Tài liệu Uftai ve tai day25977

MINISTRY OF EDUCATION AND TRAINING VIETNAM ACADEMY OF SCIENCE AND TECHNOLOGY GRADUATE UNIVERSITY SCIENCE AND TECHNOLOGY ---------------------------- VAN HONG THIEN BUILDING PHYLOGENETIC TREES FOR THE ARACEAE IN SOUTHERN VIETNAM BASED ON MORPHOLOGICAL AND MOLECULAR MARKERS Major: Ecology Code: 62 42 01 20 SUMMARY OF BIOLOGY DOCTORAL THESIS Ho Chi Minh City – 2017 The work was realized in Graduate University of Science and Technology, Vietnam Academy of Science and Technology. Advisor 1: Dr. Lưu Hong Truong Advisor 2: Dr. Nguyen Phi Nga Examiner 1: Assoc. Dr. Nguyen Van Ket Examiner 2: Assoc. Dr. Truong Thi Dep Examiner 3: Dr. Vu Ngoc Long The thesis will be evaluated by doctoral committee at Graduate University of Science and Technology, Vietnam Academy of Science and Technology on …… 2017. The thesis is available at: - Library of Graduate University of Science and Technology - National Library of Vietnam LIST OF ABBREVIATIONS CI : Consistency index DNA : Deoxyribonucleic acid HN : Herbarium of Institute of Ecology and Ecological Resources HNU : Herbarium of University of Science, Vietnam National University, Ha Noi IGS : Intergenic spacer ISI : International Scientific Indexing K : Herbarium of Kew MEGA : Molecular Evolutionary Genetics Analysis P : Herbarium of Paris PAUP : Phylogenetic Analysis Using Parsimony PCR : Polymerase Chain Reaction PHH : Herbarium of University of Science, Vietnam National University Ho Chi Minh City RC : Rescaled consistency index RI : Retention index SGN : Herbarium of Southern Institute of Ecology VNM : Herbarium of Institute of Tropical Biology Sect. : Section 1 INTRODUCTION 1. The necesstiy of the thesis Araceae in Vietnam were first systemerized by the French botanist Gagnepain in 1942 (Gagnepain, 1942). Pham-Hoang (2000) in the book "An Illustrated Flora of Vietnam" (Volume III) as well as Nguyen (2005) in the "Checklist of Plant Species of Vietnam" (Volume III, edited by Nguyen Tien Ban) provided the brief information on the taxonomy of species in Araceae. Most recently, the monograph work on Araceae in Vietnam was conducted by Nguyen (2006), according to which the author made a list of 116 species belonging to 23 genera in Vietnam. However, much of the research samples in Nguyen Van Du's work (2006) were collected in the northern regions of Vietnam. In addition, since 2006, there have been 13 new species and 4 new records of Araceae in Vietnam (mostly in the South) which were published by many authors, thereby, increasing the total number of species of Araceae in Vietnam up to 133 (Nguyen 2007 & 2008; Hetterscheid, 2006 & 2013; Nguyen & Croat, 2010; Gong et al., 2012; Bruggeman et al., 2013; Nguyen & Bui, 2013; Nguyen et al., 2013; Luu et al., 2013 & 2014; Hoang et al., 2016; Van et al., 2016). Moreover, many species and genera in Araceae have currently changed and have not been unified in the classification (Nguyen, 2006; Gusman & Gusman, 2006; Cusimano et al., 2011; Hetterscheid & Claudel 2012; Nauheimer et al., 2011; Nauheimer & Boyce, 2013). Thus, the knowledge of Araceae in Vietnam, especially in the South of Vietnam, is still incomplete, many species are propably unknown or unclear in terms of classification. Moreover, in Vietnam, no reseach has been conducted on the use of morphological characteristics and especially molecular markers for the purpose of constructing the phylogenetic tree for systematizing and providing taxonomic information of Araceae. For the above reason, we decided to implement the project: “Building phylogenetic trees for the Araceae 2 in southern Vietnam based on morphological and molecular markers”. 2. Goal of the subject Building phylogenetic trees for the Araceae in southern Vietnam based on morphological and molecular markers, thereby (1) indicating the evolutionary relationship between taxa in Araceae; (2) re-classifying and/or confirming the exact identification for ambiguous species; (3) recording the new species of Araceae for the flora in Vietnam. 