Degani, C., Institute of Horticulture, Agricultural Research Organization, Volcani Center, Bet-Dagan 50250, Israel
Lahav, E., Institute of Horticulture, Agricultural Research Organization, Volcani Center, Bet-Dagan 50250, Israel
El-Batsri, R., Institute of Horticulture, Agricultural Research Organization, Volcani Center, Bet-Dagan 50250, Israel
Gazit, S., Inst. Plant Sci. Genet. in Agric., Hebrew University of Jerusalem, Rehovot 76100, Israel

Single trees of several avocado (Persea americana Mill.) cultivars were caged with a beehive for the production of selfed progeny. Isozyme analysis was used to identify undesirable outcrosses in the planted progenies. Outcrossing rates were found to be highly variable, ranging from 0 to 0.86. Short accidental breaches in the net's integrity were suspected to be the main cause for the haphazard appearance of hybrids. Indeed, when 'Tova', which consistently had progenies with a high rate of hybrids (averaging 0.58) was caged very meticulously, outcrossing rate was consistently low, averaging only 0.07. Apparently, short periods of accidental exposure to open pollination can result in a high percentage of outcrossed progeny, due to the higher survival rate of outcrossed fruitlets compared to selfed ones. The residual low outcrossing rate found with meticulous caging probably occurred through the penetration of floating foreign pollen from adjacent trees. Thus, the production of pure selfed progeny in avocado requires meticulous caging and the absence of foreign cultivars around the caged trees. In general, the fact that net caging was not fully effective in excluding foreign pollen should encourage the performance of parentage analysis to confirm the purity of progenies produced in net cages.