Am. J. Bot.
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via ISI Web of Science (10)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Ushimaru, A.
Right arrow Articles by Kikuzawa, K.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Ushimaru, A.
Right arrow Articles by Kikuzawa, K.
Agricola
Right arrow Articles by Ushimaru, A.
Right arrow Articles by Kikuzawa, K.
(American Journal of Botany. 1999;86:436-446.)
© 1999 Botanical Society of America, Inc.

Variation of breeding system, floral rewards, and reproductive success in clonal Calystegia species (Convolvulaceae)1

Atushi Ushimaru2 and Kihachiro Kikuzawa

Center for Ecological Research, Kyoto University, Kyoto 606-8502, Japan

Received for publication July 18, 1997. Accepted for publication July 16, 1998.


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 LITERATURE CITED
 
The reproductive biology of four Japanese Calystegia species (Convolvulaceae) was studied to examine the effects of clonality and population structure on reproductive success. Calystegia soldanella, C. hederacea, and C. japonica are self-incompatible, while C. sepium is self-compatible but needs pollinator services for self-pollination. The showy, bisexual flowers of Calystegia offer pollen and nectar that attract many kinds of insects such as bees and syrphid flies. Clones of C. soldanella often formed mats just above the high tide line on beaches and produced a great number of seeds. Calystegia hederacea and C. japonica were distributed as patches of separate clones and often failed to transfer pollen grains among clones. Fruit and seed sets in C. hederacea and C. japonica were not limited by pollinators but by the number of compatible pollen grains. Although C. sepium clones were also distributed in patches, high and stable fecundity was achieved by self-compatibility and pollinator attraction. We suggest that self-compatibility in C. sepium has evolved under fertilization limitation caused by geitonogamy or facilitated selfing.

Key Words: Calystegia • clonal plant • Convolvulaceae • self-compatibility • fertilization limitation


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 LITERATURE CITED
 
Breeding systems in flowering plants are often associated with their life forms (Silander, 1985 ; Barrett, Harder, and Worley, 1996 ). Many annual plants are self-compatible (SC) and often exhibit autogamy or cleistogamy since they may thus maintain stable fecundity when pollinator availability is unpredictable (Baker, 1965 ; Lloyd, 1980 ; Cruzan and Barrett, 1993 ; but see Barrett, Harder and Worley, 1996 ). In many taxa, however, self-incompatible (SI) species are predominantly perennial (Stebbins, 1975 ). Most tree species exhibit outcrossing in tropical rain forests (Bawa, 1974 ; Chan, 1981 ; Bawa and Beach, 1983 ; Kress and Beach, 1994 ; Murawski, Dayanandan, and Bawa, 1994 ). Thus, outbreeding may be advantageous in perennial polycarpic plant species (Silander, 1985 ; Barrett and Eckert, 1990 ; Karoly, 1992 ; Barrett, Harder, and Worley, 1996 ) because they have extended opportunities to reproduce.

Most clonal plants are polycarpic perennials, with breeding systems believed to be the results of their clonal growth (Handel, 1985 ; Silander, 1985 ; Back, Kron, and Stewart, 1996 ). As clonal plants get larger, individual flowers become surrounded by other flowers of the same genet and geitonogamy results (Handel, 1985 ). Although in some clonal species, dichogamy (separation of male and female function chronologically; see Lloyd and Webb, 1986 ) and sequential flowering prevent geitonogamous pollination within a single inflorescence (Wyatt, 1982 ; Back, Kron, and Stewart, 1996 ), geitonogamy may be unavoidable in a large clone (Back, Kron, and Stewart, 1996 ). Geitonogamy occurs more frequently in clumped ("phalanx") species whose neighboring culms are often members of the same clone than in spreading ["guerilla": guerilla and phalanx are defined by Lovett Doust (1981) ] species whose nearest neighbors are often members of another clone (Silander, 1985 ). When the plant species are SI, geitonogamy decreases fitness (Handel, 1985 ; Richards, 1986 ; Klinkhamer and de Jong, 1993 ). Silander (1985) found that many "phalanx" species are SC or predominantly selfing, while most "guerilla" species are SI according to Stebbins' (1950) data. This suggests that clonal architecture is associated with mating system strategies in clonal plant species.

Even in "guerilla" species, however, the number of clones in a population or distance between nearest neighbors may affect fecundity. Rubus saxatilis, which is an insect-pollinated, SI, clonal shrub, tends to have low fruit set in low-density populations (Eriksson and Bremer, 1993 ). Genet size has little effect on fruit set in Rubus saxatilis. Thus, not only clonal architecture but also population structure may have a great effect on reproductive success in clonal plants.

Here we investigate the reproductive ecologies of four Calystegia species to examine the effects of clonality and population structure on mating success. In four Japanese Calystegia species, all of which are insect-pollinated, clonal, herbaceous vines, fruit set in natural populations varies (Kitamura, Murata, and Hori, 1972 ). Calystegia hederacea and C. japonica rarely produce fruits in natural populations, while C. soldanella and C. sepium set fruits. Clones of C. soldanella dominate just above the high tide line on coastal beaches and dunes, where other plant species cannot exist because of frequent disturbance and high salinity (Ishikawa, Furukawa, and Oikawa, 1995 ). The other three Calystegia species are often found as patches of separate clones spreading vegetatively by slender rhizomes and climbing on other plants. This difference in population structure may lead to variation in reproductive output among species.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 LITERATURE CITED
 
