Systematic Botany (2014), 39(1): pp. 75–89
Copyright 2014 by the American Society of Plant Taxonomists
Date of publication 02/05/2014
Testing Deep Reticulate Evolution in Amaryllidaceae Tribe Hippeastreae (Asparagales)
with ITS and Chloroplast Sequence Data
Alan W. Meerow,
Douglas E. Soltis,
and Pamela S. Soltis
Florida Museum of Natural History, University of Florida, Gainesville, Florida 32611, U. S. A.
Department of Biology, University of Florida, Gainesville, Florida 32611, U. S. A.
Departmento de Silvicultura, Facultad de Ciencias Forestales, Universidad de Chile, Casilla 9206, Santiago, Chile.
USDA-ARS-SHRS-National Germplasm Repository, 13601 Old Cutler Road, Miami, Florida 33158, U. S. A.
Author for correspondence (email@example.com)
Communicating Editor: Chrissen E. C. Gemmill
Abstract—The phylogeny of Amaryllidaceae tribe Hippeastreae was inferred using chloroplast (3
ycf1, ndhF, trnL
) and nuclear
(ITS rDNA) sequence data under maximum parsimony and maximum likelihood frameworks. Network analyses were applied to resolve
conflicting signals among data sets and putative scenarios of reticulate evolution. All analyses of all regions consistently revealed two major
clades, which are formalized as subtribes: A) Traubiinae, formed by Traubia, Placea, Phycella, Rhodolirium,andFamatina maulensis, character-
ized by x = 8, rare polyploidy, and a capitate stigma, and B) Hippeastrinae, including Eithea, Habranthus, Haylockia, Hippeastrum, Rhodophiala ,
Sprekelia, Zephyranthes ,andFamatina pro parte, characterized by a range of basic chromosome numbers (x = 6–11) and frequent polyploidy
and aneuploidy. No clear morphological features diagnose the latter clade, which contains ca. 90% of the tribe’s species diversity. Our
phylogenetic results question the monophyly of all genera in the tribe and show widespread cytonuclear discordance within the mainly
Neotropical Hippeastrinae, further supporting putative ancient hybridization(s) preceding the radiation of this major clade. In contrast, the
Traubiinae, endemic to Chile and Argentina, show a tree-like pattern of evolution, consistent with the apparent absence of allopolyploidy in
this clade. A brief description, circumscription, and geographic distribution are provided for each subtribe.
Keywords—cpDNA, incongruence, monocots, networks, nrDNA, reticulation.
Amaryllidaceae sensu stricto (s. s.; Amaryllidaceae subfam-
ily Amarylloideae sensu Chase et al. 2009; Asparagales) have a
worldwide distribution and comprise 14 tribes and approxi-
mately 70 genera (Meerow and Snijman 1998; Meerow et al.
1999, 2000) of mostly bulbous geophytes, many with a long
history of cultivation as ornamentals (Huxley et al. 1992;
Brickell and Zuk 1997). This clade is diagnosed and distin-
guished from the closely related Agapanthus L’He
Alliaceae by its unique alkaloidal chemistry, inferior ovary,
and hollow style (Meerow and Snijman 1998; Rudall et al.
2002; Judd et al. 2008). Phylogenetic analyses of DNA
sequences show that American Amaryllidaceae (except for
New World Crinum L. species) form a well-defined clade
(Ito et al. 1999; Meerow et al. 1999) composed of two major
clades (Meerow et al. 2000): 1) the Hippeastroid clade, formed
by tribes Griffinieae and Hippeastreae; and 2) the Andean
tetraploid clade, formed by tribes Eustephieae, Clinantheae,
Hymenocallideae, and Eucharideae.
Tribe Hippeastreae sensu Meerow et al. (2000) is com-
posed of 10–13 genera and ca. 180 species (Table 1; Meerow
and Snijman 1998; Meerow et al. 2000; Meerow 2010;
Oliveira 2012), with a major center of diversification in
central Chile and western/Andean Argentina and a second
in eastern Brazil and northeastern Argentina (Meerow and
Snijman 1998; Arroyo-Leuenberger and Dutilh 2008).
Although the tribe’s greatest species richness is in South
America, Habranthus Herb. and Zephyranthes Herb. show
another center of diversity in Mexico and extend to the
Greater Antilles and both the southwestern and southeast-
ern United States (Meerow and Snijman 1998). Despite the
systematic attention that Hippeastreae has received because
of its horticultural importance, generic relationships within
the tribe continue to be debated (e.g. Traub 1963; Ravenna
2003; Flagg et al. 2010), mostly due to the lack of unequiv-
ocal diagnostic morphological characters and pervasive
homoplasy, the latter a common feature of Amaryllidaceae
s. s. (Meerow et al. 1999).
Phylogenetic analyses based on nrDNA ITS sequences
(Meerow et al. 2000; Meerow 2010) have helped to elucidate
relationships within the tribe and indicated that certain
genera, such as Rhodophiala C. Presl, Habranthus, and
Zephyranthes, are non-monophyletic. However, these studies
lacked good representation of Chilean-Argentinean endemic
groups, such as Famatina Ravenna, Phycella Lindl., Placea
Miers, Rhodolirium Phil., and Traubia Moldenke (Ravenna
2003). Based on recombination break point analyses of ITS,
Meerow (2010) suggested an ancient reticulation event in
Hippeastreae, with subsequent diversification of daughter
clades. The “deep reticulation” hypothesis points to the line-
ages leading to Rhodophiala bifida and to Hippeastrum Herb.
as the parents in a putative hybridization event that gave rise
to a lineage that includes Eithea Ravenna, Rhodophiala pro
parte (p. p.), and Zephyranthes from South America, the
Caribbean, and the southeastern U. S. A. If concerted evolu-
tion has acted upon ITS following a reticulation event
(Wendel et al. 1995a; A
lvarez and Wendel 2003; Soltis et al.
2008), comparison with a phylogeny derived from organellar
genomes may also detect reticulate patterns, with incongru-
ent placements of hybridizing taxa/clades in different gene
trees (Soltis and Kuzoff 1995; Funk and Omland 2003; Linder
and Rieseberg 2004; McBreen and Lockhart 2006).
Several chloroplast DNA (cpDNA) markers have also been
analyzed in Amaryllidaceae (matK: Ito et al. 1999; rbcL and
: Meerow et al. 1999; ndhF : Meerow 2010),
but with low sampling of hippeastroid taxa. Of the cpDNA
markers previously explored in the Hippeastreae, ndhF
(Meerow 2010) and trnL
(including the trnL
intron and trnL
intergenic spacer; A. Meerow
unpublished data) provided the strongest phylogenetic signal.
However, with only 28 and 23 species (including tribe
Griffinieae) represented per region, respectively, the taxon
sampling was insufficient for comparison with the ITS tree.
The ndhF results, based on slightly greater sampling,
appear incongruent with the ITS phylogeny (Meerow 2010),
further supporting the hypothesis of ancient reticulation
in Hippeastreae. However, the ndhF tree remains weakly
supported and with low resolution. Therefore, tests of the
deep reticulation hypothesis in Hippeastreae require a well-
resolved cpDNA gene tree with appropriate taxon sampling
for comparison with the nuclear-encoded ITS gene tree.
A cpDNA region with much potential for phylogenetic
inference at the genus and species levels is ycf1, the longest
open reading frame in the plastome but of unknown function
(Drescher et al. 2000; Kleine et al. 2009). The 3
half of this
gene, embedded in the small single-copy region (SSC), shows
high variability (Timme et al. 2007; Neubig et al. 2009; Parks
et al. 2009; Drew and Systma 2011; Majure et al. 2012). We
here explore the utility of this coding region along with ndhF
, with the goal of obtaining a cpDNA
topology with greater resolution and support.
Over the last 60 yr, an increased awareness of the impor-
tance of reticulate mechanisms of evolution has accumulated,
both at shallow and deep levels (e.g. Anderson 1949; Stebbins
1950; Grant 1981; Rieseberg and Soltis 1991; Doolittle 1999;
Mallet 2005, 2007; Arnold and Fogarty 2009). In plants,
hybridization is considered a powerful and widespread process
(e.g. Rieseberg 1997; Arnold 1997, 2006; Soltis and Soltis 2009).
However, such reticulation results in patterns that cannot be
represented as bifurcating phylogenetic trees (Sneath 1975),
and this limitation continues to plague phylogeny recon-
struction in many groups of plants.
The inference of phylogenetic networks is an alternative
method to reconcile conflicting data sets that may represent
a non-tree-like evolutionary history (Linder and Rieseberg
2004; Vriesendorp and Bakker 2005; Huson and Bryant 2006;
MacBreen and Lockhart 2006; Nakhleh 2011). Unrooted net-
works have been highlighted as an appropriate method for the
visualization of conflicting signals among different data sets
(e.g. Huson and Bryant 2006; Wa
gele and Mayer 2007;
Morrison 2010). On the other hand, methods to estimate
rooted phylogenetic networks, such as hybridization net-
works, constitute an attractive approach because such a net-
work can be explicitly interpreted as an evolutionary history
that has involved reticulation (Huson et al. 2005; Klo
Huson 2008; Huson and Scornavaca 2010; Huson et al. 2010).
In this study we sought to: 1) enhance taxon sampling of tribe
Hippeastreae relative to previous studies (Meerow et al. 2000;
Meerow 2010) by incorporating an increased representation
of Chilean-Argentinean endemic taxa, additional Habranthus
and Zephyranthes species, and critical species that had not
been sampled before in phylogenetic studies, such as
Sprekelia howardii; 2) test further the monophyly of non-
monotypic genera in tribe Hippeastreae and infer intergeneric
relationships; 3) improve the resolution of the cpDNA tree by
increasing the sampling for ndhF and trnL
exploring the highly variable coding region, ycf1; 4) assess
the scope of incongruence between the ITS and cpDNA trees
by comparing relative branch support values between the
trees; and 5) represent the putative reticulate history of
Hippeastreae by applying a method of phylogenetic network
inference that accommodates reticulation.
Materials and Methods
Taxon Sampling—A total of 100 taxa were sampled (Appendix 1),
including most genera of Hippeastreae and nine outgroup taxa from
tribes Griffinieae, Lycorideae, Pancratieae, and Cyrtantheae (Meerow
et al. 1999, 2000). Two accessions each were analyzed for Rhodophiala
ananuca and Rhodophiala splendens; two infraspecific taxa were sampled
for Rhodophiala bifida (i.e. subspecies bifida and granatiflora). Material was
primarily obtained from A. W. M.’s bulb and DNA collections (Meerow
et al. 2000; Meerow 2010) and the bulb collection of N. G. for Chilean-
Argentinean groups. Each specimen used in this study has a herbarium
voucher, mainly in FLAS, FTG, NA, CONC, B, or K (Appendix 1).
Approximately 65 new samples of Hippeastreae were included in this
study compared to earlier investigations.
Molecular Methods—Total genomic DNA was extracted following a
CTAB method (Doyle and Doyle 1987; Cullings 1992) using
herbarium material and silica-gel-dried leaves collected in the field or
from cultivated plants. Extractions from fresh leaves frozen in liquid
nitrogen were also performed for certain accessions (e.g. Sprekelia
howardii, both Rhodophiala splendens, Zephyranthes treatiae); however, the
best-quality DNA was obtained from 10–15 mg of silica-dried leaves. Ad
hoc modifications to the CTAB protocol included the use of 2-mL
Eppendorf tubes, 1.2 mL of CTAB buffer per sample, 2.5–3 hr of incuba-
tion at 55
C, a second chloroform step, and resuspension in 50 mLof
O for RNA work (Fisher Scientific, Fair Lawn, New Jersey).
For silica-dried samples, DNA yields of 200–700 ng/mL were obtained.
Nineteen DNA extractions from A. W. M.’s DNA bank were used in cases
where no additional leaf material was available.
This study generated 309 new sequences for ITS, 3
. We also used previously published ITS, ndhF,and
sequences (Meerow et al. 1999, 2000; Shi et al. 2006;
Meerow 2010; Snijman and Meerow 2010). Amplifications of the target
loci were performed via the polymerase chain reaction (PCR) in a
Biometra T3 (Whatman Biometra, Goettingen, Germany) or AB 2720
(Applied Biosystems, Foster City, California) thermocycler. The PCR
amplifications of ITS and trnL
were performed using GoTaq
(Promega, Madison, Wisconsin), whereas for ndhF and 3
ycf1 we used
high-fidelity DNA polymerase (Finnzymes OY, Espoo, Finland)
because it enables longer and better-quality amplifications. These
amplicons produce superior sequences, especially when there are long
homopolymer runs (Fazekas et al. 2010), and consequently minimize the
number of sequencing reactions necessary for long (i.e. > 2 kb) AT-rich
regions. All PCR products were checked visually on 1.5% agarose gels.
