Abstract
The enzymatic latch hypothesis proposes that oxygen (O 2 ) limitation promotes wetland carbon (C) storage by indirectly decreasing the activities of hydrolytic enzymes that decompose organic matter. Humid tropical forest soils are often characterized by low and fluctuating redox conditions and harbor a large pool of organic matter, yet they also have the fastest decomposition rates globally. We tested the enzymatic latch hypothesis across a soil O 2 gradient in the Luquillo Experimental Forest, Puerto Rico, USA. Enzyme activities expressed on a soil mass basis did not systematically decline across a landscape O 2 gradient, nor did phenolics accumulate, the proposed mechanism of the enzymatic latch. Normalizing enzyme activities by C concentrations did suggest a decline in several enzymes as mean soil O 2 decreased. However, relationships between hydrolytic enzymes and reducing conditions were scale‐dependent: enzymes displayed neutral to strongly positive relationships with reducing conditions and phenolics when comparing samples within sites, and enzyme activities in 18‐d anaerobic incubations generally exceeded those in aerobic soils despite a fourfold increase in phenolics. In summary, although O 2 availability and the activities of some enzymes appeared to be related at landscape scales after accounting for differences in organic matter, reducing conditions and phenolic compounds did not appear to constrain soil hydrolytic enzyme activity at the scale of soil microsites, challenging the enzymatic latch hypothesis. Hydrolytic enzymes can be resilient to periodic anaerobiosis and may actually stimulate O 2 consumption at the microsite scale. We suggest a critical re‐examination of mechanisms and the scale dependence of couplings between O 2 and decomposition in terrestrial soils.