Am. J. Bot. Join the BSA
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Submit a response
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via ISI Web of Science (17)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Zomlefer, W. B.
Right arrow Articles by Judd, W. S.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Zomlefer, W. B.
Right arrow Articles by Judd, W. S.
Agricola
Right arrow Articles by Zomlefer, W. B.
Right arrow Articles by Judd, W. S.
(American Journal of Botany. 2001;88:1657-1669.)
© 2001 Botanical Society of America, Inc.

Systematics

Generic circumscription and relationships in the tribe Melanthieae (Liliales, Melanthiaceae), with emphasis on Zigadenus: evidence from ITS and trnL-F sequence data1

Wendy B. Zomlefer2,5, Norris H. Williams3, W. Mark Whitten3 and Walter S. Judd4

2Department of Botany, 2502 Plant Sciences, University of Georgia, Athens, Georgia 30602-7271 USA; 3Florida Museum of Natural History, Department of Natural History, P.O. Box 117800, University of Florida, Gainesville, Florida 32611-7800 USA 4Botany Department, University of Florida, P.O. Box 118526, Gainesville, Florida 32611-8526 USA

Received for publication August 29, 2000. Accepted for publication February 27, 2001.


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 LITERATURE CITED
 
The circumscription and relationships of genera within the tribe Melanthieae (29 representative taxa) were evaluated using parsimony analyses of ITS (nuclear ribosomal) and trnL-F (plastid) DNA sequence data, alone and in combination. Proposed new generic circumscriptions, strongly supported by the tree statistics and topologies in all analyses, are correlated with potential morphological synapomorphies at the proper level of universality. Based on the molecular cladograms, Stenanthium is biphyletic, and the traditional Zigadenus s.l. (sensu lato) is polyphyletic. Amianthium and Schoenocaulon are distinct entities; the Veratrum complex is conservatively treated as one large monophyletic genus (including Melanthium). Although some generic relationships are not highly resolved, the analyses provide strong support for Zigadenus glaberrimus as sister to the rest of the tribe, and Amianthium muscitoxicum as closely related to Veratrum s.l. As a result of these analyses, seven genera (some with novel circumscription) are recognized within the tribe Melanthieae: Amianthium, Anticlea, Schoenocaulon, Stenanthium, Toxicoscordion, Veratrum, and Zigadenus.

Key Words: ITS • Liliales • Melanthiaceae • Schoenocaulon • Stenanthium • trnL-F • Veratrum • Zigadenus


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 LITERATURE CITED
 
Recent studies have resulted in reevaluation of monocot phylogeny, particularly of the "petaloid lilioid monocots," so named because these plants have conspicuous petaloid tepals and therefore superficially resemble true lilies (Lilium spp.). The variously circumscribed Melanthiaceae, which have included such familiar genera as Veratrum and Zigadenus, constitute a lilioid segregate that has undergone recircumscription in the light of recent morphological and molecular studies. As defined here, the Melanthiaceae sensu the Angiosperm Phylogeny Group (APG, 1998 ) comprise 11–16 genera (~154–201+ spp.) of predominately woodland and/or alpine perennial herbs occurring mainly in the temperate to Arctic zones of the Northern Hemisphere (with one species of Schoenocaulon extending into South America). Table 1 summarizes the genera and monophyletic tribes of this variously circumscribed and little-studied family; notable in this most recent circumscription is the inclusion of the traditional segregate family Trilliaceae (discussed further below).


View this table:
[in this window]
[in a new window]
  		Table 1. Critical characters for the tribes of the Melanthiaceae s.l. sensu APG (1998). Data simplified from summary tables in Zomlefer (1996, 1997a). Potential autapomorphies (*) and synapomorphies (**) are followed by asterisks

 
Batsch (1802) first described the Melanthiaceae (comprising Melanthium, Veratrum, Helonias, and Narthecium), distinguishing them from the Liliaceae by the apically divergent carpels (i.e., carpels free toward apex or the styles free). Subsequent changes in circumscription and familial rank, and transfers of component genera to and from various tribes, have resulted in a very complex taxonomic history, described in detail (with summary charts) by Ambrose (1975) , Goldblatt (1995) , and Zomlefer (1997a, b, c) . The melanthioid genera (Melanthiaceae sensu Tamura, 1998 , in Table 1) usually have been placed (often as a subfamily) within the Liliaceae s.l. (e.g., Engler, 1888 ; Hutchinson, 1934, 1959, 1973 ; Takhtajan, 1980 ; Thorne, 1983 ; Cronquist, 1988 ). Distantly related genera historically included within the melanthioid group typically included Aletris, Metanarthecium, Narthecium, Pleea, and/or Tofieldia (Gray, 1837 ; Gates, 1918 ; Zomlefer, 1997b, c ). To acknowledge the variation within the melanthioid assemblage, the genera were often grouped into several variously delineated tribes, usually the Chionographideae, Heloniadeae (Helonieae), Narthecieae, Melanthieae (or Veratreae), Tofieldieae, and/or Xerophylleae (e.g., Watson, 1879 ; Baker, 1880 ; Bentham, 1883 ; Engler, 1888 ; Hutchinson, 1934, 1959, 1973 ; Schulze, 1978 ).

Critical studies by Dahlgren and associates (Dahlgren and Clifford, 1982 ; Dahlgren, Clifford, and Yeo, 1985 ) established numerous segregate families formerly included within the broadly defined, polyphyletic Liliaceae. The Melanthiaceae sensu Dahlgren and associates, divided into six tribes, comprise a useful reference point for a current understanding of the melanthioid taxa. Of concern here is the monophyly of their Chionographideae, Melanthieae, and Xerophylleae (all as delineated in Table 1), as well as their tribes Narthecieae s.l. (Aletris, Helonias, Heloniopsis, Lophiola, Metanarthecium, Narthecium, Nietneria, Ypsilandra), and Tofieldieae (Harperocallis, Isidrogalvia, Pleea, Tofieldia).

The hypothesis of the Melanthiaceae sensu Dahlgren and associates has been evaluated by cladistic analyses of morphological (Goldblatt, 1995 ) and molecular (Chase et al., 1993, 1995a, b, 2000 ; Duvall et al., 1993 ; Soltis et al., 2000 ) data. These studies resolve the family, the tribe Heloniadeae, and their broadly defined tribe Narthecieae (a combined Narthecieae s.s. and Heloniadeae of other authors) as polyphyletic, and the tribes Tofieldieae, Melanthieae, Xerophylleae, and Chionographideae, as monophyletic (Zomlefer, 1997a, b, c, 1999 ). In addition, the Tofieldieae and part of the Narthecieae s.l. (Aletris, Lophiola, Metanarthecium, Narthecium, Nietneria) compose two lineages quite isolated from each other and from the rest of the Melanthiaceae in the Liliales: the Tofieldieae (or Tofieldiaceae) resolve as a member of the Alismatales, and the Narthecieae (or Nartheciaceae), as a member of the Dioscoreales (Chase et al., 1995a, 2000 ; Caddick et al., 2000 ). Thus, neither tribe Tofieldieae nor tribe Narthecieae is associated with the Melanthiaceae or any other family in the Liliales (Zomlefer, 1997b, c ; Reveal and Zomlefer, 1998 ), and they have not been included in this study.

