Fargesia decurvata is closely allied with F. dracocephala and differs in 5 major characters (i.e. the culm sheath blade base shape, the width of the culm sheath blade base, the auricle shape, and the lower surface of leaf blade) in Fargesia. It is difficult to distinguish these two species because of existing of transitional statements of characters. The aims of this paper are to (i) investigate whether the variation of the characters is continuous or not; (ii) reveal whether the publishment of F. dracocephala was the result of discontinuous sampling of F. decurvata or not.
Ten populations of F. decurvata and F. dracocephala were investigated in their entire distribution (including type localities). The statements of 5 major characters were measured from 693 annual and 693 perennial culms of 231 individuals in 10 populations, and analyzed at population, individual and culm levels. UPGMA cluster analysis was carried out based on 29 characters from 10 populations of F. decurvata and F. dracocephala and 2 populations of F. qinlingensis as outgroup. The ITS sequences were also sequenced and analyzed.
Five major characters exhibited great variation not only at population level, but at individual level within a population, even the culm level within an individual and in different parts of the same culm. Cluster analyses showed that 10 populations of F. decurvata and F. dracocephala were not divided into two species, but they were well separated with outgroup. There was no difference in floral organ between F. decurvata and F. dracocephala. MP and NJ trees based on ITS sequences showed the same results with the cluster analysis on morphological characters. All the facts indicated that the publishment of F. dracocephala was the result of discontinuous sampling of F. decurvata, and F. dracocephala should be treated as the synonym of F. decurvata.
Citation: Zhang Y-Q, Wang X-M, Wu A-L, Ren Y (2014) Merging Fargesia dracocephala into Fargesia decurvata (Bambusoideae, Poaceae): Implications from Morphological and ITS Sequence Analyses. PLoS ONE 9(7): e101362. https://doi.org/10.1371/journal.pone.0101362
Editor: Shuang-Quan Huang, Central China Normal University, China
Received: January 28, 2014; Accepted: June 5, 2014; Published: July 2, 2014
Copyright: © 2014 Zhang et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: Funding came from the Giant Panda International Cooperation Project of the State Forestry Administration of China (WH0630). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
The correct definition of species is the most basic work for further researches. The type and other voucher specimens play an important role in the process of defining a species. However, because the herbarium specimens are often collected from the limited location(s) and/or limited individual(s) or sometimes just from parts of an individual, the continuity of variation of character(s) within a species would be artificially isolated. The artificial isolation of character(s) might lead to the result that a virtual species will be defined as different ones –.
The problems are particularly prominent in the definition of most of bamboo species. Because the bamboos typically have to go through over decades or even more than a century of vegetative growth before flowering, and generally died off within a few years after flowering, thus the chance to get the flowers and caryopses is very little when the specimens were collected, the defining and identifying of the bamboo species have been mainly depended on vegetative characters , . Each of the bamboo is distributed in a wide range and grows in diverse habitat, but the specimens can be collected in limited locations and habitat, many vegetative variations might be ignored due to the limited sampling. In additional, only a small part of the culms, branches or rhizomes can be collected as the specimens. As a result, the specimen-based description and definition of bamboo species would inevitably cause the continuous variations of the characters to be fragmented.
Fargesia is a large genus in Bambusoideae (Poaceae) with about 90 species distributed mainly in the alpine areas at altitude of 1400–3800 m in south-west China, Vietnam and adjacent Himalayas , . There are 78 species in China and 61 of them were nominated and published without the description of reproductive organs and many of them are narrow distribution species –. In the field survey of Fargesia species as the main food of giant panda, we found that some bamboo samples from this genus could not be accurately identified because of the existing of transitional or intermediate characters between or among species. Among these species, F. decurvata Lu (Fig. 1) and F. dracocephala Yi are a pair of species which have wider distribution and are most difficult to identify. Therefore, we considered that these two species might be the good materials to study the definition of bamboo species.
A. Bamboo clump. Bar = 30 cm. B. Middle parts of a bamboo shoot. Bar = 1 cm. C. Branch and leaves. Bar = 0.8 cm. D. Yong infructescence. Bar = 1.5 cm. E. Close up of a node. Bar = 1 cm.