Study species
We studied four perennial, clonal vine species of Calystegia (Convolvulaceae), C. soldanella, C. hederacea, C. japonica, and C. sepium (Ohwi, 1965 ). All members of the genus Calystegia in Japan have an extensive capacity for rhizomatous growth (>1 m per year; A. Ushimaru, personal observation). Their elongated rhizomes have many shoot buds and one clone often has many aerial shoots; thus the growth pattern is intermediate between the guerilla and phalanx strategies. All species produce rosy or white, bisexual, funnel-shaped, herkogamous (herkogamy: spatial separation of stigma and stamen in a flower; see Webb and Lloyd, 1986 ) flowers in leaf axils that are usually pollinated by insects. Flowers of all Calystegia open at sunrise and last for a day. In most cases, single flowers open sequentially from the bottom to the top of their shoots (Kitamura, Murata, and Hori, 1972 ) so that geitonogamy within a single shoot does not occur. Each flower has an incompletely two-locular ovary with two ovules in each locule, a filiform style, and five anthers. The seeds are dispersed by water in C. soldanella and by gravity in the other three species (Okamoto, 1992 ; A. Ushimaru, personal observation). In Japan, Calystegia seeds are rarely eaten by animals but are sometimes damaged by fungi (A. Ushimaru, personal observation). However, seeds of one species, C. japonica, were strongly preyed on by weevils at Tajimagahara Nature Reserve (Ishikawa, Toquenaga, and Matubayashi, 1994 ). Several flower characteristics, such as petal size and color, style length, filament color, and nectary color, often vary among clones. Leaf morphologies sometimes vary among clones. Flower type and leaf type were constant at each site in every year, and these characteristics were the basis for clone identification. A brief description of the four species follows.

Calystegia soldanella (L.) Roem. et Schult., a creeping herb, often colonizes coastal habitats throughout temperate and subtropical Eurasia and the Pacific. In Japan, it commonly occurs on the seashore of Hokkaido, Honshu, Sikoku, and Kyushu and the beaches of Lake Biwa.

Calystegia hederacea Wall., a herbaceous vine, is found throughout temperate southeast Asia and Japan. In Japan, it is a common weed in cultivated fields and in sunny disturbed sites.

Calystegia japonica Choisy., a climber, is distributed in temperate parts of Japan, Korea, and China. This species is sometimes classified as C. sepium var. japonica (Choisy) Makino (Makino, 1895 ; Iwatuki et al., 1993 ). In Japan, it occurs on sunny and disturbed riversides and roadsides. Seed production is infrequent in both Japan and Korea (Kitamura, Murata and Hori, 1972 ; Kim and Chung, 1995 ).

Calystegia sepium (L.) R. Br. (bindweed), a climber, occurs over temperate regions of the Northern Hemisphere. In Japan it occurs along roadsides and on banks of mountain rivers that are often disturbed by human activity or flooding.

Study sites
The locations of study sites for each species are shown in Table 1. We studied C. soldanella at three sites around Lake Biwa, Shiga, and at Hazaki Beach, Ibaragi. We established 2–6 1-m2 plots in each site. Except for YO1, we observed several clones at each site.


View this table:
[in this window]
[in a new window]
  		Table 1. Summary of study sites and investigations (except breeding system tests and germination test for the four Calystegia species. The symbols a, b, c, and d indicate that the investigation was carried out in 1993, 1994, 1996, and 1997, respectively. Plot, PV, Nec., and pollen refer to number of plots in each site, pollinator visit observation, nectar measurement, and pollen counting, respectively. Seedlings were carefully searched for at each site.

 
We studied four patches of C. hederacea in the north campus of Kyoto University, Kyoto. Two 1-m2 plots were placed in the Botanical Garden of the Faculty of Science (BG-A, B; near BG-B there are flowering clones of Calystegia planted in pots), a 1-m2 quadrat in front of the Research Institute for Fundamental Physics (FP), a 2-m2 plot around the Phytotron (PH), and a 4-m2 plot in the Experimental Garden of the Faculty of Agriculture (EG). In addition to these sites, three other patches were observed at Kibune, Kyoto, at Tukuba, Ibaragi, and at Misato, Gunma. With the exception of sites EG and KB3, we observed only one clone per site. Two clones were observed at EG, and we named the clone that has larger flowers EG-L and the other EG-S, while at KB3 there was a patch of C. sepium near the patch of C. hederacea.

Patches of C. japonica were studied in riparian areas of the Uji River at Yodo (YD) and the Katura River at Matuo (KA), Kyoto. Tajimagahara nature reserve (TA), where C. japonica produces many fruits, is located in the floodplain of Arakawa River at Tajimagahara, Saitama. This species is often distributed as patches of separate clones, but the size of each clone often exceeds 9 m2. There were three patches of C. japonica at YD, separated by ~100 m. We established one 2-m2 plot in each patch. At both KA and TA, C. japonica dominated relatively large area (>20 m2 in KA and >500 m2 in TA). Therefore, we established four 2-m2 plots at KA and seven 2-m2 plots at TA within these patches. At TA several clones with distinct flowers were observed.

We studied four sites (1, 2, 3, and 4) where C. sepium had colonized the roadside at Kibune (KB), Kyoto. Sites 1, 2, 3, and 4 included one, more than three, one and two patches (1, 2-A, 2-B, 2-C, 2-D, 3, 4-A and 4-B), respectively. KB2-B was next to KB2-A. We placed one plot per patch.

Pollination experiments
At Lake Biwa and Hazaki, artificial self-, outcross-, and open-pollination experiments were conducted on C. soldanella. The breeding systems of the other three species were determined in a cultivated field inside the Botanical Garden, Kyoto University. Their seeds were collected from EG, BG, KA, TA, KB1, and KB2 in 1993. In 1994, they were planted in pots and grown in order to examine breeding systems. For crossing experiments, flower buds were bagged on the afternoon before opening (Table 2). The pollination treatment was repeated the next morning and then flowers were bagged again. We used flowers from the same clone to test for geitonogamy. Bagged flowers without artificial pollination were used to assess autogamy. We removed stamens from bagged flowers to examine apomixis. Open-pollination assessed the effectiveness of pollinators. Cross-pollinations between congeneric species (C. hederacea, C. japonica, and C. sepium) were carried out in order to examine hybridization.


