AOBPreview originally published online on January 5, 2004
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Annals of Botany 93: 167-175, 2004
© 2004 Annals of Botany Company
Pollination Ecology and Pollination System of Impatiens reptans (Balsaminaceae) Endemic to China
1 College of Life Sciences, Hunan Normal University, Changsha 410081, China
* For correspondence. E-mail lkming{at}sohu.com
Received: 13 June 2003;; Returned for revision: 1 August 2003; Accepted: 1 October 2003: Published electronically: 5 January 2004
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
• Background and Aims China is one of the centres of geographical distribution of Impatiens L. Studies of the pollination ecology of this genus in China have, until now, been unreported. Impatiens reptans, a species endemic to China, was studied. The aims were to examine the pollination ecology and pollination system of this species, to compare its pollination ecology with other Impatiens species growing in Sumatra and Japan, and to discuss possible reasons for its limited distribution.
• Methods The pollination ecology of I. reptans was studied by carrying out continuous observations within three naturally growing populations. Its pollination system was studied using different pollination methods, marking and counting pollen grains, assessing pollen viability and observing pollinator behaviour.
• Key Results The flowering phase of the protandrous I. reptans lasted for 89 d. The life span of an individual flower was 3·6 d. Primary pollinators were honey-bees and bumble-bees. Secondary pollinators were diurnal hawk moths and butterflies. Bombus briviceps and Bombus sp. were nectar gatherers. The mean nectar sugar concentration was 29·5 %, and the mean value of sucrose/glucose + fructose was 0·82. The proportion of seed set ranged from 0·857 to 0·873. Distances that seeds were ejected ranged from 0·58 to 1·17 m. Percentage seed germination under controlled conditions was 23·1. Pollen viability was highest on the day of anthesis and thereafter decreased. Ratios of pollen : ovules ranged from 958·8 to 970·6.
• Conclusions Impatiens reptans reproduces by means of cross-pollination. Its dependence on a specialized habitat, a narrow environmental niche, a low percentage of seed germination, and habitat loss could be reasons for its limited distribution and endemism.
Key words: Impatiens reptans, Balsaminaceae, pollination ecology, pollination system.
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
Impatiens L. is the largest genus in the family Balsaminaceae. There are more than 900 species worldwide, most of which occur in mountains in subtropical and tropical regions from India to south-east Asia. Only a few species are found in temperate zones (Grey-Wilson, 1985). The species found in the Sino-Himalayan region are conspicuously diversified (Hooker, 1904–1906, 1908, 1910–1911). Two hundred and forty species have been found in China (Chen, 2001; Jin and Ding, 2002; Huang et al., 2003), which represents approx. 25 % of Impatiens species worldwide. Most are found in south-west China, although some species have been discovered in nearly every province. The genus has some species with single flowers, but most have racemes, corymbs or subumbels. Flower numbers range from few to many, and flower size and colour are diverse. The shape of bracts, sepals, petals and especially labellums vary. The spur’s form, length and angle vary also (Hooker, 1875). The geographical distribution of Impatiens is very localized and endemic. Most of the Chinese Impatiens species are endemic to the country or are restricted to a number of provinces.
A study of the reproductive biology of Impatiens was fundamental to investigations into its systematics and evolutionary biology (Ornduff, 1969; Anderson, 1995). Detailed studies on the pollination ecology of a few Impatiens species have been done (Rust, 1977, 1979; Schemske, 1978, 1984; Heinrich, 1979a; Kato, 1988; Kato et al., 1991; Lu, 2000). In temperate zones, some Impatiens species are pollinated by bumble-bees and hummingbirds (Rust, 1977, 1979; Heinrich, 1979; Kato et al., 1989). Out of the 109 morphologically diverse Impatiens species which grow in Africa, 58 are pollinated by butterflies, three by moths, 27 by birds and 21 by bees (Grey-Wilson, 1980). In Sumatra and Japan, hawk moths and bees act as pollinators to at least some species (Kato, 1988; Kato et al., 1991). Most Impatiens species reproduce by cross-pollination (Abrahamson and Hershey, 1977; Schemske, 1978, 1984; Simpson et al., 1985; Schmitt and Ehrhardt, 1987; Waller and Knight, 1989; Schmitt and Gamble, 1990; Twasuda and Yahara, 1994; Lu, 2000, 2002). Waller (1980) has suggested that ‘environmental chasmogamy’ more accurately describes the pollination system in Impatiens capensis. Despite the abundance of Impatiens species in China, studies of pollination systems and pollination ecology have not been reported.