Fargesia decurvata was described by Lu  based on specimen collected from Hejiaping, Changyang County, Hubei Province (Lu Jionglin 78120, Type in Herb. Honan Col. Agr. and PE). He considered that this species differs from other Fargesia species in following aspects: the base of the sheath blade is as wide as the sheath tongue or nearly so, the culm sheath tongue is promontory as a bow and strongly decurrently on both sides, the culm sheath has dense gray small bristles on the back and dense coarse wools at the base. There is no description of the reproductive organs. Thereafter Yi published F. aurita, which is based on specimens collected from Fengdu County, Sichuan Province (Yi Tongpei 75410, Type in Herb. Forestry School of Sichuan Province) , but the name was treated as a synonym of F. decurvata by Yi himself .
Fargesia dracocephala was described by Yi based on specimens collected from Guangwu Mountain, Nanjiang County, Sichuan Province (Yi Tongpei 75540, Type in Herb. Forestry School of Sichuan Province), Qianjiaping Forest Farm, Pingli County, Shaanxi Province and Shennongjia, Hubei Province . He considered that this species differs from F. robusta in following aspects: the internodes are nearly solid and the hollows are 1–2.7 mm in diameter, the culm sheath is nearly hairless or with sallow thin bristles on the back, the leaf sheath is elliptic and with setae at apex, the leaf blade has 3–4 pairs of secondary veins. He described the morphology of flowers later .
Fargesia decurvata and F. dracocephala were considered as closely allied species both in Flora Reipublicae Popularis Sinicae (FRPS)  and Flora of China  and belong to Fargesia Sect. Fargesia Ser. Yunnanenses. In the field investigation on the giant panda’s food, we found that it is difficult to identify F. decurvata and F. dracocephala by the key characters because there are many transitional forms. When the previous references on these species were checked, we found that the description and illustration of two species were confused, especially in aspects of the shape of the leaf auricle and the position of the oral setae on the auricle. The only description on the auricle was “auriculae parvae” (auricle small) when Lu described F. decurvata for the first time . Based on the observation on the type (Lu Jionglin 78120, PE), the auricles are mostly short strike-like with oral serae at margins or at margins and apex. When Yi (1985b) published F. aurita , which was treated as the synonym of F. decurvata , the auricles were described as “auriculis foliorum ellipticis” (leaf auricle elliptical), and there was no description on the seta position. However, figure in FRPS showed clearly that the setae are at the apex of the auricle. Thus, the shape of the leaf auricle should be short strike-like or elliptical with serae at the margins and/or apex. Later, the auricles of the species were described as “nearly circular” in FRPS  and Li et al. followed this description in Flora of China  and the setae were described as at the margins of the auricles , but the auricles seems not nearly circular but nearly short rectangle with setae at apex in Pl. 132, f. 17 in FRPS . When F. dracocephala was described for the first time , the auricles were described as “auriculis foliorum ellipticis” (leaf auricle elliptical) and no description of the seta position, but it seems that the auricles are not elliptical but short banding with setae at apex from the illustration Fig. 4–6 in the paper of Yi .
In FRPS, Yi  described the auricles of the species as long elliptical but the shape of the auricles in the illustration Pl. 132, f.6 is the same with that in the paper of Yi , and this description was followed in Flora of China . It seems that the variation of characters is various and might be continuous between two species. So we suspected that F. decurvata and F. dracocephala might be the representatives of two forms of a series morphological variation of vegetative organs. The publishing of the two species might be due to the discontinuous sampling of the same species.
In order to verify this question, the samples of F. decurvata and F. dracocephala from ten populations, which basically covered the total distribution of two species, were collected. The key characters plus 25 other characters of the two species were measured and analyzed to determine whether the variations of these characters are continuous or not. Furthermore, in order to confirm and enhance the variation pattern of vegetative morphologic characters, the sequences of the internal transcribed spacer (ITS) between the ribosomal RNA genes were used in the present studies because ITS has been one of the most frequent examples of DNA segments in plant systematic analysis. Although use of this multicopy region has several pitfalls, which have been summarized by , an indisputable practical advantage is the current abundance of information concerning ITS sequences in many taxa –, enabling comparisons among different studies and numerous taxa. Here we attempted to integrate both morphological and molecular (ITS) data to assess the variation pattern of these characters between F. decurvata and F. dracocephala.