The ITS region (ITS1–5.8S-ITS2) was amplified using primers AB101
and AB102 (Douzery et al. 1999) and the following PCR protocol: initial
denaturation at 94
C for 2 min preceding 30 cycles of denaturation (94
for 30 s), annealing (60
C for 45 s), and extension (72
C for 1 min), with a
final extension at 72
C for 7 min. The PCRs were done in 25-mL volumes
containing 2.5 mL of 5x buffer, 2–2.5 mL of 25 mM MgCl
,1mL of 2.5 mM
dNTPs, 2 mL of 1 M betaine, 2 mL of a 5 mM solution of each primer, 0.1 mL
of Taq polymerase (0.5 U), and 10 ng of DNA template.
region was amplified in one reaction using
primers ‘c’ and ‘f’ (Taberlet et al. 1991) and the following PCR protocol:
initial denaturation at 95
C for 5 min, six –1
C touchdown cycles (94
for 1 min, 58–53
C for 1 min, 72
C for 2.5 min), 29 regular cycles (94
1 min, 52
C for 1 min, 72
C for 2.5 min), and a final extension at 72
12 min. The PCRs were prepared as for ITS.
Table 1. Genera of Amaryllidaceae tribe Hippeastreae.
Genus # Species Distribution
Eithea 1–2 SE Brazil
Famatina 4 Andes of Mediterranean Chile and Argentina
Habranthus 40 Neotropics, including Mexico
Haylockia 1 Uruguay, NE Argentina, S Brazil
Hippeastrum 60 Neotropics, mostly in South America
Phycella 6 Mediterranean Chile
Placea 6 Mediterranean Chile
Sprekelia 2 Central Mexico
Rhodolirium 5 Andes of Mediterranean Chile and Argentina,
Coastal Atacama Desert
Rhodophiala 8 Mediterranean Chile, Atacama Desert;
only R. bifida in Uruguay, NE Argentina,
S Brazil, Paraguay
Zephyranthes 50 Neotropics, including Caribbean, Mexico,
southern U. S. A.
Tocantinia 1–2 Central Brazil
Traubia 1 Coastal Central Chile
76 SYSTEMATIC BOTANY [Volume 39
Amplifications of the ndhF locus were performed in two reactions,
one using the primer combination 32F and 1101R (Terry et al. 1997)
for the 5
end, and the second using newly designed primers 950F
) and 2071R (5
) for the 3
end. The PCRs were per-
formed in 25-mL volumes containing 5 mLof5
Phusion HF buffer,
0.5 mL of 25 mM MgCl
, 0.5 mL of 2.5 mM dNTPs, 1 mL of a 5 mM solution
of each primer, 0.25 mL of Phusion DNA polymerase (0.5 U), and 20 ng
of DNA template. The PCR protocol consisted of an initial denaturation at
C for 2 min, 35 cycles of denaturation (98
C for 10 s), annealing (60
for 15 s), and extension (72
C for 30 s), with a final extension at 72
3 min. Certain PCR products that gave poor direct sequencing results
ndhF: Famatina maulensis, Rhodophiala ananuca-1, Zephyranthes
ndhF: Habranthus sp.) were cloned using the StrataClone
PCR cloning kit (Agilent Technologies, La Jolla, California), following
the manufacturer’s instructions, and 3 or 4 colonies were sequenced per
reaction using primers T3 (5
Whole sequences of 3
ycf1 from three species of Amaryllidaceae (i.e.
Amaryllis belladonna L., Scadoxus cinnabarinus (Decne.) Friis & Nordal,
Crinum asiaticum L.) were provided by Dr. J. Leebens-Mack (University
of Georgia) and Drs. P. R. Steele and J. C. Pires (University of Missouri-
Columbia), and primers were designed from the alignment of these
sequences. Approximately 2,600 bp of 3
ycf1 were amplified using
primers Am1-F (5
–GAATCTTGATGC CAATA TGCG– 3
). Alternatively, a 2,300 -bp
amplicon was obtained using Am1s-F (5
) and Am3s-R (5
for samples that failed to amplify with the first primer combination. The
PCR protocol and reaction mix were as for ndhF, with the following
differences: annealing temperature of 56
C for Am1s-F/Am3s-R, cycle
extension time of 1 min, and 30 ng of template DNA per reaction.
The PCR products were submitted for Sanger sequencing to the DNA
Sequencing Core Laboratory at the University of Florida Interdisciplinary
Center for Biotechnology Research (ICBR). Standard procedures and ABI
Prism BigDye Terminator cycle sequencing protocols were followed. For
all regions except 3
ycf1, the same primers applied for amplification were
used for sequencing. For 3
ycf1, we employed six sequencing reactions
using prim ers Am1s-F, Am1-R (5
), Am2-R (5
), Am3-F (5
), and Am3s-R.
Sequences and Data Sets—Sequences were assembled and edited with
Sequencher 4.2.2 (Gene Codes Corp., Ann. Arbor, Michigan) and
Geneious Pro 5.4 (Drummond et al. 2010). Nucleotide alignments were
obtained with MAFFT (Katoh et al. 2002, 2009) and the G-INS-i algorithm,
recommended for sequences with global homology. The 1PAM nucleo-
tide scoring matrix, default gap opening penalty (1.53), and default
gap-offset value of 0 were applied to all loci. Final alignments were
obtained by manual editing in Geneious Pro. Sequences were submitted
to GenBank (Appendix 1), and alignments and trees to TreeBASE
(study number TB2:S13712). Pseudogenes of protein-coding loci were
identified by the presence of premature stop codons and unusually high
substitution rates (Gojobori et al. 1982; Balakirev and Ayala 2003) from
amino-acid alignments in Geneious Pro and were excluded from all sub-
sequent analyses (for detailed results and discussion on pseudogenes see
Data were explored using several alternative process partitions
(Bull et al. 1993). The ITS matrix is the largest with 102 sequences;
nine Hippeastrum species were not included in the cpDNA sampling.
Regarding the cpDNA loci, the ndhF matrix contains 89 sequences (only
half of the gene was obtained for Rhodophiala aff. advena due to
failure to amplify the 5
portion in this sample), while trnL
contains 90 sequences. The 3
ycf1 matrix includes 88 sequences. Each of
the four cpDNA regions was analyzed independently to allow inference
of the respective gene trees (Table 2). However, independent analyses of
intron and trnL
spacer (Figs. S1, S2) resulted in
poorly resolved trees due to few variable characters, and consequently
these two regions were combined in the final analyses. Taxa with identi-
cal sequences were included in the trees a posteriori by manually mod-
ifying the tree files and including them with the taxon with which they
share the same sequence. Only those 89 accessions for which at least two
of the three cpDNA regions were non-pseudogenic were included in the
combined cpDNA matrix and correspond to the cpDNA1 set (Table 2);
however, analyses were run with only 85 sequences because four species
had cpDNA sequences that were identical to those of other taxa.
To test for neutral sequence evolution, an assumption of phylogenetic
methods (e.g. Wheeler and Honeycutt 1988; Prychitko and Moore 1997;
Castresana 2000), a detection analysis of positive selection was performed
for ndhF and 3
ycf1. The analyses did not detect any evidence of positive
selection in these coding regions (for details see Supplemental Material).
Additionally, the recombination signal detected by Meerow (2010) in
ITS was reevaluated with the expanded data set (for details see
Phylogenetic Analyses—Indels that were potentially phylogenetically
informative were coded with SeqState (Mu
ller 2005) using simple indel-
coding (Simmons and Ochoterena 2000) and appended manually to the
alignment; autapomorphic and doubtfully homologous gaps were treated
as missing data. Maximum parsimony (MP) searches were conducted for
each data matrix including coded gaps with TNT ver. 1.1 (Goloboff et al.
2008) with sectorial search and parsimony ratchet (Nixon 1999) runs of
100 iterations each from 2,000 initial random addition sequences. The
consistency index (CI) and retention index (RI) were calculated using the
Stats.run script in TNT. Parsimony jackknife analyses (MP-JK; Farris et al.
1996) were conducted using TNT with the removal probability set to 37%
(Farris et al. 1996) and following the recommendations given by Simmons
and Freudenstein (2011). Five thousand MP-JK pseudoreplicates were
performed with 100 random addition tree-bisection-reconnection (TBR)
searches (each with no more than 10 trees held) per pseudoreplicate. JK
support was expressed as frequency differences (GC; Goloboff et al.
2003), an approach that estimates support of a given clade relative to the
clade that contradicts it most frequently. Resampling values greater that
80% were considered strong support, 70–80% were designated moderate
support, and less than 70% as weak. All MP analyses were run while
collapsing branches with a minimum length of zero to increase tree-
search efficiency (Davis et al. 2005). A strict consensus tree was calculated
using all most parsimonious trees found per analysis. All trees were
visualized and rooted with the three Cyrtanthus W. Aiton species forming
a monophyletic group (Meerow et al. 1999; Snijman and Meerow 2010) in
Dendroscope ver. 2.6.1 (Huson et al. 2007).
Maximum likelihood (Felsenstein 1973) analyses were conducted for
each data matrix using RAxML ver. 7.2.8 and 7.3.0 (Stamatakis 2006),
excluding gap-coded characters. For the trnL
analyses, models were partitioned by locus to increase model fit by
accommodating locus-specific variation (e.g. Castoe et al. 2004; Brandley
et al. 2005). ML searches were conducted using the new rapid hill-
climbing algorithm with 2,000 independent searches starting from ran-
domized parsimony trees with the GTRGAMMA model and four discrete
Table 2. Data-matrix and parsimony-tree statistics for each analysis. “CI” = consistency index on the most parsimonious tree(s) for all characters.
“RI” = retention index.
The cpDNA2 dataset differs from cpDNA1 only in the exclusion of Habranthus andalgalensis and Zephyranthes longistyla.
% Missing / %
# Most parsimonious
trees CI RI
ITS rDNA 102 91 646 296 (45.8) 0.1 / 0.2 6 194 0.62 0.87
ycf1 88 74 2,190 195 (8.9) 0.2 / 2.0 8 4 0.83 0.94
ndhF 89 76 1,959 87 (4.4) 1.5 / 0.0 0 1 0.86 0.94
90 49 861 35 (4.1) 0.2 / 1.1 6 11 0.93 0.96
cpDNA1 89 85 5,010 320 (6.4) 3.5 / 1.0 14 14 0.84 0.94
87 83 5,010 313 (6.3) 3.6 / 1.0 14 14 0.85 0.95
2014] GARCIA ET AL.: DEEP RETICULATION IN HIPPEASTREAE 77
rate categories. Likelihood bootstrap analyses (ML-BS; Felsenstein 1985)
were conducted with 5,000 pseudoreplicates for each data set, and sup-
port was expressed as absolute frequencies, as calculated by RAxML.
Incongruence was determined by inspection and assessment of branch
support (Soltis and Kuzoff 1995; Mason-Gamer and Kellogg 1996; Pirie
et al. 2009). Individual and combined gene topologies were compared by
eye, and incongruence was invoked only when conflicting nodes were
supported by at least 70% MP-JK and ML-BS values in both trees being
compared. Species determined to have incongruent placements among
individual cpDNA gene trees were then removed from the combined
cpDNA matrix, forming the cpDNA2 set (Table 2). ITS and cpDNA
matrices were not combined due to major topological conflicts between
them, following the conditional combination approach (Johnson and
Chromosome numbers, when known, were labeled on the trees and
network in the printed version to show karyological variation in the
tribe’s phylogeny. Chromosome counts were obtained from the literature
ndez 1970; Naranjo 1969, 1974; Naranjo and Andrada
1975; Flory 1977; Flory and Coulthard 1981; Arroyo 1982; Williams 1982;
Greizerstein and Naranjo 1987; Davin
a and Ferna
ndez 1989; Grau and
Bayer 1991; Palma-Rojas 2000; Naranjo and Poggio 2000; Flagg et al. 2002;
Baeza et al. 2007, 2009; Cisternas et al. 2010; Oliveira 2012; J. Dutilh, pers.
comm.; S. Arroyo-Leuenberger, pers. comm.) and from the first author’s
Hybridization Networks—Network analyses were performed to exam-
ine conflicting signals among ITS and cpDNA data sets and to represent
the reticulate evolution of Hippeastreae. A hybridization network
(HybNet) is defined as a rooted phylogenetic network N on a set of
taxa c that represents two rooted, bifurcating (or multifurcating) trees,
T1andT2, and has a minimum number of reticulations (Huson et al. 2010;
Huson and Scornavaca 2010). Networks were constructed from MP strict
consensus trees, in which nodes with support below a certain threshold
were collapsed to discriminate incongruence attributed to hybridization
from that caused by stochastic error (McBreen and Lockhart 2006;
Pirie et al. 2009; Huson et al. 2010). Four HybNets that represent different
levels of phylogenetic uncertainty were estimated. The first three net-
works were computed from the MP strict consensus trees resulting from
analyses of cpDNA1 and an ITS matrix reduced to the same 89 taxa where
nodes were not collapsed when both corresponding MP-JK and ML-BS
values were ³90% (HybNet1), ³80% (HybNet2), and ³70% (HybNet3).
The most liberal scenario, HybNet4, was estimated from the MP strict
consensus trees resulting from analyses of cpDNA2 and an ITS matrix
of 87 taxa; nodes with both MP-JK and ML-BS ³70% were not collapsed.
The resulting trees were used as input for Splitstree 4.10 (Huson and
Bryant 2006), and HybNets were computed using “RECOMB2007” (Huson
pper 2007), Cyrtanthus herrei as outgroup, with default settings.
ITS Tree—The ITS data set for 102 accessions was easily
alignable, comprising 646 nucleotide positions including
gaps (0.3%) and missing data (0.1%). The ITS alignment
included 244 bp of the ITS1 region, 163 bp of the 5.8S gene,
and 239 bp of the ITS2 region. Indels longer than a single
nucleotide were only observed in the outgroup taxa. Some
species of Hippeastrum, Rhodophiala, Placea, Phycella, and
Zephyranthes from Mexico and Texas have sequences that
are identical or nearly so to other congeneric species. The
number of parsimony-informative characters, lengths of
aligned matrices, and other characteristics for all data sets
are found in Table 2.
