NO EVIDENCE OF TROPHIC CASCADES IN AN EXPERIMENTAL MICROBIAL-BASED SOIL FOOD-WEB

Trophic-dynamic theories predict the biomass and productivity of troph ic levels to be partially top-down regulated in food webs, and that th e top-down regulation will manifest itself as cascading trophic intera ctions. We tested the two principal predictions deduced from these the ories: trophic cascades of(1)biomass regulation and (2) productivity r egulation occur in food webs. We created three food webs with either o ne, two, or three trophic levels in soil microcosms containing a steri lized mixture of leaf litter and humus. Twenty species of bacteria and fungi formed the first trophic level, a bacterivorous nematode (Caeno rhabditis elegans) and a fungivorous nematode (Aphelenchoides sp.) the second level, and a predatory nematode (Prionchulus punctatus) the th ird level. We sampled the microcosms destructively four times during a 5-mo experiment for estimations of the biomass of each of the trophic levels. CO2 evolution was analyzed once or twice a week,; and NH4+-N concentration in the soil was measured at the end of the: experiment. Glucose was added to the microcosms every second week to provide energ y for the microbes. The biomass of microbivores was clearly regulated by the predator. The: abundance of bacteria was not affected by the fo od chain length, and the abundance of fungi was higher in the presence of nematodes than in the pure microbial community. Net mineralization of N and C was highest in the food chains with two trophic levels, at an intermediate level in the presence of predators, and lowest in the pure microbial communities. Microbial production (estimated on the ba sis of microbial respiration) was higher in the food webs with two and three trophic levels than when the microbes were growing alone. Wheth er the biomass of the second trophic level was reduced by the predator or not had no effect on microbial biomass or microbial productivity. Therefore, although the microbivore biomass and mineralization of both C and N were regulated by the predator, our experiment did not provid e evidence of cascading trophic interactions regulating the microbial biomass and productivity in decomposer food webs. The facts that micro bes were able to compensate totally for the consumed biomass by increa sing their turnover rate and that the microbes did not behave as a uni form trophic level prevented a trophic cascade of biomass regulation f rom occurring in our soil food web. Similarly, since microbial product ivity did not depend on the biomass at the second trophic level, neith er did a trophic cascade of productivity regulation take place.