View this table:
[in this window]
[in a new window]
  		Table 2. Summary of experimental treatments performed to determine breeding system.

 
Germination of seeds
We compared germination rates of selfed and outcrossed seeds for C. sepium to assess inbreeding depression at germination. Twenty-one selfed and 20 outcrossed seeds were collected from several plants and were planted in pots filled with vermiculite and watered once a week. Germination was observed at 2 wk after planting.

Pollen tube growth
Selfed and crossed flowers of C. hederacea, C. japonica, and C. sepium were collected from the cultivated field 8 h after artificial pollination, pistils were stained with cotton blue, and pollen germination was assessed under a microscope.

Flowering and fruiting phenology
In several plots, we counted and marked all flowers at anthesis in each 5-d period and noted capsule production after 45 d for each species (Table 1).

Fruit and seed set under natural conditions
All flowers were marked within each plot and the numbers of flowers and capsules were counted for four species. At TA, a subsample of flowers was randomly chosen and monitored. We recorded fruit set for marked flowers, collected all capsules in each plot 45 d after anthesis, and counted the number of seeds per fruit. Seed set was calculated as total number of seeds per total number of ovules. Total number of ovules was estimated by multiplying the total number of flowers by 4.

Nectar
The standing volume of nectar was measured with microcapillary tubes, and sugar content (mass/mass) was estimated with a portable refractometer. This measurement was made for about ten flowers for 1 d at YO for C. soldanella and at KU for C. hederacea in 1997 and at KA for C. japonica and at KB for C. sepium in 1996. Flowers were marked and bagged before anthesis, and nectar was then collected at 1700. The average sugar production per flower was calculated.

Pollinator visitation
Pollinators were collected from all sites in 1993 and 1994. We observed flowers of the four Calystegia species for insect foragers at MA-A, IM-A, BG-A, KA-A, TA-C, and KB2-A. This observation was carried out for 1 d at each plot in 1994. We randomly chose and marked about ten flowers in each plot to count insect visitation to each flower for 15 min per hour throughout the day. Insects that carried pollen and made contact with anthers and stigmas were easily recognized by naked eye and recorded. The mean number of pollinator visits per flower each day was calculated as a multiple of the average number of pollinator visits per flower during observations.

Pollen
The mean number of pollen grains per ovule was calculated. About ten flowers were collected from IM, KU, KA, and KB. Five anthers in each flower were stored in 1.5-mL 70% ethanol. Pollen grains easily separated from anthers in ethanol, after which the anthers were removed and 3.5 mL water was added to each vial. We estimated the numbers of pollen grains per flower (in a solution) using a hematocytometer. In most cases Calystegia species have four ovules, therefore, the P : O (pollen : ovule) ratio was defined as the average number of pollen grains per flower divided by 4. In addition, we collected open-pollinated flowers at 1700 and the number of pollen grains attached on the stigma was counted under a microscope to examine pollen delivery.

Seedlings
Each site was carefully searched for seedlings in 1993 and 1994 to study seedling recruitment for every species.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 LITERATURE CITED
 
Pollination experiments
Results of pollination experiments are summarized in Table 3. In C. soldanella, C. hederacea, and C. japonica, few fruits and seeds were set following self-pollination, suggesting C. soldanella, C. hederacea and C. japonica are largely self-incompatible. In C. sepium no significant differences in fruit and seed production were recorded between self-pollinated and outcrossed flowers. Also, germination rates of both outbred and inbred seeds were almost the same (0.95 and 1.00, respectively) in C. sepium. The pollen grains of C. sepium were rarely deposited on stigmas in bagged flowers (eight flowers were examined). Further, few fruits and seeds were set in autogamy treatment. Thus C. sepium is self-compatible and not autogamous.

Fruit and seed set in open-pollinated flowers of C. soldanella were significantly lower than in outcrossed ones, while the other three species had similar reproductive output in the two treatments.

Hybrids (F1) were obtained from crossing among three of the Calystegia species. Most F1 seeds germinated and grew, and subsequently they flowered and set seeds.

Pollen tube growth
No pollen tube growth was observed in stigmas of self-pollinated flowers of either C. hederacea or C. japonica (N = 14 and 13, respectively), whereas tube growth was observed in 100 and 81.8% of cross-pollinated stigmas of these species (N = 9 and 11, respectively). Both self- and outcross-pollen tubes grew in C. sepium (N = 2 and 2, respectively).

Flowering and fruiting phenology
Calystegia soldanella produced fruits throughout the entire period of anthesis except at YO1-C (Fig. 1). The flowering periods of C. soldanella were shorter than in other Calystegia species.



View larger version (45K):
[in this window]
[in a new window]
  		Fig. 1. Flowering and fruiting phenology of Calystegia species. Lines show the number of flowers and bars the number of fruits developed. Date of first flowering is given beneath the sites monitored in a particular year. Letters above each figure show the sites and clones where observations were carried out (cf. Table 1 ).

 
The C. hederacea clone set few fruits (Fig. 1). Two clones at the Kyoto University Experimental Garden (EG-S and EG-L) set fruits only when their flowering overlapped (Fig. 1).

Flowering phenology of C. japonica varied between years and/or sites (Fig. 1). Peaks of different plots were not synchronized with each other at KA. Small numbers of fruits were produced during the flowering season (Fig. 1).

Although the blooming phenology of C. sepium was not synchronized among plots, fertilizations occurred throughout the flowering period in each plot (Fig. 1).

Fruit and seed set under natural conditions
Fruit set in C. soldanella varied from 0.0 to 72.1% under natural conditions (Table 4). Outcrossed flowers had significantly higher fruit and seed set than selfed flowers (Tables 3, 4). At YO1, fruit set in this species was extremely low (almost 0%), while at YO2 near to YO1, ~20–50% of flowers set fruits and seeds.