3. The main contents of the study Collecting plant materials; observation of specimens in hebariums; identification and decription of materials; amplification and sequencing of trnL-trnF IGS and matK chloroplast DNA regions; data processing methods. CHAPTER 1: LITERATURE REVIEW 1.1. Introduction to Aracea Araceae is a family of Monocotyledon of the Magnoliophyta with about 2824 species of 107 genera (Govaert & Frodin, 2002) and was first described in 1789 by Jussieu (Jussieu, 1789). Araceae has very featured morphological characteristics, which there is a spathe varied in shape, size and color. Spathe is responsible for protecting spadix and is usually divided into two parts, limb and tube. Another characteristic of the Araceae is spadix; spadix is divided into two forms: bisexual and unisexual, unisexual spadix is the one that carries only famale or male flowers, while bisexual spadix is a type that carries both male and female flowers. Spadix of many species in Araceae in general and most of the species in southern Vietnam in particular have a sterile portion located between the male and female part or at the end. In addition, flowers of Araceae are divided into two types, including bisexual and unisexual flowers. According to many studies, bisexual flowers often appear in subfamilies of ancient classifications such as Pothoideae, 3 Monsteroideae and Lasioideae; while the unisexual flowers are more common in the subfamily of higher evolutionary levels, such as Aroideae (Mayo et al., 1997; Nguyen, 2006; Boyce et al., 2012). 1.2. Research status with taxanomic Araceae 1.2.1. The studies on taxanomic Aracea in the world Araceae was first described in 1789 by Jussieu, in which he categorized all climbing species into a group called Pothos L., while terrestrial species were classified as Arum L. and Dracontium L. (Jussieu, 1789). Schott (1860) set up the first classification system for the Araceae. This system was then modified and developed by many researchers (Hooker, 1883; Engler, 1876b & 1820b; Grayum, 1990; Bogner & Nicolson, 1991; Mayo et al., 1997). In the field of molecular marker application research in Araceae, the first work was done by French et al. (1995). Later, further research into the construction of the phylogenetic tree for Araceae were also conducted by several authors, for example, Barabe et al. (2002) and Cabrera et al. (2008) built the phylogenetic tree for Araceae based on a combination of two non-coding regions of the chloroplast DNA. Cusimano et al. (2011) and Nauheimer et al. (2012) used phylogenetic tree of Cabrera et al. (2008) and some more genera to form a phylogenetic tree for Araceae based on chloroplast markers. 1.2.1. Studies on the taxonomic Aracea in Vietnam Gagnepain (1942) first described Araceae in Vietnam in the "Flore Général de L'Indochine". Many of the later studies of Araceae in Vietnam were carried out by a number of authors (Pham-Hoang, 1993 & 2000, Nguyen, 2005). Nguyen (2006) reported that in Vietnam there were 116 species belonging to 23 genera. From 2006 to date, there have been 13 new species and 4 new records of Araceae published in Vietnam. Among them, genus Arisaema have four new species and three new records, followed by genus Amorphophallus with four new species and one new records, 4 Typhonium and Alocasia with two new species and one new species for Rhaphidophora (Hetterscheid, 2006; Nguyen, 2008; Nguyen & croat, 2010; Gong & Li, 2012; Bruggeman et al., 2013; Hetterscheid & Claudel, 2013; Luu et al., 2013 & 2014; Nguyen et al., 2013; Nguyen et al., 2015; Nguyen & Quang, 2015; Nguyen et al., 2016; Van et al., 2016). CHAPTER 2. CONDITION AND METHOD 2.1. Research subject The species of Araceae is distributed in southern Vietnam, from Da Nang City. 2.2. Research Method 2.2.1. Field survey method The field survey method is implemented in the following steps: (1) Listing the number of species available in southern Vietnam; (2) selecting sample location; (3) recording the distribution of species; (4) documentary photography; (5) collecting samples for dried specimens; (6) DNA sample collection. 2.2.2. Method in the laboratory 2.2.2.1. Method of morphological characterization includes steps: (1) observation of specimens in hebariums (HN, VNM, HNU, SGN, PHH, P and K); (2) species identification by morphological comparison method; (3) building morphological characteristics of species in southern Vietnam. 2.2.2.2. Molecular method includes the steps: (1) selection of molecular markers in research of building phylogenetic tree; (2) total DNA extraction; (3) PCR reaction; (4) Purification of PCR products and sequencing. 