search
Contact Us
HOME

  • Crossref logo
  • Crossref Similarity Check logo
Journal Search Engine

to

Search Advanced Search Adode Reader(link)
Download PDF Export Citaion korean bibliography PMC previewer
ISSN : 1229-3857(Print)
ISSN : 2288-131X(Online)
Korean Journal of Environment and Ecology Vol.35 No.1 pp.48-67
DOI : https://doi.org/10.13047/KJEE.2021.35.1.48

Distribution Characteristics, Population Structure and Dynamics of the Endangered Plant, Viola websteri Hemsl.

Hyun-Hee Chae2*, Young-Chul Kim3, Myoung-Hai Kwak4, Gi-Heum Nam5
2Department of Biology, Gangneung-Wonju National University, 7, Jukheon-gil, Gangneung-si, Gangwon-do 25457, Korea
3Research Center for Natural Science, Gangneung-Wonju National University, 7, Jukheon-gil, Gangneung-si, Gangwon-do 25457, Korea
4Plants Resource Division, Biological Resources Research Department, National Institute of Biological Resource, 42, Hwangyeong-ro, Seo-gu, Incheon 22689, Korea
5Plants Resource Division, Biological Resources Research Department, National Institute of Biological Resource, 42, Hwangyeong-ro, Seo-gu, Incheon 22689, Korea
a

본 연구는 2018년 식물자원 유전자 다양성 연구(4단계 1차년도) 지원으로 연구되었음(NIBR201803102).


*Corresponding author: qhffba@hanmail.net
29/10/2020 01/01/2021 11/01/2021

Abstract


Plant species exhibit current characteristics as a result of interactions with environmental conditions. The plants of Viola sp. have selected chasmogamous flowers with vigorous vegetative propagation or development of cleistogamous flowers as an adaptation strategy. Viola websteri is distributed on the Korean peninsula and the eastern part of Jilin Province, China. The center and edge of the distribution are expected to exhibit different population-dynamics. It is necessary to investigate the cause of its current limited distribution even though V. websteri has a mixed-mating strategy. Firstly, We examined the vegetation environment of habitats and evaluated its characteristics. Growth characteristics were examined through plant phenology. We then evaluated the population structure, characteristics of chasmogamous flowers, and productivity of cleistogamous flowers. Moreover, we compared population sizes between 2014 and 2018. Most habitats were located in deciduous broadleaf mixed forests adjacent to valleys. V. websteri produced chasmogamous flowers with self-incompatibility in April-May and cleistogamous flowers in June-September. The cleistogamous flower production is a strategy ensuring seed production under uncertain environmental fluctuations; these were approximately twice as numerous as chasmogamous flowers. The population structure was distinguished into stable and very unstable regions. There were sites where the population experienced a sharp decline in the 2018 compared to that of 2014. This large decline was found in the edge populations. The habitats had different microsites depending on the natural disturbances of drought and the matrix constituting the habitat, thus supporting various plants. Ensuring the production of seeds through cleistogamous flowers, it was determined that rapid seedling re-establishment and population replenishment were possible when the natural disturbance factor was removed. Environmental factors did not equally affect all populations or individuals. Therefore, it was expected that it would be able to persisted in a long time, despite the rapid decrease in the number of individuals in the population regionally. Local extinction and re-establishment are likely to repeat according to environmental change. We propose the additional population investigation based on this works are required. We also suggest a need to assess the long-term population dynamics and the genetic characteristics of chasmogamous flowers and cleistogamous flowers to establish and implement effective conservation strategies.


CONSERVATION , REAR EDGE POPULATION , REPRODUCTION TRAITS , CLEISTOGAMOUS , PHENOLOGY

멸종위기야생식물인 왕제비꽃(Viola websteri Hemsl.)의 분포특성과 개체군 구조 및 동태

채현희2*, 김영철3, 곽명해4, 남기흠5
2강릉원주대학교 생물학과 박사과정
3강릉원주대학교 자연과학연구소 전임연구원
4국립생물자원관 식물자원과 연구관
5국립생물자원관 식물자원과 연구사

초록


식물 종은 주어진 환경조건과의 상호작용의 결과로 현재와 같은 특성을 나타낸다. 제비꽃속의 식물은 적응전략으로 개방화와 더불어 왕성한 영양번식체계를 선택하였거나 또는 폐쇄화를 발달시키는 전략을 선택하였다. 왕제비꽃은 제비꽃 속의 식물로 세계적으로 한반도와 중국 지린(길림)성의 동부에만 분포한다. 분포의 중심과 가장자리는 서로 다른 개체군 동태를 나타낼 것으로 예상된다. 왕제비꽃이 복합적인종자생산전략(Mixed-mating strategy)을 소유함에도 현재와 같은 제한적인 분포를 나타내는 원인을 알아볼 필요가 있다. 먼저 우리는 분포지의 식생환경을 조사하여 그 특성을 평가하였다. 식물계절학을 통해 생장특성을 알아보았다. 또한 개체군의 구조, 개방화의 결실특성, 폐쇄화의 생산성을 평가하였다. 그리고 2014년과 2018년 사이의 개체군크기 변화를 비교하였다. 분포지의 대부분은 계곡에 인접한 낙엽활엽수혼효림의 식생하부에 위치하였다. 4∼5월 에 자가불화합성의 개방화를 생산하였고 6∼9월까지 폐쇄화를 생산하였다. 폐쇄화의 생산은 불확실한 환경변화에서 종자의 생산을 보증하기 위한 전략으로 평가되며 개방화에 비해 약 2배의 생산량을 나타내었다. 개체군의 구조는 안정적인 지역과 매우 불안정한 지역이 구분되었다. 2014년에 비해 2018년의 조사에서 개체군의 급격한 감소를 경험한 지역이 존재하였다. 이러한 큰 폭의 감소는 가장자리 개체군에서 크게 나타나는 것으로 평가되었다. 왕제비꽃 분포지는 가뭄이라는 자연적인 교란 요인과 분포지를 구성하는 기질에 따른 다양한 미소입지의 존재로 다양한 식물을 부양하였다. 이러한 불확실한 조건에서 폐쇄화를 통해 종자의 생산을 보증하고 그에 따라 자연적인 교란요인이 제거되었을 때 빠르게 유묘의 재정착과 개체군의 보충이 가능한 것으로 판단되었다. 그리고 환경요인은 모든 개체군 및 개체군 내의 모든 개체에 동일하게 작용하지 않는 것으로 추정되었다. 따라서 지역적으로 급격한 개체수의 감소에도 불구하고 완전한 소멸에 이르지 않을 것이고 환경의 변화에 따라 국지적인 소멸과 재정착이 반복될 것으로 예상되었다. 우리는 본 연구를 통해 수집된 종의 특성에 대한 정보를 바탕으로 추가적인 분포지의 조사가 필요함을 제안하였다. 또한 효과적인 보전 전략의 수립과 시행을 위해서는 보다 장기적인 개체군 동태와 개방화와 폐쇄화의 유전적 특성을 평가하는 것이 필요함을 제안하였다.