The ITS topology agrees with previous studies involving
tribe Hippeastreae (Meerow et al. 2000; Meerow 2010).
Hippeastreae comprises two strongly supported major clades
(Fig. 1): A) Traubia, Placea, Phycella, Rhodolirium,andFamatina
maulensis,andB)Rhodophiala, Habranthus, Haylockia Herb.,
Hippeastrum, Sprekelia Heist., Zephyranthes, and the remainder
of Famatina. The ITS tree suggests that the only monophyletic
non-monotypic genera in tribe Hippeastreae are Hippeastrum
(60 spp. total, 20 sampled) and Sprekelia (2 spp.).
cpDNA Trees—Of the three cpDNA regions analyzed,
ycf1 was the longest and most variable with almost 9% of
its sites being parsimony-informative (Table 2); this region
provided the highest resolution among species (Fig. S3).
Despite ndhF having twice the length of trnL
both had approximately the same proportion of all parsimony-
informative positions (4.4% and 4.1%, respectively) and
yielded similar resolutions (Figs. S4, S5). As in ITS, several
species of Hippeastrum, Phycella, Placea, and North American
Zephyranthes showed identical or almost identical sequences
for the different cpDNA regions.
The concatenated cpDNA data sets contained a total of
5,010 nucleotide positions, of which approximately 6% were
parsimony-informative. Additionally, 14 informative indels
were considered in the MP analyses (Table 2). The cpDNA1
data set contains 89 sequences, while the cpDNA2 data set
contains 87 (Table 2), due to the exclusion of Habranthus
andalgalensis and Z. longistyla, which show strongly supported
conflicts among individual cpDNA regions (Figs. S1, S2, S3).
Topological differences between cpDNA1 and cpDNA2
analyses involve Clade B, particularly the placements of
Habranthus andalgalensis and Z. longistyla . Five branches with
weak support in the cpDNA1 tree change to moderate or
strong support when these two species are excluded, except
the branch leading to a clade containing most Zephyranthes
(excluding Z. bifolia and Z. filifolia) and S. howardii, which still
has low support in the cpDNA2 tree (Fig. 2). The only topo-
logical difference between MP and ML analyses of the
cpDNA1 data set involves the position of Z. longistyla, which
is placed by ML as sister to the clade formed by most
Habranthus, S. formosissima, Z. bifolia, and Z. filifolia with weak
support. This placement contrasts with that of the MP strict
consensus for cpDNA1 (Fig. 2), where Z. longistyla is recov-
ered as sister to H. andalgalensis within a clade formed by
South American-Carribbean-Mexican Zephyranthes, which
corresponds to the clade in which H. andalgalensis is also
placed by ML (not shown).
The cpDNA analyses (Fig. 2) agree with ITS in recovering
a few strongly supported major monophyletic groups,
including tribe Hippeastreae, Clade A, and Clade B. Overall,
the topology of Clade A does not conflict strongly between
ITS and cpDNA; however, the internal resolution of Clade B
is incongruent when comparing the two trees (Figs. 1, 2). The
topology of Clade A is very similar to that given by 3
(Fig. S3), except that R. laetum and Traubia are resolved in the
MP strict consensus as forming a basal grade sister to a
subclade formed by the rest of the clade (Fig. 2).
Within Clade B, all analyses agree on the sister relationship
of Habranthus immaculatus and Haylockia amer icana. The cpDNA
tree contains a major subclade within Clade B that excludes
Habranthus immaculatus, Haylockia americana, and Hippeastrum
brasilianum. Within the latter, four subclades can be identified
in the cpDNA 2 analyses (Fig. 2): 1) core-Rhodophiala (i.e. exclud-
ing R. bifida); 2) core-Hippeastrum (i.e. excluding Hippeastrum
reticulatum and Hippeastrum brasilianum); 3) R. bifida sister
to Habranthus p. p.- S. formosissima-Z. bifolia-Z. filifolia;and
4) Eithe a and Hippeastrum reticulatum sister to Zephyranthes
p. p.-S. howardii.
Most cpDNA clades are absent in Clade B of the ITS tree,
except for R. bifida and core-Rhodophiala, although the
latter has weak support (Fig. 1). Other clades present in
both ITS and cpDNA trees involve clusters of a pair or
few species, such as Habranthus brachyandrus-Habranthus
78 SYSTEMATIC BOTANY [Volume 39
Fig. 1. Strict consensus of 194 most-parsimonious trees derived from ITS nrDNA sequence data. Values above branches correspond to parsimony
jackknife frequency difference (GC MP-JK) values > 50%, followed by maximum-likelihood bootstrap absolute frequency (ML-BS) values > 50% for all
nucleotide characters. Clades subtending terminals in bold denote groups of identical sequences. A and B denote major clades. Outgroups are not
shown. “*” = 100% JK or BS; “-“ = JK or BS < 50%.
2014] GARCIA ET AL.: DEEP RETICULATION IN HIPPEASTREAE 79
Fig. 2. Strict consensus of 14 most-parsimonious trees derived from cpDNA1 data set. Values above branches correspond to parsimony jackknife
frequency difference (GC MP-JK) values > 50%, followed by maximum-likelihood bootstrap absolute frequency (ML-BS) values > 50% for all nucleotide
characters in cpDNA1 matrix. Numbers in bold below certain branches denote MP-JK/ML-BS values that change dramatically in cpDNA2 analyses.
The branch leading to Habranthus andalgalensis and Zephyranthes longistyla is dashed to indicate that it is absent in the MP strict consensus derived
from cpDNA2 data set. Clades subtending terminals in bold denote groups of identical sequences. A and B denote major clades. Outgroups are not
shown. “*” = 100% JK or BS; “-“ = JK or BS < 50%.
80 SYSTEMATIC BOTANY [Volume 39
sp.-Habranthus robustus, Z. albiella-Z. puertoricensis, Z. candida-
Z. seubertii, and Z. andina-Z. challensis (i.e. Z. andina sensu lato
(s. l.); Arroyo-Leuenberger and Leuenberger 2009).
Hybridization Networks—HybNet2 is illustrated in Fig. 3 ,
and the three remaining scenarios are provided as supple-
mental figures (Figs. S5, S6, S7). All networks are nearly iden-
tical regarding Clade A; however, there is an increasing
complexity in Clade B from more conservative to more lib-
eral networks that is reflected by a higher number of vertices
and edges necessary to reconcile both input trees (HybNet1:
nvertices = 135, nedges = 137; HybNet2: nvertices = 173,
nedges = 202; HybNet3: nvertices = 181, nedges = 214;
HybNet4: nvertices = 203, nedges = 256).
In general, the method infers a tree-like pattern of evolu-
tion at the base of Clade A, despite the presence of putative
reticulation involving Phycella and Placea species. The diver-
sification of Clade B seems to be affected by reticulate evolu-
tion from its base throughout its stem, except in the most
conservative scenario (Fig. S5), in which only Zephyranthes
species from Mexico and Texas are unequivocally inferred
as being the result of two independent hybridization events.
Both events would have had Z. citrina as donor for ITS, but
the lineages that include Z. rosea and Z. albiella-Z. candida,
respectively, would have been cpDNA donors in each event.
The most liberal scenarios (Fig. S6 and S7) suggest that
Hippeastrum brasilianum and Rhodophiala bifida are among the
earliest-diverging lineages in clade B; HybNet2 (Fig. 3) places
Z. andina s. l. at the same level as R. bifida. Based on the
inspection of HybNet2, three major clusters are identified
in Clade B: (1) Hippeastrum, including H. reticulatum and
H. brasilianum, (2) core-Rhodophiala, and (3) the reticulate
complex formed by Habranthus, Sprekelia, and Zephyranthes.
There are also four isolated lineages, some of which might
belong to the base of the Habranthus-Sprekelia-Zephyranthes
complex: (1) Rhodophiala bifida, (2) Zephyranthes andina s. l.,
(3) Eithea blumenavia, and (4) Zephyranthes cearensis.
Intergeneric Relationships—The main and best-supported
finding of this study is the presence of two major clades
within Amaryllidaceae tribe Hippeastreae. Clade A comprises
approximately 20 mainly Chilean endemic taxa and repre-
sents only 10% of the species richness of tribe Hippeastreae
(Meerow and Snijman 1998; Ravenna 2003). This clade is
characterized by a consistent haploid chromosome number
Fig. 3. Hybridization network derived from strict consensus trees resulting from analyses of cpDNA1 and equivalent ITS matrix (89 terminals)
collapsed at nodes with GC MP-JK and ML-BS < 80% (HybNet2). A and B denote major clades. Outgroups are not shown. Gray lines denote reticulation
nodes and black lines are tree nodes. Branch lengths have no direct meaning; they have been modified to make the diagram more easily readable.
Grayscale shading indicates base chromosome numbers.
2014] GARCIA ET AL.: DEEP RETICULATION IN HIPPEASTREAE 81
of x = 8 (Esponda Ferna
ndez 1970; Grau and Bayer 1991;
Naranjo and Poggio 2000; Baeza et al. 2007, 2009; Cisternas
et al. 2010), lack of polyploidy (except in a single tetraploid
Phycella cytotype; Palma-Rojas 2000), and a capitate stigma
(Ravenna 2003). Within the tribe, the only case where 2n =16
has been reported outside of Clade A is in one of two
cytotypes of Rhodophiala bifida; this species can also exhibit
2n = 18, which is the characteristic diploid number of the
core-Rhodophiala clade (Naranjo and Poggio 2000).
On the other hand, Clade B is characterized by several
basic chromosome numbers x = 6, 7, 9, 10, 11 (Naranjo 1974;
Naranjo and Andrada 1975; Arroyo 1982; Greizerstein and
Naranjo 1987; Davin
a and Ferna
ndez 1989; Naranjo and
Poggio 2000) and frequent polyploidy and aneuploidy (Flory
1977). No unequivocal morphological features diagnose
Clade B (Meerow 2010); although most species in this group
have a trifid to shortly trilobed stigma, a capitate stigma
similar to that found in Clade A is present in Famatina
herbe rtiana (Hunziker 1985), Tocantinia mira Ravenna (Ravenna
2000), and certain Hippeastrum species (Arroyo-Leuenberger,
pers. comm.; Oliveira 2012). Tocantinia Ravenna is the only
genus in the tribe not sampled in this study due to the lack of
material; however, an undescribed species morphologically
similar to T. mira was shown to be phylogenetically close to
Eithea and the Habranthus-Sprekelia-Zephyranthes complex by
Network Scenarios—Apparent reticulations in our net-
works should represent conflict rather than uncertainty,
because weakly supported clades have been collapsed, and
thus these events can be interpreted explicitly as hybridiza-
tions (McBreen and Lockhart 2006; Huson and Bryant 2006;
Pirie et al. 2009). However, as mentioned by Huson et al.
(2005), this type of analysis does “not necessarily constitute
evidence for reticulate evolution.” Therefore, alternative
explanations for the observed pattern should be considered
(Wendel and Doyle 1998).
An important caution for the approach used here is the
reduction of stochastic error present in the data sets through
the use of thresholds of resampling values (Vriesendorp and
Bakker 2005; McBreen and Lockhart 2006). Because every
conflict between the input trees will be magnified in the
resulting network, cases of inferred reticulation events that
are not real could be implied. Thresholds of relative support
are used despite the fact that this might result in an under-
estimation of incongruities, because some of the discarded
edges could be true relationships with low signal in a partic-
ular data set (Nakhleh 2011). More resolved input trees tend
to increase the complexity of the resulting network, as
reflected by the higher number of reticulation nodes that are
needed to compute the networks following HybNet1. For
example, the cpDNA2 consensus tree is much more resolved
than for cpDNA1 (or when the support threshold is raised)
due to the significant increase of support values in four inter-
nal branches of Clade B that have MP-JK and ML-BS <50%
in the cpDNA1 tree. In this respect, the threshold used for
HybNet1 might be too stringent for the phylogenetic signal of
cpDNA1, because it implies losing almost all resolution in
the respective input tree, and the resulting network is there-
fore a large polytomy with low structure (Fig. S4).
Our approach considers various levels of phylogenetic
uncertainty and, consequently, infers different degrees of
reticulate evolution within Hippeastreae, particularly within
Clade B. However, the most stringent criteria might be
excessive for the cpDNA data set, despite being 5 kb long
and a clear improvement relative to the previous cpDNA tree
of Hippeastreae (Meerow 2010). Another source of underes-
timation of reticulate evolution in Hippeastreae might come
from cases where a hybrid species inherited cpDNA and
fixed ITS sequences from the same parent (Sang et al. 1997).
In these cases the hybridization would go undetected when
comparing topologies derived from cpDNA and ITS, but
could be solved by the inference of various independent
nuclear gene trees (Linder and Rieseberg 2004; Small
et al. 2004).