2.2.3. Data processing methods 2.2.3.1.Phylogenetic tree based on morphological characteristics: building phylogenetic tree for 103 taxa, including 101 species, 1 subspecies and 1 form in southern Vietnam based on 33 morphological characteristics by PAUP software (Swofford, 5 2002), according to the Maximum Parsimony method with 1000 bootstrap replicates (Felsenstein, 1985). 2.2.3.2. Phylogenetic tree based on molecular markers: building phylogenetic tree for 70 taxa, including 64 species, 1 subspecies and 1 form which were collected in southern Vietnam and 4 taxa from Genbank data by PAUP * software (Swofford, 2002) and MEGA6 (Tamura et al., 2013) by Maximum parsimony, Neighbor joining and Maximum likelihood methods. CHAPTER 3. RESULT AND DISCUSSION 3.1. Results of the construction of the Phylogenetic tree based on morphological characteristics The results on the phylogenetic tree in Figure 3.1 showed that the resolution was not high because the branches tended to start from the same root. The morphological relationships of 101 species, 1 subspecies and 1 form were basically divided into 4 subfamilies: Pothoideae (I), Monsteroideae (II), Lasioideae (III) and Aroideae (IV). 3.1.1. Subfamily Pothoideae and Monsteroideae The phylogenetic tree in Fig. 3.1 showed that the two subfamily of Pothoideae and Monsteroideae with bisexual flower (3-tribe Potheae, Anadendreae and Monstereae) grouped together with bootstrap as 83% and were separated from subfamily Aroideae with unisexual flower . This result was consistent with the view of the classification system established by Mayo et al. (1997). 3.1.2. Subfamily Lasioideae In contrast to the subfamily Pothoideae and Monsteroideae, another subfamily with bisexual flower, Lasioideae, had no distinction with the Aroideae subfamily in the phylogenetic tree in Figure 3.1. Accordingly, despite belonging to the group with bisexual flower, subfamily Lasioideae has terrestrial or aquatic living form similar to most of tribe of the subfamily Aroideae (Mayo et al., 1997; Li et al., 2010; Boyce et al., 2012); thus, in the phylogenetic tree in 6 Figure 3.1, the subfamily Lasioideae was the intermediate group between the group with bisexual flower (Pothoideae and Monsteroideae) and the subfamily with unisexual flower (Aroideae). 3.1.3. Subfamily Aroideae Figure 3.1 showed that taxa of the subfamily Aroideae tended to be grouped into one group and could be distinguished from the bisexual group, especially the two subfamily Pothoideae and Monsteroideae. Accordingly, the subfamily Aroideae shared many morphological characteristics and could be distinguished from the subfamilies Pothoideae, Monsteroideae and Lasioideae in the characteristics of unisexual flowers, absent perigone, tuber or rhizome, their often appearrance on terrestrial, swamps or streams (Mayo et al., 1997; Pham-Hoang, 2000; Nguyen, 2006; Li et al., 2010; Boyce et al., 2012). In addition, the results of the categorization of taxa in subfamily Aroideae on the phylogenetic tree (Figure 3.1) basically showed that there was a division between the tribe and genus, whereby the arrangement of each case in the subfamily Aroideae is as follows: - Tribe Homalomeneae In the south, Homalomeneae has one genus Homalomena. In the phylogenetic tree in figure 3.1, species belonging to the genus Homalomena were grouped together on a branch with a bootstrap value of 79%. However, taxa belonging to the genus Homalomena originated from the same branch (Fig. 3.1), which did not show the difference between taxa in the genus. Therefore, this issue will be clarified in the analysis of the phylogenetic tree based on the molecular markers (see page 15 for details). 7 Figure 3.1. One of most-parsimonious tree obtained based on 33 morphological characteristics of 101 species, 1 subspecies and 1 form in in southern Vietnam with Acorus verus (Acoraceae) as outgroup. The bootstrap values of 50% or more than from 1000 replicates are shown above the nodes. Note: I. Subfam. Pothoideae, II. Subfam. Monsteroideae, III. Subfam. Lasioideae, IV. Subfam. Aroideae. 