보전 , 후방 가장자리 개체군 , 번식전략 , 폐쇄화 , 식물계절학
    Ministry of Environment(MOE)
    NIBR201803102

    INTRODUCTION

    Plants possess different species characteristics as a result of their interaction with abiotic and biotic environmental factors (Hutchinson, 1957;Pianka, 1970;Harper, 1977;Pearman et al., 2007). When characteristics of a plant species are in accord with environmental factors, it not only has a wide distribution range but also has a low risk of extinction because they have large population sizes(Hampe and Petit, 2005;Eckert et al., 2008;Hunter and Gibbs, 2011). On the other hand, populations in regions with substantial environmental changes, depending on the suitability of the given environmental conditions, may experience not only dramatic increases and declines but also temporary extirpations(Hanski, 1982;Lammi et al., 1999;Hampe and Petit, 2005;Pearson et al., 2009). Plants, especially those living in an environment with a wide range of variations, have evolved to relief these impacts(Hutchinson, 1957;Pianka, 1970;Reznick et al., 2002). Typically, they chose different life spans(Harper, 1977;Garcíia et al., 2008), life-types(Raunkiær, 1907), and reproduction strategies(Lord, 1981;Yang and Kim, 2016;Koontz et al., 2017). Therefore, the reproduction characteristics of a plant species are closely related to the environmental characteristics of the habitat selected of the species(Harper, 1977;Crawley, 1986). Consequently, it is important to understand the local characteristics of plant habitats for the purpose of conservation(Dupré and Ehrlén, 2002;Mayberry and Elle, 2010).

    Competition for resources is inevitable in all life forms on earth(Hutchinson, 1957;Crawley, 1986). The resources owned by living things are also limited and require trade-offs in all cases(Pianka, 1970;Harper, 1977;Crawley, 1986). The plants of Violaceae possess special strategies to adapt to a given environmental contexts(Newell et al., 1981;Schellner et al., 1982;Ranua and Weinig, 2010). In particular, reproductive strategies are considered as the result of very dramatic choices(Ranua and Weinig, 2010;Yang and Kim, 2016). Typically, this is the production of cleistogamous(CL) flowers(Lord, 1981). However, among the plants of Violaceae, species that possess the strategy of developing ramets as a vegetative reproduction actively utilize the stolon to produce a small number of chasmogamous(CH) flowers, but not CL flowers(Newell et al., 1981;Schellner et al., 1982). Production of CL flowers, which achieve maximum efficiency under limited resources, is a strategy for producing seeds using relatively few resources(Lord, 1981;Yang and Kim, 2016;Koontz et al., 2017). It is possible to produce as many seeds as possible at a low cost, and maintain populations and expand habitats. Therefore, it has a low risk of extinction compared to species producing a small number of seeds and having a limited distribution range by means of sexual reproduction(Lienert, 2004). Nevertheless, it has been reported that some plants in the genus Viola are endangered by natural or artificial factors(Eckstein and Otte, 2005;Moora and Jõgar, 2006). This is because of the extinction of habitats, the progression of succession, competition and inbreeding depressions. Understanding the causes of species endangerment, despite having a unique reproduction strategy, is essential for the establishment and implementation of conservation plans(Newell, 1982).

    Viola websteri Hemsl. is distributed on the Korean peninsula in East Asia and the eastern part of Jilin Province in China(Flora of China, 2020). It is assessed as being Endangered(EN) at the global level. In Korea, the Ministry of Environment has protected the species after designating it as an endangered wild plants grade II in 2005(Ministry of Environment(MOE), 2012, 2015). V. websteri has a strategy of continuously producing CL flowers, as well as CH flowers(Jang et al., 2010). This study assessed the following to determine the causes of the limited distribution of V. websteri even though it has a mixed-mating strategy. First, the vegetation characteristics of the habitats were examined. Second, the structure of each population was analyzed. Third, the productivity of CH flowers and CL flowers were compared. Fourth, population sizes assesed in 2014 and 2018 were compared. Finally, based on the collected information, the direction of additional research needed for the establishment and promotion of appropriate conservation plan will be discussed.

    MATERIALS AND METHODS

    1. Species

    V. websteri is distributed in China and the Korean peninsula. It is distributed only in the eastern part of Jilin Province in Northeast China and in North and South Korea on the Korean Peninsula(Korea National Arboretum(KNA), 2012;MOE, 2012, 2015;Flora of China, 2020;Kang et al., 2017). In accordance with the IUCN Red List Criteria, China, North Korea, and South Korea all classify the species as Endangered(EN)(KNA, 2012;MOE, 2012). V. websteri is limitedly distributed in East Asia, and the central region of the Korean peninsula is the southernmost limit of its global distribution.

    It is a perennial plant of the Violaceae, distributed mainly on the edges of mountainous areas and slopes of valleys(KNA, 2012). The rhizome robust, with dense white rootlets. Stem erect, usually simple, 30-40cm tall, puberulous, without leaves in lower part but stipulate(KNA, 2012;Flora of China, 2020). Its lanceolated or broadly lanceolated leaves are alternate and 5-12 × 1.2-5cm, abaxially subglabrous or puberulous along veins, base cuneate, decurrent to petiole, margin remotely serrate(Flora of China, 2020). Flowers bloom white or light purple in April-May, solitary in leaf axils. The flower has five petals, without hairs on the wings. Fruits from CH flowers are capsules and ripen in June-July. As the fruits of CH flowers mature, CL flowers develop(National Institute of Biological Resources(NIBR), 2017;Flora of China, 2020).

    2. Sites

    By 2017, a total of 22 populations were identified in 13 regions through field surveys (Figure 2). Two more regions were additionally identified and surveyed in 2018-2019 and 26 populations(sub-populations) were identified overall(Table 1). The habitats distributed north to Yeoncheon-gun(YN), Gyeonggi-do, south to Boeun-gun(BN), Chungcheongbuk-do, and east to Samcheok city(SD), Gangwon-do(Figure 2). The total population size surveyed by 2019 was assessed at 14,909 and the population with the most individuals was identified as Hoengseong-gun(HB), Gangwon-do(Table 1). The lowest elevation was 157m in YN, whereas the highest elevation was 1,005 m in SD, located on the ridge of Baekdudaegan(a mountain range). The average elevation of habitats was 537.6m. A total of 13 regions surveyed by 2017 were selected for the study, and the populations(subpopulations) with the largest number of individuals in each site were selected for the study(Table 1). The 2018 study was conducted on the selected populations, and one vegetation study of 4㎡ and 25㎡ plot was conducted for the population of Gapyeong-gun(GS), Gyeonggi-do in 2019(Table 1). In 2018, 16 plots were surveyed and in 2019, 4㎡(total of 17 plots) and 25㎡(total of 17 plots) vegetation study was conducted and used for analysis.

