Botanical Studies (2010) 51: 81-88.
reproductive biology
Floral morphs, pollen viability, and ploidy level of Oxalis corymbosa DC. in Taiwan
Meng-Ying TSAI1, Su-Hwa CHEN1,2, and Wen-Yuan KAO1'2'*
1Institute of Ecology and Evolutionary Biology, National Taiwan University, Taipei 10617, Taiwan
2Department of Life Science, National Taiwan University, Taipei 10617, Taiwan
(Received July 22, 2008; Accepted July 22, 2009)
ABSTRACT. Oxalis corymbosa, an exotic species in Taiwan, is a tristylous plant, i.e., with long- (LS), mid-(MS) and short-styled (SS) floral morphs, producing capsules in its native habitat. However, in Taiwan, this species was never observed producing fruits, and it reproduces vegetatively. In this study we investigated floral morphs, observed fruit and seed production of 23 populations, assessed pollen viability, and analyzed the ploidy level of some populations of O. corymbosa, in an effort to understand the factors limiting sexual reproduction of this species in Taiwan. No fruit or seed production were observed in any population. The study of morph proportions revealed that ratios of SS, MS, and semi-homostylous (SHS) individuals varied among the populations, and LS individuals appeared to be absent in this region. Furthermore, two types of anthers were observed, yellow anthers with pollen grains and white anthers without pollen grains. In each population, individuals with MS morph and bearing yellow anthers were dominant. SS and SHS individuals were less frequent than MS morph and almost always bore white and male-sterile anthers. The result of fluorochromatic reaction essay revealed that these populations had very low pollen viability (< 4%). Flow cytometry analysis of cell nuclei suggested that individuals with MS and SS floral morph had same ploidy level. These results indicate that absence of LS floral morphs, the unequal proportion of floral morphs, low pollen viability, and the presence of male sterile morphs may all contribute to the absence of sexual reproduction of O. corymbosa in Taiwan.
Keywords: Flow cytometry; Heterostyle; Male-sterile; Oxalis corymbosa; Sexual reproduction.
Hermaphrodite plants in a population displaying floral morphs with reciprocal differences in style and anther height are called heterostylous (Darwin, 1877). A heterostylous species bears two (distyly) or three (tristyly) floral morphs generally growing in the same population (Barrett, 1992). In distylous plants, a high stigma position is combined with a low anther position in long-styled morphs while short-styled morphs combine a low stigma position with a high anther position. In tristylous species, a stigma and two sets of stamens occupy three levels inside the flower (Darwin, 1877). In short-style morphs (SS), stigmas are located in the lowest position of the flower with the two sets of stamens above the stigmas. In mid-style morphs (MS), the stigmas are located in the middle position between the two whorls of anthers. In long-style morphs (LS), stigmas are located in the highest, with two sets of stamens occupying the middle and lower positions. The floral morphs are genetically controlled (Ganders, 1979). Distyly is found in at least 25 angiosperm families (Lloyd and Webb, 1992), while tristyly is much less
common and is reported in only six flowering families, i.e., Oxalidaceae, Lythraceae, Pontederiaceae, Amaryllidaceae, Connaraceae, and Linaceae (Barrett, 1993; Thompson et al., 1996; Barrett et al., 1997).
Heterostylous plants generally combine a strong self- and morph-incompatibility system, and successful pollination (or legitimate pollination) only occurs between anthers and stigmas located at the same level in the flower (Darwin, 1865). Hence, heterostyly is thought to be a modification in the flower morphology and functioning suited to promote outcrossing (Darwin, 1877; Kohn and Barrett, 1992; Lloyd and Webb, 1992). It has been suggested that one of the forces driving the evolution of this morphological adaptation is the avoidance of self-interference between stigmas and anthers within the flowers and the enhancement of accurate pollen transfer between morphs (Lloyd and Webb, 1992). Due to strong self- and morph-incompatibility systems, each floral morph is usually incapable of producing seed after self-pollination and consequently can avoid the deleterious results of self-fertilization (Barrett, 1992; Darwin, 1900). However, species having such breeding systems might not be able to achieve legitimate pollination when introduced into new areas due to the lack or scarcity of compatible
Corresponding author: E-mail:
Botanical Studies, Vol. 51, 2010
floral morphs (Barrett and Forno, 1982; Ater, 2005; Castro
et al., 2007; Lou et al., 2006).
