Ecology Letters (2011) 14: 1313–1317 This is an errata to Cleveland et al. (2011). Figure 1 includes a point in Mexico. However, original data from this lowland forest site were from a secondary (not primary) forest, and thus they were not included in the final lowland forest analysis. In the Methods section, we stated, ‘We also included some data published in previous syntheses (Elser et al. 2007; Townsend et al. 2007; LeBauer & Treseder 2008; Quesada et al. 2009), and examined relevant references cited therein for additional data’. However, the Quesada et al. (2009) citation was incorrectly reported in the references section; the correct reference is included in the list below. We regret this error given the importance of the soil P data reported in Quesada et al. (2010) to the relationships depicted in Fig. 3. We also note that data from the RAINFOR project (http://www.geog.leeds.ac.uk/projects/rainfor) were an important part of this synthesis, and we are grateful to all those involved in the RAINFOR project. A list of all references used to assemble the database is provided below, and all data (including references) are available at: http://knb.ecoinformatics.org/knb/metacat/nceas.964/nceas. Relationships among P cycling variables and ANPP in lowland tropical forests. (a) Foliar P vs. ANPP; (b) total soil P vs. ANPP; (c) total soil P vs. foliar P. In the Abstract and Methods sections, we stated that surface soil P values represented measured concentrations from 0 to 10 cm depths. However, surface soil P concentrations used in the analyses represented concentrations from depths of < 30 cm. In the Results section, we had classified lowland forests as those at a threshold elevation of 1000 m, and that this threshold corresponded to a mean annual temperature (MAT) breakpoint of 20.7 °C. However, in the final analysis, two warm sites (24 °C MAT, elevation 1100 m) were included with ‘lowland forests’, and a cool (19.7 °C MAT) montane forest site at an elevation of 800 m was included with ‘upland forests’. In the process of preparing the database for public distribution, we discovered several data reporting errors, and in a few cases, we were unable to relocate the original sources of the data used in the analyses. While none of these issues affected the overall conclusions of the article, they did lead to subtle changes in the relationships reported in Table 1, 3, 4, and Figure S2 (see Supporting Information), all of which have been reproduced correctly below. Total soil P vs.: (a) decomposition (k); and (b) soil respiration in lowland tropical forest. R-values represent Pearson correlation coefficients. Figure S2 ANPP vs. MAT (A and B) and MAP (C and D) in the upland and lowland forest sites in the database. As a service to our authors and readers, this journal provides supporting information supplied by the authors. Such materials are peer-reviewed and may be re-organized for online delivery, but are not copy-edited or typeset. Technical support issues arising from supporting information (other than missing files) should be addressed to the authors. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.