    3. Data collections

    Among the 14 populations selected as the selected sites, 1–2 plots with the highest density of V. websteri were selected(Table 1). A 25m2(5m × 5m) plot was installed at the selected sites. In the 25m2 plots, 4m2(2m × 2m) plots were installed at the points with the highest density of V. websteri. A 100m2(10m × 10m) plots of the extended quadrat was installed centering on the 25m2 plot. Plants observed in each of the established plots were recorded in layers and the coverage of the species was assessed in accordance with physiological vegetation surveys(Braun-Blanquet, 1964). The observed plants were listed by 'A synonymic list of vascular plants in Korea'(KNA and The Plant Taxonomic Society of Korea, 2007). The coverage values of observed species were not converted to ratings and the percentage values were recorded. The results of the study were listed and used to determine the significance of each plots and layer. The data were also used for Detrended correspondence analysis(DCA) to compare the vegetation differences among habitats(Lepš and Šmilauer, 2007).

    The population structures was assessed using the 4m2 plots installed at the point with the highest density of V. websteri within each population. To assess the population structures, the number of seedlings, heights, the number of leaves, CH flowers, CL flowers, and capsules produced from CH f lowers w ere measured and r eco rded(Table 2 ). The population structure was measured from May 1, 2018, in BN, until June 10, in SD, at the highest elevation above sea level. The measured results were classified according to individual class criteria and used to assess population structure(Table 2). The criteria for class to evaluate the population structure were determined by reflecting the growth characteristics observed in the habitats since 2008. Seedlings germinated from April to May of the year of the study, not only the cotyledon was observed, but also the first 1–2 leaves. The colonized and surviving individuals after germination were less than 20cm in height with 4–6 leaves in the following year and did not produce CH flowers or CL flowers. Individuals with one stem and seven or more leaves produced CH flowers and CL flowers. It has also been observed that age is difficult to measure from the class of producing CH flowers and CL flowers. This is because the size of an individual is reversed when the plant is damaged by natural or artificial influences. Thus, there were some plants with a large number of stems and leaves not producing CH flowers and CL flowers. In this study, class criteria were set by using the number of stems and leaves(Table 2).

    To collect the necessary information for phenology, YO, NB, JI and JM were selected as populations that could avoid artificial or natural influences and be periodically observed(Table 1). Among the sites, a fence was installed in the habitat of NB to prevent artificial damage by controlling human access. We set up time-lapse cameras within this population and captured approximately 40 days of footage with one frame per min. The captured images were used to supplement the observations. Emergence of stems, blooming, fruiting, producing CL flowers, and defoliation were observed, recorded, and photographed(digital camera, Canon EOS 6D Mark II). After combining and comparing the collected data, phonologic diagram were prepared.

    Self-compatibility was assessed for CH flowers in NB, where the protection facility was installed. Twenty individuals with similar sizes were selected from the population. Ten of the selected individuals were blocked insect pollinators using an insect screen with openings of 0.5 mm in diameter, and the remaining ten individuals were left for insect pollinators to visit freely. The experiment to block pollinators was conducted on May 8, 2018, when the blooming of CH flowers began in the population. At this time, the CH flowers that were fully bloomed were excluded because they might have been visited by insect pollinators. The experiment lasted for 20 days until May 28, 2018, when the maturing of CH flowers was completed, and then the screen was removed. At the completion of the protecting experiment for insect pollinators, the number of CH flowers, fruition, and the number of stems, height, and the number of leaves of the individuals selected as study targets, as well as the number of newly produced CL flowers were recorded. The measured values were used to compare the differences between individuals that had protected insect pollinators and those that had been left unprotected.

    The productivity between CH flowers and CL flowers were compared in the YO population. Fifteen individuals of different sizes were selected within the population. Each individual was given a number and a numbered marker was placed around the individual. On May 20, 2019, the number of stems of each plant, the height of each stem, number of leaves on the stems, number of CH flowers, and the number of capsules of the CH flowers were examined. Additionally, the number of CL flowers that began to produce was assessed. For the same individuals, stem height, number of leaves, and number of CL flowers produced were assessed on August 9, 2019. The collected data were used to compare the productivity of CH flowers and CL flowers. They were also used to compare the results collected in 2018 with those from 2019 within the same population.

    4. Statistical analysis

    DCA was performed using the vegetation data collected within the population(Canoco 4.53, Microcomputer Power, USA; Lepš and Šmilauer, 2007) to determine whether there was vegetational characteristics correlation among habitats. A comparison of the results of protecting and un protecting insect pollinators was performed using two sample t-tests(p<0.05) between the samples. In addition, a one-way ANOVA(p<0.05) was used to compare the productivity of the CH flowers and CL flowers, and the Tukey's honest significant difference(HSD) test (p<0.05) was performed for a post-hoc test. The number of leaves, height, number of CH flowers, number of CL flowers in the population structure plots were averaged, and the standard deviation(SD) was calculated. A comparison between the 2018 and 2019 measurement results was performed with two sample t-tests(p<0.05). Pearson correlation coefficients were calculated for the plants collected in the 4m2 population structure plots for height, number of leaves, number of CH flowers, rate of matured capsules, and number of CL flowers. All statistical analyses were performed using SYSTAT 12(Systat software Inc., USA).

    RESULTS AND DISCUSSION

    1. Vegetation difference of among habitats

    The results of the DCA varied depending on the size of the plots installed during this study(Figures 3 and 4). Among 16 of the 100m2 plots, the species observing in the tree layer were placed into 6 taxa with frequency(F) of 3 or more and an importance value(IV) of 5 or more(Table 3;Bray and Curtis, 1957). For the IV, the highest was for Acer pictum var. mono(12.77), followed by Quercus mongolica(10.87), Cornus controversa(9.57), Malus baccata(7.71), Larix kaempferi(7.14) and Fraxinus rhynchophylla(6.01)(Table 3). In the subtree layer, the high IV were for Morus bombycis(12.52), A. pictum var. mono(10.87), Acer tataricum subsp. ginnala(9.95), Acer pseudosieboldianum(8.95), C. controversa(5.52) and Ulmus davidiana var. japonica(5.04) in order. Species constituting the overstory vegetation of the V. websteri habitat were not distinguished by one or two species but were distributed under various tree layers. Such plants represent under the species constituting valleys and valley slopes(Jang et al., 2010). As can be seen from the results, the main habitats of V. websteri were constituted of the vegetation around valleys and under the deciduous broadleaf mixed forests formed at the slopes of ridges contacting the upper part of the valley(Jang et al., 2010).