Oxalis corymbosa DC. (section Ionoxalis, Oxalidaceae), a species native to South America (Lourteig, 1980), was introduced into Taiwan as an ornamental plant (Wu et al., 2004). This species belongs to a tristylous genus with strong self-incompatibility and has been reported to have sexual reproduction and to produce capsules in its native environment (Barker, 1965; Lourteig, 1980). In Taiwan, O. corymbosa has never been observed to produce fruits, and it reproduces vegetatively by bulbils (Huang and Liu, 1993). The bulbils persist in the soil for many years and are easily dispersed by anthropogenic activities, namely by moving soils for cultivation (Wu et al., 1978). As in Taiwan, this species as been described as reproducing only asexually in other areas in which it has been introduced, like North America, Europe (Denton, 1973; Lourteig, 1980), China (Lou et al., 2006), and Malaysia (Veldkamp, 1971).
Several factors have been suggested for the lack of or low seed production in Oxalis, and these include unequal proportions or the absence of reciprocal floral
morphs (Barker, 1965; Denton, 1973; Castro et al., 2007),
geographical separation between reciprocal floral morphs (Lou et al., 2006), scarcity of pollinators (Marco and
Arroyo, 1998), ovule abortion (Guth and Weller, 1986),
unviable pollen (Carniel, 1969), and meiotic problems due to polyploidy, odd ploidies, or aneuploidy (Baker, 1965; Castro et al., 2007). For example, in the Mediterranean region, populations of O. pes-caprae have long- and short-styled morphs, but the different floral morphs rarely grow in mixed populations and have different ploidies being unable to produce viable offspring (Castro et al., 2007). Previous studies developed in three populations of O. corymbosa in a close geographic range in Taiwan revealed that most of the populations were composed by the mid-styled morph (Wu et al., 1978). Additionally, O. corymbosa has also been reported with various chromosome numbers, 2n = 14 (Naranjo et al., 1982; Xu et al., 1992), 28 (Roy et al., 1988; Luo et al., 2006), 35 (Barker, 1965), corresponding to different ploidy levels, i.e., 2x, 4x and 5x, respectively. In light of these observations, it was our objective to understand the factors limiting sexual reproduction of O. corymbosa in the exotic range across Taiwan. For this, we expanded the survey and performed a large-scale study across Taiwan in order to (1) investigate the composition and distribution of floral morphs, (2) measure the pollen viability, and (3) assess the ploidy level of wild populations of O. corymbosa. Fruit and seed production were also assessed.
materials and methods
Oxalis corymbosa is a tristylic perennial herb with a tap-root and numerous bulbils. The flower is composed of five sepals, five petals (of a rose-purple color), ten statements in two rounds, and one round of five stigmas
displayed in reciprocal heights (i.e., short-styled, SS, mid-styled, MS and long-styled morphs, LS; Figure 2). Furthermore, a semi-homostylous morph (SHS) with a long whorl of anthers and a short whorl of anthers and stigmas coinciding in height (Figure 2C) was also described in Taiwan (Wu et al. , 1 978). In Taiwan, it generally flowers in late winter to spring. Twenty-three natural populations of O. corymbosa in Taiwan were investigated between December 2005 and March 2007. Populations grow in disturbed places such as cultivated ground, roadsides, wasteland, parks and schools. We divided the administrative districts of Taiwan into four regions (northern, central, southern, and eastern regions), and all the populations were randomly selected including twelve populations (no. 1-12) from the northern region (Taipei, and Yilan), four populations (no. 13-16) from the central region (Chiayi), three populations (no. 17-19) from the southern region (Tainan, Kaohsiung, and Pintung), and four populations (no. 20-23) from the eastern region (Hualien, and Taitung) (Figure 1).
Figure 1. Map of Taiwan showing location of 23 populations of O. corymbosa investigated in this study.
TSAI et al. ― Floral morphs, pollen viability, and ploidy level of Oxalis corymbosa in Taiwan
In each population, 9-193 individuals were selected randomly and their floral morphs and fruit and seed production were recorded. In a preliminary survey, we found some O. corymbosa plants with flowers bearing white anthers (Tsai and Kao, personal observation), and thus, during the survey, this character was also recorded.
