The Eastern Cape is an area in which Apis mellifera capensis, A. m. scutellata and their hybrid are known to naturally occur. I investigated the mandibular gland profiles and ovarial development of queenright workers from 4 localities. Their queens were then removed to determine how these aspects changed upon queen loss. In addition, drifted bees were analysed in the same way to determine how these factors changed once they had gained entry to a foreign hive. The queenright bees, form all 4 localities were found to have 9HDA as the most abundant of the 5 fatty acids measured and all localities had small percentages of 9ODA in their mandibular gland secretions. This resulted in relatively high queenright 9ODA:10HDA and 9HDA:10HDAA ratios. Despite this the percentage of bees with undeveloped ovaries was consistent with their queenright status. The mean values of these 2 ratios were significantly higher in the bees from East London and Cradock than those from Port Elizabeth. Steynsburg's bees were intermediate in this regard. Upon queen loss, the bees from all 4 localities had an increase in the percentage of 9ODA but the other compound changes in varying ways. East London's bees were the only ones not to become significantly more queen-like after queen loss. After 14 days without a queen, the mean values of these 2 ratios were much higher in the bees from Port Elizabeth than those from the other localities. Certain individuals from Port Elizabeth had values of these ratios that exceeded those found in A. m. capensis queens. Port Elizabeth was the only locality to display any surrogate queens and exhibited the highest increase in the number of bees with partially or fully developed ovaries. While certain individuals from the other localities had values of these ratios that exceeded these values reported in A. m. scutellata queens, Steynsburg's bees were the only ones that did increase in in terms of the number of bees with developed ovaries subsequent. The bees from Steynsburg were shown to suppress the mandibular gland and ovarial development of drifters from Port Elizabeth while bees from East London did not.

From an apicultural point of view the Cape fynbos is under-utilised and our knowledge of its utilization by the Cape honeybees is incomplete. The key aim of this study was to test the hypothesis that the Cape honeybees utilize the fynbos species as the preferred source of nectar and pollen. Subsidiary aims included distinguishing vegetation communities in the area, identifying pollen and nectar sources, the relationship between brood population and seasonal pollen collection patterns, examining the effect of meteorological factors on pollen collection. The study site was on Rivendell Farm within the Eastern Cape Albany district: an area of high species richness. A checklist of vascular plant species was produced revealing 97 families, 271 genera and 448 species. A classification by two-way indicator species (TWINSPAN) recognized seven vegetation communities: Forest, Bush clumps, Acacia savanna, Grassland, Grassy fynbos, Fynbos and Shrubland. Direct field observations of the foraging of Cape honeybees identified 54 nectar and pollen source plant species. Honeybee pollen loads trapped from four colonies of hives identified 37 pollen source plants of which Metalasia muricata, Eucalyptus grandis, Eucalyptus camaldulensis, Erica chamissonis, Helichrysum odoratissimum, Helichrysum anomalum, Crassula cultrata and Acacia longifolia were the predominant pollen source plants. It was also found that 60% of pollen yield derived from fynbos vegetation. The pollen source plants came from both Cape endemic and from nonendemic species. Thus we reject the hypothesis that Cape honeybees selectively forage fynbos species as a preferred source of pollen and nectar. The examination of the effect of temperature, wind-speed and temperature on pollen collection activity of honeybees revealed that: a temperature range of between 14°C to 26°C was optimal for pollen collection; wind speeds of up to 4m/s were conducive for pollen collection; relative humidity was found to have no significant influence on pollen collection. Pollen collection and brood rearing patterns are positively correlated with flowering intensities, but we found in our Eastern Cape study site that brood rearing was not limited to the spring flowering season but did extend to the end of summer. In order to determine the available nectar yield of common plant species hourly secretion of nectar volumes was measured for 24 hours to determine the variation of available nectar during different times of the day. In all nectar producing species the nectar volume was high in the early morning and declined as the day progressed. We found that the volume of available nectar was affected by prevailing temperature and humidity around the flowers.