Although endemic to China, Impatiens reptans is found in only two districts: Taoyuan County, Hunan Province, and Guiyang City, Guizhou Province. Not only are these two populations separated by approx. 1000 km, but their distributive range at each site is very small. How this distribution pattern was formed is unknown. No information is available on the identity of the pollinators of I. reptans or on their relationships among the species. In this paper, the pollination ecology and pollination system of I. reptans in Taoyuan County, Hunan Province was studied and compared with the pollination ecology of other Impatiens species growing in Sumatra and Japan. The reason for the limited distribution and endemism of the species is also discussed.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
Plant material
Impatiens reptans Hook. f. is an annual herb, whose stem-base is a creeper. The stem is usually 30 cm long, although it can grow to 50 cm or longer. Leaves are alternate, ovate or ovate-elliptical with a short pedicel and three or four lateral veins. There are two or three flowers in the axils of the upper leaves. The length of flower pedicels is 5–10 cm, at the bases of which occur ovate-lanceolate persistent bracts. The diameter of the yellow flowers is approx. 2·4 cm. Each flower has three sepals and five petals. The two smaller green, lateral sepals are ovate and shaped like a sickle. The third sepal is saccate, with a slightly incurvate spur. The length of this spur, which contains abundant nectar to attract pollinators, is approx. 2 cm. At the rear of the round, 1-cm-long vexil there is a narrow, keel-shaped protuberance. The lateral petals below the vexil form into two pairs of connate wings with two slits. The length of each wing is approx. 2 cm. Five united stamens are adnate to the top of the pistil, which is made up of five connate carpels. The erect ovary is spindly. The length of the cylindrical capsule and the oblong seeds is, respectively, approx. 2 cm and 2·9–3·0 mm (see Figure 2A and B for illustrations of the characteristics of the floral parts). Specimens were deposited in the herbarium of Hunan Normal University.
|
The study site was located on a hill with yellow soil near the town of Lingjintan, Taoyuan County, Hunan Province (121–357 m a.s.l.). The climate is subtropical with an annual average temperature of 16·5 °C. The I. reptans plants were growing in a wet area, covering 4 km2 beside a stream. Field observations were made and experimental samples collected from 26 Sept. to 2 Nov. 2001 and from 18 Sept. to 11 Dec. 2002, respectively, from three populations (A, B and C) 500–1100 m apart. There were tens of thousands of plants within the whole area, of which 57–302 were sampled from each population (Table 1).
|
Flower phenology
In 2002, ten buds were selected from each population and tagged with thin, red thread. The flower phenology of these buds was observed and recorded subsequently every 3 h. The number of flowering individuals within each population was recorded every 5 d.
Nectar collection and analysis
During the early morning (0500 h) of 5 Nov. 2002, when insects were inactive, ten flowers from each population were sampled. The nectar from each flower was extracted and its volume measured using a calibrated 10 µl micropipette. The nectar was put into a 1·5 ml Eppendorf tube and taken to the laboratory for analysis using paper chromatography. Concentrations of nectar sugar and of sucrose/fructose + glucose were determined.
Pollinator behaviour
During 15–17 Oct. 2001 and 2–4 Nov. 2002 (both peak periods of flowering), continuous observations of pollinator behaviour were conducted on the three populations from approx. 0530 h. This resulted in each population being observed in total for between 8 and 11 h. The length of time each pollinator visited an individual flower, pollinator visitation rates (visits per flower per hour), and the time interval between visits by members of the same pollinator species to individual flowers during fine weather were recorded. The number of flowers within a population that an individual pollinator visited at each visitation was also recorded. All hawk moth and bumble-bee visitation variables were tested for normality and transformed to attain homogeneity for analysis of variance (ANOVA) among the three populations. Records of visits by domesticated honey-bees and butterflies were not analysed. Insect behaviour was recorded by colour photography. Insects were captured using an insect net. The location of pollen deposition was determined by stereoscopic microscope examination of pinned insects. Night-time visitations by pollinators were recorded once, for between 40 and 60 min at 1800 h, 2200 h, 0200h and 0500h.
