Aims Figs and fig wasps are classical materials for research on plant-animal coevolution. There are both monoecious and dioecious figs. The evolutionary mechanism of Ficus has attracted many biologists. Most recently, Chen (2005) deduced that Ficus evolved from monoecious to dioecious, based on comparison of style distribution, pollinators' ovipositor length, and reproduction success in monoecious and dioecious figs. Our objective was to test the duduction and further study Ficus evolution.
Methods We collected figs in the female floral phase and measured the length of styles and pollinators' ovipositors in both monoecious figs and dioecious figs using dissecting microscopy. Figs in the male florets phase were also collected. We counted (i) seeds, galls, bladders, and unparasitized ovaries for monoecious figs, (ii) galls, bladders, and unparasitized ovaries for dioecious male figs, and (iii) seeds and unparasitized ovaries for female figs.
Important findings Ovipositor length was significantly longer, and the coefficient of variation (CV) of style length was higher in the two monoecious figs than in dioecious figs. Style length had unimodal distributions in the two monoecious figs, and their CVs were higher than pollinators' ovipositor length CVs. The style length had biomodal distributions in the two dioecious fig species, and the CVs of style length were also higher than those of the ovipositor length. Monoecious figs were more effective in producing galls, based on the proportions of ovaries that pollinators' ovipositors could reach, but dioecious figs were more effective in producing seeds. Therefore, the deduction of Chen (2005) is not completely correct, and additional research is needed to resolve the Ficus evolution question.