Anthers from flowers with different floral morphs were collected to assess pollen viability. We randomly collected 10, 3, 3, and 6 mid-styled individuals from Taipei (population 3), Chiayi (population 13), Kaohsiung (population 18), and Taitung (population 23), respectively, and 7 short-styled individuals from Taipei (population 3) for the analysis (Table 1). As O. corniculata, the native congener of O. corymbosa, reproduces sexually with success in Taiwan (Huang and Liu, 1993), we examined the pollen viability of O. corniculata to provide a male-fertility standard. The fluorochromatic reaction (FCR) was used to determine the vigor of pollen grains (Heslop-Harrison and Heslop-Harrison, 1970; Nepi et al., 2005). Pollen grains, from five mature anthers of an individual, were released into a microslide with a drop of FDA (flurescein diacetate)-sucrose mixture and observed under a fluorescence microscope (Leitz Diaplan, filter wavelength = 459-490 nm) at 200x magnification. Pollen grains showing bright fluorescence under the microscope were denoted as active. The percentage of active pollen (pollen viability) was then calculated as the number of pollen grains with bright fluorescence divided by the total pollen grains observed. Due to varied pollen grain numbers (0-398 pollen grains) in each individual of O. corymbosa, we counted all the pollen grains of each microslide.
The DNA ploidy level of MS and SS individuals was assessed using flow cytometry. All individuals were collected from Taipei (population 3 ) and had three replications. Due to the rarity of SHS morph, no individuals were found during the experiment, and thus we were not able to analyze this floral morph. The fresh young leaves of each plant were chopped in G-buffer (containing 45 mM MgCl2-6H2O, 30 mM Sodium citrate, 20 mM MOPS [3-(N-morpholine)-propanesulfonate], 1% Triton-X 100 and 0.5% p-mercaptoenthanol; pH= 7.0) to release the nuclei and were then filtered by a 30//m nylon filter.
The nuclear suspensions were stained with propidium iodide (PI, 50 [ig/ mL), a fluorescent dye, containing RNase in an ice-bath. Samples were then analyzed with a hemacytometer (EPICS Elite ESP Flow Cytometer, Beckman Coulter Inc., USA) with an argon Laser set at 488 nm excitation. The PI fluorescence was collected in a 600-640 nm emission by a PMT4 (photomultiple tube). In each sample at least 6,000 nuclei were analyzed. The WinMDI v.2.8 free software was used to estimate the DNA peak mean using the chicken erythrocyte nuclei (CEN) solution as a standard, with DNA index ( = peak mean of sample / peak mean of standard). The analysis was conducted at the Plant Cell Biology Center, Institute of Plant and Microbial Biology, Academia Sinica, Taipei, Taiwan.
One-way ANOVA was used to analyze differences in floral morph and pollen viability among O. corymbosa populations. Afterwards, the data was analyzed by Duncan's new multiple range test (SAS Software V8.1, USA). The significant difference was defined as P<0.05.
In the 23 populations of O. corymbosa from the areas studied, no fruit or seed production was observed in any population.
In the studied area two floral morphs were observed: MS and SS morphs (Figure 2A and B, respectively). In addition to MS and SS morphs, individuals bearing flowers with a semi-homostylous morphology (SHS) (Figure 2C) were also found in the survey. These flowers had stigmas at the same height as the lower whorl of stamens (Figure 2C). No LS individuals were observed. The proportion of morph for the 23 populations is presented in Figure 3. Oxalis corymbosa populations were polymorphic, with each population containing at least two floral morphs. Of the 23 populations investigated, all had MS morphs, 21 had SHS morphs, and 16 had SS morphs. MS morph is always the dominant form.
Both yellow and white anthers were observed within all floral morphs found in Taiwan, despite the fact that they were present in variable proportions (Figure 3). In
Table 1. Percentage of viable pollen of upper and lower stamen in mid- (MS) and short-styled (SS) floral morphs of Oxalis corymbosa populations and of homostylic (Homo) O. corniculata in Taiwan.
Population no.