    The results of DCA using vegetation data collected from 25m² plots were classified into those distributed around valleys of forest vegetation and those not(Figure 3). The results of DCA using all the observed species, including V. websteri(Figure 3A), and the results of DCA using observed species excluding V. websteri showed a similar disposition(Figure 3B). In particular, DCA results using all the observed species clearly showed that the species were distributed in the following five distinctive regions: the ridge of Baekdudaegan in SD and JI; slopes located on the upper parts of the valleys of mountains in DS; slopes adjacent to valleys in JM; the ridges of mountains in GS; and under the vegetation adjacent to the valley(Figure 3A). Therefore, V. websteri was found to be distributed in the understory of the deciduous broadleaf mixed forests adjacent to the valleys of mountainous areas(within 10m to 50m), and rarely on the slope and ridges of mountain areas.

    The results of the DCA using vegetation data collected from the 4m2 plots were different from those of the DCA using vegetation data collected from 25m2 plots(Figures 3, 4). It was not distinguished by the distance between the habitats, the habitats adjacent to the valley were classified as above and the habitats located on the ridge of the mountainous area. In particular, a relatively large difference was found between the plots installed in the same population(Figure 3A; between YY1 and YY2; NB1 and NB2; BN1 and BN2; Figure 3B; between YY1 and YY2; NB1 and NB2; BN1 and BN2). These results indicated that there were various habitats for the species or spaces where diverse plants could be distributed at a small scale within the habitats of V. websteri. As shown in the vegetation data collected from the 100m2 plots, among the plants observing in the herbaceous layer, 1 to 3 species did not dominate but various plants were distributed together(Table 3). In particular, the abundance of herbaceous plants observing in different sizes of 16 plots reached 210 taxa in 4m2 plots, 291 taxa in 25m2 plots, and 343 taxa in 100m2 plots. The average of species observing in different sizes of plots was 72.6(±18.4) taxa in 100m2 plots, 54.2(±14.9) taxa in 25m2 plots, and 28.0(±8.6) taxa in 4m2 plots(Table 4). The reasons for the distribution of such diverse plants was determined to be the existence of various habitats(β-diversity) within their distribution, and the natural disturbance that prevent the principle of competitive exclusion, whereby the diversity of species decreases when dominated by one or two species(Pickett, 1980;Wilkinson, 1999;Roxburgh et al., 2004).

    2. Growth status and population structure

    Data were collected to assess the growth status and structure in 16 populations (subpopulations) in 13 regions(Table 4). The number of individuals distributed in the 4m2 plots in each population ranged from a minimum of three individuals(YN) to a maximum of 34 individuals(JU) with an average of 13.0(±7.3) individuals(Table 4). On the other hand, the number of seedlings observed ranged from a minimum of 0(SD) to a maximum of 49(JU), with an average of 15.3(±16.7), exhibiting a large gap. An average of 3.9(±2.2) stems emerged on the individuals observed in the plots. The average height of the stems was 24.1(±7.1)cm. There was significant positive correlation between the average height and the number of CH flowers produced per individual(r=0.61, p=0.01), the number of capsules produced from CH flowers per individual(r=0.76, p<0.001), and the number of CL flower produced per individual(r=0.58, p=0.02)(Newell et al., 1981;Schellner et al., 1982;Ranua and Weinig, 2010).

    An average of 32.7(±16.9) leaves was observed per individual, and an average of 8.6(±2.0) leaves on each stem. There was a significant positive correlation between the number of leaves of each stem and the number of CH flowers produced per individual(r=0.63, p=0.01), the number of fruits produced from CH flowers per individual(r=0.77, p<0.001), and the number of CL flowers produced per individual (r=0.60, p=0.02)(Newell et al., 1981;Schellner et al., 1982). The number of CH flowers produced per individual was 3.5(±4.0)(Table 4). CH flowers were observed to success an average of 1.5(±1.3) capsule. Although 82.6(±25.8)% of the individuals that produced CH flowers produced at least one capsule from the CH flowers, only an average of 48.8(±25.2)% of CH flowers succeeded in bearing capsules. The fruition rate of CH flowers showed a obvious difference among the populations, with the lowest being 25.0% in JI and the highest at 95.7% in NB1. In particular, the fruition rate NB2, in the same population showed a obvious difference at 35.2%, but there was no significant difference between BN1(54.5%) and BN2(50.0%) in BN(Table 4). The average number of CL flowers observed at the time of this population structure study was 1.6(±1.6), and it was determined that the CL flowers began to develop at the completion of the maturing of CH flowers. In particular, as the number of capsules per individual produced from CH flowers increased, the number of CL flowers produced per individual(r=0.63, p=0.01) also showed a significant positive correlation. In other words, as the size of the plant grows, not only the production of the CH flowers but also the production of CL flowers increases(Newell et al., 1981;Schellner et al., 1982;Ranua and Weinig, 2010).

    The structure using the class of individuals(Table 2) observed in 16 sites of 4m2 plots in 13 regions exhibited a substantial difference depending on the study plots(Figure 5). Stable populations with a high proportion of seedlings and a high ratio of mature individuals capable of producing seeds were known in HB, NB1, NB2, YO, JI and JU(Figure 5). On the other hand, the population in YN was not only small in size, but also extinction was in progress because of the artificial affects. Upper class of individuals(more than class 6) were not observed in BN1 and BN2. Artificial or natural factors had recently led to the disappearance of upper-class individuals and the population was determined to be in the process of re-establishment of young individuals and restoration. In addition, in the population of SD, it was assumed that there had recently been no establishment of seedlings and the results were consistent with the absence of any seedlings observed in the 4m2plot installed for the assessment of population structure(Figure 5, Table 4). Although this study did not assess the lifespan of the individual, it was assumed that the disappearance of mature individuals occurs within the population, and the maintenance of the size of the population requires the production of seeds(seeds produced by CH flowers and CL flowers), supply into the population, and re-establishment of seedlings. In other words, the higher the proportion of upper-class individuals, the larger the number of seeds produced, and consequently, the greater the number of seedlings observed in the population and the stability of the population is maintained(Newell et al., 1981;Schellner et al., 1982;Ranua and Weinig, 2010).

    3. Self-compatibility evaluation

    V. websteri did not success the fruition of CH flowers when insect pollinators were protected(Table 5). There were no statistical differences in the heights and number of leaves selected to assess capsule maturity following the protecting of insect pollinators(Table 5). There was also no statistical difference on June 9, 2018, when the number of CL flowers produced after the fruition of CH flowers was measured. In addition, 98% of the CH flowers, for which insect pollinators were not protected, succeeded in producing capsules(Table 5). Therefore, it was determined that there was no effect on individual differences. However, in this experiment, a small number of CH flowers produced capsule in individuals with insect pollinators protected. These results indicate that as the experiment was conducted in the habitat, there was a gap in the insect screen because of the effects of rainfall, wind, and wild animals (rodents), which made insect access by pollinators possible.