Pollination systems
Five pollination systems described by Dafni (1992) were conducted in the field during 2002. Artificial pollination was effected by brushing the recipient flower stigmas with stamens from the pollen donors. The treatments were: bagged (B); emasculation and hand-outcrossing (O); emasculation and hand-selfing (S); natural pollination (N); and emasculation (E). Treatment B was conducted to examine autonomous self-pollination by enclosing intact flowers, before anthesis, in small plastic bags. Treatment O was conducted to examine xenogamy through artificial pollination using pollen from flowers 5 m away, but within the same population. Treatment S was conducted to examine geitonogamy through artificial pollination using pollen from other flowers on the same plant. Treatment N acted as the control. Treatment E was conducted to determine whether or not stigmas remained receptive before the stamens dropped by removing the stamens from a flower that had been open for 1 d then bagging it. Twenty healthy flowers from each treatment within all three populations were chosen randomly. After 25 d, fruits and seeds were collected, and their numbers recorded. In addition, the weight of each seed was recorded.
Pollen : ovule ratio and proportion of seed set
To determine pollen : ovule ratios of single flowers during 2002, 20 flowers, from all three populations, were collected randomly during early morning, before their androecia dehisced, and preserved in FAA solution. The androecium from each flower was hydrolysed in HCl solution to make 1 ml of pollen grain suspension. The number of pollen grains in a 5 µl sample of the suspension was counted under a light microscope. From these data, the mean number of pollen grains per flower was calculated. Young pistils were collected from flowers from all three populations and dissected under a stereoscopic microscope to determine the mean number of ovules per flower. The proportion of seed set was calculated from the number of seeds per flower produced by natural pollination and the number of ovules per flower.
Pollen viability
Two forms of pollen viability are apparent in entomophilous species. First, a rapid decline in species with transient pollen, and second, a gradual decline in species with longer-living pollen. Each form plays a different role (Pacini et al., 1997). To assess pollen viability in I. reptans, 15 flowers from population A were sampled. Five flowers were sampled at 0800 h on the day they opened, five at 0800 h on the day after opening, and five on the day when androecia were on the point of dropping (usually the third day after opening). Pollen grains from each flower were mixed with three drops of 0·5 % TTC solution on slides (Oberle and Watson, 1953). They were then incubated at a constant temperature of 15 °C for 15 min. The total number and the number of stained pollen grains within five separate areas on each slide were counted under a light microscope. From these figures, the percentage of stained pollen grains was calculated.
Pollen dispersal
On 6 Nov. 2002 (fine weather), four flowers from each population were selected as pollen donors. The androecium of each flower had dehisced on the morning of sampling. The pollen grains were stained with a solution of methylenum caeruleum (1 %) and the flowers bagged for 3 h. When the plastic bags were removed, insect pollinators were able to visit the flowers. After a further 3 h, styles were examined with a magnifier (x20) for the presence of blue pollen grains. The distance insects carried stained pollen grains was also measured.
Seed dispersal and percentage seed germination
Twenty mature capsules were sampled randomly from all three populations in the field. The distance to which seeds were ejected from the capsules was measured, and a mean value and standard deviation calculated. In the laboratory, 39 seeds from capsules from all three populations were extracted and soaked overnight to soften the seed coat. They were then placed on filter paper that had been dampened with sterile tap water and incubated at 25 °C and a 12 h day. Percentage seed germination was recorded.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
Flower phenology
During 2002, Impatiens reptans in all three populations flowered from 14 Sept. with a peak from early/mid-October to mid-November and a late flowering phase from late November to 11 Dec. Plants were in flower for a total of 89 d (Fig. 1). The average life span of each individual flower was 3·6 d (n = 30, s.d. = 0·382). On average, the male phase of the protandrous I. reptans was 74 h (n = 30, s.d. = 1·18) and the female phase 12·5 h (n = 30, s.d. = 2·15).
|
Nectar analysis
The concentration of nectar sugar in samples collected from all three populations in 2002 ranged from 25·8 to 31·5 % (mean = 29·5 % ± 1·52 %, n = 15). The mean value of sucrose/glucose + fructose was 0·82. The volume of nectar secreted per flower was 1·5–3·5 µl (mean = 2·86 ± 0·77, n = 15).
