| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
2Department of Forest Science, Faculty of Agriculture, Hokkaido University, Kita-ku, Sapporo 060–8589, Japan; and 3Tomakomai Research Station, Hokkaido University Forests, Tomakomai 053–0035, Japan
Received for publication September 2, 1997. Accepted for publication September 8, 1998.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
Key Words: bumble bee • display size • fruit set • geitonogamy • masting • size-structured population • spatial distribution • Styracaceae • Styrax obassia.
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
; Stephenson, 1981
; Haig and Westoby, 1988
; Ayre and Whelan, 1989
). In natural plant populations both of these factors could affect fruit and/or seed set to different degree in individuals differing in size or physiological condition, such as light condition (Lawrence, 1993
; Cunningham, 1996
). Recently the occurrence of pollen limitation has been recognized to be more common than previously thought (Burd, 1994
). Both quality of pollen (genetically compatible pollen) and quantity of pollen deposited on stigma may be affected by behavioral response of pollinators to floral abundance of individual plant and clump of plants (Barrett, Harder, and Cole, 1994
). Considering plants with a self-incompatible (SI) mating system, fruit set is crucially affected by pollen supply from other conspecific individuals. Thus, in a wild population of SI (self-incompatible) entomophilous-pollinated species, variations in local floral density, due to heterogeneous spatial distribution of individual plants, may affect the fruit set of individuals (House, 1992
). Variations of individual plant size, that is, variations of flowering display size of individual plants, are also thought to affect pollinator behavior through attraction effect and thus deposition of compatible pollen and geitonogamous pollen (Klinkhamer and de Jong, 1993
). Whether the pollinator visitation rate per individual flower increases or decreases as a function of floral display size is still controversial (Willson and Price, 1977
; Schaffer and Schaffer, 1979
; Geber, 1985
; Andersson, 1988
), but it may be reasonable to assume that a plant with a large number of flowers receives more geitonogamous pollination (de Jong, Waser, and Klinkhamer, 1993
). In an empirical study, de Jong et al. (1992)
suggested negative effects of geitonogamy on seed set in large plants for Ipomopsis aggregata, an SI species.
Because populations of tree species are typically size structured, the flowering display size of individual trees varies to a huge extent with plant size. Furthermore, the spatial distribution of individual trees of a species is complicated by the existence of populations of other species, disturbances, differences in topography, and so on. Particularly in subcanopy tree species, where even mature individuals may not reach the canopy layer, individuals may not receive sufficient light for the maturation of fruits, as described in shrub species of Lindera benzoin, in which fruit set is limited by light availability rather than pollen (Niesenbaum, 1993
). Also these light conditions seem to vary considerably among individuals due to vertical structure of the forest (Yoda, 1974
; Yoda, Nishioka, and Dhanmanonda, 1983
).
Thus, for subcanopy tree species in the forests that show a SI mating system, it is predicted that local flowering density will show a positive effect due to pollinator attraction, and floral abundance of individual trees will show a negative effect due to increased geitonogamy. Fruit set is predicted to be positively affected by the light resource availability because fruit set of trees with insufficient light condition would be resource limited. In this study we examine the hypothesis described above for S. obassia, a bumble-bee pollinated tree species, with reference to its mass-flowering pattern. This species is a subcanopy tree mixed in cool-temperate deciduous forests and shows an SI mating system, whose flowers from artificial selfed pollination show reduced fruit set due to fruit abortion (Tamura and Hiura, 1998
). It is also known to show considerable fluctuation of flowering intensity among years (Abe, 1995
).
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
). Each reproductive shoot bears one apical inflorescence, which is a raceme consisting of 1–33 white flowers. Flowering occurs from late June to early July after leaf elongation. The corolla is 2 cm long. The flowering duration of the inflorescence is 
7–10 d and that per flower is 4–6 d (Tamura and Hiura, 1998
). The most frequent insect visitors to S. obassia are Bombus spp. and other insect visitors belong to Syrphidae and Halictidae. Fruits are orbicular-ovate drupe with 2 cm in length and 1 cm in width, containing usually one, rarely two or three seeds. Pollinated ovaries continue to swell and the seed coat hardens in mid-August. The pericarp of mature fruit dehisces in mid-September. The fruits contain a strong defence chemical, saponin, and insect predators are rare. Mature seeds dropped from infructescences are dispersed by the wood mice Apodemus speciosus and A. argenteus. Seeds in the fruits that are on twigs are dispersed by bird, mostly Parus varius, which has a food-hoarding habit.