    4. Phenology

    V. websteri differed in terms of the time of emergence of stems above ground because of changes in spring atmospheric temperature each year(Author's observation, unpublished). In 2018, when the study was conducted, the emergence of stems was observed for all the individuals on approximately April 1 in YO(Figure 6A). Stem growth and leaf emergence proceeded rapidly with increasing temperatures, and flowering began on April 25, 25 day after stem emerged(YO). Regarding the flowers, apart from the case of bracts on the lower part of the stem, they appeared on the part grown from the 5th-6th axillary and although there were differences according to individual, each stem had 1 to 5 CH flowers. Therefore, from the first blooming of a CH flower, it took approximately 15 day until the upper CH flower emerged and the flowering was finished. The flowering period during which the first and last flowering individuals were present in the habitat was approximately 25 days(Figure 6, Right). At the end of the CH flower emerging and blooming, stem growth ceased, and CL flowers began to format the axillary in the top of the stem. From mid-July, when the capsule ripened and the rainy season ended, additional growth of stems occurred, and CL flowers were formed on newly growing axillary(Figure 6, Right). Although the forming of CL flowers was observed until October 15, the formation of CL flowers was observed only in a small number of individuals after September(Figure 6, Right). Afew CL flowers were also observed in individuals without CH flowers. Although not shown in the diagrams of phenology, damage by herbivores was observed from early June to late June when the CH flowers were maturing. damage by herbivores varied by population, especially in habitats with low-human interference(Author's observation, unpublished). Individuals that suffered the by herbivore had secondary growth and produced additional CL flowers at the axilla of the stems at this time(Author's observation, unpublished).

    5. Productivity comparison of CH flowers and CL flowers

    In the phenology, after the fruiting and ripening of the CH flowers, the continuous formation and maturation of the CL flowers were observed. Nevertheless, the 2018 study did not evaluate the quantity of the formation of additional CL flowers. Accordingly, the formation of CL flowers was observed after the formation and maturation of the CH flowers in the population of YO in 2019(Culley, 2002).

    The individuals selected for the study were located in the 4m2population study plot in 2018. Each individual had 1 to 4 stems and 1.73(±0.96) stems on average(Table 6). The mean stem height was 47.62(±8.39)cm at the time of CH flower maturity and 50.08(±8.25)cm when measured on August 9, the time when the degree of CL flower formation was examined, when they exhibited stem growth of approximately 2cm(Table 6). There was no statistical difference(df=50, p=0.29). On the other hand, the number of leaves averaged 11.40(±0.92) per stem at the time of CH flower maturity and 14.19(±3.19) per stem on August 9 when CL flower formation was evaluated(Table 6). The number of leaves showed a clear increase(df=50, p<0.001). The number of CL flowers was 0.96(±0.45) per stem at the time of CH flower maturity and 4.04(±1.82) per stem on August 9(df=50, p<0.001). Clear differences were observed because the CL flowers were produced continuously from mid-June to late July after the CH flowers had matured.

    The number of resulting CH flowers was 3.50(±1.70) per stem, and 95.6% of these CH flowers were successful in fruiting, with 3.3(±1.77) capsules per stem(Table 6). There was a clear difference between the total number of CL flowers measured(5.00±2.08) and the number of CH flowers produced(3.50±1.70) and the number of capsules that came to fruition(3.35±1.77)(one-way ANOVA, df=2, p<0.01). There were also differences between the total number of CL flowers produced and the number of CH flowers produced per stem(p＜0.05) and the number of capsules that came to fruition(p＜0.01), indicating that the production of CL flowers was more active(Figure 7A).

    The percentage of CH flower fruitions showed a significant difference between 2018 and 2019(Figure 7B). The number of leaves per individual selected to compare the fruition rate of CH flowers was markedly different between 2018(14.5±2.0) and 2019(11.0±0.9) (df=47, p<0.001). Nevertheless, the number of CH flowers formed per stem did not differ between 2018 (4.3±1.7) and 2019 (3.6±1.6) (df=46, p=0.19). On the other hand, the number of fruiting capsules showed a clear difference between 2018(1.3±1.2) and 2019(3.5±1.7) (df=46, p<0.001). Thus, the rate of fruition not only varies from year to year in the habitat of V. websteri but also depends on the sites of the distributed population(Table 4).

    6. Population dynamics

    After 2014, when the detailed evaluation of the distribution status of V. websteri was made, the re-evaluation for this survey revealed that there were populations size with large changes(Figure 8). As representative examples, there was a sharp decrease in the population in BN, CS and YY(Figure 8). Population reduction at a small scale was observed in JI, JM and HB. On the other hand, population size increased in DS, NB, HG, SD and YO(Figure 8). These decreases and increases in population size were considered to be related to recent environmental fluctuation. As shown in the results of the evaluation of the population structure, the population size decrease was observed in all the populations showing the unstable population structure. In addition, only a few seedlings were found to have successfully re-established in the subsequent observations of the populations with high seeding rates. From these results, it was determined that seedling re-establishment and re-establishment did not succeed in many populations. The failure of seedling re-establishment is thought to have been caused by drought in spring and early summer, which lasted from 2016 to 2018 in recent years. Sustained droughts can also affect the growth of mature individuals. Large plant biomass increases the total amount of resources required(Harper, 1977). If the lack of moisture persists, which is critical to the plant's life-sustaining activities, it may also affect mature individuals, leading to a reduction in the size or death of the plant(Harper, 1977). The drought sustained in the spring is not expected to be at the same level throughout the Korean peninsula, and accordingly, it is thought to have different levels of effects according to the population extent(Moora and Jõgar, 2006). As a result in populations greatly affected by the drought, comparatively large reductions in population size would be observed and in populations distributed in regions capable of offsetting the drought effects, little or no effect was observed(Figure 8).

    7. Concluding of Results

    The sites of V. websteri habitat is known to be distributed in valleys, slopes, and ridges in mountain area (KNA, 2012;MOE, 2012;Jang et al., 2010;Song et al., 2010). According to the results of this study, most were distributed in valleys and slopes adjacent to valleys in mountain areas, and three populations(GS, JI, SD) were distributed in mountain ridges. Some documents describe the habitat of V. websteri to be distributed in damp valleys and slopes(NIBR, 2012;2017). However, the sites of V. websteri distribution is considered to be the location formed by the accumulation of aggregates during the past valley formation underneath the vegetation adjacent to the valleys(Author's observation, unpublished). Although this study did not analyze the matrix constituting the habitat, the habitats formed in the valleys are considered to have a high ratio of rock, gravel, and sand. Jang et al.(2010) found that the proportion of sand in soil(less than 2mm) averaged 59.49% in the seven study areas, reporting a high proportion of coarse particles. As the ratio of sand was high, this was examined to be the reason for the decrease in the field capacity(Jang et al., 2010). In addition, in the vegetation environment and population structure study of habitats, a high proportion of sand was reported in habitats in JI and SD, where rock and gravel were not found as a matrix constituting habitats(Jang et al., 2010). Therefore, it is thought to exhibit a similar matrix composition as valley habitats. In reviewing the matrix observed in the habitat of V. websteri and the results of prior studies, it was evaluated that the environment with high moisture and a high content of soil moisture cannot be formed under the condition of a high sand ratio. Sites with a high proportion of sand are well drained, such as the habitat of V. websteri, and can inevitably be much affected by drought(Chae et al., 2017). However, depending on the location of the rocks and gravel that constitute the habitat matrix and the micro-sites formed among the rocks and the rocks and gravel, the effect on the dry conditions could vary. Periodic drying season occurring at these sites are considered to serve as an ecological process in which various plants can coexist; thus, preventing the operation of the principle of competitive exclusion(Rejmánek et al., 2004;Chae et al., 2017). As shown in the results of this study, which is considered to be consistent with the finding that V. websteri habitat is in communities with a high level of plant richness(Jang et al., 2010;Song et al., 2010).