Morph (replicates)
Upper stamen
Lower stamen
O. corymbosa
SS (7)
0.88 ±0.01aB
0.95 ±0.01aB
0.91 ±0.01B
MS (10)
4.32 ±0.02aB
5.12 ±0.02aB
4.72 ±0.02B
MS (3)
1.29 ±0.00aB
5.61 ±0.03aB
3.45 ±0.01B
MS (3)
6.98 ±0.03aB
3.69 ±0.01aB
5.33 ±0.01B
MS (6)
3.51 ±0.02aB
1.78 ±0.01aB
4.26 ±0.02B
O. corniculata
Homo (3)
64.04 ±0.05aA
45.83 ±0.09aA
54.93 ±0.07A
*Values, mean (%) s.e., different capital superscripts reveal significant differences within anther level (column) at P = 0.05; different lower-case superscripts reveal significant differences within floral morph (row) at P = 0.05.
Botanical Studies, Vol. 51, 2010
MS morph, the percentage of individuals with yellow anthers significantly exceeded that of those with white anthers, except in population 14. In contrast, SS and SHS morphs bore only white anthers, with the exception of the populations 3, 9 and 23. Thus, among the 16 populations containing SS morph, 14 bore only white anthers while those in the remaining two populations (3 and 23) had both individuals with yellow and individuals with white anthers. Among the 21 populations containing SHS morphs, 18 bore only white anthers while those in populations 3, 9 and 23 had both individuals with yellow and those with white anthers.
Further observations of the anthers revealed that white anthers did not produce pollen, indicating the complete male sterility of the flowers that bore them. Accordingly, most of individuals with SS morphs or SHS morphs had male sterile flowers. The percentage of viable pollen in yellow anthers from MS and SS morphs of flowers is presented in Table 1. In general, pollen viability was very low for all anther levels and floral morphs. No significant difference was found in the pollen viability between pollen grains from the two sets of anthers. Flowers with MS morph sampled from the four regions had similar pollen viability ranging from 2 to 7%. The percentage of active pollen from the SS morph flower sampled from Taipei in northern Taiwan was even lower (ca. 1%; Table 1). Oxalis corniculata, the standard of male-fertile species, had about 55% of active pollen, revealing that the staining technique was working.
Histograms produced by flow cytometric analysis revealed that the DNA ploidy level (given as DNA index) was the same among all the morphs and populations analyzed (Figure 4; Table 2). No significant differences
Figure 2. Pictures of flowers, with petals and sepals removed, of a mid-styled floral morph with yellow anther (A), a short-styled floral morph with white anther (B) and a semi-homostylous floral morph with white anther (C) of O. corymbosa found in Taiwan.
Figure 3. Percentage of individuals carrying mid- (MS), short- (SS) and semi-homo stylous (SHS) floral morph with yellow (Y) or white sterile (W) anther in the 23 populations of O. corymbosa in Taiwan.
TSAI et al. ― Floral morphs, pollen viability, and ploidy level of Oxalis corymbosa in Taiwan
were observed between the nuclear DNA index of O. corymbosa SS and MS individuals, nor between flowers of the same morph with yellow or white anthers (Table 2).
T his is the first report on the composition and distribution of the floral morphs of O. corymbosa in the invasive area of Taiwan. In this survey, flowers of O. corymbosa with MS, SS or SHS morphs were observed growing in mixed populations across Taiwan while no LS morph flowers were found on the island. The reason for
the absence of the LS genotype in Taiwan is not clear but these observations are in accordance with the information on O. corymbosa specimens in Taiwan herbaria where no LS morph flowers have been recorded (Tsai and Kao, personal observation). LS individuals are also absent from North America and China (Denton, 1973; Lou et al., 2006). Accordingly, it is possible that the LS genotype of O. corymbosa has not been introduced in these regions. For the tristylous species of the Oxalidaceae, Lythraceae, and Pontederiaceae, a genetic model of floral morph determination has been proposed, in which the diallelic factor S, s interacts epistatically with M, m. The S allele factor controls the expression of short style versus non-short. The M allele factor controls mid-long locus. When S and M are recessive (ssmm), the long morph expresses. This model was put forward by Barlow (1923) and von Ubisch (1926). It was substantiated by Fisher and Mather (1943), and it has been accepted for many other tristylous species (Lewis and Jones, 1992). Therefore, an absence of long morph could be also due to the lack of sexual reproduction which limits the segregation of the LS morph. Alternatively, different genotypes may also have morphological or physiological differences resulting in different survival rates (Agren and Ericson, 1996), and the LS genotype may not have been able to adapt to the new habitat in Taiwan.