In the network analyses, only a few lineages within Clade
B show a tree-like pattern, most notably core-Hippeastrum
and core-Rhodophiala, although each exhibits putative hybrid-
ization between congeners. Wherever there is a polytomy
in either tree, the hybridization network represents clades
found in one and not contradicted by the other (semistrict
consensus). This is the case for certain clusters that are pre-
sent in all estimated networks, such as Z. rosea-Z. atamasco-Z.
treatiae-Z. simpsonii (i.e. Zephyranthes from Cuba and the
southeastern U. S. A.), Z. carinata-Z. clintiae-Z. orellanae-Z.
drummondii-Z. morrisclintii (i.e. Zephyranthes Mexico/
Texas-1), and Z. macrosiphon-Z. pulchella-Z. chlorosolen-
Z. smalli (i.e. Zephyranthes Mexico/Texas-2).
Causes of Cytonuclear Discordance—When considering
the factors that may lead to incongruence listed by Wendel
and Doyle (1998), it is possible to dismiss some causes and
identify the more likely phenomena responsible for the pat-
tern found near the base and along the stem of the
Hippeastreae phylogeny. Most cytonuclear discordances
were robustly supported under alternative methods of anal-
ysis (MP and ML), and reticulation near the base of Clade B
was consistently inferred through different thresholds of
support values taking into account both optimization criteria
(except for the most conservative threshold as discussed
above). By considering a model-based approach such as ML,
we have taken into account possible effects of rate heteroge-
neity among sites and common unequal base frequencies
(Maureira-Butler et al. 2008). Similarly, relative branch
lengths inferred for both cpDNA and ITS (not shown) do not
suggest effects of long-branch attraction (Felsenstein 1978) or
heterotachy (Kolaczkowski and Thornton 2004) in our data.
Sequencing errors are also unlikely to have caused such
widespread cytonuclear discordance, because all amplicons
were sequenced in both directions and checked thoroughly.
Sequences were mostly generated from the same DNA
extraction for each locus examined, with only a few excep-
tions (see Appendix 1); furthermore, it is unlikely that
misidentified accessions, if present, were a major cause
Some phenomena that may contribute to the inferred pat-
tern of cytonuclear incongruence include insufficient taxon
sampling, lack of phylogenetic signal, and rapid diversifica-
tion (Wendel and Doyle 1998). Although our taxon sampling
is an improvement compared to previous studies of the
group (Meerow et al. 1999, 2000; Meerow 2010), increased
sampling (intra- and interspecific) of South American
Habranthus and Zephyranthes and of Mexican Habranthus
could improve phylogenetic inferences for Z. mesochloa and
Z. andina from South America (Greizerstein and Naranjo
1987) and for Habranthus immaculatus. Regions of the
Hippeastreae gene trees that seem to be affected by low phy-
logenetic signal/resolution and/or rapid radiations include
82 SYSTEMATIC BOTANY [Volume 39
1) Placea-Phycella, 2) core-Rhodophiala, 3) core-Hippeastrum,
4) North American Zephyranthes, and 5) base of Clade B.
Identical or nearly identical sequences in both cpDNA and
ITS in the first four cases may have resulted from recent
rapid radiations. At the base of Clade B, ancestral hybridiza-
tion(s) preceding an adaptive radiation (Seehausen 2004), as
inferred by our results and additional evidence (see below),
cannot be ruled out, further obscuring the inference of phy-
logenetic relationships at this level.
Potential problems also arise in the use of ITS in a reticu-
lating group. As described earlier, segregation of ITS types in
a hybrid and concerted evolution in an allopolyploid can
yield a single ITS type even in a species of reticulate origin.
Meerow (2010) detected a signal of recombination in his ITS
data set for this clade, which has been reevaluated and con-
firmed in this study (see Supplemental Material). When
recombination takes place, the evolutionary histories of sites
at either side of a recombination breakpoint will show diver-
gent phylogenetic signal and consequently fail to be modeled
by a bifurcating tree (Nakhleh 2011). Although the latter
point argues in favor of our decision to represent the overall
relationships of Hippeastreae as a network, putative recom-
bination in ITS might be introducing errors in the gene tree
inference step. Recombination in a hybrid derivative between
ITS types of the two parental species could yield a single
“new” ITS sequence that could subsequently become fixed.
Interlocus recombination within a genome can also yield
novel ITS sequences. Although a single PCR band was
detected for all ITS reactions, two 35S arrays have been
detected through fluorescence in-situ hybridization in sev-
eral Hippeastreae taxa, including species of Placea (Baeza
and Schrader 2004), Phycella, Zephyranthes, and Sprekelia
a, unpublished data). Potential problems with
paralogy in ITS were ruled out in the first study dealing with
this marker in American Amaryllidaceae (Meerow et al. 2000);
however, the Eurasian clade, sister to the American clade,
shows the highest levels of ITS paralogy so far detected within
Amaryllidaceae s. s. (Meerow et al. 2006).
Ancestral polymorphism with subsequent incomplete lin-
eage sorting (ILS) generates patterns of incongruence similar
to those produced by hybridization, and these two phenom-
ena are usually difficult to distinguish from each other (e.g.
Doyle 1992; Seelanan et al. 1997; Wendel and Doyle 1998;
Sang and Zhong 2000; Buckley et al. 2006). Nei and Kumar
(2000) suggest that incongruence due to incomplete sorting of
ancestral alleles is more probable when (i) the time between
speciation events or tree-splitting events measured in num-
ber of generations is short (i.e. short branch lengths) and (ii)
when the effective population size (N
) is high. As discussed
by Maureira-Butler et al. (2008), most current methods to
estimate effective population sizes require large amounts of
data and are designed to deal with few species (e.g. Yang
1997; Rannala and Yang 2003; Wall 2003), which makes them
inappropriate when analyzing large numbers of taxa.
Although ILS is generally more important to population and
species-level studies, it may occasionally affect tree topolo-
gies at higher levels of divergence (e.g. Rivers et al. 1993;
Richman et al. 1996; Meerow et al. 2006). Even though this is
a phenomenon that has been mostly reported for nuclear loci
with allelic variation, there are examples in Coreopsis (Mason-
Gamer et al. 1995), Phacelia (Levy et al. 1996), and Solanum
(Castillo and Spooner 1997) of ILS affecting cpDNA evolu-
tion, and in Zea for ITS (Buckler and Holtsford 1996).
The most widely used methods for distinguishing between
hybridization and ILS (e.g. Buckley et al. 2006; Maureira-
Butler et al. 2008; Pirie et al. 2009; Pelser et al. 2010; Joly et al.
2009; Joly 2012) require a time-calibrated phylogeny to per-
form coalescence simulations – analyses that are premature
for Hippeastreae, given the limited sampling of Amaryllidaceae
in dating studies involving monocots (Bremer 2000; Janssen
and Bremer 2004) and the paucity of dating studies in
Amaryllidaceae in particular. Moreover, the markers used in
this study are not ideal to implement these methods because
patterns resulting from concerted evolution of ITS could
potentially mislead the analysis (Joly et al. 2009) and cpDNA
usually does not have the polymorphism neccesary to gener-
ate ILS. Given that we did not explicitly test for the sorting of
ancestral polymorphisms in our data, this phenomenon
cannot be ruled out as an explanation for cytonuclear discor-
dance, especially at shallower levels of the gene trees (i.e.
among closely related species). However, it seems unlikely
that ancestral polymorphism has had a great effect at the
base of Clade B, because the incongruences involve branches
that are distant within this clade (Seehausen 2004).
Hybridization was likely involved in the early diversifica-
tion of Clade B, given that the deep nodes involved in the
cytonuclear discordance of this clade are well supported
(i.e. >80% MP-JK and ML-BS) in both trees. Additional evi-
dence supporting deep reticulation in Clade B includes the
recombination signals detected in ITS and putative old
events of chloroplast capture inferred from the comparison
of ITS and cpDNA-based trees. Our cpDNA tree exhibits
close relationships between species that are not apparently
related based on morphology and that are geographically
distant, which is a common pattern for ancient events of
cytoplasmic introgression (Soltis and Kuzoff 1995; Wendel
et al. 1995b; Wendel and Doyle 1998). The two most striking
cases of this phenomenon in our cpDNA data set include
(1) the sister-relationship of Habranthus immaculatus from
Mexico and Haylockia americana from northeastern Argentina
and Uruguay, and (2) the clade formed by Zephyranthes bifolia
from the Dominican Republic with Zephyranthes filifolia from
Argentina, Sprekelia formosissima from Mexico, and most
South American Habranthus. Another line of evidence that
supports deep reticulation in Clade B is a mosaic pattern of
morphological variation (Meerow 2010). This mosaic pattern
of characters explains in part the historical difficulties in clas-
sifying this group at the generic level (e.g. Traub 1963).
Reports of several artificial intergeneric hybrids within
Clade B further suggest that genetic barriers are easily crossed
among lineages in this group, perhaps as a consequence of
deep reticulation that was involved in its early diversification.
These intergeneric hybrids include: 1) xRhodobranthus Traub
(Rhodophiala bifida x Habranthus); 2) xZephybranthus T. M.
Howard (Zephyranthes x Habranthus;=Sydneya Traub, nomen
illegitimum); 3) xSprekanthus Traub (Sprekelia x Habranthus);
4) xSprekelianthes Lehmiller (Zephyranthes x Sprekelia); 5)
Sprekelia x Hippeastrum (“Hippeastrelia”); and 6) xHowardara
Lehmiller, a trigeneric hybrid (Hippeastrum x
Zephyranthes) (Traub 1958; Cage 1969; Pradham 1970; Flory
1977; Howard 1990; Lehmiller 2003–2004, 2010). Additionally,
Schulz (1954) mentioned a successful cross between
Hippeastrum and Zephyranthes, but she did not provide further
details or a taxonomic name for this intergeneric hybrid.
Flory (1977) mentioned that hybridization has and is prob-
ably still playing an important role in the evolution of
2014] GARCIA ET AL.: DEEP RETICULATION IN HIPPEASTREAE 83
Amaryllidaceae and supported this assertion using examples
corresponding mainly to hippeastroid taxa from Clade B. In
that same review, several examples of natural and artificial
hybrids between Zephyranthes species were provided (Flory
1977). It has been suggested that most Zephyranthes with
chromosome numbers 2n > 20 and in multiples of five or six
are of allopolyploid origin (Greizerstein and Naranjo 1987;
a and Ferna
ndez 1989). Hippeastrum hybrids are rou-
tinely obtained for horticultural reasons (e.g. Narain 1987;
Meerow 2009), and an allopolyploid origin for species with
chromosome numbers in multiples of 11 and 2n > 33 has also
been suggested (Naranjo and Andrada 1975). There are a few
examples of polyploid species in core-Rhodophiala (Naranjo
and Poggio 2000), but it is not clear whether they are of auto-
or allopolyploid origin. However, the constancy of chromo-
some number in multiples of nine is a noteworthy fact for
this latter lineage.
Deep Reticulation—In this study we have used the term
“deep reticulation” to refer to cases where an ancient hybrid-
ization event has preceded the radiation of a group
(Seehausen 2004). This term has been coined to distinguish
cases of reticulation involving relatively close tips or species,
i.e. most cases of reticulate evolution reported (e.g. Rieseberg
and Soltis 1991; Rieseberg et al. 1996; Vriesendorp and
Bakker 2005), from those cases where hybridization involved
two more distant lineages, usually resulting in a burst of
diversification and morphological divergence.
Additional cases of deep reticulation have been detected
in other angiosperms, including the Heuchera L. group in
Saxifragaceae (Soltis and Kuzoff 1995), tribe Veroniceae in
Plantaginaceae (Albach and Chase 2004), the Geinae group
in Rosaceae (Smedmark et al. 2003, 2005), tribe Opuntioideae
in Cactaceae (Majure et al. 2012), and several groups within
Asteraceae (Hawaiian silverswords alliance in Barrier et al.
1999; Pilosella Hill group in Fehrer et al. 2007; subtribe
Machaerantherinae in Morgan et al. 2009; tribe Senecioneae
in Pelser et al. 2010) and Poaceae (tribe Triticeae in Kellogg
et al. 1996; subtribe Loliinae in Catala
n et al. 2004; tribe
Danthonieae in Pirie et al. 2009; tribe Arundinarieae in
Triplett and Clark 2010; Zhang et al. 2012). Seehausen (2004)
also reported cases of ancient hybridization preceding radia-
tions in certain groups of animals, including Darwin’s
finches and African cichlids, and further suggested that
hybridization might be a frequent process associated with
adaptive radiations, an evolutionary trigger likewise
suggested for polyploidy (e.g. Soltis et al. 2009; Braasch and
Postlethwait 2012; Soltis and Soltis 2012).
Most network analyses suggest ancient reticulation within
Clade B, while Clade A shows a predominant tree-like pat-
tern of evolution. Given our current phylogenetic framework
for the group and the distribution of basic chromosome num-
bers in the lineages involved (i.e. x = 8 or 9 in Rhodophiala
bifida, x = 9 in core-Rhodophiala and Eithea, x =6inHabranthus
and Zephyranthes, x =11inHippeastrum; Fig. 3), we hypothe-
size that the putative reticulation event(s) that preceded the
radiation of Clade B most likely consisted of diploid hybrid-
ization(s) (Rieseberg 1997; Cronn and Wendel 2003;
Rieseberg and Willis 2007). However, allopolyploidizations
are likely to have been involved in the more recent diversifi-
cation of the Habranthus-Sprekelia-Zephyranthes complex, as
suggested by polyploid series of taxa based mostly on x =6
and cytogenetic evidence (Flory 1977; Greizerstein and
Naranjo 1987; Davin
a and Ferna
Taxonomic Implications—Our current results will certainly
have implications for the generic classification of the tribe, as
no non-monotypic genus is monophyletic according to either
ITS or cpDNA. Given our present phylogenetic framework
and resolution, it is premature to suggest a complete revised
classification at the generic level for tribe Hippeastreae.