The remaining tribes (Chionographideae, Melanthieae, and Xerophylleae plus the Heloniadeae s.s. [sensu stricto]) comprise the Melanthiaceae as defined by Tamura (1998) and as provisionally treated by Zomlefer (1997a ; Zomlefer and Perkins, 1999 ; see Table 1). However, according to analyses of molecular data (rbcL: Chase et al., 1995a ; combined trnL-F, rbcL: Rudall et al., 2000 ; combined rbcL, atpB, 18S rDNA: APG, 1998 ; Chase et al., 2000 ; Soltis et al., 2000 ), the family is paraphyletic with Trillium (Trilliaceae) embedded within it, as the sister group of Xerophyllum (Xerophylleae; summary in Zomlefer, 1996 ). This surprising relationship between the Trilliaceae and Melanthiaceae is supported by other molecular studies by Davis (1995 ; cpDNA: including a few melanthioid genera) and Kato et al. (1995 ; rbcL of Trilliaceae). Based on these data, the Trilliaceae have been submerged within the Melanthiaceae s.l. (sensu lato) by the APG (1998) . Currently, this expanded circumscription has the morphological support of extrorse anthers and ovaries often with three distinct styles (Rudall et al., 2000 ) and merits further study. Alternatively, other botanists have suggested segregate familial status for the well-defined "tribal" clades (see Table 1): Chionographidaceae, Heloniadaceae s.s. or s.l., Xerophyllaceae (Takhtajan, 1994, 1997 ; Zomlefer, 1997a ) along with the segregate Trilliaceae, leaving a very restricted Melanthiaceae s.s. consisting of only tribe Melanthieae.

These tribes (or segregate families) are defined by several possible chromosomal, anatomical, and morphological autapomorphies listed in Table 1 (characters indicated with one asterisk) and are linked by several synapomorphies (characters indicated by two asterisks). Although the monophyly of these taxa has, thus, been fairly well established, the relationships among them are somewhat unresolved (summary in Zomlefer, 1999 ). The tribes Heloniadeae and Chionographideae have been linked as clades in morphological (Goldblatt, 1995 ) and various molecular analyses (Chase et al., 1995a ; Fuse and Tamura, 2000 ; Rudall et al., 2000 ). The tribe Melanthieae is linked morphologically as a clade with the monogeneric Xerophylleae (raphides/styloids combination, bulbs, and susceptibility to certain rust fungi). However, as mentioned previously, molecular data of higher level taxa indicate Trillium (Trilliaceae) as the sister group to Xerophyllum (Xerophylleae) and the Trilliaceae/Xerophylleae clade as sister to the Heloniadeae/Chionographideae clade (Chase et al., 1995a ; Rudall et al., 2000 ). Additional study is needed to reconcile these preliminary morphological and molecular results and to establish clear intertribal relationships.

Tribe Melanthieae
Botanists have long recognized the tribe Melanthieae (or Veratreae), the focus of this study, as a cohesive and natural group. As listed in Table 1 (summary in Zomlefer, 1997a ), the tribe is validated by several synapomorphies (a unique type of alkaloid called veratrum alkaloids, unusual anther dehiscence) and potential synapomorphies (conspicuous tepal nectaries, andromonoecism, incompletely fused carpels maturing into a "ventricidal" capsular fruit type [splitting along the ventral sutures], base chromosome number of x = 8). The level of universality of the potential synapomorphies needs to be assessed, as they may have evolved within the tribe or may be basal (and subsequently lost in some members). As mentioned previously, the Melanthieae also share several features with the Xerophylleae.

Until our current analyses, the circumscriptions of the constituent core genera of the Melanthieae (Amianthium, Schoenocaulon, Stenanthium, Veratrum-Melanthium, Zigadenus s.l.) have not been the subjects of rigorous character analyses, and their intergeneric relationships were also unresolved. The few descriptive unpublished dissertations on some of these groups lack a phylogenetic perspective and, thus, do not address the monophyly of the taxa under study: Melanthium s.l. pro parte (Bodkin, 1978 ), Schoenocaulon (Frame, 1990 ), Veratrum s.l. pro parte, (Zimmerman, 1958 ), and Zigadenus (s.l.: Walsh, 1940 ; Preece, 1956 ; pro parte: Schwartz, 1994 ). Ambrose (1975, 1980) included a few representatives [Amianthium (Zigadenus) muscitoxicum, Stenanthium gramineum, Veratrum-Melanthium (three spp.), Zigadenus (five spp.)] in a phenetic overview of the Melanthiaceae s.l. (Liliaceae subfam. Melanthioideae). His results were ambiguous for the circumscriptions and relationships of these genera: Veratrum-Melanthium cluster as a distinct subunit, Zigadenus glaberrimus is isolated from the rest of the Melanthieae, and the other four species of Zigadenus cluster variously with Amianthium or Stenanthium.

Zigadenus
In particular, Zigadenus (however previously circumscribed) represents a heterogeneous, nonmonophyletic grouping, i.e., not defined by synapomorphies (Zomlefer, 1997a ). Zigadenus was originally established by Michaux (1803) for Z. glaberrimus, a distinctive and phenetically isolated species in the complex. The subsequent complex taxonomic history of the genus is summarized in Rydberg (1903) , Gates (1918) , Preece (1956) , and Zomlefer (1997a) . Taxonomists have long grappled with defining this assemblage, often by erecting a number of variously defined segregate genera: Amianthium (Gray, 1837) , Anticlea (Kunth, 1843) , Oceanoros (Small, 1903 ), Toxicoscordion (Rydberg, 1903 ), and Tracyanthus (Small, 1903 ). Contemporary botanists have typically accepted only one segregate, a monotypic Amianthium, with the remaining ~19 species maintained in Zigadenus s.l. (see Utech, 1986 ). Based on the segregates, Preece (1956) proposed four sections for Zigadenus s.l. but the monophyly and relationships of these subgroups of Zigadenus have not been assessed, and clear morphological synapomorphies for these groups have not been readily apparent.

The intrageneric taxa of Zigadenus have not been studied cladistically, although phenetic studies of a few species provide some basic information on character variation. However, hypotheses of phylogenetic relationships derived from phenetic approaches (as in Schwartz, 1994 ) are inappropriate. In phenetic analyses of morphological characters, Zigadenus glaberrimus is the most isolated species of those studied in the complex (Ambrose, 1980 ; Schwartz, 1994 ). When analyzed with distance Wagner methods, Zigadenus glaberrimus, Z. leimanthoides, and Amianthium muscitoxicum had very different patterns of cpDNA restriction site fragments compared to five other species (Schwartz, 1994 ).