View this table:
[in this window]
[in a new window]
  		Table 4. The range and median of fruit and seed sets under natural conditions for the four Calystegia species. The number of samples and the values of fruit and seed set refer to Appendix 1.

 

View this table:
[in this window]
[in a new window]
  		Table 3. Results of pollination treatments in four Calystegia species. Percentage fruit and seed set and average number of seed per fruit (±SE) were recorded after pollination treatments within and between species. The number of fruits is shown in parentheses for N < 4. — indicates that no data were recorded. A species name indicates an interspecific cross. Chi-square tests were performed between treatments. * indicates fruit and seed sets for its treatment are significantly (P < 0.01) lower than that for outcrossed treatment. ** indicates that fruit and seed sets for autogamy or apogamy treatments are significantly (P < 0.01) lower than that for self-pollination treatment.

 
Fruit and seed set were low in C. hederacea in natural conditions (Table 4). The clones with highest fruit production were adjacent to a flowering conspecific or congeneric clone (Appendix 1). Further, in the cultivated field where many clones were planted, fruit and seed set for open pollinated flowers were much higher than fruit and seed set in natural conditions (Tables 3, 4).

Fruit and seed set of C. japonica were 0.0–56.7 and 0.0–13.0%, respectively. Open-pollinated flowers in the cultivated field produced 65% more fruit and 50% more seed than under natural conditions (Tables 3, 4). Calystegia japonica produced few fruits and seeds at YD and KA (Appendix 1). On the other hand, this species achieved relatively high fruit set (15.7–56.7%) and seed set (2.4–27.0%) at TA, and these values were significantly higher than those at the other two sites (Mann-Whitney U = 0, 4, P = 0.0017 and 0.0087, respectively).

Flowers of C. sepium produced more fruits and seeds than those of the other three species (Table 4). The median fruit (67%) and seed set (46.3%) were close to those in the cultivated field (Tables 3, 4). The variation of fruit set among plots was relatively small. In isolated patches such as KB1 and KB3, C. sepium plants set as many fruits as those in patches that had a neighbor patch.

Pairwise species comparisons revealed that fruit and seed set in C. sepium were significantly higher than in all other species (Table 5), while C. hederacea set fewer fruits and seeds than the others. As for seed set, only between C. soldanella and C. japonica was there no significant difference (Table 5).


View this table:
[in this window]
[in a new window]
  		Table 5. Results of Mann-Whitney U tests for differences in fruit and seed sets of pairs of Calystegia species under natural pollination conditions. Results for fruit and seed sets are shown in the upper right cells and the lower left ones, respectively.

 
Flower density (NFL/m2) is one of the indicators of clone size (because larger clones have more aerial shoots that produce more flowers per area) for the four species. Kendall rank correlation coefficient tests were performed for relationships between the number of flowers per plots and fruit and seed set for four Calystegia species. Fruit and seed set in C. hederacea were negatively correlated with flower density ({tau} = -0.473, P = 0.043 and {tau} = -0.527, P = 0.024, respectively). However, as mentioned above, small clones (EG-L, BG-B, and KB-C) had a neighbor conspecific or congeneric clone. Fruit and seed sets in the other three species were not correlated with the flower density.

Nectar
The total nectar per flower, sugar concentration, and mean sugar production per flower differed significantly among the four species (ANOVA and Fisher's PSLD procedure were applied; F = 28.7, df = 3, P = 0.0001 for nectar volume; F = 4.4, df = 3, P = 0.0088 for sugar concentration; F = 29.7, df = 3, P = 0.0001 for sugar production) (Table 6). Calystegia japonica and C. soldanella secrete ~6 times as much nectar as C. hederacea, and about 2 times that of C. sepium. The sugar concentration of C. sepium is significantly higher than that of the other species. The rankings of sugar production (sugar mass per flower) were C. japonica > C. soldanella > C. sepium > C. hederacea.


View this table:
[in this window]
[in a new window]
  		Table 6. Results of pollen count and nectar measurement. The number of pollen grains in anthers and on stigmas after open-pollination was counted. The minimum and maximum values and the number of samples are shown in parentheses. The P:O (pollen:ovule) ratio and the percentage of pollen grains deposited on stigmas wre calculated for each species. Sugar production per flower for the four species were measured. Size of corolla is described by Ohwi (1965) or Kitamura, Murata, and Hori (1972).

 
Pollinator visitation
The primary pollinators of Calystegia species are several kinds of bees, which comprised 56.7% of the total visitor species. Sixteen out of 33 bee species were caught on flowers of more than one Calystegia species. Thus the pollinator systems of the four species are similar to each other. Calystegia soldanella flowers were most frequently visited by honey bees at MA and IM, whereas smaller halictid and megachilid bees frequently visited flowers of C. japonica and C. hederacea. Several types of pollinators were observed visiting C. sepium flowers: bumble bees, smaller bees, and syrphid flies (Table 6).

ANOVA and Fisher's PSLD procedure were performed on the number of pollinator visits during observation (NPO) in the different Calystegia species. Significant differences in NPO were only found between C. hederacea (BG) and C. japonica (TA) and between C. sepium (KB-B) and C. japonica (TA) (F = 4.93, df = 5, P = 0.0007). There was no relationship between NPO and fruit set ({tau} = -0.154, P = 0.303), when the Kendall rank correlation coefficient test was applied to the fruit set and NPO data for four species at these sites where pollinator visits were observed.

Pollen
The mean P:O (pollen : ovule) ratios for each Calystegia species are 2264–3759 (Table 7). ANOVA and Fisher's PSLD procedure were applied for the comparison of P:O ratios among species. Calystegia soldanella and C. japonica had significantly larger P:O ratios than C. hederacea and C. sepium. The number of pollen grains per flower in C. hederacea was slightly larger than that in C. sepium.