8 - Tribe Aglaonemateae In the history of research on Aglaonemateae, there were different opinions on the genus Aglaodorum (with the only species Aglaodorum griffithii). Some said that Aglaodorum griffithii had many morphological characteristics similar to Aglaonema such as flower structure, morphology of leaf and petiole, etc., and it was thus placed in genus Aglaonema (Schot, 1856; Hooker, 1883; Ridley, 1925). However, many authors said that there were some differences between these two genera and suggested that Aglaodorum griffithii was independent of the Aglaonema species Schot, 1856; Mayo et al., 1997; Pham-Hoang, 2000; Nguyen, 2005 & 2006; Boyce et al., 2012). Based on the phylogenetic tree in Figure 3.1, Aglaodorum griffithii was not on the same branch with the Aglaonema species. This result supported the view of many researchers who have recently said that Aglaodorum and Aglaonema were the two separate genera. - Tribe Thomsonieae Previously, Tribe Thomsonieae had two genera, Amorphophallus and Pseudodracontium (Serebryanyi, 1995; Mayo et al., 1997; PhamHoang, 2000; Nguyen, 2006). However, Hetterscheid & Claudel (2012) said that characteristic features of genus Pseudodracontium were present in species belonging to genus Amorphophallus. In addition, many molecular evidences also showed a very close genetic relationship between the two genera này (Cabrera et al., 2008; Grob et al., 2002 & 2004; Nauheimer et al., 2012); Hetterscheid & Claudel (2012) thus removed the genus Pseudodracontium and transferred all of taxa of genus Pseudodracontium to Amorphophallus. The results in the pedigree tree of Figure 3.1 showed that the species of genus Pseudodracontium, including A. fallax, A. macrophyllus, A. lacourii, A. lanceolatus, A. pseudoharmandii and A. sp3. were previously grouped together but their relationship with Amorphophallus species was unclear. In addition, the resolution of the species in tribe Thomsonieae in phylogenetic tree (Figure 3.1) was not high, so it was 9 difficult to see the species classification clearly. Therefore, this issue will be clarified in the analysis of phylogenetic tree based on molecular markers (see page 15 for details). - Tribe Arisaemateae Genus Arisaema was first described in 1831 (Martius, 1831). Schott (1860) divided into four groups: Trisecta, Pedatisecta, Radiatisecta and Peltatisecta. Many classification systems of genus Arisaema have later proposed by several authors (Nakai et al., 1950; Hara, 1971; Murata 1984 & 1991; Murata et al., 2013). Gusman & Gusman (2002) introduced a new classification system for the genus including 14 sections and 15 sections in the second edition in 2006. Murata et al. (2013) introduced a new classification system by merging sect. Fimbriata and Lobata into the sect. Attenuata and Pistillata, respectively; in addition, the author added a new section Odorata to form a classification system of 14 sections for genus Arisaema. Recently, Ohi-Toma et al. (2016) split sect. Fimbriata out of sect. Attenuata based on molecular markers and thereby introducing the classification system for Arisaema with 15 sections. In Southern Vietnam, Arisaema has three sections: Anomala, Fimbriata and Sinarisaema. Based on the phylogenetic tree in Figure 3.1, Arisaema species tended to rank together. Of which, 2 species of A. consanguineum subsp. consanguineum and A. kerrii from sect. Sinarisaema were grouped into one group with a bootstrap value of 87%. However, the taxa of the two sections of Anomala and Fimbriata were not completely separated into two groups, so the division of Arisaema into three sections in the south was not yet evident in the phylogenetic tree, which will be clarified in the analysis of phylogenetic tree based on molecular markers (see page 18 for details). - Tribe Colocasieae In southern Vietnam, Colocasieae consisted of four genera: Alocasia, Colocasia, Remusatia and Leucocasia (Mayo et al., 1997; 10 Nauheimer et al., 2011). The phylogenetic tree of Figure 3.1 showed a partial relationship between taxa in tribe Colocasieae, whereby Alocasia evrardii (which was moved to genus Colocasia in this study) had intermediate characteristics between genus Alocasia and Colocasia in which the ovule was parietal and basal placentation but they were close to the genus Colocasia because it had the characteristic of having a lot of ovules, so it tended to be closer to C. esculenta in the phylogenetic tree (Figure 3.1). In addition, Remusatia vivipara had other important characteristics compared to Alocasia and Colocasia in which spadix without appendix, hence, the species is also classified as a branch. The phylogenetic tree based on morphological characteristics in Figure 3.1 partly reflected the relationship between the levels of classification in Araceae in southern Vietnam. However, the resolution shown on the phylogenetic tree was low, leading to less clustering if the bootstrap value was significant (over 50%) (Salemi & Vandamme, 2003). For example, Figure 3.1 showed that some species of the genus Amorphophallus did not have clear clustering with species of genus Sauromatum and Typhonium (tribe Areae). Similarly, tribe Anadendreae and Monstereae tended to group together. In addition, among species in many genera, the resolution was not high and therefore, there was no relationship between species in the genus. This result was similar to the phylogenetic tree of Cusimano et al. (2011) whose study also showed that the phylogenetic tree also had the branch with low significance level (bootstrap >50%). Therefore, most of the recent researches on phylogenetic systems in plants in general and Araceae in particular were based on molecular markers (Grob et al., 2002 & 2004; Jung et al., 2004; Renner & Zhang, 2004; Renner et al., 2004; Cabrera et al., 2008; Sedayu et al., 2010; Wong et al., 2010; Cusimano et al., 2011; Nauheimer et al., 2011 & 2012; Nauheimer & Boyce, 2013; Ohi- 11 Toma et al., 2016). Therefore, in this study, we constructed the phylogenetic tree based on molecular markers (Figure 3.2) to solve many problems in Araceae which could not be clarified on the basis of morphologial characteristics. 3.2. Results of the construction of the phylogenetic tree based on melecular markers. Based on the classification in the phylogenetic tree in Figure 3.2, we provided the case-by-case observations as belows: Subfamily Pothoideae, Monsteroideae and Lasioideae Figure 3.2 showed that the subfamily Lasioideae, Pothoideae and Monsteroideae were located closely together on the phylogenetic tree. This result was consistent with many previous studies on the phylogenetic system based on molecular markers for Araceae (Cabrera et al., 2008; Cusimano et al., 2011; Nauheimer et al., 2012). Accordingly, these three subfamily (6 genera Pothos, Anadendrum, Raphidophora, Epipremnum, Scidapsus and Lasia) were located close to each other and placed outside in comparison with the subfamily Aroideae because they shared the same important morphological chacteristic of bisexual flower. Therefore, this result was consistent with the evolutionary view of Mayo et al. (1997) that taxa of bisexual flower were an ancient group while the taxa of unisexual flower were the later evolutionary groups. In this group, the subfamily Lasioideae (Lasia and Pycnospatha genus) were individually classified as a branch due to the characteristic of a habitat on marsh or soil, while the other branch was the group of tribe Potheae, Monstereae and Anadendreae which live on plant or stone (Mayo et al., 1997; Pham-Hoang, 2000; Nguyen, 2006; Li et al., 2010; Boyce et al., 2012). Subfamily Aroideae * Tribe Aglaonemateae - Aglaonema was founded by Schott in 1829 (Schott, 1829). Nicolson (1969) first divided the genus into two sections based on 12 morphological characteristics. Nguyen (2006) and Boyce et al. (2012) accepted Nicolson's view of division. At present, no research has yet been done on the use of molecular markers for the purpose of constructing the phylogenetic tree for Aglaonema. Most studies focussed on building the phylogenetic system for the whole family at the level of genus, not going to the details of species of the genus (Cabrera et al., 2008; Cusimano et al., 2011; Nauheimer et al., 2012). The results of the arrangement of the Aglaonema species shown in the phylogenetic tree (Figure 3.2) showed that there was no clear separation between species of the two sections. Therefore, in this study, we did not use the Nicolson sectioning system (1969). - Aglaonema griffithii was first described by Schott (1856). However, Schott (1858) found that, there were some differences between Aglaonema griffithii and Aglaonema species, so he founded the genus Aglaodorum and transferred the Aglaonema griffithii to this genus with its scientific name of Aglaodorum griffithii. Hooker (1883) and Ridley (1925) put Aglaodorum griffithii into genus Aglaonema. At present, Aglaodorum griffithii was still accepted by many authors as the only Aglaodorum species (Mayo et al., 1997; Pham-Hoang, 2000; Nguyen, 2005 & 2006; Boyce et al., 2012). Recently, studies on the construction of phylogenetic tree by molecular markers also showed that Aglaonema and Aglaodorum were always grouped together in the phylogenetic tree and were termed as “sister group" (Cabrera et al., 2008; Cusimano et al., 2011; Nauheimer et al., 2012). The phylogenetic tree in Figure 3.2 showed that the Aglaodrum griffithii, Aglaonema simplex, and A. costatum f. immaculatum are grouped together (bootstrap values = 88%) and nested in Aglaonema clade with Aglaonema cochinchinense as sister (bootstrap values = 100%). Thus, based on similarities in morphological and genetic characteristics, in this study we classified Aglaodorum griffithii into Aglaonema as suggested by Schott (1856). 