Current investigation
We assessed the probable taxonomic utility of certain sequences (overview in Soltis and Soltis, 1998 ) based on preliminary analyses (Zomlefer and Perkins, 1999 ). In this current study, we examined generic circumscription and relationships in the tribe Melanthieae by performing parsimony analyses of sequence data from two regions: the trnL (UAA)-trnF (GAA) intergenic intron and spacer region (treated as a single matrix designated trnL-F, plastid) and the internal transcribed spacer region ITS-1, 5.8S, and ITS-2 (referred to as ITS, nuclear ribosomal DNA or nrDNA). The trnL-F region is mainly noncoding (Taberlet et al., 1991 ) and often taxonomically useful at intrageneric levels (e.g., Gielly and Taberlet, 1996 ; Gielly et al., 1996 ; Kim, Hart, and Mes, 1996 ). The ITS nrDNA region, two noncoding spacer regions flanking the 5.8 S gene, has often been used for determining generic and species relationships; the sequence is typically too divergent at higher levels (summary in Baldwin et al., 1995 ).

Each molecular matrix was analyzed separately and the results compared to assess congruence of topology. Studies have indicated that fewer equally parsimonious trees (with greater bootstrap support for internal nodes) are often produced from combined data than when each set is analyzed independently (Barrett, Donoghue, and Sober, 1991 ; Donoghue, 1994 ; Kron and Judd, 1997 ; Chase and Cox, 1998 ; Soltis et al., 1998 ; Kron, Judd, and Crayn, 1999 ; Mishler, 2000 ; Whitten, Williams, and Chase, 2000 ).


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 LITERATURE CITED
 
Plant material
Table 2 is a list of species examined and voucher information. All material was fresh or silica dried (Chase and Hills, 1991 ) and collected specifically for this project by W. B. Zomlefer and colleagues cited in the acknowledgments. Taxa within the Melanthieae were carefully chosen to represent variation within genera, especially within Zigadenus.


View this table:
[in this window]
[in a new window]
  		Table 2. List of taxa examined and voucher location

 
Molecular techniques
DNA was extracted from plant material according to the methods of Doyle and Doyle (1987) , scaled down to 1.0-mL extraction volumes. DNA was precipitated overnight at –20°C with 0.65 volumes of isopropanol, centrifuged, washed twice with 70% ethanol, and dried. The pellet was resuspended in 75 µL of Tris-EDTA buffer and stored at –20°C. Amplification of DNA was generally performed using 50-µL reactions with 35 cycles, 2.5 mmol/L MgCl2, 1.0 mol/L betaine, and a hot start, using Promega (Promega, Madison, Wisconsin, USA) or Epicentre (Epicentre Technologies, Madison, Wisconsin, USA) buffers and Taq polymerase. Annealing temperatures for amplification were 55°–58°C for trnL-F. For ITS, a touchdown thermal cycling program was used; the initial annealing temperature was 76°C, decreasing 1°C per cycle for 15 cycles, followed by 21 cycles at 59°C. Amplification and sequencing primers were those of Sun et al. (1994) for ITS and Taberlet et al. (1991) for trnL-F. PCR products were purified using QIAquick columns (Qiagen, Santa Clarita, California, USA) and underwent dye terminator cycle sequencing with ABI (Foster City, California, USA) reagents (5-µL reactions).

The final sequencing gel runs were completed by the DNA Sequencing Core Facility (DSEQ, UF) with ABI 377 and ABI 373A automated sequencers. Both strands were sequenced to assure accurate base calling. Individual sequences were edited and assembled using the ABI software packages "Sequence NavigatorTM" and "AutoAssemblerTM" on an Apple PowerMac computer and aligned manually. Gaps were coded as missing values. The ends of matrices were trimmed to exclude sequencing artifacts, and five regions of ambiguous alignment were excluded from the trnL-F matrices for the family (totaling 196 bp) and three regions for the tribe Melanthieae (48 bp). Sequences are deposited in GenBank (trnL-F: AF303663–AF303701; ITS: AF303702–AF303731). The aligned data matrices are available from the authors (WBZ and WMW; wendyz@dogwood.botany.uga.edu; whitten@flmnh.ufl.edu) and also archived at http://ajbsupp.botany.org/v88/zomlefer-melanthiaceae.txt (trnL-F matrix for Melanthiaceae) and http://ajbsupp.botany.org/v88/zomlefer-melanthieae.txt (trnL-F and ITS matrices for tribe Melanthieae).

Search strategies
Four matrices were analyzed for this study: trnL-F for Melanthiaceae (36 taxa + outgroups) and trnL-F, ITS, and combined trnL-F/ITS for the tribe Melanthieae (29 taxa). Cladistic analyses were performed using PAUP* version 4.0b4a (Swofford, 2000) with all characters weighted equally. The outgroups for the Melanthiaceae (Lilium, Liliaceae; Smilax, Smilacaceae) were based on studies by Chase et al. (2000) , and the functional outgroup for the tribe Melanthieae (Zigadenus glaberrimus) was determined from our initial trnL-F analyses of the family (Fig. 1; see RESULTS for further discussion). The separate trnL-F analysis for the family (Fig. 1) was run using the heuristic search option (MULTREES, SPR, 1000 random replicates, holding ten trees per replicate). All analyses for the tribe Melanthieae (trnL-F, ITS, and combined trnL-F/ITS; Figs. 2 and 3) employed the branch-and-bound search option (MULTREES, SPR, 1000 random replicates, holding ten trees per replicate).



View larger version (45K):
[in this window]
[in a new window]
  		Fig. 1. Bootstrapped branch-and-bound consensus tree for the Melanthiaceae with trnL-F data. Numbers above clades = bootstrap values. Numbers below clades = branch lengths from one (of 60) most parsimonious trees (heuristic: CI = 0.817, RI = 0.883). Numbers below clades in parentheses = decay values. Branch lengths and decay values given only for clades where topology is identical to those on the bootstrap tree. This analysis establishes Zigadenus glaberrimus as the functional outgroup for tribe Melanthieae. H = Heloniadeae; C = Chionographideae; P = Parideae (Trilliaceae); X = Xerophylleae. Outgroups (Smilax smallii, Smilacaceae; Lilium catesbaei, Liliaceae) based on Chase et al. (2000)

 


View larger version (39K):
[in this window]
[in a new window]
  		Fig. 2. Bootstrapped branch-and-bound consensus tree for tribe Melanthieae with trnL-F data. Numbers above clades = bootstrap values. Numbers below clades = branch lengths from one (of nine) most parsimonious trees (branch-and-bound: tree statistics in Table 3 ). Numbers below clades in parentheses = decay values. Branch lengths and decay values given only for clades where topology is identical to those on the bootstrap tree. Am. = Amianthium, Sch. = Schoenocaulon, Stenan. = Stenanthium, Zig. s.s. = Zigadenus s.s

 
Trees were evaluated on the basis of tree length, consistency index (CI), and retention index (RI) as calculated by PAUP* (Table 3). Branch-and-bound bootstrap consensus trees (Figs. 1–3) were generated for all data sets (1000 replicates; Felsenstein, 1985 ). We assessed congruence of the separate data sets for tribe Melanthieae by comparison of tree statistics (CI and RI, Table 3) and the topology of the bootstrap consensus trees, designating the following categories of bootstrap support: unsupported (<50%), weak (50–74%), moderate (75–84%), and strong (85–100%). In addition, decay values were generated with AutoDecay 4.0.1 (1000 random addition sequence replicates; Eriksson, 1998 ).


