© CSIRO 2004 10.1071/FP03218 1445-4408/04/050491 Functional Plant Biology , 2004, 31, 491–503 www.publish.csiro.au/journals/fpb CSIRO PUBLISHING Cyanogenesis in tropical Prunus turneriana: characterisation, variation and response to low light Rebecca E. Miller A,B , Roslyn M. Gleadow A and Ian E. Woodrow A A School of Botany, The University of Melbourne, Parkville, Vic. 3010, Australia. B Corresponding author; email: rem@unimelb.edu.au Abstract. This study characterised three aspects of cyanogenesis in the late successional tropical rainforest species Prunus turneriana (F.M.Bailey) Kalkman. First, all tissues were found to be highly cyanogenic, containing combinations of the cyanogenic glycosides (R)-prunasin, (S)-sambunigrin, and amygdalin. Second, the progeny of a single parent tree varied markedly and continuously in their cyanogenic glycoside content, indicating that this variation is genetically based. Third, we investigated resource allocation to cyanogenic glycosides in light treatments representative of rainforest understorey and gap environments. Contrary to our hypothesis that under low light, photosynthetic gain would be maximised by the reallocation of nitrogen from defence to the photosynthetic system, we found no difference in cyanogenic glycoside concentration, or the proportion of nitrogen allocated to cyanogenic glycoside, between high and low light. However, within the plant, shade affected a significant change in distribution of cyanogenic glycosides between young and old leaves. There was an increased allocation of cyanogenic glycosides to old, expanded and photosynthetically productive leaves, a pattern which appears inconsistent with predictions of optimal defence theories, and the results of other studies. We suggest that such a strategy may be advantageous for seedlings of tree species that can only reach a reproductive stage following the creation of a canopy gap. Keywords: amygdalin, cyanogenic glycoside, defence, light, polymorphism, prunasin, rainforest, regeneration, Rosaceae, sambunigrin. Introduction Within the heterogeneous environment of the tropical rainforest, light is the single most variable resource and therefore, the most critical resource limiting growth and survival (Chazdon 1988; Denslow et al. 1990). For example, in Costa Rican tropical montane forest, plants in the deep shade of the understorey receive as little as 2% full sunlight, increasing to 3–11% sunlight at gap edges (Denslow et al. 1990). Even within large forest gaps, daily photon flux density rarely exceeds 50% full sunlight (Chazdon 1992). Hence, the deeply shaded conditions of the understorey represent a challenge for photosynthesis and the acquisition of sufficient carbon to support growth and survival. Rain- forests are characterised by large numbers of shade-tolerant species, including some that tolerate deep shade during part of their development, but require increased irradiance in tree fall gaps for growth and survival (e.g. Denslow 1980; Piñero and Sarukhan 1982; Augspurger 1984; Denslow et al. 1990). The main focus of research on shade-tolerant species and their response to light has been the acclimation of the photosynthetic apparatus (e.g. Langenheim et al. 1984; Chow et al. 1988; Reich et al. 1991; Chazdon 1992; Turnbull et al. 1993). However, this is just one side of the carbon acquisition equation because in the long term, photosynthetic gain depends just as much upon leaf lifespan, especially in the rainforest environment where herbivore densities are relatively high (Coley 1983; Coley et al . 1985; Reich et al . 1991). It is intriguing, therefore, that many of the attributes that increase leaf lifespan, such as thick leaves and high concentrations of resource-demanding defence chemicals, tend to decrease the capacity for photosynthesis (Mooney and Gulmon 1982; Gulmon and Mooney 1986). As a consequence there has been increasing interest in how leaf defence strategies, particularly chemical defence, respond to environmental variation (e.g. Denslow et al. 1990; Höft et al. 1996; Ralphs et al. 1998; Hägele and Rowell-Rahier 1999). Most of the research on rainforest plants under varying light conditions has focussed on carbon-based defence chemicals. Underlying much of this research was the hypothesis of Bryant et al. (1983), who proposed that under higher light, elevated photosynthesis and carbohydrate Abbreviations used: A max , maximum carbon assimilation rate; dw, dry weight; LAR, leaf area ratio; PPFD, photosynthetic photon flux density; RGR, relative growth rate; SLA, specific leaf area.