However, it seems appropriate to formalize the two main
clades, A and B, which were consistently inferred by our
analyses and have strong statistical support, as subtribes.
Therefore, we proceed to describe briefly and recircumscribe
their compositions as follows. No character evolution optimi-
zations have been performed over the phylogeny of the tribe;
however, in the descriptions we have highlighted characters
that are potential synapomorphies for each clade, given the
current knowledge on morphological variation in Hippeastrea e
and its sister tribe, Griffinieae
RAUBIINAE D. Mu
ll.-Doblies & U. Mu
Repert. 107 (5–6, Short commun.): 6. 1996. Traubieae
Moldenke, Plant Life 19: 55. 1963.—TYPE: Traubia
Bulbous perennial herbs, terrestrial. Leaves linear or lorate,
annual, sometimes histeranthous, usually flaccid with
lamellose parenchyma within. Scape hollow; spathe bracts 2,
always free. Flowers 1–7 per inflorescence, pedicellate (but
shortly pedicellate in Rhodolirium andicola and R. montanum,
i.e. Rhodolirium s. s.); perigone slightly zygomorphic,
infundibuliform to tubular, tepal-tube obsolete to short (i.e.
less than ¼ of the perigone length). Paraperigone sometimes
present in the form of appendages in pairs near the base
of tepals, oblong-subulate and inconspicuous, or long and
crenate at the apex, bifid, or trifid, otherwise connate and
forming a 3–6-lobed declinate paraperigonal corona. Fila-
ments filiform, declinate-ascending or straight, 2- or 4-seri-
ate. Stigma capitate (but obscurely trilobed in Rhodolirium
andicola); style declinate-ascending or straight. 2n = 16
(except in tetraploid Phycella cytotype mentioned in text).
Composition—This subtribe corresponds to Clade A and it
is composed of Placea, Phycella (including Famatina maulensis),
Rhodolirium s. l., and Traubia.
Distribution—Mostly restricted to habitats with Mediter-
ranean climate in Central Chile. Rhodolirium laetum is found
in the Coastal Atacama Desert, and Rhodolirium andicola
and R. montanum inhabit high-Andean habitats, also in
IPPEASTRINAE Walp., Ann. Bot. Syst. 3: 616. 1852.—TYPE:
Hippeastrum Herb., nom. cons. (1821).
Zephyranthinae Baker, J. Bot. 16: 162. 1878. Zephyranthaceae
Salisb., Gen. Pl.: 133. 1866. Zephyrantheae Hutch., Fam.
Fl. Pl. 2: 130. 1934.—TYPE: Zephyranthes Herb., nom.
Sprekelieae Nakai, Chosakuronbun Mokuroku [Ord. Fam.
Trib. Nov.]: 235. 20 Jul 1943.—TYPE: Sprekelia
Habranthinae Traub, Plant Life 7: 43. 1951.—TYPE:
Habranthus Herb. (1824).
Tocantinieae Ravenna, Onira, Bot. Leafl. 5(3): 9. 2000.—TYPE:
Tocantinia Ravenna (2000).
Bulbous perennial herbs, terrestrial, rarely epiphytic (three
species of Hippeastrum). Leaves linear, lorate, or lanceolate
84 SYSTEMATIC BOTANY [Volume 39
(but oblanceolate-petioled in Eithea), annual or persistent,
frequently histeranthous, usually firm textured and moder-
ately canaliculated within. Scape hollow; spathe bracts 2,
free or fused. Flowers 1–13 per inflorescence, sessile to pedi-
cellate; perigone actinomorphic to highly zygomorphic, tubu-
lar, campanulate, or infundibuliform, tepal-tube obsolete to
long (i.e., more than ½ of the perigone length). Paraperigone
if present consisting of basal appendages, diminutive, mem-
branous, bristle-like, and forming a fimbriate-lacerate or
callose ring, partially adnate to the perigone throat, and sur-
rounding the stamen fascicle. Filaments filiform, declinate-
ascending or straight, 2- or 4-se riat e. Stigma trifid to obscurely
trilobed (but capitate in Famatina herbertiana, Tocantinia mira,
and certain Hippeastrum spp.); style declinate or straight. 2n
(most common) = 12, 14, 18, 20, 22, 24, 44, 48, 60.
Composition—This subtribe corresponds to Clade B and it
is composed of Eithea, Habranthus, Hippeastrum, Rhodophiala
(including Famatina p. p.), Sprekelia, Tocantinia,andZephyranthes
(including Haylockia (Zephyranthes) americana).
Distribution—Mostly present in subtropical and tropical
South America, Greater Antilles, Mexico, and southern
United States. Core-Rhodophiala spp. are found in Mediterra-
nean Chile, including lowlands and high-Andean habitats
(also in Argentina), and the Atacama Desert.
Acknowledgments. We thank Silvia Arroyo-Leuenberger (B), Gerald
Smith (HPU), Marcelo C. Baeza (CONC), Mark Whitten (FLAS), Jim
Shields (Shields Gardens), and Francisco Jime
nez (National Botanic
Garden, Dominican Republic) for providing leaf samples; Julie Dutilh
for providing unpublished chromosome counts; Kurt Neubig for advice
and initial exploration of 3
ycf1 in Hippeastreae; Jim Leebens-Mack,
Roxanne Steele, and Chris Pires for Amaryllidaceae 3
or for help with figures; Fulbright and Conicyt (Chile) for
a Doctoral Fellowship in Science and Technology, and the American
Society of Plant Taxonomists for a 2010 Graduate Student Research Grant
to N.G. Development of some of the sequences used in this paper was
supported by NSF Grants DEB-968787 and 0129179 to A. W. M. We also
thank two anonymous reviewers and Walter Judd for helpful comments
to improve this manuscript.
Albach, D. C. and M. W. Chase. 2004. Incongruence in Veroniceae
(Plantaginaceae): evidence from two plastid and a nuclear ribosomal
DNA region. Molecular Phylogenetics and Evolution 32: 183–197.
lvarez, I. and J. F. Wendel. 2003. Ribosomal ITS sequences and plant
phylogenetic inference. Molecular Phylogenetics and Evolution 29:
Anderson, E. 1949. Introgressive hybridization. New York: Wiley.
Arnold, M. L. 1997. Natural hybridization and evolution. New York:
Oxford University Press.
Arnold, M. L. 2006. Evolution through genetic exchange. New York:
Oxford University Press.
Arnold, M. L. and N. D. Fogarty. 2009. Reticulate evolution and marine
organisms: the final frontier? International Journal of Molecular Sciences
Arroyo, S. C. 1982. The chromosomes of Hippeastrum, Amaryllis and
Phycella (Amaryllidaceae). Kew Bulletin 37: 211–216.
Arroyo-Leuenberger, S. C. and J. Dutilh. 2008. Amaryllidaceae. Pp. 203–
226 in Cata
logo de la plantas vasculares del Cono Sur (Argentina, Sur de
Brasil, Chile, Paraguay y Uruguay). Volumen 1. eds. F. O. Zuloaga, O.
Morrone, and M. J. Belgrano. Monographs in Systematic Botany from the
Missouri Botanical Garden, Volume 107. St. Louis: Missouri Botanical
Arroyo-Leuenberger, S. C. and B. E. Leuenberger. 2009. Revision
of Zephyranthes andina (Amaryllidaceae) including five new
synonyms. Willdenowia 39: 145–159.
Baeza, C. M. and O. Schrader. 2004. Karyotype analysis of Placea
amoena Phil. (Amaryllidaceae) by double fluorescence in situ
hybridization. Caryologia 57: 200–205.
Baeza, C. M., E. Ruiz, and M. Negritto. 2007. El nu
de Phycella australis Ravenna (Amaryllidaceae). Gayana Bota
Baeza, C. M., C. Mariangel, E. Ruiz, and M. Negritto. 2009. El cariotipo
fundamental en Rhodolirium speciosum (Herb.) Ravenna y R. andicola
(Poepp.) Ravenna (Amaryllidaceae). Gayana Bota
nica 66: 99–102.
Balakirev, E. S. and F. J. Ayala. 2003. Pseudogenes: are they “junk” or
functional DNA? Annual Review of Genetics 37: 123–151.
Barrier, M., B. G. Baldwin, R. H. Robichaux, and M. D. Purugganan. 1999.
Interspecific hybrid ancestry of a plant adaptive radiation: Allopoly-
ploidy of the Hawaiian silversword alliance (Asteraceae) inferred from
floral homeotic gene duplications. Molecular Biology and Evolution
Braasch, I. and J. H. Postlethwait. 2012. Polyploidy in fish and the teleost
genome duplication. Pp. 341–383 in
Polyploidy and genome evolution,
eds. P. S. Soltis and D. E. Soltis. Berlin: Springer-Verlag.
Brandley, M. C., A. Schmitz, and T. W. Reeder. 2005. Partitioned Bayesian
analyses, partition choice, and the phylogenetic relationships of
scincid lizards. Systematic Biology 54: 373–390.
Bremer, K. 2000. Early Cretaceous lineages of monocot flowering plants.
Proceedings of the National Academy of Sciences USA 97: 4707–4711.
Brickell, C. and J. D. Zuk. (eds.). 1997. The American horticultural society
A-Z encyclopedia of garden plants. New York: DK Publishing.
Buckler, E. S. and T. P. Holtsford. 1996. Zea systematics: Ribosomal ITS
evidence. Molecular Biology and Evolution 13: 612–622.
Buckley, T. R., M. Cordeiro, D. C. Marshall, and C. Simon. 2006. Differen-
tiating between hypothesis of lineage sorting and introgression in
New Zealand alpine cicadas (Maoricicada Dugdale). Systematic Biology
Bull, J. J., J. P. Huelsenbeck, C. W. Cunningham, D. L. Swofford, and P. J.
Waddell. 1993. Partitioning and combining data in phylogenetic
analysis. Systematic Biology 42: 384–397.
Cage, J. M. 1969. Bigenetic hybrid of Sprekelia and Habranthus. Plant Life
Castillo, R. O. and D. M. Spooner. 1997. Phylogenetic relationships of wild
potatoes, Solanum Series Conicibaccata (Sect. Petota). Syst ematic Botany
Castoe, T. A., T. M. Doan, and C. L. Parkinson. 2004. Data partitions and
complex models in Bayesian analysis: the phylogeny of the
gymnophthalmid lizards. Systematic Biology 53: 448–469.
Castresana, J. 2000. Selection of conserved blocks from multiple align-
ments for their use in phylogenetic analysis. Molecular Biology and
Evolution 17: 540–552.
n, P., P. Torrecilla, J. A. Lo
guez, and R. G. Olmstead.
2004. Phylogeny of the festucoid grasses of subtribe Loliinae and
allies (Poeae, Pooideae) inferred from ITS and trnL-F sequences.
Molecular Phylogenetics and Evolution 31: 517–541.
Chase, M. W., J. L. Reveal, and M. F. Fay. 2009. A subfamilial classifica-
tion for the expanded asparagalean families, Amaryllidaceae,
Asparagaceae and Xanthorrhoeaceae. Botanical Journal of the Linnean
Society 161: 132–136.
Cisternas, M., L. Araneda, N. Garcı
a, and C. M. Baeza. 2010. Karyotypic
studies of the Chilean genus Placea (Amaryllidaceae). Gayana Bota
Cronn, R. and J. F. Wendel. 2003. Cryptic trysts, genomic mergers, and
plant speciation. The New Phytologist 161: 133–142.
Cullings, K. W. 1992. Design and testing of a plant-specific PCR primer
for ecological and evolutionary studies. Molecular Ecology 1: 233–240.
a, J. R. and A. Ferna
ndez. 1989. Karyotype and meiotic behavior
(Amaryllidaceae) from South America. Cytologia
Davis, J. I., K. C. Nixon, and D. P. Little. 2005. The limits of conventional
cladistic analysis. Pp. 119–147 in Parsimony, phylogeny, and genomics,
ed. V. A. Albert. Oxford: Oxford University Press.
Doolittle, W. F. 1999. Phylogenetic classification and the tree of life.
Science 284: 2124–2128.
Douzery, E. J. P., A. M. Pridgeon, P. Kores, H. P. Linder, H. Kurzweil, and
M. W. Chase. 1999. Molecular phylogenetics of Diseae (Orchidaceae):
a contribution from nuclear ribosomal ITS sequences. American Journal
of Botany 86: 887–899.
Doyle, J. J. 1992. Gene trees and species trees: Molecular systematics as
one-character taxonomy. Systematic Botany 17: 144–163.
Doyle, J. J. and J. L. Doyle. 1987. A rapid DNA isolation procedure for
small quantities of fresh leaf tissue. Phytochemical Bulletin 19: 11–15.
Drescher, A., S. Ruf, T. Calsa, H. Carrer, and R. Bock. 2000. The two
largest chloroplast genome-encoded open reading frames of higher
plants are essential genes. The Plant Journal 22: 97–104.
