Electronic Telegram No. 5733 Central Bureau for Astronomical Telegrams Mailing address: Hoffman Lab 209; Harvard University; 20 Oxford St.; Cambridge, MA 02138; U.S.A. e-mail: cbatiau@eps.harvard.edu (alternate cbat@iau.org) URL http://www.cbat.eps.harvard.edu/index.html Prepared using the Tamkin Foundation Computer Network (3221) CHANGSHI V. Bao, Royal Astronomical Society of New Zealand, reports the discovery of an apparent satellite of minor planet (3221) based on an occultation of the star UCAC4 325-092673 (which has catalogued Gaia magnitudes G = 11.3 and R = 10.5) on 2026 Aug. 5.533 UT. This main-belt asteroid had a magnitude of 15.9, which was 4.6 magnitudes fainter than the target star. Bao observed from Auckland, New Zealand, using a 28-cm telescope (+ ZWO ASI662MM with a GPS 1PPS disciplined personal-computer clock running at a cadence of 0.02s). The recording showed two consecutive occultations of durations 0.15 and 2.08 s, separated by 10.81 s, corresponding to chord lengths of 0.6 and 8.3 km. The flux of the first drop fell to the brightness level of the comparison star UCAC4 325-092706, which has catalogued Gaia magnitudes 12.6 (G) and 11.9 (R), and the flux for the second drop was to the limiting magnitude of the recording, which was approximately 13.4 (G) and 12.9 (R). The depths of the two drops (1.3 and 2.1 in G, 1.4 and 2.4 in R) are sufficiently great to exclude a double star from being the explanation for the light curve. The NEOWISE-catalogued diameter of the asteroid is 6.0 +/- 0.6 km; the Gaia DR3 magnitude measurements suggest a rotational light variation in the order of 0.30 magnitude; and Carry et al. (2024, A.Ap. 687, A38) derive an oblateness of 0.3 +/- 0.2. On this basis, the main body has been modeled to fit the light curve as an ellipse having an area corresponding to the NEOWISE- catalogued diameter with axes of 8.3 km x 4.4 km, oriented parallel to the chords. Fresnel modeling of the drop from the apparent satellite indicates that the chord was tangential to the limb of the apparent satellite. The modeling indicates that the minimum size of the apparent satellite is an ellipse of size 2.0 km x 1.0 km, with the maximum diameter for a circular body being about 3 km. The separation of the satellite was 0".057 in p.a. 105 +/- 4 degrees, with a separation of 47.8 km. L. Liberato (GaiaMoons, Observatoire de la Cote d'Azur) advises that, for this minor planet, the binary-search analysis in the Gaia astrometry (Liberato et al. 2026, A.Ap.; accessible via website URL https://doi.org/10.1051/0004-6361/202659515) detected a period of 8h.5 with a very low statistical confidence, and that while this period is much shorter than the expected orbital period of this apparent satellite, it is not contradictory to the satellite evidence from this occultation. The sky-plane configuration showing the observed occultation chords is posted at http://www.cbat.eps.harvard.edu/iau/cbet/005700/CBET5733_Fig1.png. Representative models of the system, and the observed light curves with superimposed Fresnel modeled light curves, are posted at website URL http://www.cbat.eps.harvard.edu/iau/cbet/005700/CBET5733_Fig2.png. D. Herald (who forwarded this report to the Central Bureau) and D. Gault, Trans-Tasman Occultation Alliance; and C. Weber, European Section of the IOTA, aided in the analysis. The reader is also referred to the end notes on CBET 5679. NOTE: These 'Central Bureau Electronic Telegrams' are sometimes superseded by text appearing later in the printed IAU Circulars. (C) Copyright 2026 CBAT 2026 August 26 (CBET 5733) Daniel W. E. Green