Use of carbon and nitrogen isotope ratios in termite research ICHIRO TAYASU* Laboratory of Forest Ecology, Division of Environmental Science and Technology, Graduate School of Agriculture, Kyoto University, Kyoto 606±8502, Japan In this paper, I review carbon and nitrogen isotopic (natural abundance levels) studies of termites. The carbon isotope ratio of CH 4 emitted from termites, together with the emission rates of CO 2 , CH 4 and H 2 , showed several trends corresponding to the kinds of symbiotic microbes and feeding habits. The fraction of methane oxidized in the nest structure was estimated by comparing carbon isotope ratio of CH 4 emitted from the nest with that produced by termites in the nest. Symbiotic nitrogen ®xation in the gut of termites has been shown to have a signi®cant contribution to the nitrogen economy in some wood-feeding termites. The carbon isotope ratio distinguishes between C4 from C3 plants, and the fractional contribution of grass in the diet can thereby be estimated. The carbon and nitrogen isotope ratios in termites are discernible among soil-feeders, fungus cultivators and wood-feeders. Wood/soil- interface feeders have intermediate values between wood- and soil-feeders, and thus carbon and nitrogen stable isotope ratios are assumed to characterize the degree of humi®cation of the material consumed by termites. It is suggested that carbon and nitrogen isotope ratios are useful indicators of the functional position of termites in the decomposition process. A similar isotope pattern has been obtained in earthworms, suggesting that isotope signatures might be useful parameters in investigating detritivorous animals in general. Key words: decomposition; detritivore; humi¢cation; Isoptera; trophic enrichment. INTRODUCTION Stable carbon and nitrogen isotope ratios have been widely applied in ecological studies. Natural abundance isotopic signatures can be used to ®nd patterns and mechanisms at the single organism level as well as to trace food webs, understand palaeodiets and follow whole ecosystem cycling in both terrestrial and marine ecosystems (Pe- terson & Fry 1987; Wada et al. 1991). Since variation in stable isotope ratios is so small, the natural abundance of 13 C and 15 N is expressed per mil (&) deviation from international stan- dards, as de®ned by the following equation: d 13 C; d 15 N R sample =R standard  1 1000; where R in d 13 C and d 15 N denotes 13 C/ 12 C and 15 N/ 14 N, respectively. Pee Dee Belemnite and atmospheric nitrogen are used as the standards. In this paper, I use the term `trophic enrichment' as the difference in isotope ratio between an animal and its putative diet, expressed as DdX (  dX animal dX diet , where dX denotes d 13 C and d 15 N). The carbon isotope ratio has been used as a dietary indicator. In a review paper, Deines (1980) reported that d 13 C values are separated by bimodal distribution (i.e. 20 to 35& for C3 plants [woody forms] and 9 to 14& for C4 plants [grasses]). In contrast, trophic enrichment of carbon isotopes is small, mostly within 1& (DeNiro & Epstein 1978; Fry & Sherr 1984). Therefore, it has been possible to characterize the carbon ¯ow of a food chain based on woody and/or herbaceous materials. Trophic enrichment of nitrogen isotopes is usually positive, averaging 3.4& (DeNiro & *Email: <tayasu@kais.kyoto-u.ac.jp> Received 14 January 1998. Accepted 3 March 1998. Ecological Research (1998) 13, 377±387