Study site
This study was conducted in 1995 and 1996 on a permanent 4-ha plot set in Tomakomai Experimental Forest (42°40'N, 141°36'E, 50–95 m elevation) in Hokkaido, northern Japan. Mean annual temperature is 6.4°C and annual precipitation is 1450 mm. From December to March, snow cover reaches a depth of 50 cm.
The study area was a mature deciduous forest stand, and the canopy height reached up to 25 m. Thirty-two tree species (dbh 
4 cm) have been recorded in the plot. Tree density and basal area of this stand were 965 stems/ha and 27.5 m2/ha, respectively. Quercus crispula, Tilia japonica, Cercidiphyllum japonicum, Acer mono, and Ostrya japonica dominate the canopy layer. During the flowering period of S. obassia, there were no flowering trees of other species in the studied plot.
Population structure and reproductive success of individual trees
We divided the 4-ha plot into 1600 subquadrats in 5 x 5 m and recorded species and dbh of all stems 4 cm in dbh in each subquadrat in 1995. In 1995 and 1996, the presence of inflorescences in each S. obassia tree was checked by observing with binoculars at the time of shoot elongation. Trees below 4 cm in dbh were also checked for the presence of inflorescences in order to confirm the size threshold for reproduction (Thomas, 1996
). Thereafter, the number of inflorescences per individual tree was counted by climbing the tree with a ladder or a rope with climbing ascenders for all reproductive individuals in this plot.
We selected 37 individual trees to provide differences in size and local density of S. obassia and counted the fruit set from one representative first-order branch of each tree. For these trees, we measured the trunk diameter just below the lowest living branch (DB) as an indicator of leaf mass based on pipe model theory (Shinozaki et al., 1964
; M. Suzuki and T. Hiura, unpublished data). We measured photon flux density (PFD) using quantum sensors (LI-190SA: LI-COR inc., Lincoln, Nebraska) at ten randomly selected points for each sample branch (3.21 ± 1.40 cm) on an overcast day in early August. At the same time, the PFD of an open area was also measured. Relative photon flux densities (RPFD) were then calculated and their geometric means used to represent the light condition for the branches. Although instantaneous measurements of relative light levels have often been considered inadequate, Parent and Messier (1996)
demonstrated that instantaneous measures of RPFD taken under overcast sky conditions give a meaningful measure of relative light availability under a forest canopy.
The flowering display size of individual trees was represented by the number of inflorescences per tree and the mean number of open flowers per inflorescence (mean inflorescence size). Since over 90% of flowers open simultaneously within an inflorescence on a peak flowering day (Tamura and Hiura, 1998
), it is reasonable to use mean inflorescence size as the floral display of inflorescence. For each of the 37 trees, the local density of inflorescence excluding each focal tree was calculated after estimating the clump size of flowering trees. Reproductive success of an individual tree was estimated in terms of fruit set (number of mature fruits per open flower) and the mean seed mass to examine the effects of floral abundance and light availability on success on a per flower basis. In addition, pollination and fertilization success were quantified as initial fruit set (fruit set 2 wk after flowering).
Flowering and fruiting
In June 1995, a mass-flowering year, 50 reproductive shoots from one first-order branch were marked for each of the 37 observed trees (a total of 1753 reproductive shoots). For all marked inflorescences, we counted the number of initial flowers, the number of open flowers every 4 d, the number of fruits at 2 wk after flowering (18–22 July), the number of fruits at 6 wk after flowering (16–20 August), and the number of fruits at 10 wk after flowering (17–21 September). We counted the open flowers of the inflorescence twice (4-d interval) for each observed tree during the flowering period. Because the flowers within and between inflorescences bloom simultaneously, it was easy to count all open flowers of inflorescence at one or two observations. The initial flower number is the sum of open flowers and scars of abscised flowers (which failed to open) in the flowering period. For individuals with < 50 inflorescences per first-order branch, two or three similar-sized first-order branches were used. During 17–21 September, all marked infructescences were harvested; the seeds were counted and the mass of seeds was determined after oven drying at 80°C for 48 h. Since 97.9% of the fruits contained only one seed (N = 2855), it is reasonable to use fruit set as a indicator of reproductive success.
Data analysis
All analysis was based on the data of S. obassia trees > 4 cm in dbh because there was only one reproductive tree < 4 cm in dbh (see Results). Spatial distribution of flowering trees was analyzed by using Morisita's index, I
(Morisita, 1959
). The quadrat size (S) with the highest value of I(S)/I(2S) is assigned to the size of the clump of flowering trees.