View this table:
[in this window]
[in a new window]
  		Table 3. Values and statistics from branch-and-bound PAUP analyses of separate and combined data matrices for the tribe Melanthieae (29 taxa) for cladograms in Figs. 2 and 3

 

   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 LITERATURE CITED
 
Analyses of trnL-F data for Melanthiaceae
To ascertain the functional outgroup of the tribe Melanthieae, the initial analyses included representatives of all tribes of Melanthiaceae (36 taxa; Fig. 1) plus outgroups Lilium (Liliaceae) and Smilax (Smilacaceae), based on Chase et al. (2000) . These analyses include only trnL-F data because ITS sequences could not be aligned unambiguously to compare these higher level taxa. The problem of aligning very divergent ITS sequences for more distant taxonomic levels has been addressed by several researchers (e.g., Bogler and Simpson, 1996 ) who have proposed elaborate computerized alignment schemes that likely constrain the data at inappropriate levels of taxonomic utility (Slowinski, 1998 ). The trnL-F data establish Zigadenus glaberrimus as the functional outgroup for the tribe Melanthieae (for subsequent analyses), reinforce the monophyly of the tribes of Melanthiaceae, and moderately support tribal relationships (including sister groups Parideae [Trilliaceae] and Xerophylleae). Notable is that four individuals of Z. glaberrimus (two shown) were sequenced to confirm its divergent position in all analyses included in this study. (All four individuals had identical sequences.)

Analyses of ITS and trnL-F for tribe Melanthieae
Based on the trnL-F analyses of the family (Fig. 1), Zigadenus glaberrimus was used as the functional outgroup for evaluating generic relationships within the tribe Melanthieae (29 representative taxa). Figures 2 and 3 are the bootstrapped, branch-and-bound, consensus trees for trnL-F, ITS, and combined trnL-F/ITS data, respectively. Due to space limitations, examples of the most parsimonious cladograms and strict consensus trees are not included here, but branch lengths (from one of a few most parsimonious trees) are given on those clades in the bootstrapped trees where the topologies match. Table 3 lists the tree statistics for all three data sets. We justify performing the combined analysis by comparison of the bootstrap consensus trees for trnL-F (Fig. 2) and ITS (Fig. 3A). These trees do not exhibit hard incongruence, i.e., the highly supported clades for the trnL-F analysis do not conflict with highly supported clades with ITS (Whitten, Williams, and Chase, 2000) . The ITS and combined consensus trees have similar topologies (see Fig. 3). The combined analysis (Fig. 3B) produced fewer trees (two) compared to trnL-F or ITS alone (nine and three trees, respectively), generally with equivalent bootstrap support (14, 12, and 15 clades with >85% support) and comparable retention indices (0.887, 0.912, and 0.891, respectively). Decay values for all major clades are mostly three or greater for the ITS analysis and mostly four or greater for the combined analysis (except for one major clade in each analysis with a decay value of two). In comparison, the decay values are lower for major clades on the trnL-F tree (including three clades with a decay value of one), which indicates much weaker support for this data.



View larger version (58K):
[in this window]
[in a new window]
  		Fig. 3. Bootstrapped branch-and-bound consensus tree for tribe Melanthieae with (A) ITS sequence data and (B) combined ITS and trnL-F sequence data. Numbers above clades = bootstrap values. Numbers below clades = branch lengths; for (A) values are from one (of three) most parsimonious trees; for (B) values are from one (of two) most parsimonious trees (branch-and-bound: tree statistics in Table 3 ). Numbers below clades in parentheses = decay values. Branch lengths and decay values given only for clades where topology is identical to those on the bootstrap tree. Branch-and-bound, most parsimonious trees with both data sets are basically equivalent, supporting the proposed generic clades (in shaded boxes). M. = Melanthium; V. = Veratrum; Z. = Zigadenus

 
Of particular interest in all analyses is the polyphyly of Zigadenus, with species of Stenanthium embedded within two clades. Amianthium muscitoxicum (Zigadenus muscitoxicum) is only distantly related to the other Zigadenus species. The remaining species of the Zigadenus complex form four strongly supported groups: (1) the strikingly divergent Zigadenus glaberrimus (the type species of the genus), (2) the Toxicoscordion clade, (3) the Stenanthium clade (including Stenanthium gramineum), and (4) the Anticlea clade (including Stenanthium occidentale). The taxa of the Veratrum/Melanthium complex form one strongly supported clade (all analyses) with two well-supported subclades—species of Veratrum s.s. and the traditional North American segregate Melanthium (plus the Asian V. maackii)—with ITS and combined analyses.

The branching order at deeper nodes is more ambiguous. Topologies differ in the position of Schoenocaulon: embedded within the tribe and sister to Amianthium with trnL-F (weak support) vs. much more strongly supported as sister to most of the tribe with ITS and combined analyses. In addition, the Toxicoscordion clade is sister to the Amianthium/Schoenocaulon/Veratrum clade in the trnL-F analyses, but Stenanthium/Anticlea is sister to Amianthium/Schoenocaulon with ITS and combined data. Support for either of these divergent topologies is rather weak to moderate (see Figs. 2 and 3A, B; bootstrap values 58, 73, and 57; decay values 1, 3, and 2, respectively).


   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
LITERATURE CITED
 
Circumscription of the genera of tribe Melanthieae
All analyses challenge the traditional delimitations of Stenanthium, Melanthium, Veratrum, and Zigadenus. Table 4 lists the proposed genera based on the clades resolved by our molecular data and their defining characters (including possible nonmolecular synapomorphies). Formal comprehensive cladistic analyses of morphological characters are required to confirm these putative synapomorphies (since some characters are homoplasious) and to identify additional ones. We recognize genera based on criteria outlined by Backlund and Bremer (1998) for general principles of classification: first and foremost, they are monophyletic but secondarily, they should have strong statistical support and also be recognizable based on morphological characters. Several of these clades, established as subgroups (sections, subgenera, or segregate genera) by traditional systematists, have additional support from an historical standpoint. Further nomenclatural consequences of our work (including recombinations) will be addressed in a companion paper (Zomlefer and Judd, unpublished data).


View this table:
[in this window]
[in a new window]
  		Table 4. Characters defining the clades (tentative genera) of tribe Melanthieae, with particular emphasis on the species of Zigadenus s.l., based on this molecular study. Data were compiled from preparatory examination of specimens (FLAS, GH, MO, and NY) by Zomlefer and from Walsh (1940); Preece (1956); Kupchan, Zimmerman, and Afonso (1961); Hitchcock et al. (1969); Cronquist et al. (1977); Sterling (1982); Wiggins (1980); Utech (1986, 1987a, b); Welsh et al. (1987); Gleason and Cronquist (1991); Turner (1992); McNeal (1993); Mullin (1994); Schwartz (1994); Hess and Sivinski (1995); Zomlefer (1997a); and Frame, Espejo, and López-Ferrari (1999). [!] = taxa sequenced for this study; * = possible autapomorphies; x = base chromosome number

 
The conventional circumscription of Schoenocaulon (Brinker, 1942 ; Frame, 1990 ), well supported by morphological synapomorphies (fibrous-tunicate bulb covered by dark brown to black scales and fibers; unbranched, spicate [to somewhat racemose] inflorescence of tiny, sessile to subsessile flowers), is reinforced by our analyses.

