Biocycling of sulfur (S) has been proposed to play an important role in the recovery of ecosystems following anthropogenic S deposition. Here, we investigated the importance of the humus layer in the biocycling of S in three forested catchments in the Ga˚rdsjo¨n area of southwestern Sweden with differing S inputs and S isotope signature values. These experimental sites consisted of two reference catchments and the Ga˚rdsjo¨n roof experiment catchment (G1), where anthropogenic deposition was intercepted from 1991 until May 2002 by a roof placed over the entire catchment area. Under the roof, controlled levels of deposition were applied, using a sprinkler system, and the only form of S added was marine SO 4 2À with a d 34 S SO 4 of +19.5&.We installed ion exchange resin bags at the interface between the humus layer and mineral soil at each of the catchments to collect SO 4 2À passing through the humus. The resin bags were installed on four occasions, in 1999 and 2000, covering two summer and two winter periods. The ions collected by each bag during these sampling periods were then eluted and their d 34 S SO 4 values and SO 4 2À concentrations determined. The most striking result is that the average d 34 S SO 4 value in the resin bags was more than 12& lower compared to that of the sprinkler water in the G1 roof catchment. There was no increasing trend in the isotope value in the resin bag SO 4 2À despite that the roof treatment has been on-going for almost 10 years; the average value for all resin bags was +7.1&. The highest d 34 S SO 4 values found in the G1 roof catchment were between +11& and +12&. However, these values were all obtained from resin bags installed at a single sampling location. Throughfall and resin bag d 34 S SO 4 values were more similar in the two reference catchments: about +7.5& in both cases. There was, however, an increase in resin bag d 34 S SO 4 values during the first winter period, from about +7& to +9&. The resin bag d 34 S SO 4 value was linearly and positively related (r 2 = 0.26, p < 0.001) to the amount of SO 4 2À extracted from the resin bags, if relatively high amounts (>50 mmol m À2 ) were excluded. High amounts of resin bag SO 4 2À seemed to be related to groundwater inputs, as indicated by the d 34 S SO 4 value. Our results suggest that rapid immobilization of SO 4 2À into a large organic S pool may alter the S isotope value and affect the d 34 S SO 4 values measured in the mineral soil and runoff.

2À during base flow conditions. Our results suggest that during the spring snowmelt, snow S and desorbing SO 4 of mainly anthropogenic origin are the two major S sources in four of the investigated streams. Two streams in forested areas also indicate that reoxidation of reduced S may be released during the spring flood. The stream in the cultivated area was found to be strongly influenced by the acid sulfuric soils independent of stream flow conditions.