As part of the project LBA-EUSTACH (European Studies on Trace gases and Atmosphere Chemistry as a contribution to the Large-Scale Biosphere-Atmosphere experiment in Amazonia), we examined the diel pattern of isoprenoid exchange in the wet season of 1999 at a remote field site in Rond¿nia, Brazil. The emission pattern of two tree species in a secondary forest was investigated by means of a dynamic branch enclosure system and was compared on the basis of climatological variables like temperature and light and physiological parameters such as assimilation, transpiration, and stomatal conductance. While the species Hymenaea courbaril was found to be a strong isoprene emitter, Apeiba tibourbou was found to exclusively emit monoterpenes and no isoprene. Species-related standard emission factors calculated on a carbon basis were 45.4 ¿g g-1 h-1 (24.9 nmol m-2 s-1) for isoprene and 3.6 ¿g g-1 h-1 (0.75 nmol m-2 s-1) for monoterpene emission, representing a nontrivial carbon loss by the plants of 1.8% and 0.1% C relative to the net carbon assimilation on a daily basis. About 90% of the total monoterpene emission from A. tibourbou was comprised of sabinene, α-pinene, ¿-pinene, myrcene, and limonene, in decreasing quantity. Despite the difference in isoprenoid emission composition, the diel emission pattern of both tree species reacted similarly toward the environmental conditions, fluctuating light and temperature, indicating that closely related metabolic controls were involved in the actual emission. Both isoprene emission and monoterpene emission exhibited a light saturation curve similar to CO2 assimilation. No isoprenoid emission was found during nighttime. The strong light dependence of the monoterpene emission by A. tibourbou suggests that this tree species does not store monoterpenes but emits them immediately upon production. The diel pattern of both the isoprene and the monoterpene emission could be adequately simulated by current isoprene algorithms. The ambient air mixing ratios of isoprenoids were clearly dominated by isoprene, with peak values of 8 ppb inside the main canopy. Vertical gradients of ambient air mixing ratios in and above a primary rain forest site illustrated the emission of isoprene by the main canopy dependent upon light and temperature but were also indicative of a potential sink at the forest floor. For monoterpenes, corresponding gradients could hardly be resolved, reflecting observed ambient air mixing ratios more than an order of magnitude lower than for isoprene. Nevertheless, a strong diel cycle of short-lived monoterpene compounds like α-pinene was found in the primary as well as in the secondary forest site, which further strengthens our finding of a strong light dependence of biogenic monoterpene emission even on a larger scale. Our findings to some extent question the applicability of the commonly used monoterpene emission algorithm to the tropics. A strong light dependence of biogenic monoterpene emissions may, if generalized for tropical tree species in common, have a strong impact on estimated global flux rates for tropical regions.