Veratrum, a problematic group of 20–45 species with a complicated taxonomic history, has been variously circumscribed with Melanthium submerged totally or in part (detailed summaries in Zimmerman, 1958 ; Kupchan, Zimmerman, and Afonso, 1961 ; Bodkin, 1978 ; Zomlefer, 1997a ). Contemporary authors generally recognize both genera, with Melanthium typically composed of two to four eastern North American species (Bodkin, 1978 ). The monophyly of Veratrum and Melanthium needs to be assessed, especially to determine whether segregation of Melanthium (however defined) creates a paraphyletic Veratrum. In addition, Veratrum s.l. has also formally (Loesener, 1926, 1927, 1928 ) and informally (Zimmerman, 1958 ) been subdivided into several subgenera and sections (all of doubtful monophyly).

Unfortunately, modern studies on the large Veratrum-Melanthium complex have been regional in scope and have not addressed the morphological variation of the entire group on a worldwide basis. The wide elliptical leaves plaited along several prominent primary veins (as in the familiar V. album or V. viride), considered distinctive for Veratrum, actually do not occur in all veratrums (Zimmerman, 1958 ; Mathew, 1989 ), and characters traditionally used to segregate Melanthium (two perigonal glands per tepal, epitepalous stamens, incurved filaments, clawed tepals) actually exhibit a wide range of variation (including intermediates) within the complex (Zimmerman, 1958 ; Kupchan, Zimmerman, and Afonso, 1961 ). In his unpublished monograph of North American Melanthium, Bodkin (1978) even mentioned that four (undesignated) Asian species of Veratrum should probably be transferred to Melanthium. Although generic division is problematical, the monophyly of the whole group is supported by the synapomorphies of floccose pubescence of vegetative parts (at least of the inflorescence) and broadly winged seeds (Zomlefer, 1997a ). In a phenetic study of the family, the three species of the Veratrum-Melanthium group formed a distinct cluster (Ambrose, 1975, 1980 ).

We sequenced nine representative species of this large complex: four of the traditional Veratrum s.s. species (plicate leaves), all four Melanthium species (sensu Bodkin), plus one Asian intermediate (V. maackii). In our analyses, the whole complex forms a monophyletic group with moderate (trnL-F) to strong (ITS; ITS + trnL-F) support. However, taxon sampling for the Veratrum-Melanthium clade is insufficient for assessing division of the complex, although the traditional Veratrum s.s. and Melanthium species (along with the Asian intermediate) form two strongly supported subclades with the ITS and combined data (Fig. 3), indicating possible recognition of two subgenera. Our preliminary investigation confirms the need for further molecular and morphological study with identification and inclusion of all morphologically intermediate taxa. Currently, we conservatively treat this group as a single broadly defined genus, Veratrum.

In a previous overview of the genera of the Melanthieae for the Generic Flora of the Southeastern United States project (Zomlefer, 1997a ), identifying synapomorphies for both Stenanthium and Zigadenus was nearly impossible, indicating that neither taxon, as traditionally defined, is likely monophyletic. The weakly defined Stenanthium s.l. [3–5 spp.; narrow (lanceolate) acuminate tepals] has sometimes been divided into two genera (see Rydberg, 1900 ; Gleason, 1952 ; Utech, 1987a, b ): a monotypic Stenanthium (S. gramineum; x = 10; inflorescence paniculate; flowers rotate with spreading tepals; tepal glands lacking) and Stenanthella (2–4 spp.; x = 8; inflorescence racemose to paniculate; flowers campanulate with reflexed tepals; one bilobed gland/tepal). Until our investigation, the relationship of Stenanthium s.l. to the rest of the melanthioids was unknown, although in phenetic analyses by Ambrose (1975, 1980) , S. gramineum clustered with several Zigadenus species.

Our analyses included Stenanthium gramineum (Stenanthium s.s.) and S. occidentale (to represent Stenanthella). The results with all data sets strongly support a markedly different circumscription of Stenanthium: the type element, Stenanthium gramineum, forms a clade with Zigadenus densus and Z. leimanthoides. These two closely related Zigadenus species are possibly conspecific (McDearman, 1984 ) and were segregated from the rest of the Zigadenus complex by Small (1903 ; Oceanoros, Tracyanthus) and Preece (1956 ; sect. "Oceanoros," not validly published). With inclusion of these taxa, Stenanthium is now defined by the putative autapomorphies of a slender (cylindrical) bulb, one obscure (or lacking) gland per tepal, and a base chromosome number of 10 (see Table 4). All members of this clade are restricted to the Coastal Plain of the southeastern United States.

Stenanthium occidentale, however, resolves with species of Zigadenus previously designated as the segregate genus Anticlea (Kunth, 1843 ; Rydberg, 1903 ) or Zigadenus sect. Anticlea (see Preece, 1956 ). The Anticlea clade also has strong support with all three molecular data sets. Within the clade, Stenanthium occidentale is well supported as sister to the other species (Fig. 3) with the ITS and combined data sets (but this relationship within the clade is unresolved with trnL-F; Fig. 2). Tentative autapomorphies include a narrowly ovoid bulb and one bilobed gland per tepal. Notable also is that these species have a much wider distribution (North America to Guatemala; Asia) than those of the Stenanthium clade.

In all analyses, the Stenanthium clade and Anticlea are sister groups (strong support with trnL-F and combined data sets and moderate support with ITS). The combined Stenanthium-Anticlea clade has the tentative synapomorphies of slim bulbs and the tendency for a half-inferior ovary.

In all analyses, the rest of the Zigadenus complex form three distantly related clades: Amianthium (monotypic), Zigadenus s.s. (monotypic), and Toxicoscordion. As mentioned in the introduction, species of the Zigadenus complex encompass a wide range of variation and have been variously divided, but only one segregate, Amianthium muscitoxicum, has gained acceptance among modern floristic botanists. This monotypic genus of the Coastal Plain of the southeastern United States has the autapomorphies of obscure/absent tepal glands, reddish to purple, somewhat fleshy seed coat, and the unique veratrum alkaloid, amianthine (Neuss, 1953 ; Kupchan, Zimmerman, and Afonso, 1961 ; Utech, 1986 ; Zomlefer, 1997a ). All three subsets of the molecular data strongly support the separation of Amianthium from other elements of the Zigadenus complex.