To examine factors affecting reproductive success of individual trees, multiple regression analysis was conducted on initial and mature fruit set. For each dependent variable, the following factors were considered as independent variables: DB, RPFD, inflorescence number, mean inflorescence size, and local inflorescence floral density. We used local inflorescence density excluding focal tree as local floral density in multiple regression analysis because inflorescence number of tree and local density of inflorescence within clump will necessarily tend to be correlated. To improve normality and reduce heteroscedasticity, analyses were performed on the following transformed data: arcsine transformation for fruit set and RPFD; log transformation for DB and inflorescence number; square-root transformation for local floral density. All analyses were carried out with SYSTAT. ver. 5.2 (SYSTAT, 1992
).
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
The size structure of the population and the reproductive status of individual trees are shown in Table 1. In 1995, all size classes contained reproductive individuals, and the ratios of nonreproductive individuals were 44.8, 10.4, and 2.3% for the 4–6, 6–8, and 8–10 cm dbh classes, respectively. All individuals 10 cm dbh flowered. In this study plot, there was only one reproductive S. obassia individual that had flowers below 4 cm in dbh (3.92 cm). The number of inflorescences per tree was significantly correlated with dbh for reproductive individuals in 1995, although there was some variation (ln inflorescence number = 2.824 ln dbh - 1.436; r = 0.564, P < 0.0001, N = 176). In contrast, in 1996, only three individuals flowered (dbh sizes were 6.1, 11.0, and 13.2 cm).
|
(S)/I(2S) for flowering trees was highest at a quadrat size of 1111 m2 (Fig. 2), which is approximately represented by a 35 x 35 m quadrat size. Therefore, we used 35 x 35 m as the denominator of the local floral density in the following analyses.
|
|
The relationship between inflorescence size and mean number of fruits per inflorescence is shown in Fig. 3. A quadratic regression showed a better fit than a linear regression for this relationship (initial fruit: y = - 0.060 + 0.414x - 0.008x2, r2 = 0.223, N = 1699; mature fruit: y = 0.178 + 0.197x - 0.004 x2, r2 = 0.122, N = 1699). Coefficients of x2 in the quadratic equations were significantly different from 0 (P < 0.0001). The rate of increase in number of fruits with inflorescence size was quite small, and the quadratic regression shows that the rate of increase in fruit number is a decelerating function of inflorescence size.
|
Reproductive success of individuals in 1995
Table 2 shows tree, flower, fruit, and seed parameters for the 37 sampled trees. Among the measured variables, inflorescence number was correlated with DB (Table 3).
|
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
The flowering display of individual trees of S. obassia consists of two parts. One is the inflorescence and the other is the group of inflorescences within the tree. Because studies on the effects of flowering display size have been concerned mostly with herbaceous plants, these studies mainly refer to the effect of the number of flowers on an inflorescence. In S. obassia, inflorescence size has a negative effect on fruit set because the number of fruits developed on an inflorescence did not increase linearly with inflorescence size. Initial fruit set, which might represent fertilization success with compatible pollen rather than resource limitation (Tamura and Hiura, 1998
), was negatively affected by inflorescence size although we did not conduct supplemental pollination experiment to confirm pollen limitation in this study. Thus geitonogamy within an inflorescence is suggested as one cause of this decelerating function of fruit development within an inflorescence. Furthermore, when considering the effects of geitonogamy, it is important to consider the existence of other inflorescences within the same individual. In this study, initial and mature fruit set ratios were negatively affected by inflorescence number per tree. By the negative effects of inflorescence size on the fruit set ratios, large floral display of individual tree did not contribute to attraction effect for the fruit set ratios. It is suggested that increased geitonogamy or reduced visitation rate to individual flowers might occur at individuals with a large floral display. Similarly, Andersson (1988)
found that proportion of flowers that set fruit decreased with inflorescence number over the whole size in Anchusa officinalis. He showed that pollinator visitation rate per flower did not increase linearly with plant size, and he also suggested the importance of geitonogamous pollination on large plants. However, we did not record the explicit behavior of the pollinators that might cause insufficient pollination of larger inflorescences, which is likely to occur as in the case of Ipomopsis aggregata (de Jong et al., 1992
).
