The resource allocation for vegetative growth and female reproduction in three tree species of subgenus Cyclobalanopsis (Quercus, Fagaceae), i.e., Q. salicina, Q. sessilifolia, and Q. acuta, were examined on a per-individual basis in two consecutive reproductive seasons, in order to test whether these trees fit the predictions of the masting hypotheses about resource matching versus resource switching. Since the three Quercus species have a biennial fruiting habit, it takes 3 years for the observation of two reproductive events. Female flower and acorn production per tree were investigated by using a seed-trap method and a numerical analysis of seed dispersal. The net production of each individual was estimated as the sum of the annual increase in the dry mass of vegetative organs and reproductive investment per tree. In the data analyses, the three species were pooled, since all 12 sample trees of the subgenus apparently showed masting in the same year, with no exceptions. Female flower and acorn production per individual tree changed considerably between years. The net production per tree increased with tree size, but did not differ between years. Therefore, the reproductive allocation (proportion of a plant's annual assimilated resources which are used for reproduction) differed dramatically between years. On the other hand, within a year, the reproductive allocation increased with increasing net production per tree. These results suggest that the switching of resource allocation between years within an individual are occurring in subgenus Cyclobalanopsis species, and the intensity of the switching increases with increasing tree size.

Reproductive success of the tall tree species Lindera erythrocarpa Makino (Lauraceae) was examined on a per individual basis using the seed-trap method and a new method of numerical analysis of seed dispersal. Female flower and seed production per tree were successfully estimated using the model, and the dependence of each variable on tree size was expressed using allometric equations. Although seed set was highest on mediumsized trees (5% at 27 cm d.b.h.), seed production per tree was highest on the largest trees. Thus, a relative measure of reproductive success, such as seed set, was not a reliable predictor of individual variations in absolute seed production. Dry matter investment in reproduction was also expressed using an allometric equation. Reproductive investment increased as net production increased. Vegetative growth of the woody organs of individual trees reached a peak of 31.2 kg/year, at which point reproductive investment was 4.41 kg/year. The usefulness of the seed-trap method and the allometric approach to evaluate differences in reproductive variables among individuals of a large tree species was confirmed.

No abstract

Many masting species switch resources between vegetative growth and reproduction in mast and non‐mast years. Although masting of oak species is well known, there have been few investigations of the relationship between vegetative growth and reproduction based on long‐term monitoring data, especially in evergreen oaks of subgenus Cyclobalanopsis. We investigated annual variations over 13 years in acorn and leaf production of three evergreen oak species in subgenus Cyclobalanopsis, genus Quercus (Fagaceae)—Q. acuta, Q. salicina and Q. sessilifolia—in western Japan. In these species, the maturation of acorns occurs in the second autumn after flowering, which is known as a biennial‐fruiting habit. We found a pattern of acorn production and masting in alternate years that was synchronized in all three species. Masting was not correlated with temperature and precipitation. Annual leaf‐fall also showed 2‐year cycle in the three oak species; peak years were synchronized between species and peak leaf‐fall alternated with acorn production in all three species. Furthermore, there was a significant negative correlation between acorn and leaf production in all three species. Data showing 2‐year cycles of acorn and leaf production and the negative correlation between them supports the hypothesis of resource switching between vegetative growth and reproduction. The 2‐year cycle might be the basic, intrinsic rhythm of resource allocation in biennial‐fruiting Cyclobalanopsis species.

Many fagaceous species mature acorns during the second autumn after flowering and are called "2-year species." In 2-year species of subgenus Cyclobalanopsis (genus Quercus), alternate bearing (2-year seed production cycle) is prominent, but this has not been reported for other 2-year species in the genera of Lithocarpus and Castanopsis. We tested the hypothesis that the difference in reproduction of 2-year species is linked to the weevil, Mechoris ursulus. After ovipositing on acorns, this insect cuts the shoots to which the acorns are attached. We examined the host preference of M. ursulus and reproductive traits of fagaceous 2-year species. Percentage infestation was remarkably high in the subgenus Cyclobalanopsis, whereasLithocarpus and Castanopsis species suffered almost no damage. Furthermore, unlike Lithocarpus and Castanopsis, most of the acorn-producing shoots in subgenus Cyclobalanopsis had no current-year shoots. The clear relationship suggests that alternate bearing in subgenus Cyclobalanopsis is an adaptation to the shootcutting behavior of M. ursulus. K E Y W O R D S2-year species, acorn production, alternate bearing, plant-predator interactions, reproductive ecology