2014] GARCIA ET AL.: DEEP RETICULATION IN HIPPEASTREAE 85
Drew, B. T. and K. J. Systma. 2011. Testing the monophyly and placement
of Lepechinia in the tribe Mentheae (Lamiaceae). Systematic Botany
Drummond, A. J. , B. Ashton, S. Bu xton, M. Cheung, A. Cooper, J. Heled, M.
2010. Geneious v5.4, Available from http://www.geneious.com.
ndez, P. 1970. Cytotaxonomy of two species of the genus
Hippeastrum (Amaryllidaceae). Cytologia 35: 431–433.
Farris, J. S., V. A. Albert, M. Ka
, D. Lipscomb, and A. G. Kluge.
1996. Parsimony jackknifing outperforms neighbor-joining. Cladistics
Fazekas, A. J., R. Steeves, and S. G. Newmaster. 2010. Improving sequencing
quality from PCR products containing long mononucleotide repeats.
BioTechniques 48: 277–285.
Fehrer, J., B. Gemeinholzer, J. Chrtek Jr., and S. Bra
utingam. 2007. Incongru-
ent plastid and nuclear DNA phylogenies reveal ancient intergeneric
hybridization in Pi losella ha wkweeds (Hieracium,Cichorieae,
Asteraceae). Molecular Phylogenetics and Evolution 42: 347–361.
Felsenstein, J. 1973. Maximum likelihood and minimum-steps methods
for estimating evolutionary trees from data on discrete characters.
Systematic Zoology 22: 240–249.
Felsenstein, J. 1978. Cases in which parsimony or compatibility methods
will be positively misleading. Systematic Zoology 27: 401–410.
Felsenstein, J. 1985. Confidence limits on phylogenies: An approach using
the bootstrap. Evolution 39: 783–791.
Flagg, R. O., G. L. Smith, and W. S. Flory. 2002. Zephyranthes. Pp. 296–303
in Flora of North America North of Mexico, Volume 26, eds. Flora of
North America Editorial Committee. New York and Oxford: Oxford
Flagg, R. O., G. L. Smith, and A. W. Meerow. 2010. New combinations in
Habranthus (Amaryllidaceae) in Mexico and Southwestern U.S.A.
Novon 20: 33–34.
Flory, W. S. 1977. Overview of chromosomal evolution in the
Amaryllidaceae. The Nucleus 20: 70–88.
Flory, W. S. and R. F. Coulthard Jr. 1981. New chromosome counts, num-
bers and types in genus Amaryllis. Plant Life 37: 43–56.
Funk, D. J. and K. E. Omland. 2003. Species-level paraphyly and
polyphyly: Frequency, causes, and consequences, with insights from
animal mitochondrial DNA. Annual Review of Ecology Evolution
and Systematics 34: 397–423.
Grant, V. 1981. Plan t s peci at ion,2nd ed. New York: ColumbiaUniversityPress.
Grau, J. and E. Bayer. 1991. Zur Systematischen Stellung der Gattung
Traubia Moldenke (Amaryllidaceae). Mitteilungen der Botanischen
nchen 30: 479–484.
Greizerstein, E. J. and C. A. Naranjo. 1987. Estudios cromoso
especies de Zephyranthes (Amaryllidaceae). Darwiniana 28: 169–186.
Gojobori, T., W.-H. Li, and D. Graur. 1982. Patterns of nucleotide substi-
tution in pseudogenes and functional genes. Journal of Molecular
Evolution 18: 360–369.
Goloboff, P. A., J. S. Farris, M. Ka
, B. Oxelman, M. J. Ramı
C. A. Szumik. 2003. Improvements to resampling measures of group
support. Cladistics 19: 324–332.
Goloboff, P. A., J. S. Farris, and K. C. Nixon. 2008. TNT, a free program for
phylogenetic analysis. Cladistics 24: 774–786.
Howard, T. M. 1990. xCoobranthus coryi T. M. Howard. A natural bigeneric
hybrid of the tribe Zephyrantheae. Herbertia 46: 119–123.
Hunziker, A. T. 1985. Estudios sobre Amaryllidaceae. VI. Sobre la
inexistencia en Argentina del ge
nero Phycella. Lorentzia 5: 13–15.
Huson, D. H., T. Klo
pper, P. J. Lockhart, and M. A. Steel. 2005. Recon-
struction of reticulate networks from gene trees. Pp. 233–249 in
Research in computational molecular biology: Proceedings of the Ninth
International Conference on Research in Computational Molecular
Biology (RECOMB), vol. 3500. Berlin: Springer.
Huson, D. H. and D. Bryant. 2006. Application of phylogenetic net-
works in evolutionary studies. Molecular Biology and Evolution
Huson, D. H., D. C. Richter, C. Rausch, T. Dezulian, M. Franz, and R.
Rupp. 2007. Dendroscope: an interactive viewer for large phyloge-
netic trees. BMC Bioinformatics 8: 460.
Huson, D. H. and T. Klo
pper. 2007. Beyond galled trees – Decomposition
and computation of galled networks. Pp. 211–225 in Research in
computational molecular biology: Proceedings of the 11th International
Conference on Research in Computational Molecular Biology (RECOMB),
vol. 4453. Berlin: Springer.
Huson, D. H. and C. Scornavaca. 2010. A survey of combinatorial
methods for phylogenetic networks. Genome Biology and Evolution
Huson, D. H., R. Rupp, and C. Scornavaca. 2010. Phylogenetic networks.
Concepts, algorithms and applications. New York: Cambridge Uni-
Huxley, A., M. Griffiths, and M. Levy (eds.). 1992. The New Royal Horti-
cultural Society dictionary of gardening. London: Macmillan.
Ito, M., A. Kawamoto, Y. Kita, T. Yukawa, and S. Kurita. 1999. Phylogeny
of Amaryllidaceae based on matK sequence data. Journal of Plant
Research 112: 207–216.
Janssen, T. and K. Bremer. 2004. The age of major monocot groups
inferred from 800+ rbcL sequences. Botanical Journal of the
Linnean Society 146: 385–398.
Johnson, L. A. and D. E. Soltis. 1998. Assessing congruence: empirical
examples from molecular data. Pp. 297–348 in Molecular systematics
of plants II: DNA sequencing, eds. D. E. Soltis, P. S. Soltis, and J. J.
Doyle. Boston: Kluwer.
Joly, S., P. A. McLenachan, and P. J. Lockhart. 2009. A statistical approach
for distinguishing hybridization and incomplete lineage sorting.
American Naturalist 174: E54–E70.
Joly, S. 2012. JML: testing hybridization from species trees. Molecular
Ecology Resources 12: 179–184.
Judd, W. S., C. S. Campbell, E. A. Kellogg, P. F. Stevens, and M. J.
Donoghue. 2008. Plant systematics: A phylogenetic approach, 3rd edi-
tion. Sunderland: Sinauer Associates.
Katoh, K., K. Misawa, K. Kuma, and T. Miyata. 2002. MAFFT: a novel
method for rapid multiple sequence alignment based on fast Fourier
transform. Nucleic Acids Research 30: 3059–3066.
Katoh, K., G. Asimenos, and H. Toh. 2009. Multiple alignment of DNA
sequences with MAFFT. In Bioinformatics for DNA Sequence Anal-
ysis, D. Posada (ed.). Methods in Molecular Biology 537: 39–64.
Kellogg, E. A., R. Appels, and R. J. Mason-Gamer. 1996. When genes tell
different stories: the diploid genera of Triticeae (Gramineae). Systematic
Botany 21: 321–347.
Kleine, T., C. Voigt, and D. Leister. 2009. Plastid signalling to the nucleus:
Messengers still lost in the mists? Trends in Genetics 25: 185–192.
pper, T. H. and D. H. Huson. 2008. Drawing explicit phylogenetic
networks and their integration into SplitsTree. BMC Evolutionary
Biology 8: 22.
Kolaczkowski, B. and J. W. Thornton. 2004. Performance of maximum
parsimony and likelihood phylogenetics when evolution is
heterogeneous. Nature 431: 980–984.
Lehmiller, D. J. 2003–2004. A new nothogeneric taxon: xSprekelianthes
(Amaryllidaceae). Herbertia 58: 123–127.
Lehmiller, D. J. 2010. xHowardara, a new trigeneric hybrid
(Amaryllidaceae). Herbertia 64: 125–135.
Levy, F., J. Antonovics, J. E. Boynton, and N. W. Gillham. 1996. A popu-
lation genetic analysis of chloroplast DNA in Phacelia. Heredity
Linder, C. R. and L. H. Rieseberg. 2004. Reconstructing patterns of reticu-
late evolution in plants. American Journal of Botany 91: 1700–1708.
Majure, L. C., R. Puente, M. P. Griffith, W. S. Judd, P. S. Soltis, and D. E.
Soltis. 2012. Phylogeny of Opuntia s. s. (Cactaceae): clade delineation,
geographic origins, and reticulate evolution. American Journal of
Botany 99: 847–864.
Mallet, J. 2005. Hybridization as an invasion of the genome. Trends in
Ecology & Evolution 20: 229–237.
Mallet, J. 2007. Hybrid speciation. Nature 446: 279–283.
Mason-Gamer, R. J., K. E. Holsinger, and R. K. Jansen. 1995. Chloro-
plast DNA haplotype variation within and among populations of
Coreopsis grandiflora (Asteraceae). Molecular Biology and Evolution
Mason-Gamer, R. J. and E. A. Kellogg. 1996. Testing for phylogenetic
conflict among molecular data sets in the tribe Triticeae (Gramineae).
Systematic Biology 45: 524–545.
Maureira-Butler, I. J., B. E. Pfeil, A. Muangprom, T. C. Osborn, and J. J.
Doyle. 2008. The reticulate history of Medicago (Fabaceae). Systematic
Biology 57: 466–482.
McBreen, K. and P . J. Lockhart. 2006. Reconstructing reticulate evolutionary
histories of plants. Trends in Plant Science 11: 398–404.
Meerow, A. W. 2009. Tilting at windmills: 20 years of Hippeastrum
breeding. Israel Journal of Plant Sciences 57: 303–313.
Meerow, A. W. 2010. Convergence or reticulation? Mosaic evolution in the
canalized American Amaryllidaceae. Pp. 145–168 in Diversity, phylog-
eny, and evolution in the monocotyledons, eds. O. Seberg, G. Petersen,
A. S. Barfod and J. I. Davis. Aarhus: Aarhus University Press.
Meerow, A. W. and D. A. Snijman. 1998. Amaryllidaceae. Pp. 83–110 in
Families and genera of vascular plants, volume 3, ed. K. Kubitzki. Berlin:
86 SYSTEMATIC BOTANY [Volume 39
Meerow, A. W., M. F. Fay, C. L. Guy, Q.-B. Li, F. Q. Zaman, and M. W.
Chase. 1999. Systematics of Amaryllidaceae based on cladistic anal-
ysis of plastid rbcL and trnL-F sequence data. American Journal of
Botany 86: 1325–1345.
Meerow, A. W., C. L. Guy, Q.-B. Li, and S.-L. Yang. 2000. Phylogeny of
the American Amaryllidaceae based on nrDNA ITS sequences.
Systematic Botany 25: 708–726.
Meerow, A. W., J. Francisco-Ortega, D. N. Kuhn, and R. J. Schnell. 2006.
Phylogenetic relationships and biogeography within the Eurasian
clade of Amaryllidaceae based on plastid ndhF and nrDNA ITS
sequences: lineage sorting in a reticulate area? Systematic Botany
Morgan, D. R., R.-L. Korn, and S. L. Mugleston. 2009. Insights into reticu-
late evolution in Machaerantherinae (Asteraceae: Astereae): 5S ribo-
somal RNA spacer variation, estimating support for incongruence,
and constructing reticulate phylogenies. American Journal of Botany
Morrison, D. A. 2010. Using data-display networks for exploratory data
analysis in phylogenetic studies. Molecular Biology and Evolution
ller, K. 2005. SeqState: Primer design and sequence statistics for phy-
logenetic DNA datasets. Applied Bioinformatics 4: 65–69.
Nakhleh, L. 2011. Evolutionary phylogenetic networks: models and
issues. Pp. 125–158 in Problem solving handbook in computational
biology and bioinformatics, eds. L. S. Heath and N. Ramakrishnan.
New York: Springer.
Narain, P. 1987. Hippeastrum hybrids. Herbertia 43: 25–27.
Naranjo, C. A. 1969. Cariotipos de nueve especies argentinas de
Rhodophiala, Hippeastrum, Zephyranthes y Habranthus (Amaryllidaceae).
Kurtziana 5: 67–87.
Naranjo, C. A. 1974. Karyotypes of four Argentine species of Habranthus
and Zephyranthes (Amaryllidaceae). Phyton 32: 61–71.
Naranjo, C. A. and A. B. Andrada. 1975. El cariotipo fundamental
en el ge
nero Hippeastrum Herb. (Amarylidaceae). Darwiniana
Naranjo, C. A. and L. Poggio. 2000. Karyotypes of five Rhodophiala spe-
cies (Amaryllidaceae). Boletı
n de la Sociedad Argentina de Bota
Nei, M. and S. Kumar. 2000. Molecular evolution and phylogenetics.
New York: Oxford University Press.
Neubig, K. M., W. M. Whitten, B. S. Carlsward, M. A. Blanco, L. Endara,
N. H. Williams, and M. Moore. 2009. Phylogenetic utility of ycf1 in
orchids: a plastid gene more variable than matK. Plant Systematics and
Evolution 277: 75–84.
