This paper examines within-population variation in realized fecundity of Pistacia lentiscus (Anacardiaceae) and considers how plant traits relevant to the interaction with avian seed dispersers do influence it. A high1y variable fraction of final-sized fruits (on average, 80.1% and 55.5% in two consecutive study years) contained empty seeds, either because of embryo abortion or parthenocarpy, but these fruits were retained and eventually consumed by birds. A total of 26 bird species consumed the fruits, with 19 of them being seed dispersers accounting for the removal of 82.5% and 83.3% of the final-sized fruit crop in two study years. Only four bird species removed 82.2% of the seeds and pulp/seed predators took 6.0%. Plant characteristics direct1y related to fecundity had strong effects on estimates of dispersal success relevant to fitness, accounting for 82.4% of its variance when holding constant the effects of fruit-design traits; the latter accounted for 1.8% of this variance. Fruit removal by frugivores had a negligible contribution to variation in realized reproductive output. This was largely attributable to the cumulative effects of the pre-dispersal phase, when great losses of potential fecundity occurred as abortion of embryos or production of parthenocarpic fruit. Variation in removal rates by frugivores did not "screen-off" these pre-dispersal effects.

Photo: Ripe fruits of Pistacia lentiscus. The seed is shown split open; from left to right: parthenocarpic fruit, with the seed only with vestigial remains; aborted seed (note the blackish embryo aborted); ripe fruit with seed including a fully expanded, viable, embryo.