In our study, some kinds of resource limitation on fruit set are also suggested. Light availability had no effect on fruit initiation, but was positively related to mature fruit set. This suggests that even pollinator visitation rate or deposition of compatible pollen affects fertilization success of individual flowers, and good light condition is needed for maturation of fruits. Another reason that suggests existence of resource limitation of fruit set for some tree comes from the effects of tree size (DB). Inflorescence number increased as DB in a mass-flowering year, but there is some variation in this relationship among individuals (r = 0.614, Table 3). Considering individuals bearing similar numbers of inflorescences but differing in DB, individuals with larger DB will have better fruit set based on the multiple regression. This phenomenon may be caused by resource limitation because available resource per inflorescence may increase with DB if inflorescence number is fixed. Tamura and Hiura (1998)
showed that neither pollen limitation nor current photosynthetic production of the reproductive shoot limited fruit set in S. obassia in their pollination and leaf removal experiment at the branch level. They conducted the experiment on some first-order blanches, not for the whole tree, and their experimental trees were grown under full sunlight in the arboretum, which suggests the importance of stored resources and/or reallocation of resources from other branches or from the trunk for maturation of fruit. Our results suggest that in a forest community, resource limitation occurs for individuals that are shaded by other tall trees. Therefore, in natural population of S. obassia, the fruit set ratio is affected by geitonogamy or pollinator visitation rate to individual flowers and also by the difference of resource level of individual trees in a focal year.
It is notable that an attraction effect at the level of a patch might exist. The local density of inflorescences affected both the initial and mature fruit sets, although significance levels were marginal. A few studies found an effect of patch size or population density on the flower visitation rate (Sih and Baltus, 1987
; House, 1992
). The visitation rates of bumble bees to individual flowers of catnip in a large patch were higher than they were in a small patch, and pollinator limitation on seed set actually decreased as patch size increased (Sih and Baltus, 1987
). This kind of pollinator behavior might have caused the patch size effect on fruit set found in S. obassia in this study.
As noted above, light resource levels also affected the abscission rate of initial fruits. However, light availability did not affect seed mass. Tamura and Hiura (1998)
concluded from a defoliation experiment at the branch level that the mean seed mass of S. obassia individual trees is determined by genetic effects or some other intrinsic trait of the trees, rather than resource limitation. Thus, resource might have a negligible effect on seed mass, however a more detailed study on seed mass on the individual shoot level is required to answer this question.
Styrax obassia showed a highly synchronous masting pattern among individual trees between years. In a mass-flowering year, it showed a relatively short flowering period and high synchrony among trees in flowering time. Its mass-flowering behavior may thus be advantageous for attracting pollinators to patches of trees. Even individuals of large floral display size would receive decreased fruit set. Krannitz and Maun (1991)
demonstrated the importance of grouping in pollination in one year of their study on the shrub Viburnum opulus. De Jong, Klinkhamer, and van Staalduinen (1992)
suggested that a self-compatible mating system and density-dependent pollination to individual plants are the key factors in explaining the evolution of mass blooming. However, it may be advantageous for even self-incompatible species to reproduce with mass flowering when flower production needs a large amount of storage matter and xenogamous pollination depends on floral display at the scale of tree patches. However, the effect of the spatial distribution of individual trees on pollination success and fruit set will need to be examined further in additional tree species. The challenge remains to test whether effects of spatial distribution on reproductive success are related to mass-flowering behavior.
|
FOOTNOTES |
|---|
4 Present address: Graduate School of Environmental Earth Science, Hokkaido University, Kita-ku, Sapporo 060–0810, Japan.
|
LITERATURE CITED |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
Andersson, S. 1988 Size-dependent pollination efficiency in Anchusa officinalis (Boraginaceae): causes and consequences. Oecologia 76: 125–130.[ISI]
Ayre, D. J., and R. J. Whelan. 1989 Factors controlling fruit set in hermaphroditic plants: studies with the Australian Proteaceae. Trends in Ecology and Evolution 4: 267–272.[CrossRef]
Barrett, S. C. H., L. D. Harder, and W. W. Cole. 1994 Effects of flower number and position on self-fertilization in experimental populations of Eichhornia paniculata (Pontederiaceae). Functional Ecology 8: 526–535.[CrossRef][ISI]
Bierzychudek, P. 1981 Pollinator limitation of plant reproductive effort. American Naturalist 117: 838–840.[CrossRef][ISI]
Burd, M. 1994 Bateman's principle and plant reproduction: the role of pollen limitation in fruit and seed set. Botanical Review 60: 83–139.
