Previous in vitro studies have suggested that tetracycline‐HCl (TTC‐HCl) is adsorbed and actively released from root dentin. The aim of the current study was to evaluate the binding to and release of TTC‐HCl from human root dentin surfaces in vivo, and to evaluate the clinical utility of TTC‐HCl irrigation as an adjunct to scaling and root planing. Experiment I utilized two contralateral mandibular single‐rooted teeth which were examined in four adults with severe generalized periodontitis. One tooth in each patient was carefully scaled and root planed, under local anesthesia, and the other used as an unsealed control. Each subgingival root surface was irrigated for 5 min with an aqueous TTC‐HCl solution at a concentration of 100 mg/ml. Gingival crevicular fluid samples were collected on paper strips for the next three weeks. The TTC‐HCl concentrations in each sample were determined by the inhibition zone of B. cereus cultured on agar plates. The TTC‐HCl concentrations in gingival crevicular fluid collected 15 min after irrigation were 3100±670 μg/ml from the scaled lesions and 4700±1300 μg/ml from the unsealed root surfaces. The antibiotic concentrations decreased logarithmically over the next 7 days; 1500±270 μ/g/ml and 1100±330μ/g/ml at 2 h. 880±350μ/g/ml and 1300±360 μ/g/ml at 6 h and 19±5μ/g/ml and 31±26 μ/g/ml at 1 week for scaled and unsealed root surfaces, respectively. Results for week two and three indicated an average of over 8 μg/ml. The TTC‐HCl concentrations in gingival crevicular fluid from scaled and unsealed root surfaces were not statistically different at any time point. The tetracycline irrigation resulted in release of tetracycline at concentrations well above therapeutic concentrations for at least 1 week. Experiment II comprised 11 patients with severe adult periodontitis. All subjects were scaled and root planed prior to baseline measurements. The patients were monitored by the following parameters: probing pocket depth (PPD), probing attachment level (PAL), gingival index (GI) and plaque index (PI). 54 contralateral teeth exhibiting residual pocket depths of 5 mm were selected. Within each pair identified for the study, teeth were randomly assigned as test or control sites. After baseline measurement, each subgingival root surface was irrigated for 5 min; either with an aqueous TTC‐HCl solution of 100 mg/ml (test), or a 0.9% NaCl solution (control). At 3 and 6 months post‐treatment, the PI was unchanged for both groups. The GI index was reduced (0.062 > p > 0.001) in a similar manner for both groups. PPD showed statistically significant (p < 0.001) mean/patient decrease of similar magnitudes, 2.3±1.0 mm (test), and ‐1.6±0.8 mm (control) at 3 months, and ‐2.1±1.1 mm (test), and ‐1.4±0.9 mm at 6 months (control), respectively. Also, PAL measurements indicated a statistically significant average gain/patient of 2.1±1.1 mm in the test group (p<0.00l) and again of 1.2±1.0 mm in the controls (p = 0.002) at 3 months, and 1.8±1.1 mm (test; p<0.001) and 1.0±0.9 mm (controls; p= 0.005) at 6 months. Comparisons of the changes, between the groups, indicated statistically greater gain of PAL in the test group at both the 3 (p= 0.042) and 6 months (p= 0.034) intervals. These results suggest that TTC‐HCl irrigation of root surfaces for long periods of time (5 min) results in a subsequent release of active antibiotic into the gingival fluid at therapeutic levels for at least 1 week. TTC‐HCl irrigation resulted in significantly greater attachment gain as compared to scaling and root planing alone over at least a month period of healing.