Studies were made on the bee visitation of 6 quince cultivars and on the foraging behaviour of honeybees at quince flowers for 3 consecutive years. The bee visitation was highly intense because both the plantation and its surrounding was overpopulated by honeybee colonies. Some 5.5 bee visits were counted in average at 50 opening quince flowers in 10 minutes and some 9.7 flowers were visited of the 50 ones observed meanwhile. This equals some 7 bee visits per flower per day per in good weather. Bee visitation, however, was variable and it was greatly different in the three consecutive years with fairly favourable weather. Some cultivars tended to be more and others less visited by honeybees than the others but these differences were not consequent at each occasion. No consequent relationship between the weather and the bee visitation of quince trees could be recognised. It was concluded that .other factors were responsible for the variable nature of the bee activity at quince. Most honeybees tended to collect pollen (51.6% in average for the 3 years) and usually much less gathered for nectar only (19.9%), and the rest of them were mixed behaviour foragers gathering for both (28.5%). There were some slight differences in the foraging behaviour of honeybees at some cultivars but these differences were not always consequent in consecutive years. Also the nectar production of flowers failed to affect the bee visitation and the foraging behaviour of honeybees definitely. For the variable nature of bee visitation and bee behaviour at quince flowers, supplementary pollination is needed to achieve as high set of fruits as high is required to a good crop at quince (at least 20-25% because the flower density is low of this fruit tree species). Since the intensity of bee visitation at the flowers is the only reliable estimate of the necessity of supplementary bee pollination further research is needed to explore the relationship between the number of honeybee visits and the consequent fruit set at quince.