Cunningham, S. A. 1996 Pollen supply limits fruit initiation by a rain forest understory palm. Journal of Ecology 84: 185–194.[CrossRef]
de Jong, T. J., P. G. L. Klinkhamer, and M. J. van Staalduinen. 1992 The consequences of pollination biology for selection of mass or extended blooming. Functional Ecology 6: 606–615.[CrossRef][ISI]
———, N. M. Wasrer, and P. G. L. Klinkhamer. 1993 Geitonogamy; the neglected side of selfing. Trends in Ecology and Evolution 8: 321–325.
———, ———, M. V. Price, and R. M. Ring. 1992 Plant size, geitonogamy and seed set in Ipomopsis aggregata. Oecologia 89: 310–315.
Geber, M. A. 1985 The relationship of plant size to self-pollination in Mertensia ciliata. Ecology 66: 762–772.
Haig, D., and M. Westoby. 1988 On limits to seed production. American Naturalist 131: 757–759.[CrossRef][ISI]
House, S. M. 1992 Population density and fruit set in three dioecious tree species in Australian tropical rain forest. Journal of Ecology 80: 57–69.[CrossRef][ISI]
Klinkhamer, P. G. L., and T. J. de Jong. 1993 Attractiveness to pollinators: a plant's dilemma. Oikos 66: 180–184.[CrossRef][ISI]
Krannitz, P. G., and M. A. Maun. 1991 An experimental study of floral display size and reproductive success in Viburnum opulus: importance of grouping. Canadian Journal of Botany 69: 394–399.[CrossRef]
Lawrence, W. S. 1993 Resource and pollen limitation: plant size-dependent reproductive patterns in Physalis longifolia. American Naturalist 141: 296–313.
Morisita, M. 1959 Measuring of the dispersion of individuals and analysis of the distributional patterns. Memories of Faculty of Science, Kyushu University, Sereis E (Biology) 2: 215–235
Niesenbaum, R. A. 1993 Light or pollen—seasonal limitations on female reproductive success in the understory shrub Lindera benzoin. Journal of Ecology 81: 315–323.
Parent, S., and C. Messier. 1996 A simple and efficient method to estimate microsite light availability under a forest canopy. Canadian Journal of Forest Research 26: 151–154.[CrossRef]
Schaffer, W. M., and M. V. Schaffer. 1979 The adaptive significance of variations in reproductive habit in the Agaveaceae. II Pollinator foraging behavior and selection for increased reproductive expenditure. Ecology 60: 1051–1069.[CrossRef][ISI]
Shinozaki, K. Yoda, K. Hozumi, and T. Kira. 1964 A quantitative analysis of plant form—the pipe model theory. II Further evidence of the theory and its application in forest ecology. Japanese Journal of Ecology 14: 133–139.
Sih, A., and M.-S. Baltus. 1987 Patch size, pollinator behavior, and pollinator limitation in catnip. Ecology 68: 1679–1690.[CrossRef][ISI]
Stephenson, A. G. 1981 Flower and fruit abortion: proximate causes and ultimate functions. Annual Review of Ecology and Systematics 12: 253–279.
SYSTAT. 1992 Statistics, version 5.2 edition. SYSTAT, Evanston, Ill.
Tamura, S., and T. Hiura. 1998 Proximate factors affecting fruit set and seed mass of Styrax obassia in a masting year. Ecoscience 5: 100–107.
Thomas, S. C. 1996 Relative size at reproductive onset in rain forest trees: a comparative analysis of 37 Malaysian species. Oikos 76: 145–154.[CrossRef][ISI]
Willson, M. F., and P. W. Price. 1977 The evolution of inflorescence size in Asclepias (Asclepiadaceae). Evolution 31: 495–511.[CrossRef][ISI]
Yoda, K. 1974 Three-dimensional distribution of light intensity in a tropical rain forest of West Malaysia. Japanese Journal of Ecology 24: 247–254.
———, M. Nishioka, and P. Dhanmanonda. 1983 Vertical and horizontal distribution of relative illuminance in the dry and wet seasons in a tropical dry-evergreen forest in Sakaerat, NE Thailand. Japanese Journal of Ecology 33: 97–100.
This article has been cited by other articles:
|
|
 |
|
  Y. MIYAZAKI, T. HIURA, E. KATO, and R. FUNADA Allocation of Resources to Reproduction in Styrax obassia in a Masting Year Ann. Bot., June 1, 2002; 89(6): 767 - 772. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Abe Flowering phenology, display size, and fruit set in an understory dioecious shrub, Aucuba japonica (Cornaceae) Am. J. Botany, March 1, 2001; 88(3): 455 - 461. [Abstract] [Full Text]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
