Gross and net O 2 production between May 1996 and February 1999 was determined in bottle incubation experiments with H 2 18 O spike and from the change in O 2 concentration. Carbon fixation rates were obtained from 14 C incubations. In general, production rates determined using the H 2 18 O-spike were about twice the primary production determined by the 14 C method, where the latter was close to net oxygen evolution. These relationships are similar to results for the open ocean. During the spring bloom, when the dinoflagellate Peridinium was abundant, the ratio of gross O 2 production to carbon fixation was about 7.5, and net O 2 production was greater than carbon fixation. The difference between O 2 gross production and carbon fixation results, at least in part, from uptake by Mehler reaction and from recycling of the 14 C tracer by dark respiration and the alternative oxidase (AOX). We used the difference in isotopic discrimination against 18 O, occurring during O 2 consumption by various biological pathways, to place constraints on the relative engagement of these pathways. We estimated the overall discrimination against 18 O in the lake from O 2 isotopic mass balance as 20.5-29‰. The only mechanism that can explain the strong overall fractionation in the lake is AOX, which strongly discriminates against 18 O (ϳ31‰). Our results show, for the first time, that uptake by AOX is widespread and quantitatively important to oxygen consumption in aquatic systems.Oxygen exchange by photosynthesis and respiration is the largest biogeochemical cycle in aquatic systems. In order to understand this cycle, it is necessary to know the gross rates of the major processes involved in oxygen production and uptake. Production of O 2 is known to occur in one process in photosystem II, but O 2 consumption in aquatic organisms is possible by several reactions. These include ordinary respiration through the cytochrome oxidase pathway, respiration by the alternative oxidase pathway, Mehler reaction, and photorespiration. The first two processes take place in light as well as in dark conditions, whereas the latter two occur only under illumination. Although the presence of the above mechanisms has been established in different studies, their quantitative importance in the overall O 2 uptake in aquatic systems is not well known, and it is necessary to assess their role in natural environments. In this respect, ordinary O 2 incubation methods, which are very useful for the assessment of photosynthetic production from light-and dark-incubation experiments (e.g., Williams and Purdie 1991), do not provide the necessary information. The main drawback of these methods is their inability to measure the rate of O 2
AcknowledgmentsWe thank Y. Geiffman and the Mekorot Watershed Unit for provision of wind data and extend special thanks to the Kinneret Limnological Laboratory skipper M. Hatab. We are grateful