Upon harvest, we measured above and below ground biomass, N and P content, rootshoot ratio, legume nodulation, and VAM infection levels. The total above-ground and root biomass of mycorrhizae-inoculated P. conjugatum seedlings were 2.5 and 2.8 times greater than those of noninoculated seedlings. In contrast, VAM-inoculated seedlings of H. aturensis produced 8.4 and 5.9 times more total above-ground and root mass than noninoculated seedlings. Mycorrhizae-inoculated E. berteroana seedlings produced 10.6 times greater shoot biomass for inoculated versus noninoculated seedlings, while E. berteroana vegetative stakes exhibited a negative growth response to VAM inoculation (an approximately 16% decrease in shoot biomass for VAM-inoculated cuttings). The difference in responsiveness between Erythrina growth forms is hypothesized to reflect the cost-benefit relationship between plant host and fungal symbiont for energy and nutrient reserves.