Nixon, K. C. 1999. The parsimony ratchet, a new method for rapid parsi-
mony analysis. Cladistics 15: 407–414.
Oliveira, R. S. 2012. Oge
nero Hippeastrum Herb. (Amaryllidaceae) no Brasil:
ncia de evoluc¸a
o reticulada e ana
lise de caracteres florais. Ph. D. Thesis.
Campinas, Brazil: Universidade Estadual de Campinas.
Palma-Rojas, C. 2000 . Car acterizacio
tica de los ge
Presl. y Phycella Lindl. (Amaryllidaceae). Pp. 73–79 in Los geo
nativos y su importancia en la floricultura,eds.P.Pen
ailillo and F.
Schiappacasse. Talca, Chile: Fundacio
n para la Innovacio
(FIA) y Direccio
n de Investigacio
n, Universidad de Talca (DIUT).
Parks, M., R. Cronn, and A. Liston. 2009. Increasing phylogenetic resolu-
tion at low taxonomic levels using massively parallel sequencing of
chloroplast genomes. BMC Biology 7: 84.
Pelser, P. B., A. H. Kennedy, E. J. Tepe, J. B. Shidler, B. Nordenstam, J. W.
Kadereit, and L. E. Watson. 2010. Patterns and causes of incongruence
between plastid and nuclear Senecioneae (Asteraceae) phylogenies.
American Journal of Botany 97: 856–873.
Pirie, M. D., A. M. Humphreys, N. P. Barker, and H. P. Linder. 2009.
Reticulation, data combination, and inferring evolutionary history:
an example from Danthonioideae (Poaceae). Systematic Biology
Pradham, U. C. 1970. An apparent Sprekelia-Amaryllis cross. Plant Life 26: 11 7.
Prychitko, T. M. and W. S. Moore. 1997. The utility of DNA sequences of
an intron from the b-fibrinogen gene in phylogenetic analysis of
woodpeckers (Aves: Picidae). Molecular Phylogenetics and Evolution
Rannala, B. and Z. Yang. 2003. Bayes estimations of species divergence
times and ancestral population sizes using DNA sequences from
multiple loci. Genetics 164: 1645–1656.
Ravenna, P. F. 2000. Tocantinia and Cearanthes, two new genera, and
Tocantinieae new tribe, of Brazilian Amarylidaceae. Onira 5: 9–43.
Ravenna, P. F. 2003. Elucidation and systematics of the Chilean genera of
Amaryllidaceae. Botanica Australis 2: 1–20. (Chile).
Richman, A. D., M. K. Uyenoyama, and J. R. Kohn. 1996. Allelic diversity
and gene genealogy at the self-incompatibility locus in the Solanaceae.
Science 273: 1212–1216.
Rieseberg, L. H. 1997. Hybrid origins of plant species. Annual Review of
Ecology and Systematics 28: 359–389.
Rieseberg, L. H. and D. E. Soltis. 1991. Phylogenetic consequences of
cytoplasmic gene flow in plants. Evolutionary Trends in Plants 5: 65–84.
Rieseberg, L. H., J. Whitton, and C. R. Linder. 1996. Molecular marker
incongruence in plant hybrid zones and phylogenetic trees.
Acta Botanica Neerlandica 45: 243–262.
Rieseberg, L. H. and J. H. Willis. 2007. Plant speciation. Science 317: 910–914.
Rivers, B. A., R. Bernatzky, S. J. Robinson, and W. Jahnen-Dechent. 1993.
Molecular diversity at the self-incompatibility locus is a salient fea-
ture in natural populations of wild tomato (Lycopersicon peruvianum).
Molecular & General Genetics 238: 419–427.
Rudall, P. J., R. M. Bateman, M. F. Fay, and A. Eastman. 2002. Floral
anatomy and systematics of Alliaceae with particular reference to
Gilliesia, a presumed insect mimic with strongly zygomorphic
flowers. American Journal of Botany 89: 1867–1883.
Sang, T., D. J. Crawford, and T. F. Stuessy. 1997. Chloroplast DNA phylog-
eny, reticulate evolution, and biogeography of Paeonia (Paeoniaceae).
American Journal of Botany 84: 1120–1136.
Sang, T. and Y. Zhong. 2000. Testing hybridization hypotheses based on
incongruent gene trees. Systematic Biology 49: 422–434.
Schulz, P. 1954. Amaryllis ... and how to grow them. New York: M. Barrows
and Co., Inc.
Seehausen, O. 2004. Hybridization and adaptive radiation. Trends in
Ecology & Evolution 19: 198–207.
Seelanan, T., A. Schnabel, and J. F. Wendel. 1997. Congruence and con-
sensus in the cotton tribe (Malvaceae). Systematic Botany 22: 259–290.
Shi, S., Y. Qui, E. Li, L. Wu, and C. Fu. 2006. Phylogenetic relationships
and possible hybrid origin of Lycoris species (Amaryllidaceae)
revealed by ITS sequences. Biochemical Genetics 44: 198–208.
Simmons, M. P. and H. Ochoterena. 2000. Gaps as characters in sequence-
based phylogenetic analysis. Systematic Biology 49: 369–381.
Simmons, M. P . and J. V. Freudenstein. 2011. Spurious 99% bootstrap and
jackknife support for unsupported clades. Molecular Phylogenetics
and Evolution 61: 177–191.
Small, R. L., R. C. Cronn, and J. F. Wendel. 2004. Use of nuclear genes
for phylogeny reconstruction in plants. Australian Systematic Botany
Smedmark, J. E. E., T. Eriksson, R. C. Evans, and B. Bremer. 2003. Ancient
allopolyploid speciation in Geinae (Rosaceae): evidence from
nuclear granule-bound starch synthase (GBSSI) gene sequences.
Systematic Biology 52: 374–385.
Smedmark, J. E. E., T. Eriksson, and B. Bremer. 2005. Allopolyploid evo-
lution in Geinae (Colurieae: Rosaceae) – building reticulate species
trees from bifurcating gene trees. Organisms, Diversity & Evolution
Sneath, P. H. A. 1975. Cladistic representation of reticulate evolution.
Systematic Zoology 24: 360–368.
Snijman, D. A. and A. W. Meerow. 2010. Floral and macroecological
evolution within Cyrtanthus (Amaryllidaceae): inferences from com-
bined analyses of plastid ndhF and nrDNA ITS sequences. South
African Journal of Botany 76: 217–238.
Soltis, D. E. and R. K. Kuzoff. 1995. Discordance between nuclear and
chloroplast phylogenies in the Heuchera group (Saxifragaceae).
Evolution 49: 727–742.
Soltis, D. E., E. V. Mavrodiev, J. J. Doyle, J. Rauscher, and P. S. Soltis. 2008.
ITS and ETS sequence data and phylogeny reconstruction in
allopolyploids and hybrids. Systematic Botany 33: 7–20.
Soltis, D. E., V. A. Albert, J. Leebens-Mack, C. D. Bell, A. H. Paterson, C.
Zheng, D. Sankoff, C. W. dePamphilis, P. K. Wall, and P. S.
Soltis. 2009. Polyploidy and angiosperm diversification. American
Journal of Botany 96: 336–348.
Soltis, P. S. and D. E. Soltis. 2009. The role of hybridization in plant
speciation. Annual Review of Plant Biology 60: 561–588.
Soltis, P. S. and D. E. Soltis (eds.). 2012. Polyploidy and genome evolution.
Stamatakis, A. 2006. RaxML-VI-HPC: maximum likelihood-based phylo-
genetic analyses with thousands of taxa and mixed models.
informatics 22: 2688–2690.
Stebbins, G. L. 1950. Variation and evolution in plants. New York:
Columbia University Press.
Taberlet, P., L. Gielly, G. Pautou, and J. Bouvet. 1991. Universal primers
for amplification of three non-coding regions of chloroplast DNA.
Plant Molecular Biology 17: 1105–1109.
2014] GARCIA ET AL.: DEEP RETICULATION IN HIPPEASTREAE 87
Timme, R. E., J. V. Kuehl, J. L. Boore, and R. K. Jansen. 2007. A compara-
tive analysis of the Lactuca and Helianthus (Asteraceae) plastid
genomes: identification of divergent regions and categorization of
shared repeats. American Journal of Botany 94: 302–312.
Terry, R. G., G. K. Brown, and R. G. Olmstead. 1997. Examination of
subfamilial phylogeny in Bromeliaceae using comparative sequenc-
ing of the plastid locus ndhF. American Journal of Botany 84: 664–670.
Traub, H. P. 1958. Crosses involving Rhodophiala, Habranthus and
Zephyranthes. Plant Life 14: 47–51.
Traub, H. P. 1963. Genera of the Amaryllidaceae. La Jolla: American Plant
Life Society, California.
Triplett, J. K. and L. G. Clark. 2010. Phylogeny of the temperate bamboos
(Poaceae: Bambusoideae: Bambuseae) with an emphasis on Arundinaria
and allies. Systematic Botany 35: 102–120.
Vriesendorp, B. and F. T. Bakker. 2005. Reconstructing patterns of reticu-
late evolution in Angiosperms: what can we do? Taxon 54: 593–604.
gele, J. W. and C. Mayer. 2007. Visualizing differences in phylogenetic
information content of alignments and distinction of three classes of
long-branch effects. BMC Evolutionary Biology 7: 147.
Wall, J. 2003. Estimating ancestral population sizes and divergence times.
Genetics 163: 395–404.
Wendel, J. F., A. Schnabel, and T. Seelanan. 1995a. Bidirectional interlocus
concerted evolution following allopolyploid speciation in cotton
(Gossypium). Proceedings of the National Academy of Sciences USA
Wendel, J. F., A. Schnabel, and T. Seelanan. 1995b. An unusual ribosomal
DNA sequence from Gossypium gossypioides reveals ancient, cryptic,
intergenomic introgession. Molecular Phylogenetics and Evolution
Wendel, J. F. and J. J. Doyle. 1998. Phylogenetic incongruence: window
into genome history and molecular evolution. Pp. 264–296 in Molec-
ular systematics of plants II: DNA sequencing, eds. D. E. Soltis, P. S.
Soltis, and J. J. Doyle. Boston: Kluwer.
Wheeler, W. C. and R. L. Honeycutt. 1988. Paired sequence difference in
ribosomal RNAs: evolutionary and phylogenetic implications.
Molecular Biology and Evolution 5: 90–96.
Williams, M. 1982. Chromosome counts for six Amaryllis taxa. Plant Life
Yang, Z. 1997. On the estimation of ancestral population sizes of modern
humans. Genetical Research 69: 111–116.
Zhang, Y.-X., C.-X. Zeng, and D.-Z. Li. 2012. Complex evolution in
Arundinarieae (Poaceae: Bambusoideae): incongruence between
plastid and nuclear GBSSI gene phylogenies. Molecular
Phylogenetics and Evolution 63: 777–797.
Appendix 1. List of taxa sampled with taxonomic authorities,
voucher information, and GenBank accession numbers (ITS rDNA, 3
region, respectively) for new sequences generated
in this study. G##: N. G.’s bulb collection accession number –: published
in previous studies; X: missing data.
Outgroups: Cyrtanthus carneus Lindl.–NBG 175/94, South Africa; –,
KC207481, –, KC217443. Cyrtanthus herrei (F. M. Leight.) R. A. Dyer–Van
Zijl 104 (NBG), South Africa; –, KC207482, –, KC217444. Cyrtanthus
obliquus (L. f.) Aiton–Meerow 3124 (NA), South Africa; –, KC207483, –,
KC217445. Lycoris radiata (L’He
r.) Herb.–Meerow 2606 (FTG), ex hort.,
except ITS from S20020910 (HZU); –, KC207484, KC217373, KC217446.
Pancratium canariense Ker Gawl.–Meerow 1142 (FTG, FLAS), ex hort.; –,
KC207485, –, –. Griffinia espiritensis Ravenna–Meerow 2107 (FTG), ex
hort.; KC207418, KC207486, KC217374, KC217447. Griffinia hyacinthina
(Ker Gawl.) Ker Gawl.–Meerow 2106 (FTG, FLAS), Brazil; –, KC207487, –,
–. Griffinia parviflora Ker Gawl.–Meerow 2108 (FTG), Brazil; KC207419,
KC207488, KC217375, KC217448. Worsleya procera (Lem.) Traub–
Meerow 2411 (FTG), ex hort.; KC207420, KC207489, KC217376, KC217449.
Ingroups: Eithea blumenavia (K. Koch & C. D. Bouche
Ravenna–Meerow 3115 (NA), ex hort.; KC207421, KC207490, KC217377,
KC217450. Famatina andina (Phil.) Ravenna. CHILE. Meerow 2408 (FTG),
(as Phycella ignea Lindl. in Meerow et al. 2000; Meerow 2010); –, X, –,
KC217451. Famatina cisandina Ravenna. CHILE. N. Garcı
(SGO); KC207422, KC207491, KC217378, KC217452. Famatina
herbertiana (Lindl.) Ravenna. ARGENTINA. B. E. Leuenberger, U. Eggli &
S. Arroyo-Leuenberger 4682 (B); –, KC207492, KC217379, KC217453.
Famatina maulensis Ravenna. CHILE. N. Garcı
a 4384/G43 (FLAS,
CONC); KC207423, KC207493, X, KC217454. Habranthus andalgalensis
Ravenna. ARGENTINA. B. E. Leuenberger & S. Arroyo-Leuenberger 5012
(B); KC207424, KC207494, KC217382, KC217455. Habranthus
brachyandrus (Baker) Sealy. ARGENTINA. Meerow 2400 (FTG); –,
KC207495, KC217383, KC217456. Habranthus immaculatus Traub &
Clint. MEXICO. Meerow 2401 (FTG); –, KC207496, –, KC217457.