Flower visitors and their behaviour
In total, 26 specimens of seven species were collected and sent to the Zoological Museum of Hunan Normal University (Table 2). The buds of I. reptans were opened usually by foraging bumble-bees, similar to Impatiens species in Japan (Kato, 1988). Most flowers were open by early morning. At dawn (approx. 0530 h), bees became active, followed soon after by hawk moths. In fine weather, the average time between visits to individual flowers by bumble- bees (Fig. 2C) and by hawk moths (Fig. 2E–F) was 4·9 min (n = 52, s.d. = 0·31) and 24·7 min (n = 17, s.d. = 0·84), respectively. Pollen loads were found on the hind legs (pollen basket) and heads of bumble-bees. They pollinated flowers by transferring pollen grains to the sigma. Pollen grains were also found on the heads and long proboscis of captured hawk moths. It seems probable that whilst foraging for nectar, hawk moths pollinated flowers by transferring pollen grains from one flower to another. The peak period of bumble-bee and hawk moth activity was, respectively, from 0700 to 1600 h and from 0830 to 1500 h each day. Hawk moths fly rapidly and visited many flowers, spending, on average, 2·4 s at each flower. Bumble-bees spent, on average, 5·3 s at each flower. Bumble-bees visited fewer flowers during a foraging visit than hawk moths (Table 3). More bumble-bees than hawk moths visited each population, which was contrary to their visitation rates (visits per flower per hour). For populations A, B and C, the average bumble-bee and hawk moth visitation rates were, respectively, 8·88 ± 2·80, 7·67 ± 2·54 and 7·11 ± 2·26, and 1·88 ± 1·45, 2·0 ± 1·22 and 1·44 ± 0·88. Hawk moth visitation rates differed significantly between the three populations (ANOVA, F = 6·15, d.f. = 53, P < 0·01). Bumble-bee visitation rates also showed differences between the three populations (ANOVA, F = 7·82, d.f. = 53, P < 0·005). Hawk moths did not select flowers of any particular size, contrary to the observations of Wilson (1995). They foraged and visited flowers in turn. The length of the flower spur had no effect on hawk moth visitation rates among the three populations (ANOVA, F = 1·23, d.f. = 44, P = 0·70). Numbers of hawk moths foraging at the same time in the different populations were significantly different (ANOVA, F = 16·98, d.f. = 71, P < 0·005), as were numbers of bumble-bees (ANOVA, F = 5·59, d.f. = 65, P < 0·01). There was little difference in the number of pollinator species within populations between 2001 and 2002. There was a strong relationship between the weather and butterfly activity. When it was sunny, butterflies actively visited I. reptans flowers, spending, on average, 2·8 min at each. However, when the weather was cloudy, butterflies were less active. On rainy days, they were completely inactive. Hawk moth numbers were also reduced on cloudy and rainy days, but bumble-bees were unaffected by bad weather. Apis cerena (Fig. 2D), a species of pollen-collecting honey-bee, was the most abundant visitor, visiting more flowers than any other species (Table 3). Individuals have a small body and a short proboscis. They are active during daylight. Cruden (1973) suggested that flowers open for more than 12 h are exposed to potential visitations by a number of different diurnal and nocturnal pollinators, but no visitors were observed at night. Nectar gatherers of Impatiens have been reported (Rust, 1977, 1979; Kato et al., 1991; Temeles and Pan, 2002). Two species of nectar gatherers, Bombus briviceps (Fig. 2G) and Bombus sp. (Fig. 2H), were identified during the study. Their common characteristic was a relatively short (approx. 2 mm) proboscis with which they were unable to extract nectar from the spur at the front of each flower. However, they bit through the spur to gain access to the nectar. In 2002, during peak flowering the percentage of flowers with bitten-through spurs in populations A, B and C was 26·8, 29·7 and 30·4, respectively, and during the late flowering period 53·3, 45·6 and 51·7, respectively. It was observed that, after the activities of nectar gatherers, it was some time before the volume of nectar returned to sufficient quantities to entice other pollinators to visit the flowers. This was probably because bumble-bees could detect both remote and proximate chemical clues left by other foraging individuals following their visits. It showed that bumble-bees had an acute sense enabling them to detect chemical clues. Bumble-bees gathered much nectar.