13 Figure 3.2. One of most-parsimonious tree obtained based on combined data set of trnL-trnF IGS and matK of 70 taxa (Araceae) in southern Vietnam with Acorus verus (Acoraceae) as outgroup. The bootstrap values of 50% or more than from 1000 replicates are shown above the nodes. CI= 0.72, RI= 0.87, and RC= 0.63. Note: taxa(*) are referenced from the GenBank database; I. Subfam. Pothoideae, II. Subfam. Monsteroideae, III. Subfam. Lasioideae, IV. Subfam. Aroideae. 14 * Tribe Homalomeneae Currently, two species of H. cochinchinesis and H. occulta still have different opinions from botanists, for example, Hu (1968) and Li (1979) identified that H. cochinchinesis was a synonym of H. occulta, Nguyen (2006) had the same view as the two authors. However, Pham-Hoang (2000) and Govaerts et al. (2002), H. cochinchinesis and H. occulta were the two distinct species. Based on the type specimens and specimens of these two species stored in herbariums (HN, VNM, HNU and P), especially fresh specimens collected, we found that H. cochinchinensis can be distinguished from H. occulta in having: spadix being 2/3 shorter than spathe, long stipitate; density flowers with many staminodes in female part; oval male part, diameter larger than female part. In addition, The results of the phylogenetic tree shown in Figure 3.2 showed that H. cochinchinesis and H. occulta did not group together, but grouped with H. pierreanum. Therefore, based on morphological and genetic differences, we agree with the views of Pham-Hoang (2000) and Govaerts et al. (2002) that H. cochinchinesis is a good species and obviously distinct from H. occulta. * Tribe Thomsonieae Amorphophallus genus were grouped together with a bootstrap value of 100%, especially, group included A. coudercii, A. fallax, A. lacourii, A. lanceolatus, A. macrophyllus, A. pseudoharmandii, A. tenuistylis, A. sp1., A. sp2. and A. sp3., of which A. tenuistylis were ranked separately, followed by three species A. coudercii, A. sp1. and A. sp2. which were grouped together, the rest was the group of six species of the previous genus Pseudodracontium. This result was in accordance with the views of Hetterscheid & Claudel (2012) and Bogner et al. (1985) suggesting that species of the genus Pseudodracontium have most relative to some species in genus Amorphophallus, including A. coudercii and A. tenuistylis. 15 In addition, A. fallax, A. macrophyllus, A. lacourii, A. lanceolatus, A. pseudoharmandii and A. sp3. had the morphological characteristics belonging to the genus Pseudodracontium. This genus was established by Brown (1882) with a new species, P. anomalum and a species from Amorphophallus, P. lacouri. Serebryanyi (1995) reported the number of species of the genus Pseudodracontium to be 7 and suggested that Pseudodracontium had the morphological characteristics that Amorphophallus species did not have or rarely had. Many researchers later identified that Pseudodracontium and Amorphophallus were the two separate genera of tribe Thomsonieae (Mayo et al., 1997; Pham-Hoang, 2000; Nguyen, 2005 & 2006). However, Hetterscheid & Claudel (2012) disagreed with the view of Serebryanyi (1995) and suggested that many Amorphophallus species also have morphological characteristics that Serebryanyi (1995) attributes to Pseudodracontium. Therefore, Hetterscheid & Claudel (2012) removed Pseudodracontium and moved all species of this genus to Amorphophallus. Boyce et al. (2012) also accepted the view of Hetterscheid & Claudel (2012). Recently, studies on the development of phylogenetic systems using molecular and morphological markers for Araceae have shown that Amorphophallus and Pseudodracontium are very close together, which are termed as "sister group" (Cabrera et al., 2008; Cusimano et al., 2011; Nauheimer et al., 2012). More specifically, Grob et al. (2002 & 2004) or Sedayu et al. (2010) constructed the phylogenetic tree for tribe Thomsonieae using molecular markers which showed that Pseudodracontium