    Plants possess an asexual reproduction strategy, as well as the usual sexual reproduction as a strategy to adapt to highly fluctuating environments and maintain the change and balance of the two reproduction systems as needed(Ranua and Weinig, 2010;Yang and Kim, 2016). Many plants also possess a mixed-mating strategy to ensure seed production and dispersal even under uncertain circumstances(Mayers and Lord, 1983;Ranua and Weinig, 2010;Koontz et al., 2017). V. websteri also possesses a mixed-mating seed production strategy of CH flowers and CL flowers(Tables 4, 5, 6; Figure 7). Because they possessed a strategy to produce CL flowers, vegetational reproduction by the development of stolon did not develop(Newell et al., 1981;Schellner et al., 1982). As plant size increases, short rhizomes develop and the number of stems emerging above ground increases(Table 4). Viola spp., which have a strong production of CL flowers, have been r eported t o be highly a daptable t o changes i n the environment, including natural disturbances, and to increase population size rapidly in disturbed spaces(Newell, 1982). In addition, Viola spp. have been reported to show increased mortality with increasing age and growing size of plants(Schellner et al., 1982). Thus, the habitat of V. websteri has a natural disturbance because of drought, and it is believed that this species possesses a strategy of producing CL flowers as a strategy to ensure seed production and the re-establishment of seedlings from these disturbances.

    Dry conditions, a natural disturbance that occurs periodically in V. websteri habitats, can interrupt the growth and expansion of competitive species and also prevent plant invasion that requires a long time and a lot of resources for establishments(Grubb, 1977;Pickett and Thompson, 1978;Pickett, 1980;Roxburgh et al., 2004). On the other hand, V. websteri itself can also be affected and sustained dry conditions could lead to rapid population size reduction(Figure 8;Newell et al., 1981). Nevertheless, V. websteri is capable of seedling growth and re-establishment faster than any other species when dry conditions, a natural disturbance, have discontinued(Grubb, 1977). In this study, the decrease and increase of individuals number were observed differently according to the populations(Figure 8). This is thought to indicate that recent droughts do not have the same magnitude of affect across the entire habitat(Newell et al., 1981). In particular, substantial decreases in the populations of BN, YN and YY, which correspond to the peripheral area of the V. websteri distribution extent is thought to be caused by the fact that the recent drought had a greater impact on these populations.

    Population size repeats the increases and decreases depending on the sites and species characteristics of the V. websteri habita (Evju et al., 2010). Thus, bo ttleneck e ffect can easily occur(Eckstein and Otte, 2005;Moora and Jõgar, 2006). In addition, in Korea, each population is located in a remote location from each other and accordingly, they are distributed with a high degree of isolation(Figure 1). Therefore, the exchange of genes between populations is considered to be impossible. Consequently, the genetic distance between the population is distant, whereas the genetic diversity within the populations is expected to be low. Recent research by Kang et al.(2017) reflected these expectations. Despite the low genetic diversity within the populations and high variation among the population, it is expected that the populations will be sustained by the reproductive strategy of V. websteri. However, if the dry conditions(drought) during spring and early summer are sustained over a long period of time as in recent years, local extinction is likely to occur from the periphery of the distribution extent.

    Low genetic diversity of endangered plant populations is used as an indicator of high extinction risk(Lammi et al., 1999;Pearson et al., 2009). However, species that possess selective reproduction strategies, such as V. websteri, or species that have developed a vegetational reproduction strategy, have retained their populations over a long period of time, despite low genetic diversity(Lienert, 2004). In addition, high genetic distances between populations can be evaluated as important processes that promote species differentiation(Hampe and Petit, 2005;Pfeifer et al., 2010). In addition, inbreeding depressions and outbreeding depressions have been reported(Eckstein and Otte, 2005). Therefore, the promotion of introduction to increase genetic diversity needs to be avoided(Eckstein and Otte, 2005).

    The collection of additional information is needed for the effective conservation of V. websteri, which shows a limited distribution on the Korean peninsula and Jilin Province in China(Newell, 1982). First, it is necessary to investigate the dynamics of the populations on a long-term basis(Crawley and Ross, 1990;Newell, 1982;Kim et al., 2018). Second, it is necessary to evaluate the germination properties of each seed produced from CH and CL flowers(Ranua and Weinig, 2010;Baskin and Baskin, 2017). Third, it is necessary to evaluate the genetic characteristics of individuals grown from seeds produced from CH and CL flowers(Jang et al., 2010;Kang et al., 2017). This is expected to serve as important data for evaluating the current genetic characteristics of the V. websteri population. Fourth, most of the seeds of the Viola spp. are known to possess the strategy of spreading their seeds as the ballistic dispersal or by the elaiosomes attached to the seeds attracting ants to spread the seeds(Myrmecochorous)(Ohkawara and Higashi, 1994). However, the two seed dispersal strategies have been reported to reach no more than 1.5m and 2m, respectively(Ohkawara and Higashi, 1994). Therefore, studies of long-distance seed dispersal(LDD) reaching several kilometers are needed(Nathan et al., 2008;Hampe, 2011;Pellerin, 2016). The study of seed dispersal should be conducted to be able to understand the formation of habitats at relatively remote distances in many subpopulations of HB, HG and YY.(Hanski, 1998;Myers et al., 2004). Finally, plant species consisting of a small number of populations are not readily extinguished if they possess the characteristics of a metapopulation(Pickett and Thompson, 1978;Hanski, 1998). Therefore, it is necessary to investigate the distribution of other populations within a certain range centering on the current habitat. In these additional distribution investigations and surveys, information, including species and environmental characteristics of the habitat obtained from this study will serve as a useful basis(Elzinga et al., 1998;Kim et al., 2016).

    ACKNOWLEDGEMENT

    본 논문은 정부(환경부)의 재원으로 환경부 국립생물자원관 의 2018년 식물자원 유전자 다양성 연구(4단계 1차년도) 지원 을 받아 수행하였다(NIBR201803102). 연구의 모든 과정은 왕제비꽃 분포지에 미치는 영향이 최소화 될 수 있도록 계획하 였고 또한 진행하였다.