No fruit or seed production was observed in any population of O. corymbosa from the studied areas, and this result is consistent with previous reports by Huang and Liu (1993). With a strong self- and morph-incompatible system, tristylous plants may have limited sexual reproduction in the habitats in which they are introduced due to the scarcity of reciprocal floral morphs, and thus resort to asexual reproduction under these conditions (Barrett, 1992). This phenomenon has been observed in several heterostylous species, including Eichhornia carssipes (Barrett and Forno, 1982), Nymphoides peltata (Wang et al., 2005), O. pes-capraae (Castro et al., 2007), and O. corymbosa (Denton, 1973; Veldkamp, 1971; Lou et al., 2006). Oxalis corymbosa populations growing in North America have only SS and SHS flowers, and no sexual reproduction was found (Denton, 1973). Sexual reproduction in O. corymbosa is also unknown in China
(Lou et al., 2006) and Malaysia (Veldkamp, 1971). In
China both MS and SS individuals were detected, but the natural populations were monomorphic, and the spatial separation of floral morphs, together with other factors, leads to no sexual reproduction in this area (Lou et al., 2006). In contrast, Taiwan populations of O. corymbosa are polymorphic, and individuals with different floral morphs display no obvious spatial separation. Although the frequencies of the different floral morphs are not equal, there are apparently legitimate mates in the vicinity. The results suggest that other factors are involved the lack of sexual reproduction in O. corymbosa.
The occurrence of O. corymbosa individuals with white anthers has been previously reported by Wu et al. (1978),
Table 2. DNA index of individuals of mid- (MS) and short-styled (SS) floral morphs with yellow or white anther in O. corymbosa. The chicken erythrocyte nuclei (CEN) solution was used as a standard.
DNA index
0.90 ± 0.02
0.90 ± 0.02
0.93 ± 0.03
0.92 ± 0.03
*Values are Mean s.e. (n = 3).
Figure 4. Histograms of mid- (MS) and short-styled (SS) floral morphs with yellow or white anthers in O. corymbosa (A-D), and chicken erythrocyte nuclei (CEN) standard solution (E). The peak mean is estimated by WinMDI v. 2.8 to calculate the DNA index in every plant.
Botanical Studies, Vol. 51, 2010
and its significance and frequency is studied in detail in the present study. This character occurred in all three morphs of O. corymbosa flowers in Taiwan, with most of the SS and SHS morph flowers carrying white anthers. The white anthers proved to be empty with no pollen in the pollen sac. Such unique character confers to the flowers complete sterility for the male function. The factors which might cause a male sterile flower and prevent pollen production are currently under study. Additionally to the presence of anthers lacking pollen, mature yellow anthers revealed very low pollen viability. Although most of MS individuals had yellow anthers with mature pollen, only about 4% of the pollen was viable. Low pollen viability was also found in O. corymbosa populations growing in China (Lou et al., 2006). Thus, lack of pollen grains and their poor quality may be considered one of the factors limiting sexual reproduction of O. corymbosa in Taiwan.
Different ploidy levels have been shown to play a major role in limited sexual reproductions in other Oxalis species (Castro et al., 2007). In this study, no differences in ploidy level were found between SS and MS individuals of O. corymbosa in Taiwan. Thus, apparently different ploidy levels between floral morphs are not among the causes responsible for the lack of sexual reproduction in this species at this region. Due to difficulties in staining the chromosomes of O. corymbosa, we used the flow cytometry technique to analyze its DNA index, and this was revealed to be an efficient way to rapidly compare ploidy levels among plants. However, future large-scale ploidy screenings are necessary to explore in detail cytotype composition within and between populations of O. corymbosa. Furthermore, the possibility of aneuploidy in this species cannot be excluded and needs to be considered in further works as a potential factor in the lack of fruit production.