harmandii and P. lanceolatum were nested within the group of species in genus Amorphophallus and Grob et al. (2002) also suggested that Pseudodracontium should be transferred to genus Amorphophallus (because Amorphophallus was the genus formly established and had more species). In this study, six Pseudodracontium species’s molecular data were most analyzed for the first time. Accordingly, the results shown 16 in Figure 3.2 showed that six species of Pseudodracontium were nested within Amorphophallus species with an maximum bootstrap value (100%). In addition, the genetic distance between these species and some Amorphophallus species were lower than the genetic distance between Amorphophallus species, for example, the genetic distance between Pseudodracontium species and A. coudercii ranged from 0.011 to 0.013 or with A. tenuistylis from 0.015 to 0.016. Meanwhile, the genetic distance between A. longicomus and A. tenuistylis, A. coudercii was 0.027 and 0.025, respectively. Accordingly, based on the study results shown in the phylogenetic tree (Figure 3.2), all species of genus Pseudodracontium were transferred to genus Amorphophallus with the scientific names proposed by Hetterscheid & Claudel (2012). * Tribe Colocasieae The results shown in the phylogenetic tree (Figure 3.2) showed that there has been some confusion in the taxonomy of some taxa of tribe Colocasieae in Southern Vietnam. For example, Alocasia evrardii was not grouped with Alocasia species but tended to be classified as Colocasia species. Alocasia evrardii was first described by Gagnepain (1942) based on specimens collected by Francois Evrard in Dalat, Lam Dong province. This species was also recorded by Pham-Hoang (2000) in the book "An Illustrated Flora of Vietnam". However, the description of Gagnepain (1942) was not valid under the international nomenclature of algae, fungi and plants (Turland, 2013). Nguyen et al. (2013) described Alocasia evrardii for the first time and classified species into genus Alocasia because of the characteristics of ovule with basal placentation. However, through the detailed morphological study, we found that A. evrardii has ovules with parietal and basal placentation. Therefore, that fact that Nguyen et al. (2013) classified this species into genus Alocasia was unstable. In addition, the results shown in the phylogenetic tree (Figure 3.2) showed that A. evrardii was not completely grouped 17 together with Alocasia species, but tended to be close to Colocasia species. Therefore, A. evrardii differed in morphological and genetic characteristics compared to Alocasia species and in our opinion, this species has been confused during classification. Therefore, in this study, we have transferred Alocasia evrardii to Colocasia with the scientific name of Colocasia evrardii. * Tribe Arisaemateae In this group, asside from A. roxburghii, the species which was widespread in the whole Indochinese, Malayan Peninsula and Vietnam (Pham-Hoang, 2000, Nguyen, 2005 & 2006, Gusman & Gusman, 2006, Boyce et al. 2012), there have been five other species, including A. chauvanminhii, A. condaoense, A. honbaense, A. pierreanum and Arisaema sp2, recently discovered in southernern Vietnam, so they have never been included in any of the phylogenetic systems. Of those species, two species of A. condaoense and A. pierreanum have been currently suspected as a synonym of A. roxburghii (Gusman & Gusman, 2006). Based on results obtained from the detailed morphological and molecular characterization, we made the two observations as belows: - A. condaoense was first described by Nguyen Van Du as a new species for science in 2000. In this publication, the author suggested that A. condaoense could be distinguished from A. roxburghii in having appendix very long exserted from spathe tube (about 4 cm) (Nguyen, 2000). However, the description of A. condaoense by Nguyen (2000) was based on based on only dried specimens with male inflorescences only (N.T. Bân & Averyanov 385, HN!, 20 June 1989) so the descriptive information lacked color information of plant parts and characteristics of female inflorescences. Thus, Gusman & Gusman (2006) suspected that A. condaoense might be a synonym of A. roxburghii and that "it is necessary to study living material to identify the status of A. condaoense". To ascertain its identity, we reexamined the questioned taxon. New fresh material (male and female