|
|
Pollination systems
Results from the pollination systems experiments are shown in Table 4. In all three populations, treatments B and S produced few seeds (1·57 and 1·75 per fruit, respectively), suggesting that I. reptans may be self-compatible. There was a significant difference between the numbers of seeds produced by treatments O and B (P < 0·05), the former producing 6·7 seeds per fruit, the highest number for the five treatments. Treatment E also produced few seeds. The average weight of individual seeds produced by treatments O and B were 8·6 mg and 11·4 mg, respectively, which were significantly different (P < 0·05).
|
Pollen : ovule ratios and proportion of seed set
In population A, the mean number of pollen grains per flower was 6450 (n = 16, s.d. = 187), the mean number of ovules per ovary was 6·7 (n = 15, s.d. = 1·15), and the mean number of seeds per fruit was 5·85 (n = 14, s.d. = 2·61). In population B, the respective values were 6600 (n = 20, s.d. = 164), 6·8 (n = 20, s.d. = 1·032), and 5·9 (n = 18, s.d. = 2·18), and in population C, 6520 (n = 15, s.d. = 168), 6·8 (n = 18, s.d. = 1·12), and 5·83 (n = 15, s.d. = 2·21). So the proportion of seed set ranged from 0·857 to 0·873, and pollen/ovule ratios from 958·8 to 970·6. According to Cruden (1977), I. reptans is dependent on cross-pollination, which is consistent with the results from the pollination systems experiments.
Pollen viability and distribution
Pollen viability was greatest on the day of anthesis, with 99·1 % of pollen grains staining red. By the following day, the percentage of pollen grains staining red had reduced to 90·7 %, and by the third day, before androecia dropped from pistils, the viability of the remaining pollen was very low with only 5·3 % of grains staining red. The percentage of pollen grains staining light red or unstained was 79·5 and 15·2 %, respectively (Fig. 3).
|
Pollinators dispersed pollen up to 37 cm from the source, with the mean distance being 27·75 cm (n = 12, s.d. = 3·46). At distances further than 29 cm, pollen dispersed was low. The conclusion, that bees (A. cerena and B. trifasciatus) are capable of dispersing pollen up to 29 cm from its source, confirmed the results of Young (2002).
Seed dispersal and percentage seed germination
As capsules matured, internal pressure caused the cell wall of the pericarp to rupture and seeds to be ejected to 0·58–1·17 m (mean =1·05 m ± 0·11 m, n = 20). Under laboratory conditions, percentage germination of seeds collected from populations A, B and C were 20·5, 23·1 and 25·6 %, respectively, with a mean of 23·1 % (n = 117).
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
It is well known that the flowers of Impatiens have enormous diversity and different pollinators. In this study, bumble-bees, honey-bees, hawk moths and butterflies pollinated I. reptans growing on a mountain in the subtropics of China. In subtropical regions of Africa the species is pollinated by hummingbirds as well as by insects (Grey-Wilson, 1980). In temperate zones, pollinators are bumble-bees and hummingbirds only (Rust, 1977, 1979; Heinrich, 1979; Kato et al., 1989). Hence, in different climatic regions, species of pollinators vary. In flowering biology, the life span of an I. reptans flower was 3·6 d, which was longer than that of I. textori (1·7 d) in Japan (Kato, 1988) and longer, or the same, as that of I. platypetala, I. korthalsii, I. talangensis and I. eubotrya in Sumatra (Kato et al., 1991). The male phase lasted six times as long as the female phase in I. reptans, which was similar to that of I. capensis (Schemske, 1978; Bell et al., 1984). The ratio of functional sex (the male phase/anthesis of individual flowers) was 0·85, greater than that of Impatiens in Sumatra (Kato et al., 1991). The mean concentration of nectar sugar in the long sepal spur was 29·5 %, and the mean value of sucrose/glucose + fructose was 0·82. According to Baker and Baker (1983, 1990), I. reptans belongs in the list of species pollinated by long-tongued bees. This was consistent with our study of captured pollinators.