In conclusion, the absence of LS floral morph, the biased floral morph frequencies, the low pollen viability, and the presence of male sterile morphs may all contribute to the absence of sexual reproduction of O. corymbosa in Taiwan.
Acknowledgments. The authors thank the personnel of the Plant Cell Biology Center at the Institute of Plant and Microbial Biology, Academia Sinica, for providing technical support in the flow cytometry analysis. We are also grateful to Han-Ling Huang and Ying-Da Luo for helping survey floral morphs in the field. The manuscript greatly benefited from the suggestions of an anonymous reviewer. This study was partially supported by a grant (NSC 97-2313-B-002-042) from the National Science Council, Taiwan.
literature cited
Agren, J. and L. Ericson. 1996. Population structure and morph-specific fitness differences in tristylous Lythrum salicaria. Evolution 50: 126-139.
Ater, M. 2005. Biologie de la reproduction d'Oxalis pes-caprae au Maroc. In J. Menendez , F. Bastida, C. Fernandez-Quin-tanilla, J. L. Gonzales, J. Recasens, M. Royuela, et al. (eds), Malherbologia Iberica: Soluciones Comunes a Problemas Comunes. Universidad de Huelva Publicaciones, Huelva.
Barker, H. G. 1965. Characteristics and modes of origin weeds. In H. G. Barker, and G. L. Stebbins (eds.), The Genetics of Colonizing Species. Academic Press, New York, pp. 147-172.
Barlow, N. 1923. Inheritance of the three forms in trimorphic species. J. Genet 13: 133-146.
Barrett, S.C.H. and I. W. Forno. 1982. Style morph distribution in New World populations of Eichhornia crassipes (Mart.) Solms Laubach (water hyacinth). Aquat. Bot. 13: 299-306.
Barrett, S.C.H. 1992. Heterostylous genetic polymorphisms: model systems for evolutionary analysis. In S.C.H. Bar­rett (ed.), Evolution and Function of Heterostyly. Springer-Verlag, New York, pp. 1-29.
Barrett, S.C.H. 1993. The evolutionary biology of tristyly. Ox­ford Surveys in Evolutionary Biol. 9: 283-286.
Barrett, S.C.H., W.W. Cole, M. Arroyo, B. Cruzan, and D.G.
Lloyd. 1997. Sexual polymorphisms in Narcissus triandrus (Amaryllidaceae): is this species tristylous? Heredity 78:
Carniel, K. 1969. Beitrage zur entwicklungsgeschichte des an-therentapetums in der Gattung Oxalis. I Oxalis rosea und O. pubescebs. Plant Syst. Evol. 116: 423-429.
Castro, S., J. Loureiro, C. Santos, M. Ater, G. Ayensa, and L. Na-varro. 2007. Distribution of flower morphs, ploidy level and sexual reproduction of the invasive weed Oxalis pes-caprae in the western area of the Mediterranean region. Ann. Bot.
99: 507-517.
Darwin, C. 1865. On the sexual relations of the three forms of Lythrum salicaria. Biol. J. Linn. Soc. 8: 169-196.
Darwin, C. (eds.). 1877. The different forms of flowers on plants of the same species. John Murray, London, 352 p.
Darwin, C. (eds.). 1900. The effects of cross and self fertilization in the vegetable kingdom. Murray, London, UK.
Denton, M.F. (eds.). 1973. A Monograph of Oxalis, Section Ionoxalis (Oxalidaceae) in North America. Publication of Michigan state university museum, Biological series 4, pp.
Fisher, R.A. and K. Mather. 1943. The inheritance of style length in Lythrum salicaria. Ann. Eugen. 12: 1-23.
Ganders, F.R. 1979. The biology of heterostyly. New. Zeal. J.
Bot. 17: 607-635. Guth, C.J. and S.G. Weller. 1986. Pollination, fertilization
and ovule abortion in Oxalis magnifica. Am. J. Bot. 73:
Heslop-Harrison, J. and Y. Heslop-Harrison. 1970. Evaluation of pollen viability by enzymatically induced fluorescence; intracellular hydrolysis of fluorescein diacetate. Stain. Tech
nol. 45: 115-120.