Zigadenus glaberrimus, the type species of the genus, is also restricted to the Coastal Plain. Historically, some botanists (e.g., Rydberg, 1903 ; Small, 1903, 1933 ; Gates, 1918 ) advocated a monotypic Zigadenus, comprising only this very distinct species. However, more contemporary botanists have not accepted a restricted circumscription of Zigadenus, although Preece (1956) proposed placement of this species in its own section in his unpublished monograph of the genus. Phenetic studies (Ambrose 1975, 1980 ; Schwartz, 1994 ) also reveal Z. glaberrimus as an isolated entity. Autapomorphies (discussed in Zomlefer, 1997a ) include an unusual chromosome number (n = 26, tentatively reported by Preece, 1956 ), a rhizome lacking a bulb, and several anatomical features (foliar stomata with two aperture lips, distinct root exodermis, bracteolate pedicels, and dense tannin-like inclusions; Ambrose, 1975 ). Our initial trnL-F analyses of the Melanthiaceae (Fig. 1) confirm the isolated position of Z. glaberrimus: the species is strongly supported as the sister to all the rest of the tribe Melanthieae (i.e., the functional outgroup), thereby confirming a monotypic circumscription of Zigadenus.

The genus Toxicoscordion comprises the remaining taxa of the Zigadenus complex. These species, segregated by Rydberg (1903) , are distributed in the midwestern United States to western North America and include the well-known poisonous "Zigadenus" species of the rangelands (see Marsh, Clawson, and Marsh, 1915 ; Marsh, Clawson, and Roe, 1926 ). In their unpublished dissertations, both Preece (1956) and Schwartz (1994) treated this group as Zigadenus sect. Chitonia. Probable autapomorphies include conspicuously clawed tepals (especially the inner three), one obovate gland per tepal, and a base chromosome number of 11. We sequenced eight out of ten of the Toxicoscordion taxa (see Table 4). Analyses of all three subsets of our molecular data resulted in Toxicoscordion as a strongly supported clade, distinct from other species in the Zigadenus complex.

Generic relationships in the tribe Melanthieae
In our study, circumscription of genera is generally more highly resolved than certain associations between them. Generic relationships are usually best supported by the ITS data (Fig. 3A). However, all analyses provide strong support for Zigadenus glaberrimus as the sister to the rest of the tribe and for the Stenanthium and Anticlea clades as sister groups. Amianthium is always linked to Veratrum. Topologies differ in the position of Schoenocaulon, but its placement as sister to most of the tribe (Fig. 3A, B: bootstrap values 89 and 80, decay values 8 and 6), rather than to Amianthium (Fig. 2: bootstrap 62, decay 1), is strongly supported. Since the ITS tree exhibits no hard incongruence with the trnL-F tree, this discrepancy between topologies likely reflects a lack of sufficient characters (sampling error) and homoplasy in the trnL-F data at this level (Mishler, 2000 ; Whitten, Williams, and Chase, 2000 ). Therefore, we conclude that the trnL-F vs. ITS/combined data sets (i.e., plastid and nuclear DNA, respectively) are likely not tracking different evolutionary or genetic phenomena such as lineage sorting, gene duplication, and introgression resulting in conflicting phylogenetic signals between data sets (Bull et al., 1993 ; Huelsenbeck, Bull, and Cunningham, 1996 ; Dolphin et al., 2000 ). The trees deviate (trnL-F vs. ITS or ITS + trnL-F), and support is consistently moderate to weak for the relationships of the Toxicoscordion clade and the Stenanthium/Anticlea clade.

Further study is needed to clarify generic relationships, including additional data appropriate at this level (e.g., matK; Zomlefer et al., unpublished data) and comprehensive phylogenetic analyses of morphological characters to distinguish potential synapomorphies linking the genera. Currently, we have identified putative synapomorphies only for the Stenanthium/Anticlea sister clades (slim bulbs, tendency for a half-inferior ovary). However, our investigation with ITS and trnL-F is an excellent foundation for focusing attention on this problematic tribe, historically lacking treatment from a phylogenetic perspective.

General conclusions
The development of a phylogenetic understanding of the lilioid monocots is crucial for the establishment of a natural system of monocot classification (Chase et al., 1995a ). As higher level relationships of the monocots are resolved, the need increases for comprehensive, lower level, generic and species studies employing rigorous phylogenetic character analyses. The tribe Melanthieae is a diverse, yet superficially similar, lilioid group whose generic circumscriptions have been more problematic than in many other monocots. In our study we have used a conservative approach by broadly defining monophyletic genera and identifying potential synapomorphies (Table 4).

Phylogenetic decisions (at any taxonomic level) based on molecular data are often not readily adopted by nonmolecular botanists actively involved with teaching, floristics, and field work. A higher level example is the circumscription of Liliaceae: despite the groundbreaking work on monocot phylogeny in many recent molecular studies (as well as previous morphological analyses by Dahlgren and associates), contemporary floristic treatments (e.g., The Flora of North America project [FNA, 1993 ]; Hickman; 1993 , Kartesz, 1994 ; Wunderlin, 1998 ; Diggs, Lipscomb, and O'Kennon, 1999 ), still often include the melanthioid genera within a variously defined, polyphyletic Liliaceae s.l. Molecular studies, therefore, especially need to include practical recommendations, including careful documentation of morphological synapomorphies for diagnosable characters ultimately used in keys and descriptions by floristic/field-oriented botanists (Sanders and Judd, 2000) .


View this table:
[in this window]
[in a new window]
  		Table 4. Continued

 

   FOOTNOTES
 
1 The authors thank the following botanists and horticulturists for providing silica-dried samples and pressed vouchers: Richard Bruce (Highlands Biological Station Botanical Garden), Roy Buck, Hillary Cherry (FLAS), Ron Determann (Atlanta Botanical Garden), Holly C. Forbes (University of California Botanical Garden), Angus Gholson, Jr., Madeleine Groves (Atlanta Botanical Garden), Ron M. Halvorson, Linda Hasselbach, Charlie Lapham, Mia Molvray (OKL), Robert Naczi (KNK), Richard G. Olmstead (WTU), Carolyn L. Parker (ALA), Dan Pittillo (WCUH), Daniela Roth, Robert Sivinski, J. Dan Skean (ALBC), Gerald Smith, Minoru N. Tamura, Fred Utech, David H. Wagner, Tom Wendt (TEX), Thomas F. Wieboldt (VPI), Barbara Wilson (OSC), Richard D. Worthington (UTEP), and William B. Wykle (Highlands Biological Station Botanical Garden). Mark W. Chase and two anonymous reviewers critiqued the manuscript, and Wesley Higgins provided valuable advice to WBZ in the lab. A very special thank-you goes to Angus Gholson, Jr., for showing WBZ populations of Amianthium muscitoxicum, Chamaelirium luteum, Melanthium virginicum, Stenanthium gramineum, Veratrum woodii, Zigadenus densus, and Z. glaberrimus in northern Florida. Hillary Cherry and G. "Stinger" Guala assisted in the field, and Kent D. Perkins (FLAS) helped to process voucher specimens. This project was initiated by WBZ, while an NSF post-doctoral associate for the Generic Flora of the Southeastern United States project (WSJ, PI, DEB-9419940). Equipment for this work was funded by NSF grants DEB-9509071(WMW, PI) and DEB-9815821 (NHW, PI). Back

5 Author for reprint requests (wendyz{at}dogwood.botany.uga.edu ). Back


   LITERATURE CITED
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 LITERATURE CITED
 
Ambrose J. D. 1975 Comparative anatomy and morphology of the Melanthioideae (Liliaceae). Ph.D. dissertation, Cornell University, Ithaca, New York, USA

