Supplement 1. Data set and list of references compiled.

File List DataSet.txt ReferenceList.txt DescriptionNotes relevant to the data set compiled: 1. Columns correspond to: 1st column: Reference 2nd column: Community type 3rd column: Ecosystem type 4th column: Net primary production (NPP); in gC m-2 yr-1 5th column: Leaf production (leaf P); in gC m-2 yr-1 6th column: Detrital production (DP); in gC m-2 yr-1 7th column: Leaf detrital production (leaf DP); in gC m-2 yr-1 8th column: Decomposition (D); in gC m-2 yr-1 9th column: Biomass of detritus consumers (DB); in gC m-2 10th column: Consumption by herbivores (H); in gC m-2 yr-1 11th column: Percentage of production consumed by herbivores (%NPP consumed) 12th column: Herbivore biomass (HB); in gC m-2 13th column: Total consumption by first-order consumers (TC); in gC m-2 yr-1 14th column: Carbon accumulation (CA); in gC m-2 yr-1 15th column: Nitrogen concentration in producer biomass (Nbiomass); in % of dry weight 16th column: Phosphorus concentration in producer biomass (Pbiomass); in % of dry weight 17th column: Decomposition rate (k); in day-1 18th column: Nitrogen concentration in producer detritus (Ndetritus); in % of dry weight 19th column: Phosphorus concentration in producer detritus (Pdetritus); in % of dry weight 20th column: Producer nitrogen concentration used in the regressions vs. Detrital Production and Decomposition (Ndp); in % of dry weight- see note 5 21st column: Producer phosphorus concentration used in the regressions vs. Detrital Production and Decomposition (Pdp); in % of dry weight- see note 5 2. Cells with a dot and no numbers denote variables not provided in the reports. 3. Entries marked with # contain our indirect estimates of net primary production, and entries marked with @ our indirect estimates of decomposition, with the highest uncertainty, which still seems unimportant for the results obtained (see text). 4. Reports of communities dominated by rooted macrophytes (i.e., freshwater macrophyte meadows, seagrass meadows and terrestrial communities) having an asterisk (*) include both the above- (leaves and stems) and belowground (roots) compartments. Reports with two asterisks (**) only include the belowground compartment, and reports with no asterisk refer to the aboveground compartment. 5. We did not find many reports with concomitant values of detrital production and nutrient concentrations in producer detritus, or values of decomposition and nutrient concentrations in producer detritus. Therefore, to test the independence between detrital production and producer nutritional quality (Fig. 8), and between decomposition and producer nutritional quality (Fig. 10), we used values of nutrient concentration in producer biomass or detritus indistinctly (columns 20th and 21st). The results should be similar to those obtained if we had only used concentrations in producer detritus, because, for a given type of ecosystem, nutrient concentrations do not differ between producer biomass and detritus (see Figs. 1b,c and 6c,d). 6. In the reports marked with ¶, decomposition rates refer to leaves. Hence, in those reports values of leaf detrital production have been used for the relationship between decomposition rates and detrital production (Fig. 11b). Values of total detrital production (both above- and belowground if marked with one asterisk, or only aboveground if marked with no asterisk) have been used for all the other relationships and Figs. (except for Megonigal and Day [1988], see note 8). 7. In the reports marked with $, decomposition rates referring to different compartments (e.g., leaves and stems; above- and belowground) have been weighted and averaged into a single value for the relationship between decomposition rates and detrital production (Fig. 11b). Similarly, in the same reports values of nutrient concentration in producer detritus referring to different compartments have been weighted and averaged into a single value for the relationships between producer nutrient concentration and detrital production (Fig. 8) and producer nutrient concentration and decomposition (Fig. 10). Thus, to avoid redundancies, those mean values have been ignored in the histograms (Figs. 6c,d, and e) and relationship between decomposition rates and detritus nutrient concentrations (Fig. 9). 8. In the terrestrial reference Megonigal and Day (1988), values of producer nutrient concentrations refer to leaves. Hence, in this report values of leaf detrital production have also been used in the relationship between detrital production and producer nutrient concentrations (Fig. 8). 9. In the terrestrial reference McNaughton (1985), values of net primary production in parentheses include both the above- and belowground compartments, whereas those with no parentheses refer to the aboveground compartment only. The aboveground values have been used in this paper.