The objective of this study was to develop an asexual embryogenic regeneration system for Rubus, even though previous attempts had proved unsuccessful (Fiola and Swartz, 1986). In an initial experiment, fruits of fieldgrown R. occidentalis 'Bristol' and 'Jewel' and R. idaeus 'Exp-72' plants were collected at the early green, late-green or early yellow, and the red stages of ripeness. Seeds were extracted and disinfested in a 0.5% sodium hypochlorite and 0.5% Alconox solution for 15 min followed by two 30-sec sterile distilled water rinses. Cotyledon explants (0.5 to 1.0 mm) were excised from embryos that had been aseptically dissected from the seeds and cultured on a modified MS medium (Murashige and Skoog, 1962) containing 200 mg casein hydrolysate/liter, 3% sucrose, and 0, 0.45, or 4.5 µM (2,4-dichlorophenoxy)acetic acid (2,4-D). For the three stages of ripeness, we used 133, 56, and60 explants for 'Bristol'; 143, 60, and 77 for 'Jewel'; and 64, 166, and 60 for 'Exp-72', with each group divided equally among growth regulator treatments. The pH of the medium was adjusted to 5.7 before adding 0.25% Gelrite (Scott Laboratories, Atlanta) and autoclaving. Cultures were incubated at 24C under cool-white fluorescent illumination (70 µmol•m-2 •s-1). About 70% of all 'Bristol' and 'Jewel' explants extracted from seed of late-green or early yellow fruit were highly embryogenic regardless of 2,4-D concentration. Most embryoids originated directly from explant tissue and not through callus. Repetitive embryogenesis (Tulecke and McGranahan, 1985) was common in all of these cultures. Even though the initiation of embryogenesis occurred sooner on explants cultured on media containing 2,4-D, embryoids developed on such media exhibited many abnormalities, including polycotyledony, "rooty" masses, and increased anthocyanin production. Sixty-nine 'Jewel' and 14 'Bristol' normal embryoids (parallelling