View this table:
[in this window]
[in a new window]
  		Table 7. Numbers of pollinator visits during observation of Calystegia species. Numbers of flowers observed are shown in parentheses under the site name. The mean number of pollinators per flower during observation (±SE) and the mean number of pollinators per flower per day were calculated.

 
The number of pollen grains on stigmas did not differ among the four species (Table 7). However, pollen grains of other species, such as Pharbitis nil (Convolvulaceae), were observed deposited on the stigmas of both C. hederacea and C. japonica. The ratio of average numbers of pollen grains received on stigmas to the average numbers of pollen grains produced per flower was ~1% for each species.

Seedlings
No seedlings of C. soldanella were found at any site around Lake Biwa. At Hazaki Beach, many seedlings of C. soldanella were observed on 27 May 1994. Seedlings often showed a clumped distribution. A few seedlings survived until 20 October (S. Ishikawa, personal communication, Tukuba University). Seedlings of the three climbing species were never encountered at any site.


   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
LITERATURE CITED
 
Variation of breeding systems and floral rewards
Breeding systems varied within Calystegia. All four species were herkogamous and needed pollinator visits for pollination, though only C. sepium was self-compatible. Self-incompatibility has been reported in other genera (Ipomoea and Merremia) of Convolvulaceae (Kowyama, Shimano, and Kawase, 1980 ; Devall and Thien, 1989 , 1993 ; Willmott and Burquez, 1996 ). The sporophytic SI system has been well studied in Ipomoea (Kowyama, Shimano, and Kawase, 1980 ; Devall and Thien, 1993 ). In C. hederacea and C. japonica, growth of self-pollen tubes was blocked on the surface of dry stigmas, suggesting that SI in Calystegia is sporophytic.

All four Calystegia species had relatively high P:O ratios, higher than the mean ratio for facultatively xenogamous plants (796.6) and lower than the mean ratio for xenogamous species (5859.2) as reported by Cruden (1977) . In addition, these ratios for Calystegia species are 20 times as large as those for species living in highly disturbed areas (Cruden, 1977 ). In Calystegia, only ~1% of pollen grains was carried to stigmas, indicating that most pollen grains were either used as rewards for insects or deposited elsewhere. In most cases, pollen grains were almost entirely removed from anthers by noon (A. Ushimaru, personal observation). The P:O ratio of C. sepium was lower than those of the three SI species, suggesting that C. sepium reproduces through selfing as well as outcrossing. The fact that no inbreeding depression was expressed at seed production and germination also is consistent with selfing in C. sepium.

Interspecific differences in nectar production were also found in Calystegia species. The ranking of nectar production in the four species was almost the same as the ranking of corolla size (C. japonica = C. sepium > C. soldanella > C. hederacea) (Table 7). Nectar production is likely to be affected by the size of the nectary, which in turn is related to corolla size. However, C. sepium with a large corolla secreted relatively small amounts of nectar. It is likely that C. sepium evolved reduced nectar production in association with selfing. Floral rewards in mixed-mating species are generally smaller than those in related outcrossing species (Banks, 1980 ; Spira, 1980 ; Haber and Frankie, 1982 ; Rathcke, 1988 ). However, SI C. hederacea produced even less nectar than C. sepium, which we cannot explain.

Variation of pollination and reproductive success under natural conditions
Fruit and seed sets varied enormously among species. High, stable fruit and seed sets throughout the entire flowering season were recorded only in C. sepium, while fruit and seed sets in SI species varied depending on site conditions.

Pollination experiments show that all Calystegia species were pollen limited under natural conditions. Here, two different types of pollen limitation should be recognized: one is pollinator limitation (lack of pollinator services) and the other is fertilization limitation (lack of compatible, intraspecific pollen on stigmas) (Garwood and Horvitz, 1985 ; Byers, 1995 ). Flowers of all four Calystegia species received many insect visits, and stigmas of each species were equally pollinated by them. The more limited reproduction of SI Calystegia species compared to C. sepium suggests that fertilization limitation occurs more frequently in SI plants than in SC plants and causes variation in reproductive success among the four Calystegia species.

Effects of clonal and population structure on reproductive successes
Fertilization limitation can be due to several factors: autogamy, facilitated selfing, geitonogamy, mating among incompatible phenotypes.

Although in Calystegia herkogamy and sequential flowering in a single shoot prevent autogamy and geitonogamy within a shoot, respectively, facilitated selfing and geitonogamy among shoots of one clone are inevitable in large clones. However, fruit and seed sets in natural Calystegia populations were not correlated with the flower density, indicating that the clone size did not affect reproductive success. In Calystegia species, slender rhizomes intermingle with each other under the ground when two or more clones coexist. Thus, clone size may not affect outcross pollination as in other "guerilla" species.

On the other hand, existence of neighboring clones had great effects on fruit and seed sets in SI Calystegia species. The lack of fruit set in C. soldanella at site YO1 was perhaps due to there being a single genet at this site. At YO2, only 50 m from YO1, fruit and seed set were relatively high. Although insect species collected from YO1 were not different from other sites, pollinators did not seem to move between sites. Similarly, clones of C. hederacea and C. japonica that have neighboring conspecific or congeneric clones set more fruits and seeds than isolated clones. At the EG plot two clones produced seeds only when their flowering was synchronized. These findings suggest that seed output in these species is limited by the lack of genetically distinct pollen grains. This is also supported by the fact that open-pollinated flowers of C. hederacea and C. japonica in our cultivated field, where many clones simultaneously flowered, produced as many fruits and seeds as outcrossed flowers. If the genets are distant from each other, outcrossing occurs infrequently among them (Richards, 1986 ; Eriksson and Bremer, 1993 ; Momose, Nagamitsu, and Inoue, in press ). In C. sepium, however, no effects of clone distribution were found, presumably due to its self-fertility.