The number of bumble-bee species that pollinated I. reptans was fewer than in Japan and Sumatra. In the study reported here, only one species of bumble-bee acted as a pollinator. The frequency of bumble-bee foraging was less than in Japan but similar to that in Sumatra. Apis cerena, a domestic species of bee, was the most frequent visitor, being active during daylight.
In Sumatra, I. platypetala was visited once at dawn by the crepuscular hawk moth (Macroglossum corythus) (Kato et al., 1991). Some hawk moths (Macroglossum spp.) also visited Impatiens flowers in Japan (Kato, 1988). In the study reported here, two species of hawk moths (M. corythus and M. variegatum) visited I. reptans flowers. Both species were active during daylight and were rapid fliers. They visited many flowers at each visitation and were, therefore, major pollinators. When the weather was fine, butterflies visited the flowers, spending an average of 2·8 min on individual flowers. Butterflies have not been observed as pollinators of Impatiens in Japan or Sumatra.
Two species of bumble-bees were identified as nectar gatherers in this study. The effect of nectar gathering on visitations by pollinators was significant. After nectar was gathered from an individual flower, it was not visited by a pollinator until the volume of nectar had returned to a sufficient level to attract pollinators. This meant that the activities of nectar gatherers significantly reduced the number of flowers visited by pollinators. Temeles (2002) has suggested that nectar gathering did not affect seed set but this conclusion was drawn during peak flowering. It was observed that, although there were hardly any pollinators present during the late phase of flowering, there were abundant nectar gatherers. Hence, it is suggested that nectar gathering might hasten the end of flowering.
Pollination systems may be seen as a biological market. Differences in the quality of the ‘output’ (e.g. nectar quantity and concentration) offered by the ‘seller’ (plant species) attracted different ‘customers’ (e.g. nectar gatherers and pollinators) (Campbell and Motten, 1985). Chittka and Schürkens (2001) reported that, after it had invaded Europe, I. glanduifera tempted bee pollinators away from native plants as it secreted much more nectar than any of the native species. It was found that, at the end of the flowering period of I. reptans, some entomophilous plants, such as Aster ageratoides Turcz, Dendranthema indicum (L.) Des Monl (Compositae) and Dipsacus asper Wall. (Dipsacaceae), entered their peak flowering phase and secreted significantly more nectar than I. reptans. Hence, the long-tongued bumble-bees that previously visited the flowers of I. reptans were attracted to visit the flowers of the abundant nectar producers. The view of Raven (Heinrich and Raven, 1972; Heinrich, 1975, 1979), that it is the principal aim of insect pollinators to forage nectar as a source of energy, and that they will select plants that secrete most, is supported by the observations reported here.
Plants of I. reptans occurred mostly in clumps either alongside a stream or in damp places or as a number of individuals within a group. Isolated individuals were rarely found. The plants had become adapted to these special habitats through natural selection and adaptation to the environment (including landform, soil and other species). Although endemic to China, environmental niches suitable for the establishment of I. reptans in the country are few. The distance to which seeds were ejected from mature capsules was short, ranging from 0·58 to 1·17 m. Most seeds were dispersed around the maternal plants. Although some seeds were deposited into the stream and swept downstream, the percentage germination of these seeds was low. Even if the seeds were able to germinate in new habitats and develop into seedlings, they were often unable to compete with native species in the struggle for survival. Hence, the spread of the species was limited. The distance to which pollen was dispersed by honey-bees and bumble-bees, the primary pollinators, was approx. 29 cm, so pollen dispersal was restricted also. As the pollinators of I. reptans are widespread, the obligatory adaptation to pollinators of limited distribution cannot be the reason for endemism of the species.
Habitat disjunction leads to lower population persistence (Ruggiero et al., 1994). I. reptans grows in rural areas. With over-exploitation of agricultural resources and improper wetland management, villagers have either planted crops in original habitats of I. reptans or converted natural ditches into artificial canals. The sides of canals are often lined with cement blocks, resulting in disjunction and the loss of wetland habitats. Under such conditions, weeds, such as Alternanthera philoxcroides (Mart.) Griseb (Amaranthaceae) and Paspalum distichum L. (Gramineae), compete for resources with I. reptans and spread rapidly, posing a direct threat to the growth of I. reptans. If habitats continue to be destroyed, populations of I. reptans will continue to decline. It is therefore concluded that adaptation for a specialized habitat, a narrow environmental niche, a low percentage of seed germination, and habitat loss could be reasons for the limited distribution and endemism of I. reptans.