Huang, T.-C. and T.-S. Liu. 1993. Oxlidaceae. In T.-C. Huang et
TSAI et al. ― Floral morphs, pollen viability, and ploidy level of Oxalis corymbosa in Taiwan
al. (eds.), Flora of Taiwan, vol. 3, 2nd ed. Editiorial Com­mittee of the Flora of Taiwan, Department of Botany, Na­tion Taiwan University, Taiper, Taiwan, pp. 397-402.
Kohn, J.R. and S.C.H. Barrett. 1992. Experimental studies on the functional significance of heterostyly. Evolution 46: 43-55.
Lewis, D. and D.A. Jones. 1992. The genetics of heterostyly. In S.C. H. Barrett (eds.) Evolution and Function of Hetero-styly. Springer-Verlag, Berlin, pp. 129-150.
Lloyd, D.G. and C.J. Webb. 1992. The Selection of Heterostyly.
In S. C. H. Barrett (ed.), Evolution and Function of Hetero-styly. Springer-Verlag, Berlin, pp. 179-207.
Lourteig, A. 1980. Flora of Panama, Part IV. 84. Ann. Mo. Bot. Gard. 67: 823-850.
Luo, S., D. Zhang, and S.S. Renner. 2006. Oxalis debilis in China: distribution of flower morphs, sterile pollen and
polyploidy. Ann. Bot. 98: 459-464.
Marco, D.E. and M.T.K. Arroyo. 1998. The breeding system of Oxalis squamata, a tristylous South American species. Bot.
Acta 111: 497-504.
Naranjo, C.A., L.M. Mola, L. Poggio, and M. Mulgura de Ro­mero. 1982. Estudios citotaxonomicos y evolutivos en es-pecies herbaceas sudamericanas de Oxlais (Oxalidaceae). I.
Bol. Soc. Argent. de Bot. 20: 183-200.
Nepi, M., M. Guarnieri, S. Mugnaini, L. Cresti, E. Pacini, and B. Piotto. 2005. A modified FCR test to evaluate pollen viabil­ity in Juniperus communis L. Grana 44: 148-151.
Roy, S.C., S. Ghosh, and A. Chatterjee. 1988. A cytological sur­vey of eastern Himalayan plants. II. Cell & Chromosome
11: 93-97.
Thompson, J.D., Y. Piller, D. Strasberg, and D. Manicacci. 1996. Tristyly in the endangered Masarene Island endemic Hugo-nia serrata (Linaceae). Am. J. Bot. 83: 1160-1168.
Veldkamp, J.E. 1971. Oxalidaceae. In C. G. J. van Steenis (ed.), Flora Malesiana ser. I. Spermatophyta 7, Wolters-Noord-hoff, Groningen, pp. 151-178.
Wang, Y, Q.F. Wang, Y.H. Guo, and S.C.H. Barrett. 2005. Repro­ductive consequences of interactions between clonal growth and sexual reproduction in Nymphoides peltata: a distylous
aquatic plant. New Phytol. 165: 329-335.
Wu, L., Y.-T. Kiang, and J. Antonovics. 1978. Field observations on Oxalis corymbosa, a widespread weedy species in Tai­wan. In Studies and essays in commemoration of the golden jubilee of Academia Sinica. Taipei, Taiwan, Republic of
China, pp. 619-626. Wu, S.-H., C.-F. Hsieh, and M. Rejmanek. 2004. Catalogue of
naturalized flora of Taiwan. Taiwania 49: 16-31.
Xu, B.-S., R.-F. Weng, and M.-Z. Zhang. 1992. Chromosome
numbers of Shanghai plants I. Invest. Stud. Nat. 12: 48-65.
Botanical Studies, Vol. 51, 2010
紫花酢漿草在台灣的花柱型分布、花粉活性及 染色體倍數之調查
蔡孟穎1 陳淑華1,2 高文媛1,2
紫花酢漿草 (Oxafc corymbosa DC.),在台灣爲外來種植物,在原生地(北美地區)具有花柱三型
性 (長花柱、中花柱和短花柱型)和結實等相關紀錄;引入台灣後,僅採用鱗莖行無性繁殖,並無結實
內有活性的花粉比率很低'少於4% 。經流測分析短花柱型和中花柱型個體之核DNA '發現兩者應具有