Materials and Methods
According to regulations of the People’s Republic of China on the protection of wild plants, permits are required only for the species included in the list of state-protected plant species at the time of collection. Neither Fargesia decurvata and F. dracocephala nor F. qinlingensis is on the list of state-protected plant species (Yu YF, A milestone of wild plants protection in China - the first list of wild plants protected by the nation, Plants 1999 (5): 3–11; Regulations of the People’s Republic of China on the protection of wild plants, http://www.people.com.cn/item/faguiku/zrzyf/U1020.html). Thus, no specific permits were required for the described field studies. During the sample collection, only a part of culms and leaf blade were collected to avoid causing any harm to the plants and their habitats.
The samples from ten populations of F. decurvata and F. dracocephala, including the populations from the type localities of two species, and two populations of F. qinlingensis Yi et J. X. Shao as the outgroup were collected from July to August in 2009 and 2010 (Table 1). Fifteen to twenty clumps (individuals) were randomly collected from each population, and the neighboring individuals were at least 5 m apart, so as to avoid resampling from the same individual. From each individual, 3–5 fresh leaves were sampled and immediately placed in silica gel, and stored at room temperature for DNA extraction. We collected seedlings of two to four years in Pop 3 and Pop 7 and floral organs in Pop 3, Pop 7 and Pop 9 among which Pop 3 can be identified as F. decurvata according to vegetative characteristics. The voucher specimens were deposited in the Herbarium of Shaanxi Normal University (SANU).
Measurement and analysis of morphological characters
Based on FRPS  and Flora of China , five major characters, i.e. the base of the culm sheath blade is decurrent or not, the base of culm sheath blade is narrower than or as wide as the apex of the culm sheath, the auricle is nearly circular or oblong with setae at the apex or the margin, and the lower surface of leaf blade is setose or nearly glabrous, are used to identify F. decurvata and F. dracocephala (Table 2 and Table S1). Meanwhile, 24 expanded characters were also selected and measured (Table 3 and Table S1).
Three annual and three perennial culms were randomly selected in each individual and the vegetative morphological characters were measured in the field. Each of the characters was measured from the basal five nodes of three culms in a clump. For the characters from the sheathes, the measurement was carried out on the annual culms (bamboo shoots), and for the characters from the culms, branches and leaves, the measurement was carried out on the culms more than three-year-old (perennial culms), the characters of branch and leaf measure from the basal five nodes which have branches. The number of the hires on the culm sheathes and leaf blades was calculated from three 10 * 4.5 visions (about 1/20.25 square millimetres) with the stereomicroscope in three different areas for each of twenty sheathes or leaves on perennial culms and for all of sheathes and leaves on seedlings. Spring and summer leaves for perennial culms were measured separately.
The cluster analysis of morphological data from ten populations of the ingroup and two populations of the outgroup was carried out with UPGMA method using NTSYSpc 2.1 software package (Exeter Software, NY, USA).
DNA isolation, amplification, sequencing and analyses
Total DNA was extracted from silica gel-dried leaves using a modified CTAB procedure . The concentration of genomic DNA was determined by electrophoresis on 1.0% agarose gels. The DNA samples were diluted by ddH2O and stored at −20°C for use. The nrDNA ITS region was amplified by primers ‘ITS4’ and ‘ITS5’ from White et al.  (1990) and Guo et al. . PCR amplifications of both chloroplast DNA fragments and ITS region were conducted in a 25-µL volume containing 1×PCR Buffer, 2 mM Mg2+, 250 µM each of dNTPs, 0.8 mM of each primer and 50–100 ng genomic DNA. The reactions were performed in an ABI Veriti™ 96-well Thermal cycler (Applied Biosystems, CA, USA), with the following program: initial denaturation at 95°C for 4 min; 35 cycles of 95°C for 45 s, 55°C for 1 min, 72°C for 1 min; and last synthesis at 72°C for 7 min. PCR products were detected on a 1% agarose gel, stained with ethidium bromide and visualized under UV light. The products were subsequently purified and sequenced by Beijing Dingguo Changsheng Biotechnology Co. Ltd. (Beijing, China).
Chromas Lite v2.01 (http://www.technelysium.com.au/chromas_lite.htm) was used to check the quality of the complete sequences. Sequence divergences between taxa and base frequencies (G+C content) were determined using MEGA v5.05 . Neighbor-joining (NJ) tree was also calculated using MEGA v5.05 . Phylogenetic analyses were performed using PAUP v4.0b10 .
Analysis of key characters in ten populations
Variation on culm sheaths.