Ambrose J. D. 1980 A re-evaluation of the Melanthioideae (Liliaceae) using numerical analyses. In C. D. Brickell, D. F. Cutler, and M. Gregory [eds.], Petaloid monocotyledons, 65–81. Academic Press, London, UK

APG [Angiosperm Phylogeny Group]. 1998 An ordinal classification for the families of flowering plants. Annals of the Missouri Botanical Garden 85: 531-553[CrossRef][ISI]

Backlund A. K. Bremer 1998 To be or not to be—principles of classification and monophyletic plant families. Taxon 47: 391-400[CrossRef][ISI]

Baker J. G. 1880 A synopsis of the Colchicaceae and the aberrant tribes of Liliaceae. Journal of the Linnean Society, Botany 17: 405-410

Baldwin B. G. M. J. Sanderson J. M. Porter M. F. Wojciechowski C. S. Campbell M. J. Donoghue 1995 The ITS region of nuclear ribosomal DNA: a valuable source of evidence on angiosperm phylogeny. Annals of the Missouri Botanical Garden 82: 247-277[CrossRef][ISI]

Barrett M. M. J. Donoghue E. Sober 1991 Against consensus. Systematic Zoology 40: 486-493[CrossRef]

Batsch A. J. 1802 Tabula affinitatum regni vegetabilis. Landes-Industrie-Comptior, Weimar, Germany

Bentham G. 1883 Liliaceae. In G. Bentham and J. D. Hooker [eds.], Genera plantarum, vol. 3, 748–836. L. Reeve, London, UK

Bodkin N. L. 1978 A revision of North American Melanthium L. (Liliaceae). Ph.D. dissertation, University of Maryland, College Park, Maryland, USA

Bogler D. J. B. B. Simpson 1996 Phylogeny of Agavaceae based on ITS rDNA sequence variation. American Journal of Botany 83: 1225-1235[CrossRef][ISI]

Brinker R. R. 1942 Monograph of Schoenocaulon. Annals of the Missouri Botanical Garden 29: 287-315[CrossRef]

Bull J. J. J. P. Huelsenbeck C. W. Cunningham D. L. Swofford P. J. Waddell 1993 Partitioning and combining data in phylogenetic analysis. Systematic Biology 42: 384-497[CrossRef]

Caddick L. R. P. J. Rudall P. Wilkin M. W. Chase 2000 Yams and their allies: systematics of Dioscoreales. In K. L. Wilson and D. A. Morrison [eds.], Monocots: systematics and evolution, 475–487. CSIRO Publishing, Collingwood, Victoria, Australia

Chase M. W. A. V. Cox 1998 Gene sequences, collaboration and analysis of large data sets. Australian Systematic Botany 11: 215-229[CrossRef][ISI]

Chase M. W. M. R. Duvall H. G. Hills J. G. Conran A. V. Cox L. E. Eguiarte J. Hartwell M. F. Fay L. R. Caddick K. M. Cameron S. Hoot 1995a Molecular phylogenetics of Lilianae. In P. J. Rudall, P. J. Cribb, D. F. Cutler, and C. J. Humphries [eds.], Monocotyledons: systematics and evolution, 109–137. Royal Botanic Gardens, Kew, UK

Chase M. W. H. G. Hills 1991 Silica gel: an ideal material for field preservation of leaf samples for DNA studies. Taxon 40: 215-220[CrossRef][ISI]

Chase M. W. D. W. Stevenson P. Wilkin P. J. Rudall 1995b Monocot systematics: a combined analysis. In P. J. Rudall, P. J. Cribb, D. F. Cutler, and C. J. Humphries [eds.], Monocotyledons: systematics and evolution, 685–730. Royal Botanic Gardens, Kew, UK

Chase M. W. et al 1993 Phylogenetics of seed plants: an analysis of nucleotide sequences from the plastid gene rbcL. Annals of the Missouri Botanical Garden 80: 528-580[CrossRef][ISI]

Chase M. W. et al 2000 Higher-level systematics of the monocotyledons: an assessment of current knowledge and a new classification. In K. L. Wilson and D. A. Morrison [eds.], Monocots: systematics and evolution, 3–16. CSIRO Publishing, Collingwood, Victoria, Australia

Cronquist A. 1988 The evolution and classification of flowering plants, 2nd ed. New York Botanical Garden, Bronx, New York, USA

Cronquist A. A. H. Holmgren N. H. Holmgren J. L. Reveal P. K. Holmgren 1977 Intermountain flora, vol. 6. Columbia University Press, New York, New York, USA

Dahlgren R. M. T. H. T. Clifford 1982 The monocotyledons: a comparative study. Academic Press, London, UK

Dahlgren R. M. T. H. T. Clifford P. F. Yeo 1985 The families of the monocotyledons. Springer-Verlag, Berlin, Germany

Davis J. I. 1995 A phylogenetic structure for the monocotyledons, as inferred from chloroplast DNA restriction site variation, and a comparison of measures of clade support. Systematic Botany 20: 503-527[CrossRef][ISI]

Diggs G. M. B. L. Lipsomb R. J. O'Kennon 1999 Shinners and Mahler's illustrated flora of north central Texas. Botanical Research Institute of Texas, Fort Worth, Texas, USA

Dolphin K. R. Belshaw C. D. L. Orme D. L. J. Quicke 2000 Noise and incongruence: interpreting results of the incongruence length difference test. Molecular Phylogenetics and Evolution 17: 401-406[CrossRef][ISI][Medline]

Donoghue M. J. 1994 Progress and prospects in reconstructing plant phylogeny. Annals of the Missouri Botanical Garden 81: 405-418[CrossRef][ISI]

Doyle J. J. J. L. Doyle 1987 A rapid DNA isolation procedure for small quantities of fresh leaf tissue. Phytochemical Bulletin 19: 11-15

Duvall M. R. M. T. Clegg M. W. Chase W. D. Clark W. J. Kress H. G. Hills L. E. Eguiarte J. F. Smith B. S. Gaut E. A. Zimmer G. H. Learn 1993 Phylogenetic hypotheses for the monocotyledons constructed from rbcL data. Annals of the Missouri Botanical Garden 80: 607-619[CrossRef][ISI]

Engler A. 1888 Liliaceae. In A. Engler and K. Prantl, Die natürlichen Pflanzenfamilien, vol. III, part 5, 10–91. Kreysing, Leipzig, Germany

Eriksson T. 1998 AutoDecay ver. 4.0.1. Program distributed by author, Department of Botany, Stockholm University, Stockholm, Sweden

Felsenstein J. 1985 Confidence limits on phylogenies: an approach using the bootstrap. Evolution 46: 59-173

FNA [Flora of North America Editorial Committee] [eds.] 1993 Flora of North America, vol. 1, 299–316. Oxford University Press, Oxford, UK

Frame D. M. 1990 A revision of Schoenocaulon (Liliaceae—Melanthieae). Ph.D. dissertation, City University of New York, New York, New York, USA

