An efficient method for the production of haploid plants or cell lines would open new avenues for citrus genetics and breeding. In view. of the limited success obtained with androgenesis as described in the literature, we have instead experimentally induced in situ parthenogenesis. Flowers of Clementines were pollinated in the field with pollen of.Meyer lemon (C. limon) irradiated at 0, 300, 600 and 900 Gray. The rate of in vitro pollen germination was not modified by irradiation up to 900 Gray. Fruit set was slightly affected only by the 900 Gray dose. However, the number per fruit and size of seeds was drastically reduced by the 300 Gray dose and higher. Embryos found in the small seeds of irradiated treatments were saved in vitro and the ploidy level of the plantlets obtained was evaluated by- flow cytometry. Some haploid plantlets were identified and some mixoploid structures (haploid/diploid) were observed. From embryos cultivated in a medium favourable to callogenesis two embryogenic haploid callus lines were obtained which have produced many secondary haploid embryos. Isoenzyme profiles of the haploid plants and calli studied were consistent with the theoretical profiles for haploid Clementine. The in vitro growth of haploid plantlets was very slow but grafting haploid shoots onto one year old C. macrophylla has resulted in the satisfactory development of haploid Clementine. Moreover these haploid embryogenic calli create the opportunity for triploid synthesis by 'diploid+haploid'somatic hybridisation and also have uses in genetic transformation programmes.