The two characters from the culm sheath vary in different nodes on a same bamboo shoot from the bottom to the apex, the blades change from short, triangular to long and ribbon or even leaf like, their bases change from as wide as the sheath apex, decurrent to narrower than sheath apex and not decurrent. Here we selected a culm from Pop 4 to illustrate the changes of the shape and base of sheath blades. The sheath blade is very small, triangular, the base is not decurrent and usually as wide as the sheath apex or nearly so in the third node (Fig. 2-A), the sheath blade is ribbon like, its base is also decurrent but is narrower than the sheath apex in the ninth node (Fig. 2-B), the sheath blade is longer, ribbon (or even leaf like in other culms of the same individual or in other populations), the blade base is not decurrent and much narrower than the sheath apex in the twenty first node (Fig. 2-C). Here we used the ratio of the width of the blade base and the width of the sheath apex to exhibit the change of the blade bases in different populations. The ratio of blade base and sheath apex is from 0.7 to 1.0 in the third node, from 0.53 to 0.82 in the ninth node, from 0.42 to 0.5 in the twenty first node. The ratio changes slowly from the first to the fifth nodes, while decreases gradually from the fifth to the seventeenth nodes, and then changes slowly again from the seventeenth to the twenty third nodes in any of ten populations (Fig. 3).
Bar = 5 mm. A. In lower part, showing blade base decurrent (arrow) and as wide as sheath apex. B. In middle part, showing blade base not decurrent (arrow) and narrower than sheath apex. C. In upper part, showing blade base not decurrent (arrow) and much narrower than sheath apex. D–L. Close up of leaf auricles and oral setae, showing the shape of leaf auricles and position of oral serae. Bar = 1 mm. D–F. From Pop 1. D–E. From an annual culm. D. Long triangular auricle with serae at upper margins and apex. E. Curving long triangular auricle with setae at margins and apex. F. Short sickle-like with setae mostly at one margin and apex from a perennial culm. G–H. From Pop 3. G. Sickle-like auricle with setae mostly at one margin and apex from an annual culm. H. Very narrow sickle-like auricle with setae mostly at one margin and apex from a perennial clam. I–J. Culms from Pop 8. I. Banding auricle with setae at upper margins and apex from an annual culm. J. Short banding auricle with setae at upper margins and apex from a perennial culm. K–L. From Pop 10. K. very small auricle with setae at whole margin from an annual culm. L. Almost without auricle but with setae on upper margin of sheath from a perennial culm. C: culm. CB: culm sheath blade. CS: culm sheath. LA: leaf auricle. LS: leaf sheath. S: oral seta.
Variation on leaf auricles.
Based on the observation on the types of F. decurvata (Lu Jionglin 78120, PE) and F. dracocephala (Yi Tong-pei 75540, Herb. Forestry School of Sichuan Province), the leaf auricles are short sickle-like, banding, short banding, but not elliptical, long elliptical or nearly circular, therefore we did not use “elliptical”, “long elliptical” and “nearly circular” to describe the shape of the auricles.
The shape of the auricles and the position of the oral setae are various in different populations, different individuals in the same population and even on a same culm, long triangular with setae at upper margins and apex (Fig. 2-D), curving long triangular with setae at margins and apex (Fig. 2-E), short sickle-like with setae mostly at one margin and apex (Fig. 2-F), sickle-like with setae mostly at one margin and apex (Fig. 2-G), very narrow sickle-like with setae mostly at one margin and apex (Fig. 2-H), banding with setae at upper margins and apex (Fig. 2-I), short banding with setae at upper margins and apex (Fig. 2-J), very small with setae at whole margin (Fig. 2-K), and almost without auricle but with setae on upper margin of sheath (Fig. 2-L). The field observations showed that the auricles and oral setae exist on all newly formed leaves while fall off on perennial leaves in some cases.
Variation on pubescence on blade back.
It seems that the pubescence on blade back is an important difference between two species among the expanded characters, so more attention was paid on this character in the present study.
The pubescence on blade back is various: glabrous on blade back, pubescent at the base of the blade back and pubescent on the whole blade back. In most of the populations, all individuals have the same status of the leaf pubescence in the same population, but in other populations different individuals have different status of the leaf pubescence in the same population (Fig. 4). The leaf pubescence is also various on the leaves from different seasons. For example, there are an average of 90.3 hires in a vision on spring leaves and an average of 125.5 hires on summer ones in Pop 3.
