The in vivo distribution of physiological concentrations of NO3and N02 labeled with 13N was studied in germfree and conventional-flora Sprague-Dawley rats after gastric intubation (gavage), intravenous (cardiac or tail vein), or intraluminal (intestinal) injection. Some in vitro studies were performed to determine the influence of the bacterial flora on ion distribution. After gavage with 13N03-, essentially all of the label passed into the upper small intestine, where most was absorbed; however, up to 24% of the 13N could reach the ileum within 1 h. Gavage with 13N02resulted in some gastric absorption of the label, but most seemed to exit the stomach via passage into the duodenum. The exit of`3N02-from the stomach was slower, and less 13N appeared to be absorbed from the small intestine than with 13N03-. Movement of label through the gastrointestinal tract could be enhanced by inducing diarrhea. Absorbed '3N was either excreted in the urine, reentered the gastrointestinal tract at various points, or was temporarily stored in the eviscerated carcass. The bacterial flora, either by incorporation or chemical alteration, appeared to have some influence on the distribution of '3N from 13N03or 13N02-. 60 to 90 min duration were possible with '3N, and the detection of physiological concentrations of these labeled ions, and their metabolites, could be rather easily and sensitively traced in rats. MATERLALS AND METHODS 13N03and 13N02production and measurement. The tandem accelerator at the University of Wisconsin was used to produce "carrier-free" "3N as '3N03 (26). Chemical determinations showed that less than 0.1% of the '3N was in the form of '3N02or '3NH4'. About 30 mCi was made by bombarding an aerated water target for 20 min with a 3-,uA beam of 11-MeV protons (28). Passage of '3N03through regenerated spongy cadmium columns was used to make '3N02-(29). Samples were measured for radioactivity by placement inside an Nal (TI) crystal "well" detector. The carcass (i.e., the rat without the internal organs specified in the tables), however, was counted at a fixed distance from the well. Corrections to compensate for this change in counting geometry were made on these carcass values to allow comparison with other samples. All samples were counted at least twice for a 10-s interval, and values were decay-corrected to the start of '3N administration with the equation C, = Co-e"', where C, = the measured counts at various times (t) after the start of dissection, Co = the counts corrected to the start of dissection, and A = ln2/half-life of '3N. Animals. GF Sprague-Dawley rats were supplied by the University of Wisconsin Gnotobiology Labora