The aim of the work reported here was to ascertain that the patterns of labeling seen in isolated bacteroids also occurred in bacteroids in intact nodules and to observe early metabolic events following exposure of intact nodules to 4CO2. Intact nodules of soybean (Glycine max L. Merr. cv Ripley) inoculated with Bradyrhizobium japonicum USDA 110 and pea (Pisum sativum L. cv Progress 9) inoculated with Rhizobium leguminosarum bv viciae isolate 128C53 were detached and immediately fed 14CO2 for 1 to 6 min. Bacteroids were purified from these nodules in 5 to 7 min after the feeding period. In the cytosol from both soybean and pea nodules, malate had the highest radioactivity, followed by citrate and aspartate. In peas, asparagine labeling equaled that of aspartate. In B. japonicum bacteroids, malate was the most rapidly labeled compound, and the rate of glutamate labeling was 67% of the rate of malate labeling. Aspartate and alanine were the next most rapidly labeled compounds. R. leguminosarum bacteroids had very low amounts of 14C and, after a 1-min feeding, malate contained 90% of the radioactivity in the organic acid fraction. Only a trace of activity was found in aspartate, whereas the rate of glutamate and alanine labeling approached that of malate after 6 min of feeding. Under the conditions studied, malate was the major form of labeled carbon supplied to both types of bacteroids. These results with intact nodules confirm our earlier results with isolated bacteroids, which showed that a significant proportion of provided labeled substrate, such as malate, is diverted to glutamate. This supports the conclusion that microaerobic conditions in nodules influence carbon metabolism in bacteroids. Conclusions from studies of carbon uptake and metabolism by isolated bacteroids need to be confirmed with bacteroids in their natural environment, i.e. in the intact nodules. A noninvasive method to establish the form of carbon received by bacteroids would be to feed "4CO2 to the leaves and analyze the compounds labeled in the nodule and bacteroids.Results from experiments of this type (15,21,22) have yielded valuable information. However, due to the variation in rates of photosynthesis, translocation, and subsequent metabolism, the method does not allow clear resolution of the order of metabolic events in nodules.The problem can also be approached by taking advantage of high PEPC2 activity in the nodules (4, 16). This approach is biased to some degree because metabolism of labeled carbon begins at OAA and malate rather than sucrose (31). However, the malate pool in the cytosol of nodules is a relatively large carbon pool (29), and the fact that PEPC activity may be equivalent to as much as 14% of nodule respiration (1 1) makes it possible rapidly to label the relatively large pool. These facts combined with the established importance of dicarboxylic acids in bacteroid carbon nutrition make this an attractive approach for the study of carbon metabolism in intact nodules. The approach has been employed by others, b...