Cluster analysis of vegetative characters
The UPGMA cluster result (Fig. 5) of five major characters plus 24 expanded ones showed that twelve populations can be divided into two groups. Group I included two populations of F. qinlingensis as the outgroup, while the remaining populations of F. decurvata and F. dracocephala formed Group II. In addition, Group II can be divided into two subgroups. Pop 6 and Pop 7 constituted the first subgroup, and the second one comprised of the other eight populations. Among Group II, according to the key characters, Pop 1 was and Pop 3 and Pop 7 were approximate to F. decurvata, and the others were approximate to F. dracocephala. Therefore, the cluster analysis on the vegetative characters showed that the populations of these two species were nested one with another.
Comparison of the floral organs
There is no difference either between any of two populations or between the flowers that we collected and the descriptions of F. dracocephala  on floral organs.
The relationship based on the ITS sequence
Alignment of all the ITS region sequences of F. decurvata and F. dracocephala resulted in a matrix of 618 positions, including ITS1, ITS2 and 5.8s. Characteristics of these sequences, including length, G+C content, number of indels and variable bases, are summarized in Table 4. The length of ITS1 and ITS2 is 213 bp, and 216 bp, respectively. ITS1, with a maximum of 2.0% divergence across all taxa (1.9% in F. decurvata and F. dracocephala), is very slightly shorter and less variable in length than ITS2, which has a maximum of 2.0% divergence across all taxa (2.0% in F. decurvata and F. dracocephala). The 5.8S is 163 bp long in all taxa. As expected, the 5.8S gene showed little variation, with 1.2% divergence across all taxa. The overall of ITS divergence was approximately 1.86% within F. decurvata and F. dracocephala.
Based on the ITS sequence parsimony informative sites, both the MP tree and the NJ tree (Figs. 6 and 7) resulted in two major clades (I and II), with more than 75% bootstrap support. Two populations of F. qinlingensis formed clade I and the populations of F. decurvata and F. dracocephala constituted clade II. Clade II was mainly divided into two subcaldes. Pop 4, Pop 5, Pop 9 and Pop 10 of approximate F. dracocephala formed the first subclade. The second subclade is constitutive of the others populations (including F. decurvata and F. dracocephala); therefore, the populations of F. dracocephala were not included in the same subclades and that of F. decurvata were nested within that of F. dracocephala. Hence, according to the phylogenetic trees based on ITS sequence data, the present study did not support the ten populations to be separated as two species.
Number above branches is bootstrap values; values <50% are not indicated; the abbreviation is same as that in Fig. 5.
On taxonomy of Fargesia decurvata and F. dracocephala
According to the previous description –, the differences between the two species are: the culm sheath apex is promontory, the blade base of culm sheath is decurrent and is as wide as the apex of culm sheath, the leaf auricle is nearly circular with setae at the margin, pubescent on lower surface of leaf blade in F. decurvata; while the blade base of culm sheath is not decurrent and narrower or much narrower than the apex of culm sheath, the leaf auricles is elliptic with setae at the apex, glabrous on lower surface of leaf blade in F. dracocephala. Among the ingroup populations in the present studies, Pop 1 was collected from the type locality and the vegetative characters match completely with those of the type specimen of F. decurvata we observed in PE (Lu Jionglin 78120), Pop 8 was collected from the type locality and the vegetative characters match with those of F. dracocephala except the shape of the leaf auricle and pubescence. Based on the comparative analyses on 10 populations, each of the above mentioned characters is continuous or has intermediate status between two species. The continuous variation of characters occurs not only among populations, but also among individuals within a population, among different culms of a same individual and even in different parts of a same culm. The result from the cluster analysis of multiple vegetative characters, which was successful in the studies of allied species , , –, showed that ten populations were not divided into two species as well. Therefore, it is impossible to distinguish these two species by using vegetative characters if the sampling goes beyond the type localities and if the key characters could not distinguish two species well, not multiple ones.
The characters of the flowers and fruits play an important role in the definition of species in angiosperms as well as in bamboos although it is difficult to obtain the flowers and fruits in bamboos –. In the present studies, the flowers were collected in Pop 3, Pop 7 and Pop 9, among them Pop 3 matches F. decurvata well in vegetative organs. The floral characters are the same in three populations and no difference with the floral description of F. dracocephala , , . Therefore, it is obvious that there is no difference on the floral organs between these two species.