Seedling recruitment is rare in most clonal plants (Harper, 1977 ; Eriksson, 1992 ; Eriksson and Frøborg, 1996). Seedlings of Calystegia species were not found except for a population of C. soldanella at HA, suggesting that seedling recruitment in C. hederacea, C. japonica, and C. sepium occurred only when the population was established. In contrast, in C. soldanella there was repeated seedling recruitment (see Eriksson, 1989). Repeated recruitment in C. soldanella may promote local genet diversity and outcrossing. On the other hand, rare seedling establishment and intensive clonal growth cause patches of the other three Calystegia species to be composed of a single genet, as in many clonal plant species (Oinonen, 1967 ; Anderson and Beare, 1983 ; Worthen and Stiles, 1986; Murawski and Hamrick, 1990 ), causing geitonogamy and little outcrossing.

Evolution of self-compatibility in Calystegia sepium
All Calystegia species except C. sepium are SI in Japan, though C. sepium in Europe is SI (A. Ushimaru, personal observation; S. Sakamoto, personal communication, Ryukoku University). Therefore it is possible that in Japan, which is at the margin of the distribution of C. sepium, this species has evolved SC from SI without morphological change from herkogamy to autogamy.

Pollinator-limited seed production is often presumed to favor the evolution of SC from SI ancestors (Stebbins, 1957; Levin, 1971 ; Wyatt, 1983, 1984 ; Dafni and Bernhardt, 1990 ; Weisler and Snow, 1992 ). However, the evolution of SC from SI under pollinator limitation usually occurs together with the change from herkogamy to autogamy (Darwin, 1877 ; Hogan, 1983; Wyatt, 1984 ; Inoue, 1988 , 1990 ; Dafni and Bernhardt, 1990 ). Calystegia sepium was visited by many kinds of insects and needs their visits even for selfing, showing that pollinator limitation is not the selective pressure for the evolution of SC.

Calystegia sepium often occurs in isolated patches, as in C. hederacea and C. japonica. The lack of cross-pollination seems to be the selective pressure favoring the evolution of SC in C. sepium. Theoretical work suggests that SC without autogamy can evolve from SI under conditions of fertilization limitation (Ushimaru, Higashi and Kikuzawa, unpublished data). Frequent selfing due to geitonogamy occurs in Iris versicolor, Microtis parviflora, and Carex platyphylla, all of which are SC, nonautogamous (dichogamous and/or herkogamous or monoecious), and clonal (Handel, 1985 ; Peakall and Beattie, 1989 , 1991 ; Back, Kron, and Stewart, 1996 ), suggesting that SC without autogamy has evolved and is maintained to reduce the negative effects of geitonogamy in some clonal plant species.

Another possible explanation for the evolution of SC in C. sepium is that the loss of genetic diversity in Japan has led to a mixed-mating strategy. Biparental inbreeding due to low genetic diversity has been known to induce the evolution of selfing (Lloyd, 1979 ; Uyenoyama, 1986 ). This idea can be examined by comparing the mating systems and genetic diversity of C. sepium populations in Japan and in Europe.

From our data, it cannot be explained why only C. sepium has evolved SC but C. hederacea and C. japonica, which often occur as isolated clones, have not. Growth rates of rhizomes and climbing shoots in C. japonica and C. hederacea are relatively higher than in C. sepium (A. Ushimaru, personal observation). This high clonal growth rate may complement infrequent seedling regeneration in C. japonica and C. hederacea. This is one possible explanation and should be examined in future research.

Our results suggest that a lack of genetically distinct pollen grains caused by the isolation of clones reduces fruit and seed production in three SI Calystegia species. Their colonizing habits, rare seeding establishment, and vigorous clonal growth may often decrease local reproductive success. Self-compatibility in C. sepium avoids fertilization limitation due to self-pollination by insects and ensures high and stable fecundity.


View this table:
[in this window]
[in a new window]
  		Appendix 1. Fruit set, seed set, and average number of sound seeds (±SE) in the field for each clone of Calystegia in 1993 and 1994. Lettera indicates that data was recorded in 1993 and no letter in 1994. — shows no data. Number of flowers is shown in parenthesis at column of fruit set. Number of fruit is shown in parenthesis if N < 4. rs means that flowers were randomly sampled in a plot.

 

View this table:
[in this window]
[in a new window]
  		Appendix 2. Flower visitors collected on Calystegia flowers. Species marked with * were caught from flowers of more than two Calystegia species.

 

   FOOTNOTES
 
1 The authors thank Hideo Tabata, David Ackerly, David W. Roubik, and Takakazu Yumoto for reviewing the manuscript; Kiyoshi Matsui for advice on statistics, Tamiji Inoue for useful comments for this study; Peter E. Scott for many valuable comments on our manuscript; Sadao Sakamoto, Shin-ichi Ishikawa, and Yukihiko Toquenaga for allowing us to use their data on Calystegia species; Iwao Kojima for cultivation of Calystegia species; and Teruyoshi Nagamitsu for identification of insects. This research was financially supported by grants from the Center for Ecological Research, Kyoto University. Back

2 Author for correspondence. Back


   LITERATURE CITED
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 LITERATURE CITED
 
Anderson, R. C., and M. N. Beare. 1983 Breeding system and pollination ecology of Trientalis borealis (Primulaceae). American Journal of Botany 72: 152–158.[CrossRef]

Back, A. J., P. Kron, and S. C. Stewart. 1996 Phenological regulation of opportunities for within-inflorescence geitonogamy in the clonal species, Iris versicolor (Iridaceae). American Journal of Botany 83: 1033–1040.[CrossRef][ISI]

Baker, H. G. 1965 Characteristics and modes of origin of weeds. In H. G. Baker and G. L. Stebbins [eds.], The genetics of colonizing species, 147–172. Academic Press, New York, NY.