|
ACKNOWLEDGEMENTS |
|---|
We thank Prof. Peng Fu, College of Life Sciences, Hunan Normal University and Dr Mei-Cai Wei, Central South Forest University for identifying the insect specimens; Prof. Xian-Chun Wang of the key of National Laboratory, Hunan Normal University for help with the component analysis of nectar; Dr Yun-Fei Deng, Kunming Institute of Botany, Chinese Academy of Sciences for advice on the research work; Prof. D. Levin, Prof. D.Waller and Dr H. Sato for their comments and valuable suggestions on the manuscript; Prof. Myong Gi Chung, Gyeongsang National University, Republic of Korea and Prof. Sheng Luan and Dr LeGong Li, UC at Berkeley, USA for their kind help in improving the manuscript; and Prof. R. W. Rust, University of Nevada, USA and Prof. Makoto Kato, Kyoto University, Japan for providing cited articles.
|
LITERATURE CITED |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
-
Abrahamson WG, Hershey GJ. 1977. Resource allocations and growth of Impatiens capensis (Balsaminaceae) in two habitats. Bulletin of the Torrey Botanical Club 104: 160–164.[CrossRef]
Anderson GJ. 1995. Systematics and reproductive biology. Monographs in Systematic Botany 53: 263–272.
Baker HG, Baker I. 1983. Floral nectar sugar constituents in relation to pollinator type. In: Jones CE, Little RJ, eds. Handbook of experimental pollination biology. New York: Van Nostrand, 117–141.
Baker HG, Baker I. 1990. The predictive value of nectar chemistry to the recognition of pollinator types. Israel Journal of Botany 39: 157–166.
Bell G, Lefebvre L, Giraldeau L, Weary D. 1984. Partial reference of insects for the male flowers of an annual herb. Oecologia 64: 287–294.[CrossRef][ISI]
Campbell DR, Motten AF. 1985. The mechanism of competition for pollination between 2 forest herbs. Ecology 66: 554–563.[CrossRef]
Chen YL. 2001. Flora Reipublicae Popularis Sinicae. Tomus 47(2) [in Chinese].
Chittka L , Schürkens S. 2001. Successful invasion of a floral market. Nature 411: 653.
Cruden RW. 1973. Reproductive biology of weedy and cultivated Mirabilis (Nyctaginaceae). American Journal of Botany 60: 802–809.[CrossRef][ISI]
Cruden RW. 1977. Pollen-ovule ratios: a conservative indictor of breeding systems inflowering plants. Evolution 31: 32–36.
Dafni A. 1992. Pollination ecology, the practical approach. Oxford: Oxford University Press.
Grey-Wilson C. 1980. Impatiens of Africa. Rotterdam: Balkema.
Grey-Wilson C. 1985. The family Balsaminaceae. In: Dassanayake MD, Fosberg FR, eds. A revised handbook to the flora of Ceylon. New Delhi: Amerind Publishing Co.
Heinrich B. 1975. Energetics of pollination. Annual Review of Ecology and Systematics 6: 139–170.
Heinrich B. 1979a. Resource heterogeneity and patterns of movements in foraging bumblebees. Oecologia 40: 235–245.[CrossRef]
Heinrich B. 1979b. Bumblebee economics. Massachusetts: Harvard University Press.
Heinrich B, Raven P. 1972. Energetics and pollination ecology. Science 176: 597–602.
Hooker JD. 1875. Flora of British India. Kent: L. Reeve, 440–483.
Hooker JD. 1904–1906. An epitome of the British Indian Species of Impatiens. Records of the Botanical Survey of India 4: 1–58.
Hooker JD. 1908. Impatiens. In: Hooker’s Icones Plantarum, 4th Series 9. London, 2851–2875
Hooker JD. 1910–1911.Impatiens. In: Hooker’s Icones Plantarum, 4th Series 10. London, 2901–2975.
Huang SH, Shui YM, Chen WH. 2003. New taxa of Impatiens from Yunnan. Acta Botanica Yunnanica 25: 261–280.