ITS sequence has been used widely in the studies of inter- and intra-specific relationships –, , . In the present study, the relationship of ten populations from F. decurvata and F. dracocephala and two populations from F. qinlingensis as the outgroup was established by using ITS sequence, with the credible bootstraps support (Figs. 5 and 6). The length and levels of variability of the ITS sequence accorded with the previous studies on bamboos , , . The tree from the ITS sequence showed that twelve populations were clustered into two clades. The first clade included two populations of F. qinlingensis, and the second comprised ten populations of F. decurvata and F. dracocephala. This result indicated that the difference among ten populations from F. decurvata and F. dracocephala was less than that between ten populations and the outgroup. In the second clade, ten populations from F. decurvata and F. dracocephala were nested one with another. Although the ten populations were included into two clades, but none of them matched any species, and the two populations (Pop 1 and Pop 8) from the type localities were included in the same clade. Therefore, the analysis on the ITS sequence did not support the separation of F. decurvata and F. dracocephala.
Based on the analyses of the morphological characters and ITS sequence, we considered that the description of F. dracocephala might be the result of discontinuous sampling of F. decurvata and F. dracocephala should be one species, and the name F. dracocephala should be treated as the synonym of F. decurvata.
Fargesia decurvata J. L. Lu in Journ. Henan Agr. Coll. 1981 (1): 74. Pl. 6. 1981; Flora Reipublicae Popularis Sinicae 9 (1): 471. Pl. 132: f. 16, 17. 1996; Flora of China 22: 93. 2006–F. aurita Yi in Journ. Bamb. Res. 4 (4): 22. pl. 6. 1985 et in ibid. 7 (7): 15. 1988–F. dracocephala Yi in Bull. Bot. Res. 5 (4): 127. f. 4. 1985 et in Journ. Bamb. Res. 7 (2): 15. 1988 et in ibid. 9 (1): 32. f. 3. 1990. adjust. flor. Descry; Flora Reipublicae Popularis Sinicae 9 (1): 469. Pl. 132: f. 1–13. 1996; Flora of China 22: 93.
On some characters used in key of species in Fargesia
The characters used to distinguish F. decurvata and F. dracocephala, such as with or without and the shape of the leaf auricle, the pubescence on leaf blade back, and the position of the oral serae, were frequently used in the key of species in Fargesia, especially in Ser. Yunnanenses, Ser. Angustissimae, and Ser. Fargesia , . Based on the present studies, these characters are unstable because they vary not only between populations, but also between individuals within a population, and even between culms of an individual or different position of a same culm. Some of the characters may vary in different developmental stages or different age. The rationality of the species based on such characters might be doubtable. Although there have been no evidence to show that many of the characters used in the key of species are continuous between or among species, but fact is that many tiny morphology characters were used in distinguishing two or even groups of species and one cannot find any biological or ecological meaning.
On sampling in definition of bamboo species
It is well know that a species is composed by population(s), and there has more or less difference(s) between populations because of the exiting of the phenotypic plasticity and genetic differentiation , , , . The phenotypic plasticity and genetic differentiation within a species might lead to the continuous variations of morphological characters. The phenotypic plasticity and genetic differentiation might be ignored and the continuous variation of the morphological characters might be isolated artificially within a species in case of discontinuous sampling. This might lead to the result that many microspecies were described and published. The result from the present studies showed that F. dracocephala is one of the variations of F. decurvata and the publication of F. dracocephala is the result of the discontinuous sampling of F. decurvata. The cases of discontinuous sampling which made the morphological characters were isolated artificially was also found in either bamboo or other taxa , , , , for example, Ph. sapida Yi (Bambusoideae, Poaceae) was proved to be one of the ecotypes Ph. flexuosa A. et C. R. cased by the environment changes after comparing a large number of specimens , ten species in Cornus schindleri complex (Cornaceae) collected from Kongding, Sichuan Province were proved to be two subspecies of C. schindleri . In fact, if four or more morphological characters, such as four key characters between F. decurvata and F. dracocephala, or some tiny morphological characters, such as pubescence situation on the leaves, which is the only difference between some species or even two groups of species in Fargesia , have to be used in the distinguish of two or more allied bamboo species, then it is reasonable to doubt that the so called allied species might be microspecies causing by the discontinuous sampling of the same species.