Banks, J. A. 1980 The reproductive biology of Erythronium propullans Gray and sympatric populations of E. albidum Nutt. (Liliaceae). Bulletin of the Torrey Botanical Club 107: 181–188.[CrossRef][ISI]

Barrett, S. C. H., and C. G. Eckert. 1990 Variation and evolution of mating systems in seed plants. In S. Kawano [ed.], Biological approaches and evolutionary trends in plants, 229–254. Academic Press, New York, NY.

———, L. D. Harder, and A. G. Worley. 1996 The comparative biology of pollination and mating in flowering plants. Philosophical Transaction of Royal Society, London 351: 1271–1280.[CrossRef]

Bawa, K. S. 1974 Breeding systems of tree species of a lowland tropical community. Evolution 28: 85–92.

———, and J. H. Beach. 1983 Evolution of sexual systems in flowering plants. Annuals of the Missouri Botanical Garden 68: 254–274.

Byers, D. L. 1995 Pollen quantity and quality as explanations for low seed set in small populations exemplified by Eupatorium (Asteraceae). American Journal of Botany 82: 1000–1006.[CrossRef][ISI]

Chan, H. T. 1981 Reproductive biology of some Malaysian Dipterocarps III. Breeding systems. Malaysian Forester 44: 28–36.

Cruden, R. W. 1977 Pollen-ovule ratio: a conservative indicator of breeding system in flowering plants. Evolution 31: 32–46.

Cruzan, M., B., and S. C. H. Barrett. 1993 Contribution of cryptic incompatibility to the mating system of Eichhornia paniculata (Pontederiaceae). Evolution 47: 925–934.[CrossRef][ISI]

Dafni, A., and P. Bernhardt. 1990 Pollination of terrestrial orchids of southern Australia and the Mediterranean region: systematic, ecological, and evolutionary implications. Evolutionary Biology 24: 193–252.[ISI]

Darwin, C. 1877 The various contrivances by which orchids are fertilized, 2d ed. John Murray, London.

Devall, M. S., and L. B. Thien. 1989 Factors influencing the reproductive success of Ipomoea pes-caprae (Convolvulaceae) around the Gulf of Mexico. American Journal of Botany 76: 1821–1831.[CrossRef][ISI]

———, and ———. 1993 Self-incompatibility in Ipomoea pes-caprae (Convolvulaceae). American Midland Naturalist 128: 22–29.[CrossRef][ISI]

Eriksson, O. 1992 Evolution of seed dispersal and recruitment in clonal plants. Oikos 63: 439–448.[CrossRef][ISI]

———. 1989 Seedling dynamics and life histories in clonal plants. Oikos 55: 231–238.[CrossRef][ISI]

———, and B. Bremer. 1993 Genet dynamics of the clonal plant Rubus saxatilis. Journal of Ecology 81: 533–542.

———, and H. Frøborg. 1996 "Windows of opportunity" for recruitment in long-lived clonal plants: experimental studies of seedling establishment in Vaccinium shrubs. Canadian Journal of Botany 74: 1369–1374.

Garwood, N. C., and C. C. Horvitz. 1985 Factors limiting and seed production of a temperate shrub, Staphylea trifolia L. (Styphyleaceae). American Journal of Botany 72: 453–466.[CrossRef][ISI]

Haber, W. A., and G. W. Frankie. 1982 Pollination of Luehea (Tiliaceae) in Costa Rican deciduous forest. Ecology 63: 1740–1750.[CrossRef][ISI]

Handel, S. N. 1985 The intrusion of clonal growth patterns on plant breeding systems. American Naturalist 125: 367–384.[CrossRef][ISI]

Harper, J. L. 1977 Population biology of plants. Academic Press, London.

Hogan, K. P. 1983 The pollination biology and breeding system of Aplectrum hyemale (Orchidaceae). Canadian Journal of Botany 61: 1906–1910.

Ishikawa, S., A. Furukawa, and T. Oikawa. 1995 Zonal plant distribution and edaphic and micrometeorogical conditions on a coastal sand dune. Ecological Research 10: 259–266.[CrossRef][ISI]

———, Y. Toquenaga, and T. Matubayashi. 1994 Occurrence of Spermophagus rufiventris Boheman (Coleoptera, Bruchidae) in Calystegia japonica Choisy (Convolvulaceae) Seeds. Japanese Journal of Entomology 62: 431–432.

Inoue, K. 1988 Pattern of breeding-system change in the Izu Islands in Campanula punctata: Bumblebee-absence hypothesis. Plant Species Biology 3: 125–128.[CrossRef]

———. 1990 Dichogamy, sex allocation, and mating system of Campanula microdonta and C. punctata. Plant Species Biology 5: 197–203.

Iwatuki, K., T. Yamazaki, D. E. Boufford, and H. Ohba. 1993 Flora of Japan, vol. IIIa, Angiospermae Dicotyledoneae Sympetalae (a). Kodansha LTD., Tokyo.

Karoly, K. 1992 Pollinator limitation in the facultatively autogamous annual, Lupinus nanus (Leguminosae). American Journal of Botany 79: 49–56.[CrossRef][ISI]

Kim, S. T., and M. G. Chung. 1995 Genetic and clonal structure in Korean populations of Calystegia japonica (Convolvulaceae). Botanical Bulletin of Academia Sinica 36: 135–141[ISI]

Kitamura, S., G. Murata, and M. Hori. 1972 Colored illustrations of herbaceous plants of Japan (Sympetalae). Revised Edition in Japanese Hoikusha Publishing Co., Ltd.

Klinkhamer, P. G., and T. J. De Jong. 1993 Attractiveness to pollinators: a plant's dilemma. Oikos 66: 180–184.[CrossRef][ISI]

Kowyama, Y., N. Shimano, and T. Kawase. 1980 Genetic analysis of incompatibility in the diploid Ipomoea species closely related to the sweet potato. Theoretical Applied Genetics 58: 149–155.