Habranthus martinezii Ravenna. ARGENTINA. Meerow 2437 (FTG); –,
KC207497, –, KC217458. Habranthus pedunculosus Herb.. ARGENTINA.
B. E. Leuenberger & S. Arroyo-Leuenberger 4496 (B); –, KC207498, –,
KC217459. Habranthus robustus Herb. ex Sweet. ex hort. G. Smith s.n.
(HPU); KC207425, KC207499, KC217384, KC217460. Habranthus
r.) Traub. ex hort. Meerow 3116 (NA); KC207426,
KC207500, KC217385, KC217461. Habranthus sp. BRAZIL. Meerow 2402
(FTG); –, KC207501, KC217386, KC217462. Haylockia americana
(Hoffmanns.) Herter. URUGUAY. M. W. Chase 3585 (K); –, KC207502,
KC217387, KC217463. Hippeastrum ambiguum Hook. BRAZIL. Meerow
3127 (NA); KC207427, X, X, X. Hippeastrum argentinum (Pax) Hunz.
ARGENTINA. Doran 1119 (MO); KC207428, X, X. Hippeastrum
brasilianum (Traub & J. L. Doran) Dutilh. BRAZIL. Meerow 2405 (FTG);
–, KC207503, KC217388, KC217464. Hippeastrum breviflorum Herb.
BRAZIL. Dutilh s.n. (UEC); KC207429, X, X, X. Hippeastrum calyptratum
(Ker Gawl.) Herb. BRAZIL. Meerow 3117 (NA); KC207430, X, X, X.
Hippeastrum cipoanum (Ravenna) Meerow. BRAZIL. Joa
o Semir 8673
(UEC),; –, X, X, X. Hippeastrum cybister (Herb.) Benth. ex Baker.
BOLIVIA. Meerow 1149 (NA); KC207431, X, X, X. Hippeastrum doraniae
(Traub) Meerow. VENEZUELA. Meerow 3128 (NA); KC207432, X, X, X.
Hippeastrum evansiae (Traub & I. S. Nelson) H. E. Moore. BOLIVIA.
Meerow 3118 (NA); KC207433, KC207504, KC217389, KC217465.
Hippeastrum leopoldii T. Moore. BOLIVIA. Ca
rdenas 6331 (MO);
KC207434, X, X, X. Hippeastrum morelianum Lem. BRAZIL. Meerow
1304 (NA); KC207435, KC207505, KC217390, KC217466. Hippeastrum
rdenas) Van Scheepen. BOLIVIA. Doran 1319 (MO);
KC207436, X, X, X. Hippeastrum papilio (Ravenna) Van Scheepen. ex
hort. Meerow 2406 (FTG); –, KC207506, –, KC217467. Hippeastrum
parodii Hunz. & A. A. Cocucci. ARGENTINA. Meerow 2434 (FTG); –,
KC207507, KC217391, KC217468. Hippeastrum psittacinum (Ker Gawl.)
Herb. BRAZIL. Meerow 1305 (NA); KC207437, KC207508, KC217392,
KC217469. Hippeastrum puniceum (Lam.) Voss. ex hort. Meerow 3125
(NA); KC207438, KC207509, KC217393, KC217470. Hippeastrum
r.) Herb. BRAZIL. Meerow 2407 (FTG); –, KC207510, –,
KC217471. Hippeastrum striatum (Lam.) H. E. Moore. BRAZIL. Meerow
3120 (NA); KC207439, KC207511, KC217394, KC217472. Hippeastrum
traubii (Moldenke) H. E. Moore. BOLIVIA. Traub 1078 (MO); KC207440,
KC207512, KC217395, KC217473. Hippeastrum sp. ex hort. Meerow 3126
(NA); KC207467, KC207542, KC217423, KC217504.Phycella angustifolia
Phil. CHILE. N. Garcı
a 3247/G57 (CONC); KC207441, KC207513,
KC217396, KC217474. Phycella australis Ravenna. CHILE. C. M. Baeza
4285a (CONC); KC207442, KC207514, X, KC217475. Phycella aff.
cyrtanthoides (Sims) Lindl. CHILE. N. Garcı
KC207443, KC207515, KC217398, KC217476. Phycella scarlatina
Ravenna. N. Garcı
a 862/G14 (CONC), Chile; KC207444, KC207516,
KC217399, KC217477. Phycella sp.– CHILE. N. Garcı
a 4165/G80 (CONC);
KC207445, KC207517, KC217400, KC217478. Placea arzae Phil. CHILE.
Meerow 3100 (FTG); –, X, –, KC217479. Placea davidii Ravenna. CHILE.
a 3031/G40 (FLAS, CONC); KC207446, KC207518, KC217401,
KC217480. Placea germainii Phil. CHILE. O. Ferna
ndez & P. Novoa 17
(JBN-Chile); KC207447, KC207519, 2C217402, KC217481. Placea lutea
Phil. CHILE. P. Guerrero et al. 222 (INIA-Chile); KC207448, KC207520,
KC217403, KC217482. Placea ornata Miers. CHILE. O. Ferna
(JBN-Chile); KC207449, KC207521, KC217404, KC217483.
ornata Miers. CHILE. N. Garcı
a 726/G17 (FLAS, CONC); KC207450,
KC207522, KC217405, KC217484. Rhodolirium andicola (Poepp.)
Ravenna. CHILE. Aedo 7230 (CONC); KC207451, KC207523, KC217406,
KC217485. Rhodolirium laetum (Phil.) Ravenna. CHILE. N. Garcı
G10 (FLAS, CONC); KC207452, KC207524, KC217407, KC217486.
Rhodolirium montanum Phil. CHILE. Montero 12473 (CONC); KC207453,
KC207525, KC217408, KC217487. Rhodolirium speciosum (Herb.)
Ravenna. CHILE. L. I. Escobar 35 (CONC); KC207454, KC207526, X,
KC217488. Rhodophiala advena (Ker Gawl.) Traub. CHILE. N. Garcı
2964 (FLAS, CONC); KC207455, KC207527, KC217410, KC217489.
Rhodophiala aff. advena (Ker Gawl.) Traub. CHILE. M. Cisternas s.n./
G36 (FLAS, CONC); KC207456, KC207528, KC217411 (3
Rhodophiala ananuca (Phil.) Traub (1). CHILE. N. Garcı
CONC); KC207457, KC207529, KC217412, X. Rhodophiala ananuca (Phil.)
Traub (2). CHILE. M. Rosas 7042/G15 (CONC); KC207458, KC207530,
KC217414, KC217492. Rhodophiala araucana (Phil.) Traub. CHILE. N.
a 4345/G83 (FLAS, CONC); KC207459, KC207531, KC217415,
KC217493. Rhodophiala bagnoldii (Herb.) Traub. CHILE. N. Garcı
G24 (FLAS, CONC); KC207460, KC207532, KC217416, KC217494.
88 SYSTEMATIC BOTANY [Volume 39
Rhodophiala bifida (Herb.) Traub subsp. bifida. ARGENTINA. Meerow
3102 (FTG); –, KC207533, –, KC217495. Rhodophiala bifida (Herb.) Traub
subsp. granatiflora (E. Holmb.) Ravenna. ARGENTINA. Meerow 3103
(FTG); –, KC207534, –, KC217496. Rhodophiala montana (Phil.) Traub.
CHILE. N. Garcı
a 243/G53 (FLAS, CONC); KC207461, KC207535,
KC217417, KC217497. Rhodophiala phycelloides (Herb.) Hunz. CHILE.
Meerow 3104 (FTG); –, X, –, KC217498. Rhodophiala splendens (Renjifo)
Traub (1). ex hort. N. Garcı
a 4372/G79 (FLAS); KC207462, X, KC217418,
KC217499. Rhodophiala splendens (Renjifo) Traub (2). CHILE. N. Garcı
4349/G84 (FLAS, CONC); KC207463, KC207538, KC217419, KC217500.
Rhodophiala tiltilensis (Traub & Moldenke) Traub. CHILE. N. Garcı
3587/G87 (CONC); KC207464, KC207539, KC217421, KC217502.
Rhodophiala sp. CHILE. N. Garcı
a 2411/G07 (SGO); KC207465,
KC207536, KC217420, KC217501. Sprekelia formosissima (L.) Herb. ex
hort. Meerow 1151 (FTG), except trnL intron and trnL
M.W. Chase 577 (K); –, KC207540, –, –. Sprekelia howardii Lehmiller.
MEXICO. Lehmiller 1940 (TAMU); KC207466, KC207541, KC217422,
KC217503. Traubia modesta (Phil.) Ravenna. . N. Garcı
a 4357/G88 (FLAS,
CONC); KC207468, KC207543, KC217424, KC217505. Zephyranthes
albiella Traub. VENEZUELA. N. Garcı
a 4375 (FLAS); KC207469,
KC207544, KC217425, KC217506. Zephyranthes andina (R. E. Fr.) Traub.
ARGENTINA. B. E. Leuenberger & S. Arroyo-Leuenberger 4546 (B); –,
KC207545, KC217426, KC217507. Zephyranthes atamasco (L.) Herb.
U. S. A. Meerow 2412 (FTG); –, KC207546, –, KC217508. Zephyranthes
bifolia M. Roem. DOMINICAN REPUBLIC. Meerow 3121 (NA);
KC207470, KC207548, KC217427, KC217510. Zephyranthes candida
(Lindl.) Herb. ex hort. Meerow 2414 (FTG); –, KC207547, –, KC217509.
Zephyranthes carinata Herb. ex hort. Meerow 2419 (FTG), (as Z. grandi-
flora Lindl. in Meerow et al. 2000; Meerow 2010); –, KC207549, –,
KC217511. Zephyranthes cearensis (Herb.) Baker. BRAZIL. Meerow 2415
(FTG); –, KC207550, KC217428, KC217512. Zephyranthes challensis
Ravenna. BOLIVIA. Thomas ET 495 (FTG); KC207471, KC207551, –,
KC217513. Zephyranthes chlorosolen (Herb.) D. Dietr. ex hort. Meerow
3122 (NA); KC207472, KC207552, KC217429, KC217514. Zephyranthes
citrina Baker. ex hort. N. Garcı
a4376(FLAS), except ITS from Meerow
2416 (FTG); –, KC207553, KC217430, KC217515. Zephyranthes clintiae
Traub. ex hort. G. Smith s.n. (HPU); KC207473, KC207554, KC217431,
KC217516. Zephyranthes drummondii D. Don in R. Sweet. ex hort.
Meerow 2417 (FTG); –, KC207555, KC217432, KC217517. Zephyranthes
filifolia Herb. ex Kraenzl. ARGENTINA. B. E. Leuenberger & S. Arroyo-
Leuenberger 4387 (B), except ITS from M. W. Chase 1836 (K); –, KC207556,
KC217433, KC217518. Zephyranthes flavissima Ravenna. BRAZIL.
Meerow 2418 (FTG); –, KC207557, X, X. Zephyranthes longistyla Pax.
ARGENTINA. B. E. Leuenberger & S. Arroyo-Leuenberger 4763 (B);
KC207474, KC207558, KC217434, KC217519. Zephyranthes macrosiphon
Baker. ex hort. G. Smith s.n. (HPU); KC207475, KC207559, KC217435,
KC217520. Zephyranthes mesochloa Herb. ex Lindl. ex hort. Meerow
2420 (FTG); –, KC207560, –, KC217521. Zephyranthes minima Herb.
ARGENTINA. B. E. Leuenberger & S. Arroyo-Leuenberger 4501 (B); –, X,
–, X. Zephyranthes morrisclintii Traub & T. M. Howard. ex hort. Meerow
2421 (FTG); –, KC207561, KC217436, KC217522. Zephyranthes orellanae
Carnevali, Duno & J. L. Tapia. ex hort. M. Whitten 3798 (FLAS);
KC207476, KC207562, KC217437, KC217523. Zephyranthes
puertoricensis Traub. ex hort. Meerow 3123 (NA); KC207477, KC207563,
KC217438, KC217524. Zephyranthes pulchella J. G. Sm. U. S. A. Meerow
2422 (FTG); –, KC207564, KC217439, KC217525. Zephyranthes rosea
Lindl. ex hort. Meerow 2429 (FTG); –, KC207565, –, KC217526.
Zephyranthes seubertii H. H. Hume. ARGENTINA. S. Arroyo-
Leuenberger 3973 (B); –, KC207566, –, KC217527. Zephyranthes simpsonii
Chapm. U. S. A. Meerow 2413 (FTG); –, KC207567, KC217440, KC217528.
Zephyranthes smallii (Alexander) Traub. ex hort. Meerow 2423 (NA);
KC207478, KC207568, KC217441, KC217529. Zephyranthes treatiae S.
Watson. U. S. A. L. C. Majure 3982 (FLAS); KC207479, KC207569,
2014] GARCIA ET AL.: DEEP RETICULATION IN HIPPEASTREAE 89