In the definition of species, after all, the specimen is numbered. The species itself is a collection of groups with a typical character, species defining is the delimitation of the characters of the group back together –, and the specimen is only a reference point on the naming. Often in the case of insufficient number of specimens, the continuous variation characters will be presented in the form of intermittence, causing difficulties and even errors in classification, and when the observed populations are too small or the observation is intermittent, it will also definitely lead to fragmentation of the continuously changing characters. So in the species defining, as Nooteboom  stated “It is always necessary to study as many collections as possible from numerous herbaria, not only from a restricted area but from the whole area where a taxon may occur”.
Conceived and designed the experiments: YR. Performed the experiments: YQZ ALW. Analyzed the data: YQZ. Contributed reagents/materials/analysis tools: YQZ XMW. Wrote the paper: YR XMW.
- 1. Grang V (1985) The evolutionary process, A critical review of evolutionary theory. New York: Columbia University Press.
- 2. Chen JK, Wang HQ (1986) The application of the concept and method of population to plant taxonomy. J Wuhan Bot Res 4: 377–383.
- 3. Tang YC, Xiang QY (1989) A reclassification of the genus Clematoclethra (Actinidiaceae) and further note on the methodology of plant taxonomy. Acta Phyt Sin 27: 81–95.
- 4. Nooteboom HP (1992) A point of view on the species concept. Taxon 41: 318–320.
- 5. Munro W (1868) A monograph of the Bambusaceae. Trans Linn Soc 26: 1–153.
- 6. Keng PC, Wen TH (1989) A preliminary study on bamboo classification according to the vegetative characters. J Bamb Res 8: 17–29.
- 7. Yi TP (1996) Fargesia. In: Volume Keng PC, Wang ZP, editors. Flora Reipublicae Popularis Sinicae. Beijing: Science Press. 9: 387–478.
- 8. Li DZ, Guo ZH, Chris S (2006) Fargesia. In: Volume Wu ZY, Raven PH, editors. Flora of China. (Poaceae) Beijing: Science Press, St Louis: Missouri Botanical Garden Press. 22: 74–96.
- 9. Lu JL (1981) The new species of Bambusoidea from China. J Henan Coll Agr 74: 70–79.
- 10. Yi TP (1985b) New taxa of bamboo from China. Bull Bot Res 5: 121–137.
- 11. Yi TP (1985a) Classification and distribution of the food bamboos of the giant panda. J Bamb Res 4: 20–45.
- 12. Yi TP (1990) Additional notes on Bambusoideae from Sichuan, China. J Bamb Res 9: 27–32.
- 13. Álvarez I, Wendel JF (2003) Ribosomal ITS sequences and plant phylogenetic inference. Mol Phylog Evol 29: 417–434.
- 14. Roalson EH, Friar EA (2000) Infrageneric classification of Eleocharis (Cyperaceae) revisited: Evidence from the internal transcribed spacer (ITS) region of nuclear ribosomal DNA. Syst Bot 25: 323–336.
- 15. Yano O, Katsuyama T, Tsubota H, Hoshino T (2004) Molecular phylogeny of Japanese Eleocharis (Cyperaceae) based on ITS sequence data, and chromosomal evolution. J Plant Res 117: 409–419.
- 16. Gallego MT, Werner O, Sérgio, Guerra J (2005) A morphological and molecular study of the Syntrichia laevipila complex (Pottiaceae) in Portugal. Nova Hedwigia 80: 301–322.
- 17. Hinchliff CE, Roalson EH (2009) Stem architecture in Eleocharis subg. Limnochloa (Cyperaceae): Evidence of dynamic morphologyical evolution in a group of pantropical sedges. Am J Bot 96: 1487–1499.
- 18. Trevor RH, Stephen AR, Grainne NC (2000) A comparison of ITS nuclear rDNA sequence data and AFLP markers for phylogenetic studies in Phyllmtachys (Bambusoideae, Poaceae). J Plant Res 113: 259–269.
- 19. Košnar J, Košnar J, Herbstová M, Macek P, Rejmánková E, et al. (2011) Natural hybridization in tropical spikerushes of Eleocharis subgenus Limnochloa (Cyperaceae): evidence from morphology and DNA markers. Am J Bot 97: 1229–1240.
- 20. Kumar MA, Sharma P (2011) Molecular and morphological characters: An appurtenance for antagonism in Trichoderma spp. Afr J Biotechnol 10: 4532–4543.
