Abstract. The oxygen minimum zone (OMZ) of the Arabian Sea is the thickest of the three oceanic OMZ. It is of global biogeochemical significance because of denitrification in the upper part leading to N2 and N2O production. The residence time of OMZ water is believed to be less than a decade. The upper few hundred meters of this zone are nearly anoxic but non-sulfidic and still support animal (metazoan) pelagic life, possibly as a result of episodic injections of O2 by physical processes. We report on discrete measurements of dissolved O2 and NO2–, temperature and salinity made between 1959 and 2004 well below the tops of the sharp pycnocline and oxycline near 150, 200, 300, 400, and 500 m depth. We assemble nearly all O2 determinations (originally there were 849 values, 695 of which came from the OMZ) by the visual endpoint detection of the iodometric Winkler procedure, which in our data base yields about 0.04 mL L−1 (~ 2 μM) O2 above the endpoint from modern automated titration methods. We acknowledge that much lower (nanomolar) O2 values have been measured recently with the STOX (Switchable Trace amount OXygen) sensor in the eastern tropical South Pacific, and that similar conditions may also prevail in the Arabian Sea OMZ. In spite of the error in O2 measurements at vanishingly low levels, we argue that the temporal trends of the historic data should still hold. We find 632 values acceptable (480 from 150 stations in the OMZ). The data are grouped in zonally paired boxes of 1° lat. and 2° long. centered at 8, 10, 12, 15, 18, 20, and 21° N along 65 and 67° E. The latitudes of 8–12° N, outside the OMZ, are treated in passing. The principal results are as follows: (1) an O2 climatology for the upper OMZ reveals a marked seasonality at 200 to 500 m depth with O2 levels during the northeast monsoon and spring intermonsoon seasons elevated over those during the southwest monsoon season (median difference, 0.08 mL L−1 [~ 3.5 μM]). The medians of the slopes of the seasonal regressions of O2 on year for each of the NE and SW monsoon seasons are −0.0043 and −0.0019 mL L−1 a−1, respectively (−0.19 and −0.08 μM a−1; n = 10 and 12, differing at p = 0.01); (2) four decades of statistically significant decreases of O2 between 15 and 20° N but an opposing trend toward an increase near 21° N are observed. The mechanisms of the balance that more or less annually maintain the O2 levels are still uncertain. At least between 300 and 500 m, the replenishment is inferred to be due to isopycnal re-supply of O2, while at 200 (or 250?) m it is diapycnal, most likely by eddies. Similarly, recent models show large vertical advection of O2 well below the pycnoclines and oxyclines. The NO2– distribution, taken as an indicator of active NO3– reduction, does not show a trend in the redox environment for a quarter of a century at a GEOSECS station near 20° N. In the entire OMZ, the regression slopes on year within seasons for the rather variable NO2– do not present a clear pattern but by other measures tended to an increase of NO2–. Vertical net hauls collect resident animal (metazoan) pelagic life in the NO2– maximum of the OMZ at O2 levels well below the lower limit of the Winkler titration; the extremely low O2 content is inferred from the presence of NO2– believed to be produced through microbial NO3– reduction. Instead of the difficult measurement by the STOX sensor, the relation between the very low O2 inferred from presence of NO2– and mesozooplankton should be studied with 100 to 150 L bottles rather than nets. The spatial (within drift stations) and temporal (daily) variability in hydrography and chemistry is large also below the principal pycnocline. The seasonal change of hydrography is considerable even at 500 m depth. Future O2 or nutrient budgets for the OMZ must not be based on single cruises or sections obtained during one season only. Steady state cannot be assumed any longer for the intermediate layers of the central Arabian Sea.