    Figure

    KJEE-35-1-48_F1.gif

    Photographies of different life stages of V. websteri, A ; adult with chasmogamous flowers, B ; chasmogamous flowers, C ; chasmogamous fruit(capsule), D ; cleistogamous flower, E ; seedling, F ; juveniles(established individuals survived from seedling).

    KJEE-35-1-48_F2.gif

    Distribution maps of V. websteri in Korea, A is selected population to investigate for the studies, black circle; selected population for this study, gray circle; observed population during 2007 to 2019.

    KJEE-35-1-48_F3.gif

    DCA ordination with collected vegetation data from 17 plots(25㎡), A was used with observed species, B was used with observed species except V. websteri in the plots(included one of 25㎡ plot in GS, 2019).

    KJEE-35-1-48_F4.gif

    DCA ordination with collected vegetation data from 17 plots(4㎡), A w as u sed with o bserved s pecies, B was used with observed species except V. websteri in the plots(included one of 4㎡ plot in GS, 2019).

    KJEE-35-1-48_F5.gif

    Comparison of population structures observed in the 14 sites(16 plots), YY and NB was established each of 2 study plots.

    KJEE-35-1-48_F6.gif

    Phenology, Left - photographs of growing features observed in the habitats, A; emergence of shoots, B; emergence of leaves, C; stem growing, D; chasmogamous flowering, E and F; chasmogamous capsule maturing and generation of cleistogamous flower, G; matured chasmogamous capsule opening, H and I; cleistogamous capsules generation and maturing, Right - Diagram of phenology, abbreviations are related to Table 1.

    KJEE-35-1-48_F7.gif

    A; comparison of ability between chasmogamous flowers(fruits) and cleistogamous fruits, B; comparison of leaf production, chasmogamous flowers production and chasmogamous fruits production between 2018 and 2019 in the habitat(YO).

    KJEE-35-1-48_F8.gif

    Variation of 13 selected population sizes between 2014 and 2018, population sizes of DS and HB are re-assessed in 2019.

    Table

    The list of selected sites and populations to study of distributional characteristics and population structures

    The classification of individuals observed in the plots for population structure

    The list of observed species in 100㎡ plots, tree layer and sub-tree layer species are observed more than 3 times and has more than 5.0 importance values

    The list of collected from 4m2 plots(16 plots) which installed to access population structure