Frame D. M. A. Espejo A. R. López-Ferrari 1999 A conspectus of Mexican Melanthiaceae including a description of new taxa of Schoenocaulon and Zigadenus. Acta Botánica Mexicana 48: 27-50

Fuse S. M. N. Tamura 2000 A phylogenetic analysis of the plastid matK gene with emphasis on Melanthiaceae sensu lato. Plant Biology 2: 415-427

Gates R. R. 1918 A systematic study of the North American Melanthaceae [sic] from a genetic standpoint. Journal of the Linnean Society, Botany 44: 131-172

Geilly L. P. Taberlet 1996 A phylogeny of the European gentians inferred from chloroplast trnL (UAA) intron sequences. Botanical Journal of the Linnean Society 120: 57-75[CrossRef]

Geilly L. Y.-M. Yuan P. Küpfer P. Taberlet 1996 Phylogenetic use of noncoding regions in the genus Gentiana L.: chloroplast trnL (UAA) intron versus nuclear ribosomal internal transcribed spacer sequences. Molecular Phylogenetics and Evolution 5: 460-466[CrossRef][ISI][Medline]

Gleason H. A. 1952 The new Britton and Brown illustrated flora of the northeastern United States and adjacent Canada, vol. 1, The Pteridophyta, Gymnospermae and Monocotyledoneae. Hafner, New York, New York, USA

Gleason H. A. A. Cronquist 1991 Manual of vascular plants of northeastern United States and adjacent Canada, 2nd ed. The New York Botanical Garden, Bronx, New York, USA

Goldblatt P. 1995 The status of R. Dahlgren's orders Liliales and Melanthiales. In P. J. Rudall, P. J. Cribb, D. F. Cutler, and C. J. Humphries [eds.], Monocotyledons: systematics and evolution, 181–200. Royal Botanic Gardens, Kew, UK

Gray A. 1837 Melanthacearum Americae septentrionalis revisio. Annals of the Lyceum of Natural History of New York 4: 105-140

Hess W. J. R. C. Sivinski 1995 A new species of Zigadenus (Liliaceae) from New Mexico, with additional comments on section Anticlea. Sida 16: 389-400

Hickman J. C. [ed.] 1993 The Jepson manual: higher plants of California. University of California Press, Berkeley, California, USA

Hitchcock C. L. A. Cronquist M. Ownbey J. W. Thompson 1969 Vascular plants of the Pacific Northwest, part 1, Vascular cryptogams, gymnosperms, and monocotyledons. Univeristy of Washington Press, Seattle, Washington, USA

Huelsenbeck J. P. J. J. Bull C. W. Cunningham 1996 Combining data in phylogenetic analysis. Trends in Ecology and Evolution 11: 152-158[CrossRef]

Hutchinson J. 1934 The families of flowering plants, vol. 2. Monocotyledons. MacMillan, London, UK

Hutchinson J. 1959 The families of flowering plants, 2nd ed., vol. 2. Monocotyledons. Clarendon Press, Oxford, UK

Hutchinson J. 1973 The families of flowering plants, 3rd ed. Oxford University Press, Oxford, UK

Kartesz J. T. 1994 A synonymized checklist of the vascular flora of the United States, Canada, and Greenland, 2nd ed., vol. 1, Checklist. Timber Press, Portland, Oregon, USA

Kato H. R. Terauchi F. H. Utech S. Kawano 1995 Molecular systematics of the Trilliaceae sensu lato as inferred from rbcL sequence data. Molecular Phylogenetics and Evolution 4: 184-193[CrossRef][ISI][Medline]

Kim J. H. H. 'T Hart T. H. M. Mes 1996 The phylogenetic position of east Asian Sedum species (Crassulaceae) based on chloroplast DNA trnL (UAA)-trnF (GAA) intergenic spacer sequence variation. Acta Botanica Neerlandica 45: 309-321

Kron K. A. W. S. Judd 1997 [1998]. Systematics of the Lyonia group (Andromedeae, Ericaceae) and the use of species as terminals in higher-level cladistic analyses. Systematic Botany 22: 479-492[CrossRef][ISI]

Kron K. A. W. S. Judd D. M. Crayn 1999 Phylogenetic analyses of Andromedeae (Ericaceae subfam. Vaccinioideae). American Journal of Botany 86: 1290-1300[Abstract/Free Full Text]

Kunth C. S. 1843 Melanthaceae [sic]. Enumeratio plantarum, vol. 4. J. G. Collae, Stuttgart, Germany

Kupchan S. M. J. H. Zimmerman A. Afonso 1961 The alkaloids and taxonomy of Veratrum and related genera. Lloydia 24: 1-26

Loesener O. 1926 Studien über die Gattung Veratrum und ihre Verbreitung. Verhandlungen Botanischen Vereins die Provinz Brandenburg 68: 105-166

Loesener O. 1927, 1928 Übersicht über Arten der Gattung Veratrum. Repertorium Specierum Novarum Regni Vegetabilis 24: 61-72, 25: 1-10

Marsh C. D. A. B. Clawson H. Marsh 1915 Zygadenus [sic], or death camas. Bulletin of the U.S. Department of Agriculture 125: 1-46

Marsh C. D. A. B. Clawson G. C. Roe 1926 Nuttall's death camas (Zygadenus [sic] Nuttallii) as a poisonous plant. Bulletin of the U.S. Department of Agriculture 1376: 1-13

Mathew B. 1989 A review of Veratrum. Plantsman 11: 35-61

McDearman W. W. 1984 Systematics of Zigadenus densus and Z. leimanthoides and phylogenetic implications of breeding systems in the Veratreae. ASB Bulletin 31: 71 (Abstract)

McNeal D. W. 1993 Zigadenus. In J. C. Hickman [ed.], The Jepson manual: higher plants of California, 1210–1211. University of California Press, Berkeley, California, USA

Michaux A. 1803 Flora Boreali-Americana, vol. 1. Levrault, Paris, France

Mishler B. D. 2000 Deep phylogenetic relationships among "plants" and their implications for classification. Taxon 49: 661-683[CrossRef][ISI]

Mullin J. M. 1994 Zigadenus. In G. Davidse, M. Sousa S., and A. O. Chater [eds.], Flora Mesoamericana, vol. 6, Alismataceae a Cyperaceae, 26. Universidad Nacional Autónoma de México, México, D.F

Neuss N. 1953 A new alkaloid from Amianthium muscaetoxicum Gray. Journal of the American Chemical Society 75: 2772-2773[CrossRef]

Preece S. J. 1956 A cytotaxonomic study of the genus Zigadenus. Ph.D. dissertation, State College of Washington, Pullman, Washington, USA

Reveal J. L. W. B. Zomlefer 1998 Two new orders for monocotyledonous plants. Novon 8: 176-177[CrossRef][ISI]

Rudall P. J. K. L. Stobart W.-P. Hong J. G. Conran C. A. Furness G. C. Kite M. W. Chase 2000 Consider the lilies: systematics of Liliales. In K. L. Wilson and D. A. Morrison [eds.], Monocots: systematics and evolution, 347–357. CSIRO Publishing, Collingwood, Victoria, Australia

Rydberg P. A. 1900 Studies on the Rocky Mountain flora