- 21. Doyle JJ, Doyle JL (1987) A rapid DNA isolation procedure for small quantities of fresh leaf material. Phytochem Bull 19: 11–15.
- 22. White TJ, Bruns Y, Lee S, Taylor J (1990) Amplification and direct sequencing of fungal RNA genes for phylogenetics. In: Innis M, Gelfand D, Sninsky J, White T editors. PCR Protocols: A Guide to Methods and Applications. Academic Press, San Diego. 315–322.
- 23. Guo ZH, Chen YY, Li DZ, Yang JB (2001) Genetic variation and evolution of the alpine bamboos (Poaceae: Bambusoideae) using DNA sequence data. Plant Res 114: 315–322.
- 24. Guo ZH, Chen YY, Li DZ (2002) Phylogenetic studies on Thamnocalamus group and its allies (Bambusoideae: Poaceae) based on ITS sequence data. Phylogenet Evol 22: 20–30.
- 25. Tamura K, Peterson D, Peterson N, Stecher G, Nei M, et al. (2011) MEGA5: molecular evolutionary genetics analysis using maximum likelihood, evolutionary distance, and maximum parsimony methods. Mol Biol Evol 28: 2731–2739.
- 26. Swofford DL (2003) PAUP*: Phylogenetic Analysis Using Parsimony (* and Other Methods). Version 4.0b10. Sinauer Associate, Sunderland, MA.
- 27. Barber HN (l955) Adaptive gene substitutions in Tasmanian Eucalyptus: l. Genes controlling the development of glaucousness. Evolution 9: 1–14.
- 28. Hsu PS (1986) The significance of ecological variation in plant taxonomy and evolution. Guihaia 6: 201–216.
- 29. Xiang QY (1989) Taxonomy of Cornus schindleri complex based on quantitative analysis of some characters. Bull Bot Res 9: 125–138.
- 30. Harper JL (1982) After description. In: Newman EI, editor. The plant community as a working mechanism. Oxford: Blackwell Scientific Publications. 11–25.
- 31. Keng PC (1982) A revision of the genera of bamboos for the world. J Bamb Res 1: 1–19.
- 32. Stapleton CMA (1997) Morphology of woody bamboos. In: Chapman GP The Bamboos. London: Academic Press. 251–267.
- 33. Tarciso VP, Filguelras S, Paiva ES (2011) A new combination in Aulonemia (Poaceae: Bambusoidea Bambuseae) based on floral analysis, anatomical features, and distribution. Britt 63: 102–112.
- 34. Yang ZY, Chao Z, Huo KK, Xie H, Tian ZP, et al. (2007) ITS sequence analysis used for molecular identification of the Bupleurum species from northwestern China. Phytomedicine 14: 416–422.
- 35. Eduardo RS, Victoria S, Teresa M (2011) Molecular phylogenetics of the Mesoamerican bamboo Olmeca (Poaceae, Bambuseae): Implications for taxonomy. Taxon 60: 89–98.
- 36. Renvoize SA, Hodkinson TR (1997) Classification of Phyllostachys. In: Volume Chapman GP, editor. The Bamboos. London: Academic Press. 19: 95–105.
- 37. Guo ZH, Li DZ (2004) Phylogenetics of the Thamnocalamus group and its allies (Gramineae: Bambusoideae): inference from the sequences of GBSSI gene and ITS spacer. Mol Phylogenet Evol 24: 1–12.
- 38. Heywood VH (1959) The taxonomic treatment of ecotypic variation. London: Syst Assoc Publ 3: 87–112.
- 39. Hsu PS, Gu DX (1987) An analysis of the variation patterns of infra and inter-populations of Rhododendron ovatum. Bull Bot Res 7: 81–94.
- 40. Lai GH (1999) A revision of some species of the genus of Phyllostachys (Bambusoideae) (I). J Wuhan Bot Res 17: 319–322.
- 41. Lai GH (2001) A revision of some species of the genus Phyllostachys (Bambusoideae) (II). Bull Bot Res 21: 182–185.
- 42. Mayr E (1942) Systematics and the origin of species. New York: Columbia University Press.
- 43. Mayr E (1982) The growth of biological thought. Cambridge: Harvard Univ Press.
- 44. Davis PH, Heywood VH (1963) Principles of angiosperm taxonomy. Edinburgh & London: Oliver & Boyd.