The effect of several male sterile cytoplasms on intercultivar canola hybrids was assessed over three years (1987-1989). The nuclear genomes of several cultivars and hybrids of canola were transfered to the nap, ogu and pol male sterile cytoplasms. Days to first flower and to days to 50% flower were increased in the pol but not in the ogu cytoplasm. Seed yield and total dry matter were reduced for both open pollinated populations and hybrids in the ogu cytoplasm. Seed yield and total dry matter were significantly lower in pol than in the nap cytoplasm for hybrids and to a lesser degree for the open pollinated populations. Relative seed oil concentration was reduced in the pol cytoplasm by 1.8 and 2.3 percent; in ogu by 1.5 and. 6.1 percent for 1988 and 1989 respectively. Relative seed protein was increased by the pol and ogu cytoplasms by 1.2 and 3.3 percent in both years. The effect of the male sterile cytoplasms on classical growth characters was also assessed in the three years. The ogu cytoplasm had consistently lower crop growth rates, net assimilation rates, relative growth rates and leaf area development than similar treatments in the nap and pol cytoplasms. Genotypes in the pol cytoplasm produced similar growth characters as those in nap, except the former produced a 10 % lower net assimilation rate early in the growing season. Hybrids in the nap cytoplasm produced greater levels of heterosis for leaf area and crop growth rates than the pol cytoplasm. B. napus hybrids were heterotic for leaf area index and leaf area duration but appeared to exhibit little or no heterosis for crop growth rates, net assimilation rates or relative growth rates. Carbon dioxide exchange rates of cultivars and hybrids in the three cytoplasms were determined in 1987, 1988 and 1989. In all three years, there were no significant differences between treatments in the nap and pol cytoplasms. As compared to nap cytoplasm treatments, ogu cytoplasm treatments reduced photosynthetic rates by approximately 20%. Genotypes in the ogu cytoplasm had approximately 65% of the leaf chlorophyll of the nap cytoplasm treatments early in the growing season and approached chlorophyll levels of the nap cytoplasm treatments later in the growing season. There were no significant differences in photosynthetic rate per unit chlorophyll between any of the cytoplasms. In the narrow range of genotypes tested there was no heterosis for photosynthetic rate or leaf chlorophyll content.