Jin XF, Ding BY. 2002. A new species of Impatiens from eastern Zhejiang, China. Acta Phytotaxonomica Sinica 40: 167–169.
Kato M. 1988. Bumblebee visits to Impatiens spp.: pattern and efficiency. Oecologia 76: 364–370.
Kato M, Itino I, Hotta M, Abbas I, Okada H. 1989. Flower visitors of 32 plant species in West Sumatra. Occasional Papers of the Kagoshima University Research Center for the South Pacific No. 16, 15–31.
Kato M, Itino T, Hotta M, Inoue T. 1991. Pollination of four Sumatran Impatiens species by hawkmoths and bees. Tropics 1: 59–73.
Lu YQ. 2000. Effects of density on mixed mating systems and reproduction in natural populations of Impatiens capensis. International Journal of Plant Sciences 161: 671–681.[CrossRef]
Lu YQ. 2002. Why is cleistogamy a selected reproductive strategy in Impatiens capensis (Balsaminaceae)? Biological Journal of the Linnean Society 75: 543–553.[CrossRef]
Oberle GD, Watson R. 1953. The use of 2,3,5-triphenyl tetrazolium chloride in viability tests of fruit pollens. Journal of the American Society for Horticultural Science 61: 299–303.
Ornduff R. 1969. Reproductive biology in relation to systematics. Taxon 18: 121–133.[CrossRef]
Pacini E, Franchi GG, Lisci M, Nepi M. 1997. Pollen viability related to type of pollination in six angiosperm species. Annals of Botany 80: 83–87.
Ruggiero LF, Aubry KB, Buskirk SW, Lyon LJ, Zielinski WJ. 1994. The scientific basis for conserving forest carnivores: American marten, fisher, lynx and wolverine in the Western United States. USDA Forest Service General Technical Report RM-251.
Rust RW. 1977. Pollination in Impatiens capensis and Impatiens pallida (Balsaminaceae). Bulletin of the Torrey Botanical Club 104: 361–367.[CrossRef]
Rust RW. 1979. Pollination of Impatiens capensis: pollinators and nectar robbers. Journal of the Kansas Entomological Society 52: 297–308.
Schemske DW. 1978. Evolution of reproductive characteristics in Impatiens (Balsaminaceae): the significance of cleistogamy and chasmogamy. Ecology 59: 596–613.[CrossRef][ISI]
Schemske DW. 1984. Population structure and local selection in Impatiens pallida (Balsaminaceae), a selfing annual. Evolution 38: 817–832.[CrossRef]
Schmitt J, Ehrhardt D. 1987. A test of the sibcompetition hypothesis for the outcrossing advantage in Impatiens capensis. Evolution 41: 579–590.[CrossRef]
Schmitt J, Gamble SE. 1990. The effect of distance from the parental site on offspring performance and inbreeding depression in Impatiens capensis: a test of the local adaptation hypothesis. Evolution 44: 2022–2030.[CrossRef]
Simpson RL, Leck MA, Parker VT. 1985. The comparative ecology of Impatiens capensis Meerb. (Balsaminaceae) in central New Jersey. Bulletin of the Torrey Botanical Club 112: 295–311.[CrossRef]
Temeles EJ, Pan IL. 2002. Effect of nectar robbery on phase duration, nectar volume, and pollination in a protandrous plant. International Journal of Plant Sciences. 163: 803–808.[CrossRef]
Twasuda M, Yahara T. 1994. Reproductive ecology of a cleistogamous annual, Impatiens noli-tangere L. occurring under different environmental conditions. Ecological Research 9: 67–75.
Waller DW. 1980. Environmental determinants of outcrossing in Impatiens capensis (Balsaminaceae). Evolution 34: 747–761.[CrossRef]
Waller DW, Knight SE. 1989. Genetic consequences of outcrossing in the cleistogamous annual, Impatiens capensis. Outcrossing rates and genotypic correlations. Evolution 43: 860–869.[CrossRef]
Wilson P. 1995. Selection for pollination success and the mechanical fit of Impatiens flowers around bumblebee bodies. Biological Journal of the Linnean Society 55: 355–383.[CrossRef]
Young HJ. 2002. Diurnal and nocturnal pollination of Silene alba (Caryophyllaceae) American Journal of Botany 89: 433–440.
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||