    The evaluation of protection for visiting of pollinators in the habitat

    Comparison of annual stem growth and fruit success in 2019, observed in the YO

    Reference

    1. Baskin, J.M. and C.C. Baskin (2017) Seed germination in cleistogamous species: Theoretical considerations and a literature survey of experimental results. Seed Science Research 2: 84-98.
    2. Braun-Blanquet, J. (1964) Pflanzensoziologie. 3. Springer-Verlag, New York, 865pp.
    3. Bray, R.J. and J.T. Curtis (1957) An ordination of the upland forest communities of southern Wisconsin. Ecological Monographs 27: 325-349.
    4. Chae, H.H. , H.U. Ji, J.H. Min and Y.C. Kim (2017) Report on the monitoring of special protection zone in Odaesan National Park-Iris odaesanensis Y.N. Lee, Paeonia obovata Maxim-. Journal of National Park Research 8: 8-23. (in Korean and English abstract)
    5. Crawley, M.J. and G.J.S. Ross (1990) The population dynamics of plants [and discussion]. Philosophical Transactions of the Royal Society B Biological Sciences 330: 125-140.
    6. Crawley, M.J. (1986) Plant ecology. Blackwell Scientific Publications, Oxford, 496pp.
    7. Culley, T.M. (2002) Reproductive biology and delayed selfing in Viola pubescens(Violaceae), an understory herb with chasmogamous and cleistogamous flowers. International Journal of Plant Sciences 163: 113-122.
    8. Dupré, C. and J. Ehrlén (2002) Habitat configuration, species traits and plant distributions. Journal of Ecology 90: 796-805.
    9. Eckert, C.G. , K.E. Samis and S.C. Lougheed (2008) Genetic variation across species’ geographical ranges: The centralmarginal hypothesis and beyond. Molecular Ecology 17: 1170-1188.
    10. Eckstein, R.L. and A. Otte (2005) Effects of cleistogamy and pollen source on seed production and offspring performance in three endangered violets. Basic and Applied Ecology 6: 339-350.
    11. Elzinga, L. , D.W. Salzer and J.W. Willoughby (1998) Measuring & monitering plant populations. U.S. Bureau of Land Management, Lincoln, USA, 496pp.
    12. Evju, M. , R. Halvorsen, K. Rydgren, G. Austrheim and A. Mysterud (2010) Interaction between local climate and grazing determine the population dynamics of the small herb Viola biflora. Oecologia 163: 921-933.
    13. Flora of China (2020) . http://www.efloras.org/florataxon.aspx?flora_id=2&taxon_id=200027868, 2020.1.15.
    14. Garcíia, M.B. , X. Picóo and J. Ehrléen (2008) Life span correlates with population dynamics in perennial herbaceous plants. American Journal of Botany 95(2): 258-262.
    15. Grubb, J.P. (1977) The maintenance of species-richness in plant communities: The importance of the regeneration niche. Biological Review 52: 107-145.
    16. Hampe, A. and R.J. Petit (2005) Conservation biodiversity under climate change: The rear edge matters. Ecology Letters 8: 461-467.
    17. Hampe, A. (2011) Plants on the move: The role of seed dispersal and initial population establishment for climate-driven range expansions. Acta Oecologica 37: 666-673.
    18. Hanski, I. (1982) Dynamics of regional distribution: The core and satellite species hypothesis. Oikos 38: 210-221.
    19. Hanski, I. (1998) Metapopulation dynamics. Nature 396: 41-49.
    20. Harper, J.L. (1977) Population biology of plants. Academic Press, London, 892pp.
    21. Hunter Jr, M.L. and J. Gibbs (2011) Fundamentals of conservation biology. Blackwell Publishing, Oxford, 497pp.
    22. Hutchinson, G.E. (1957) Concluding remarks. Cold Spring Harbor Symposia on Quantitative Biology 22: 415-427.
    23. Jang, S.K. , K.S. Cheon, K.A. Kim, J.H. Jang and K.O. Yoo (2010) Ecological characteristics of Viola westeri Hemsley habitats. Korean Journal of Plant Resources 23: 261-273.(in Korean and English abstract)
    24. Kang, J.S. , X. Zhang and M.H. Kwak (2017) Isolation and characterization of microsatellite markers for Viola websteri (Violaceae) and cross-species amplification within the Genus Viola. Plant Molecular Biology Reporter 35: 145-153.
    25. Kim, Y.C. , H.H. Chae, B.R. Hong, H.K. Oh, K.H. Lee and K.S. Lee (2016) The status of endangered plants distributed in the middle eastern area of Korea and evaluation of the risk factors. Korean Journal of Environment and Ecology 30: 291-307. (in Korean and English abstract)
    26. Kim, Y.C. , H.H. Chae, G.H. Nam and K.S. Lee (2018) Adapting strategy of biennial endangered Halenia corniculata (L.) Cornaz to environmental stress and its implications for conservation plan. Journal of Plant Biology 61: 177-185.
    27. Koontz, S.M. , C.W. Weekley, S.J. Haller Crate and E.R. Menges (2017) Patterns of chasmogamous and cleistogamous, a mixed-mating strategy in an endangered perennial. AoB PLANTS 6: 1-9.
    28. Korea National Arboretum(KNA) (2012) Rare plants in Korea. Korea National Arboretum, Gyeonggi-do, Korea, 412pp.(in Korean)
    29. Korean National Arboretum(KNA) and The Plant Taxonomic Society of Korea (2007) A synonymic list of vascular plants in Korea. Korea National Arboretum, Pocheon.
    30. Lammi, A. , P. Siikamäki and K. Mustajärvi (1999) Genetic Diversity, Population Size, and Fitness in Central and Peripheral Populations of a Rare Plant Lychnis viscaria. Conservation Biology 13(5): 1069-1078.
    31. Lepš, J. and P. Šilauer (2007) Multivariate analysis of ecological data using CANOCO. Cambridge University Press, New York, 269pp.
    32. Lienert, J. (2004) Habitat fragmentation effects on fitness of plant populations-a review. Journal for Nature Conservation 12(1): 53-72.
    33. Lord, E.M. (1981) Cleistogamy: A tool for the study of floral morphogenesis, function and evolution. The Botanical Review 47: 421-449.
    34. Mayberry, R.J. and E. Elle (2010) Conservation of a rare plant requires different methods in different habitats: Demographic lessons from Actaea elata. Oecologia 164: 1121-1130.
    35. Mayers, A.M. and E.M. Lord (1983) Comparative flower development in the cleistogamous species Viola odorata. Ⅰ. A growth rate study. American Journal of Botany 70: 1548-1555.
    36. Ministry of Environment(MOE) (2012) Law of wild animal and plants protection. Ministry of Environment.
    37. Ministry of Environment(MOE) (2015) Law of wild animal and plants protection. Ministry of Environment.
    38. Moora, M. and Ü. Jõgar (2006) Competitive responses of the rare Viola elatior and common Viola mirabilis. Plant Ecology 184: 105-110.
    39. Myers, J.A. , M. Vellend, S. Gardescu and P.L. Marks (2004) Seed dispresal by white-tailed deer: Implications for long-distance dispersal, invasion, and migration of plants in eastern North America. Oecologia 139: 35-44.
    40. Nathan, R. , M. Schurr, O. Spiegel and O. Steinitz (2008) Mechanisms of long-distance seed dispersal. Trends in Ecology and Evolution 23: 638-647.
    41. National Institute of Biological Resources(NIBR) (2012) Red data book of endangered vascular plants in Korea. National Institute of Biological Resources, Incheon, 390pp.(in Korean)
    42. National Institute of Biological Resources(NIBR) (2017) An overview of endangered wild organisms. National Institute of Biological Resources, Incheon, 593pp.(in Korean)
    43. Newell, S.J. (1982) Demography of Newfoundland population of three Viola species. Canadian Journal of Botany 61: 2581-2591.
    44. Newell, S.J. , O.T. Solbrig and D.T. Kincaid (1981) Studies on the population biology of the Genus Viola: Ⅲ. The demography of Viola blanda and Viola pallens. Journal of Ecology 69: 997-1016.
    45. Ohkawara, K. and S. Higashi (1994) Relative importance of ballistic and ant dispersal in two diplochorous Viola species(Violaceae). Oecologia 100: 135-140.
    46. Pearman, P.B. , A. Guisan, O. Broennimann and C.F. Randin (2007) Niche dynamics in space and time. Trends in Plant Science 23: 149-158.
    47. Pearson, G.A. , A. Lago-Leston, M. Valente and E.A. Serrão (2009) Frayed at the edges: Selective pressure and adaptive response to a biotic s tressors a re m ismatched i n low diversity edge populations. Journal of Ecology 97: 450-462.
    48. Pellerin, M. , M. Picard, S. Saïid, E. Baubet and C. Baltzinger (2016) Complementary endozoochorous long-distance seed dispersal by three native herbivorous ungulates in Europe. Basic and Applied Ecology 17: 321-332.
    49. Pfeifer, M. , N.G. Passalacqua, S. Bartram, B. Schatz, A. Croce, P.D. Carey, H. Kraudelt and F. Jeltsch (2010) Conservation priorities differ at opposing species borders of a European orchid. Biological Conservation 143: 2207-2220.
    50. Pianka, E.R. (1970) On r-and K-Selection. The American Naturalist 104: 592-597.
    51. Pickett, S.T.A. and J.N. Thompson (1978) Patch dynamics and the design of nature reserves. Biological Conservation 13(1): 27-37.
    52. Pickett, S.T.A. (1980) Non-equilibrium coexistence of plants. Bulletin of the Torrey Botanical Club 107(2): 238-248.
    53. Ranua, V.A. and C. Weinig (2010) Mixed-mating strategies and their sensitivity to abiotic variation in Viola lanceolata L.(Violaceae). The Open Ecology Journal 3: 83-94.
    54. Raunkiær, C. (1907) Planterigets Livsformer og deres Betydning for Geografien. Gyldendalske Boghandel-Nordisk Forlag, København and Kristiania, 132pp.
    55. Rejmánek, M. , E. Rejmánková and W. Holzner (2004) Species diversity of plant communities on calcareous screes: The role of intermediate disturbance. Preslia 76: 207-222.
    56. Reznick, D. , M.J. Bryant and F. Bashey (2002) r- and K-selection revisited: the role of population regulation in life-history evolution. Ecological Society of America 83:1509-1520.
    57. Roxburgh, S.H. , K. Shea and J.B. Wilson (2004) The intermediate disturbance hypothesis: Patch dynamics and mechanisms of species coexistence. Ecology 85: 359-371.
    58. Schellner, R.A. , S.J. Newell and O.T. Solbrig (1982) Studies on the population biology of the Genus Viola: Ⅳ. Spatial pattern of ramets and seedlings in three stoloniferous species. Journal of Ecology 70: 273-290.
    59. Song, J.M. , G.Y. Lee, N.Y. Kim and J.S. Yi (2010) Vegetation structure and site environment of natural habitat of an endangered plant, Viola westeri. Journal of Korean Forest Society 99: 261-276. (in Korean and English abstract)
    60. Wilkinson, D.M. (1999) The disturbing history of intermediate disturbance. Oikos 84: 145-147.
    61. Yang, Y.Y. and J.G. Kim (2016) The optimal balance between sexual and asexual reproduction in variable environment: A systematic review. Journal of Ecology and Environment 40: 12.
    Copyright © Korean Society of Environment & Ecology (KSEE). All right reserved.