Dissolved organic matter (DOM) contributes to organic carbon either stored in mineral soil horizons or exported to the hydrosphere. However, the main controls of DOM dynamics are still under debate. We studied fresh leaf litter and more decomposed organic material as the main sources of DOM exported from the forest floor of a mixed beech/oak forest in Germany. In the field we doubled and excluded aboveground litter input and doubled the input of throughfall. From 1999 to 2005 we measured concentrations and fluxes of dissolved organic C and N (DOC, DON) beneath the Oi and Oe/Oa horizon. DOM composition was traced by UV and fluorescence spectroscopy. In selected DOM samples we analyzed the concentrations of phenols, pentoses and hexoses, and lignin-derived phenols by CuO oxidation. DOC and DON concentrations and fluxes almost doubled instantaneously in both horizons of the forest floor by doubling the litter input and DOC concentrations averaged 82 mg C l−1 in the Oe/Oa horizon. Properties of DOM did not suggest a change of the main DOM source towards fresh litter. In turn, increasing ratios of hexoses to pentoses and a larger content of lignin-derived phenols in the Oe/Oa horizon of the Double litter plots in comparison to the Control plots indicated a priming effect: Addition of fresh litter stimulated microbial activity resulting in increased microbial production of DOM from organic material already stored in Oe/Oa horizons. Exclusion of litter input resulted in an immediate decrease in DOC concentrations and fluxes in the thin Oi horizon. In the Oe/Oa horizon DOC concentrations started to decline in the third year and were significantly smaller than those in the Control after 5 years. Properties of DOM indicated an increased proportion of microbially and throughfall derived compounds after exclusion of litter inputs. Dissolved organic N did not decrease upon litter exclusion. We assume a microbial transformation of mineral N from throughfall and N mineralization to DON. Increased amounts of throughfall resulted in almost equivalently increased DOC fluxes in the Oe/Oa horizon. However, long-term additional throughfall inputs resulted in significantly declining DOC concentrations over time. We conclude that DOM leaving the forest floor derives mainly from decomposed organic material stored in Oe/Oa horizons. Leaching of organic matter from fresh litter is of less importance. Observed effects of litter manipulations strongly depend on time and the stocks of organic matter in forest floor horizons. Long-term experiments are particularly necessary in soils/horizons with large stocks of organic matter and in studies focusing on effects of declined substrate availability. The expected increased primary production upon climate change with subsequently enhanced litter input may result in an increased production of DOM from organic soil horizons.