Icarus 148, 589–593 (2000) doi:10.1006/icar.2000.6535, available online at http://www.idealibrary.com on NOTE Slowly Rotating Asteroid 1999 GU 3 Petr Pravec and Lenka ˇ Sarounov´ a Astronomical Institute, Academy of Sciences of the Czech Republic, CZ-25165 Ondˇ rejov, Czech Republic E-mail: ppravec@asu.cas.cz Lance A. M. Benner, Steven J. Ostro, Michael D. Hicks, Raymond F. Jurgens, Jon D. Giorgini, Martin A. Slade, and Donald K. Yeomans Jet Propulsion Laboratory, California Institute of Technology, Pasadena, California 91109 David L. Rabinowitz Yale University Physics Department, P.O. Box 208121, New Haven, Connecticut 06520-8121 Yurij N. Krugly Astronomical Observatory, Kharkiv National University, Sumska Street 35, Kharkiv 61022, Ukraine and Marek Wolf Astronomical Institute, Charles University Prague, V Holeˇ soviˇ ck´ ach 2, CZ-18000 Praha, Czech Republic Received April 13, 2000; revised September 11, 2000 Optical and radar observations reveal that 1999 GU 3 is sub- kilometer-sized object with a synodic period of 9.0 days, low vi- sual and radar albedos, and colors more consistent with the or- dinary chondrites than the vast majority of main-belt asteroids. c  2000 Academic Press Key Words: asteroids, rotation;photometry;radar. 1. Introduction. Studies of the distribution of asteroid spin rates vs diameter have shown that there is a significant excess of slow rotators with periods >30 h at diameters below ≈50 km (Dermott et al. 1984, Binzel et al. 1989, Fulchignoni et al. 1995, Pravec and Harris 2000). Asteroid collisional evolution models explain small asteroids as fragments generated in catastrophic disruptions or cratering of larger asteroids and predict their spin-up but not significant numbers of slow rotators (Harris 1979, Farinella et al. 1992). Mechanisms causing spin- down of small asteroids have been suggested (see Discussion below) but none of the hypotheses has been proven to satisfactorily explain the basic observational data. The largest known slow rotator is C-type main belt Asteroid 253 Mathilde, which has a rotation period of 17.41 days and a mean diameter of 53 km (Mottola et al. 1995, Veverka et al. 1997). The best-studied slow rotator is S-type near- Earth Asteroid 4179 Toutatis, which radar observations have shown to be an elongated body in a nonprincipal axis (NPA) rotation state characterized by periods of 5.37 days (rotating about the long axis) and 7.42 days (precession of the long axis about the angular momentum vector; Hudson and Ostro 1995, Ostro et al. 1999a). Toutatis’ NPA rotation causes its lightcurve to appear nonperiodic but it contains characteristic frequencies (corresponding to the observed synodic periods of 7.3 and 3.1 days) that are related to the periods of the NPA rotation (Spencer et al. 1995, Kryszczy´ nska et al. 1999). Lightcurves of several other asteroids with long periods have shown deviations from simple periodicity that are suggestive of NPA rotation: 1689 Floris-Jan (Harris 1994), 288 Glauke and 3288 Seleucus (Harris et al. 1999), and 3691 Bede and 1997 BR (Pravec et al. 1998). Thus, although we cannot say that all slowly rotating asteroids are in NPA rotation states, such states appear to be common among them. The size distribution of very slow rotators extends from Mathilde (≈50 km) down to 1997 BR (≈1.5 km, Pravec et al. 1998). Here we present observations of the ≈0.5 km object 1999 GU 3 that reveal a very slow rotation of 9 days. 2. Optical observations. We observed 1999 GU 3 photometrically from 1999 April 14.3 to May 19.9. The observations were made at Ondˇrejov Observa- tory, Kharkiv Observatory, and Table Mountain Observatory. We used telescopes with diameters of 0.6– 0.7 m that are equipped with CCDs. The observations were made and reduced in the standard way as described by Pravec et al. (1996) and Rabinowitz (1998). The measurements were calibrated in the Johnson–Cousins system and calibrated using Landolt (1992) standards. Most observations were made through the R filter, with additional measurements in B, V, and I on April 14 and 15 at Table Mountain Observatory. The consistency of the R data calibra- tions from all three stations is about 0.02 mag. The times have been corrected for light travel time, and magnitudes have been reduced to unit geocentric and heliocentric distances. Table I summarizes the optical observations. Figure 1A plots lightcurve measurements reduced to a phase angle of 60 ◦ (assuming a linear phase parameter of 0.031 mag/deg; see below) vs time. The relatively densely covered part of the lightcurve obtained between April 14 and 24 shows a long-period, large amplitude variation. If we assume that there are two maxima/minima pairs per cycle, then this 10-day interval appears 589 0019-1035/00 $35.00 Copyright c  2000 by Academic Press All rights of reproduction in any form reserved.