Stable carbon isotope analysis of tree rings has become a widely used proxy in environmental and palaeoclimatological studies. In those studies, alpha-cellulose has often been the preferred material because of its singular composition and its immobility in wood. However, cellulose extraction is a time-consuming procedure and since the development of on-line isotope ratio mass spectrometers has become the time-limiting step in the isotopic analysis of wood samples for dendrochemical purposes. In this study we evaluate the necessity of cellulose extraction for isotopic analysis of tree rings in a tropical mangrove tree, Rhizophora mucronata Lam. Comparison between the delta(13)C of extracted alpha-cellulose and bulk wood material revealed a highly significant linear relationship (delta(13)C(bulk) (wood)=0-92 (+/- 0.08)*delta(13)C(alpha-cellulose)-2.91 (+/- 2.04); p < 0.001) for alpha-cellulose values between -24 parts per thousand and -27 parts per thousand. However alpha-cellulose was on average 0.97 +/- 0.03 parts per thousand enriched in C-13 as compared to bulk wood. The slope of the regression was not significantly different from one (p < 0.05). Furthermore, no significant difference was found between either the delta(13)C(bulk) (wood) - delta(13)C(alpha-cellulose) Slopes for earlywood and latewood or between the slopes for samples from trees growing in contrasting environmental conditions. These results indicate that bulk wood can be used instead of alpha-cellulose when measuring stable carbon isotopes in the sapwood of R. mucronata in the context of a dendrochronological investigation. (c) 2005 Elsevier B.V. All rights reserved.