Kress, W. J., and J. H. Beach. 1994 Flowering plant reproduction systems. In L. A. McDade et al. [eds.], La Selva, 161–182. The University of Chicago Press. Chicago, IL.

Levin, D. A. 1971 Competition for pollinator service: a stimulus for the evolution of autogamy. Evolution 26: 668–669.[CrossRef]

Lloyd, D. G. 1979 Some reproductive factors affecting the selection of self-fertilization in plants. American Naturalist 113: 67–79.[CrossRef][ISI]

———. 1980 Demographic factors and mating patterns in Angiosperms. In O.T. Solbrig [ed.], Demography and evolution in plant populations, 67–88 Blackwell, Oxford.

———, and C. J. Webb. 1986 The avoidance of interference between the presentation of pollen and stigma in angiosperms. I. Dichogamy. New Zealand Journal of Botany 24:135–162.

Lovett-Doust. L. 1981 Population dynamics and local specialization in a clonal perennial (Ranunculus repens). I. The dynamics of ramets in contrasting habitats. Journal of Ecology 69: 743–755.[CrossRef]

Makino, T. 1895 In Japanese. Botanical Magazine 9: 311–316.

Momose, K., T. Nagamitsu, and T. Inoue. In press Thrips cross-pollination of Popowia pisocarpa (Annonaceae) in a lowland dipterocarp forest in Sarawak. Biotropica

Murawski, D. A., and J. L. Hamrick. 1990 Local genetic and clonal structure in the tropical terrestrial bromeliad Achmea magdalenae. American Journal of Botany 77: 1201–1208.

———, B. Dayanandan, and K. S. Bawa. 1994 Outcrossing rate of two endemic Shorea species from Sri Lankan tropical forests. Biotropica 26: 23–29.[CrossRef][ISI]

Ohwi, J. 1965 Flora of Japan. Smithsonian Institution. Washington, DC.

Oinonen, E. 1967 Sporal regeneration of bracken (Pteridium aquilinum (L.) Kuhn.) in Finland in the light of the dimensions and the age of its clones. Acta Forestalis Fennica 83: 1–96.

Okamoto, M. 1992 Dispersal characteristics exhibited in fruit morphology. Acta Phytotaxonomics and Geobotany 43: 155–166.

Peakall, R., and A. J. Beattie. 1989. Pollination of the orchid Microtis parviflora R. Br. by flightless worker ants. Functional Ecology 3: 515–522.

———, and ———. 1991 The genetic consequence of worker ant pollination in a self-compatible, clonal orchid. Evolution 45: 1837–1848.[CrossRef][ISI]

Rathcke, B. J. 1988 Interactions for pollination among coflowering shrubs. Ecology 69: 446–457.[CrossRef][ISI]

Richards, A. J. 1986 Plant breeding systems. George Allen and Unwin, London.

Silander, J. A., Jr. 1985 Microevolution in clonal plants. In J. B.C. Jackson, L. W. Buss, and R. E. Cook [eds.], Population biology and evolution of clonal organisms, 107–152. Yale University Press, New Haven, CT.

Spira, T. P. 1980 Floral parameters, breeding system and pollinator type in Trichostema (Labitae). American Journal of Botany 67: 278–284.[CrossRef][ISI]

Stebbins, G. L. 1950 Variation and evolution in plants. Columbia University Press, New York, NY.

———. 1957 Self-fertilization and population variability in the higher plants. American Naturalist 91: 337–354.[CrossRef][ISI]

Uyenoyama, M. K. 1986 Inbreeding and the cost of meiosis: the evolution of selfing in populations practicing biparental inbreeding. Evolution 40: 388–404.[CrossRef][ISI]

Webb, C. J., and D. G. Lloyd. 1986 The avoidance of interference between the presentation of pollen and stigma in angiosperms. II. Herkogamy. New Zealand Journal of Botany 24: 163–178.[ISI]

Weisler, S. L., and A. A. Snow. 1992 Potential for the loss of self-incompatibility in pollen-limited populations of Mayapple (Podophyllum peltatum). American Journal of Botany 79: 1273–1278.[CrossRef][ISI]

Willmott, A. P., and A. Burquez. 1996 The pollination of Merremia palmeri (Convolvulaceae): can hawk moths be trusted? American Journal of Botany 83: 1050–1056.[CrossRef][ISI]

Worthen, W. B., and E. W. Stiles. 1988 Pollen-limited fruit set in isolated patches of Maianthemum canadense Desf. in New Jersey. Bulletin of the Torrey Botanical Club 115: 299–305.[CrossRef][ISI]

Wyatt, R. 1982 Inflorescence architecture: how flower number, arrangement, and phenology affect pollination and fruit-set. American Journal of Botany 69: 585–594.[CrossRef][ISI]

———. 1983 Pollinator-plant interactions and the evolution of breeding systems. In L. Real [ed.], Pollination biology, 51–95. Academic Press, New York, NY.

———. 1984 The evolution of self-pollination in granite outcrop species of Arenaria (Caryophyllaceae). I. Morphological correlates. Evolution 38: 804–816.[CrossRef][ISI]




This article has been cited by other articles:


Home page
 Am. J. Bot.Home page
O. Totland and M. Sottocornola
Pollen limitation of reproductive success in two sympatric alpine willows (Salicaceae) with contrasting pollination strategies
Am. J. Botany, June 1, 2001; 88(6): 1011 - 1015.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via ISI Web of Science (10)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Ushimaru, A.
Right arrow Articles by Kikuzawa, K.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Ushimaru, A.
Right arrow Articles by Kikuzawa, K.
Agricola
Right arrow Articles by Ushimaru, A.
Right arrow Articles by Kikuzawa, K.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS