全 文 ：Journal of Systematics and Evolution 46 (3): 375–390 (2008) doi: 10.3724/SP.J.1002.2008.08065
(formerly Acta Phytotaxonomica Sinica) http://www.plantsystematics.com
Phylogeny and putative hybridization in the subtribe Paranepheliinae
(Liabeae, Asteraceae), implications for classification, biogeography,
and Andean orogeny
1Akiko SOEJIMA* 2Jun WEN 3Mario ZAPATA 4Michael O. DILLON
1(School of Science, Osaka Prefecture University, Sakai, Osaka 599-8531, Japan)
2(Department of Botany, National Museum of Natural History, MRC-166, Smithsonian Institution, Washington, DC 20013-7012, USA)
3(Museo de Historia Natural, Universidad Antenor Orrego, Trujillo 1075, Peru)
4(Department of Botany, The Field Museum, Chicago, IL 60605, USA)
Abstract The nuclear ribosomal ITS region and the chloroplast trnL-trnF (trnLF) intergenic region were se-
quenced for 45 accessions of Paranephelius and six accessions of Pseudonoseris, the two genera of the subtribe
Paranepheliinae (Liabeae, Asteraceae) distributed in the alpine regions of the Andes. This data set was used to
estimate relationships between these genera and within each genus to aid in evaluating morphological variation
and classification. Our results with both ITS and trnLF markers support the monophyly of subtribe Paranephelii-
nae, and place Pseudonoseris discolor as the first diverged taxon sister to the clade containing Paranephelius.
Pseudonoseris szyszylowiczii exhibited intraspecific divergence supporting intergeneric hybridization between
Pseudonoseris and Paranephelius. Within Paranephelius, genetic divergence is low and not adequate to fully
resolve phylogenetic relationships at the species level, but two genetically and morphologically recognizable
groups were revealed by the ITS data. Several accessions possessing multiple ITS sequences represent putative
hybrids between the two groups. These putative hybrids have caused some taxonomic confusion and difficulties in
establishing species boundaries in Paranephelius. The divergence time estimates based on ITS sequences indi-
cated that the stem of subtribe Paranepheliinae dates to 13 million years ago, but the diversification of the crown
clade of the extant members began in the early Pleistocene or late Pliocene, perhaps associated with the uplift of
the Andes and the climatic changes of global cooling.
Key words Andes, Asteraceae, biogeography, hybridization, ITS sequences, Liabeae, Paranepheliinae, Peru,
speciation, trnL-trnF.
Hybrid speciation in plants and the potential in-
sights gained from molecular studies on this mode of
speciation have been reviewed recently (Hegarty &
Hiscock, 2005). Hybridization in general has been
considered an important process in evolutionary
biology (Mallet, 2007). Most examples in nature are
interspecific hybridizations, whereas intergeneric
hybrids have only been rarely documented in As-
teraceae with examples reported from a few tribes,
e.g., Cichorieae (Fehrer et al., 2007) and Gnaphalieae
(Smissen et al., 2007). Hybridization has not been
reported in the Liabeae.
The Andean Cordillera forms a continuous, high
elevation chain (+3000 m), largely unbroken for over
7500 km along the Pacific side of South America,
from Venezuela to Tierra del Fuego. In general, the
Andean alpine environments are wetter and more
fragmented in the north, which is referred to as
“páramo” distributed from Venezuela to northern Peru
(Cuatrecasas, 1968; Luteyn, 1999). To the south,
between central Peru and northwestern Argentina, the
high-elevation communities are termed “puna” and are
more xeric and continuous. In northern Peru, between
the páramo and puna, transitional alpine communities
above 3000 m are locally known as “jalca” formations
(Weberbauer, 1936, 1945; Sánchez & Dillon, 2006),
and are interpreted as drier than páramo but wetter
than puna (Sánchez, 1976; Bazán-Zurita et al., 1998;
Luteyn, 1999). This region is well known for its high
levels of endemism and taxon diversity in plant and
animal groups originated through the climatic changes
and continuous uplift of the Andean Cordillera begin-
ning in the late Tertiary (Simpson & Todzia, 1990).
Above 1000 m in elevation throughout the Andean
Cordillera, the Asteraceae are well represented with
the estimated diversity of over 300 genera and nearly
3500 species (Funk et al., 2007). The diversity of
Asteraceae increases above 3000 m, where it becomes
the largest family of flowering plants with ca. 110
genera and over 950 species recorded between

———————————
Received: 29 April 2008 Accepted: 3 May 2008
* Author for correspondence. E-mail: soejima@b.s.osakafu-u.ac.jp; Tel. &
Fax: +81-72-275-9751.
Journal of Systematics and Evolution Vol. 46 No. 3 2008 376
Venezuela and northern Peru (Dillon, 2005). The tribe
Liabeae is entirely neotropical in its distribution from
Mexico to South America and has its greatest generic
and species diversity in northern Peru (Robinson,
1983; Funk et al., 1995, 1996). The tribe consists of
three well-supported subtribes: Liabinae, Munnoziinae
and Paranepheliinae (Kim et al., 2003). Among the
three subtribes, only Paranepheliinae, consisting of
two genera (Paranephelius Poepp. ex Endl. and
Pseudonoseris H. Rob. & Brettell), is endemic to the
alpine region of Andes.
The subtribe Paranepheliinae is distributed from
northern Peru to northwestern Argentina (Dillon,
2005; Fig. 1). The Huancabamba Depression (HD in
Fig. 1), a region implicated as a biogeographic barrier
in many groups (Ayers, 1999; Weigend, 2002), repre-
sents the northern limit of its distribution. Department
of Cajamarca, the region directly south of the Huan-
cabamba Depression is one of the centers of biodiver-
sity in the Andes. Many plant groups show high levels
of species diversity in this region, e.g., Iochrominae
(Solanaceae, Smith & Baum, 2006) and Nasa Wei-
gend (Loasaceae, Weigend, 2002; Weigend et al.,
2004). This region is geographically complex and its
extreme species diversity is considered to be a result
of isolation and adaptive radiation in the small frag-
mented ranges and habitats within this area (Young et
al., 2002; Sánchez-Baracaldo, 2004; Sánchez &
Dillon, 2006). The subtribe Paranepheliinae also has
its center of diversity in this region.
Six of the seven named species of Paranephelius
along with Pseudonoseris striata (Cuatrec.) H. Rob. &
Brettell and Ps. szyszylowiczii (Hieron.) H. Rob. &
Brettell are found in this region. Within Dept. Caja-
marca reside the type localities for P. ferreyrii H. Rob.
(near Cumbemayo) and P. jelskii (Hieron.) H. Rob. &
Brettell (near Cutervo). Paranephelius wurdackii H.
Rob. was described from the Calla Calla region in
western Dept. Amazonas directly east of Dept. Caja-
marca and the Río Marañón. The type localities for P.
uniflorus Poepp. & Endl. and P. ovatus Wedd. are
thought to be in central Peru, and the type locality of
P. bullatus A. Gray ex Wedd. is likely in Dept. Junin,
south of La Libertad, an area where José Pavón, the
collector of the type, is known to have explored in the
late 1780’s. Paranephelius ovatus is a wide-ranging
and morphologically variable taxon recorded from 13
departments from northern to southern Peru and
extending into northern Bolivia and northern Argen-
tina. Paranephelius uniflorus has been cited from
northern to southern Peru and northern Bolivia; and P.
asperifolius (Muschl.) H. Rob. & Brettell is described






















Fig. 1. Distribution map of Paranephelius (1–7) and Pseudonoseris
(8–10). 1, P. ovatus; 2, P. uniflorus; 3, P. asperifolius; 4, P. wur-
dackii; 5, P. ferreyrii; 6, P. bullatus; 7, P. jelskii; 8, Ps. striata; 9, Ps.
szyszylowiczii; 10, Ps. discolor. HD=Huancabamba depression.

from Bolivia and is reported from northwestern
Argentina.
Paranephelius is a distinctive genus possessing
several apomorphic characters for the Liabeae (Funk
et al., 1996) such as acaulescent habit with a basal
rosette of leaves, showy capitulescences of one or a
few, large, essentially sessile radiate capitulae, each
capitula with 21 or more yellow ray florets (Fig. 2:
A–O) with tomentum on the outer surfaces, and long
tubular, 5-lobed disc florets. Though the genus has
long been recognized as distinct, delineation of spe-
cies has been difficult due to random and confusing
suites of character variations, such as the variable
combinations of leaf shape, texture, and pubescence
(Fig. 3), and characters of the phyllaries. Since these
characters have been used in the delineation of spe-
cies, variation at the population level has made classi-
fication difficult.
Pseudonoseris is a genus of three species with
narrow distributional ranges in the central Andes of
Peru with one species, Ps. discolor (Muschl.) H. Rob.
& Brettell, confined to lower elevations of the “ceja de
la montaña” of Dept. Puno in southern Peru and the
other two species, Ps. striata and Ps. szyszylowiczii,
from the western versant of the Depts. Cajamarca and
Lambayeque in northern Peru, a disjunction of over
SOEJIMA et al.: Speciation and hybridization of Paranephelius in Andes

377
















































Fig. 2. Morphological variation of Paranephelius. A, Sessile capitula with acute phyllaries (ASA 17557). B, Short peduncles arising from basal
rosette (ASA 17557). C, Basal rosette with multiple capitula (ASA 17557). D, Basal leaves (ASA 17575). E, Sessile capitula with densely tomentose,
acute phyllaries. F, Abaxial tomentose leaf surface (ASA 17557). G, Basal leaves (ASA 17578). H, Fleshy roots (ASA 17578). I, Habit with solitary
capitulum (ASA 17573). G, Basal leaf (ASA 17573). K, Fleshy roots and stolons (ASA 17573). L, Sessile capitula with oblong or ovate phyllaries
(ASA 17573). M, Individuals, several are products of stolons (ASA 17573). N, Capitulum of bullate-leaved population (ASA 17572). O, Basal leaves
with adaxial (left) and abaxial (right) surfaces (ASA 17572).
Journal of Systematics and Evolution Vol. 46 No. 3 2008 378
1200 km (Fig. 1). It possesses short or rosulate leafy
stems, scapose capitulescences with 2–4 pedunculate
capitula (Fig. 4: A–G), stipitate glandular phyllaries,
and 15–20, red to orange ray florets. Pseudonoseris
has been suggested to be the closest relative of
Paranephelius, differing from the latter in possessing
latex, erect and branching capitulescences (Robinson,
1977) and a suite of pollen ultrastructural characters
(Feuer & Dillon, 1982; Robinson & Marticorena,
1986). Because the subtribe Paranepheliinae is re-
stricted to the Andean Cordillera from high-elevation
or alpine environments to adjacent, lower-elevation
environments on both slopes (e.g., ceja de la mon-
taña), its evolutionary history may have been influ-
enced by orogeny of the Andes. In this study, we
examined the genetic variation and status of the
subtribe Paranepheliinae using nuclear ribosomal ITS
sequences and a chloroplast intergenic region. These
data allowed for testing the current classification and
morphological characters used in delineation of
species against the observed genetic differentiation.
1 Material and methods
1.1 Plant materials
Forty-five accessions of Paranephelius, six ac-
cessions of Pseudonoseris, and five outgroup species
representing genera within tribe Liabeae were sampled
in this study. Collections were made in the field and
additional accessions were derived from sampling
herbarium collections at HAO, CPUN, F, MO, NY,
and US. Populations of Paranephelius were field
sampled in Departments of Amazonas, Cajamarca,
and La Libertad (Appendix I), including the type
localities of P. ferreyrii and P. wurdackii (Robinson,
1977). Sampling included populations along altitud-
inal gradients, and in areas of secondary sympatry
where morphological variability was evident. Some
sequences used in the analyses were retrieved from the
data bank of DDBJ (DNA Data Bank of Japan) (Ap-
pendix II). For the accessions of Pseudonoseris, about
20 PCR products from a single individual were ampli-
fied by cloning and then sequenced. All new se-
quences generated in this study have been registered
with DDBJ (DNA Data Bank of Japan, Appendix I).
1.2 DNA extraction and PCR amplification
Total DNA was extracted from dried leaves using
a modified CTAB method of Doyle and Doyle (1987).
The primers used for amplification were ITS1 and
ITS4 (White et al., 1991) for the internal transcribed
region of the nuclear ribosomal DNA (ITS), and trnL
and trnF (Taberlet et al., 1991) for the intergenic
region between trnL and trnF in the chloroplast DNA
(designated as trnLF hereafter). Amplification reac-
tions followed Wen and Zimmer (1996); 95 ℃ for 3
min, followed by 38 cycles at 94 ℃ for 20 s, 50 ℃
for 30 s, and 72 ℃ for 40 s, and 72 ℃ for 5 min.
PCR products were sequenced in both directions by
cycle-sequencing using the Big-Dye version 3 chem-
istry (Perkin-Elmer), with a Prism 3100 Genetic
Analyzer (ABI).
Sequences were edited and aligned with Se-
quencher version 4.1 (Gene Codes), Clustal X version
1.81 and SeqPup/PPC version 0.6, followed by man-
ual alignment.
1.3 Phylogenetic analyses
The ITS and trnLF sequence data were analyzed
phylogenetically with PAUP* (version 4.0b10, Swof-
ford, 2003), using maximum parsimony and
neighbor-joining (Saitou & Nei, 1987) methods in
which gaps were treated as missing data or as new
characters. Maximum parsimony analyses were
performed by heuristic searches with 1000 random
sequence additions and tree bisection-reconnection
(TBR) branch swapping. No character was weighted.
The bootstrap support (BS) for the clades (Felsenstein,
1985) revealed in the maximally parsimonious tree(s)
(MPTs) was examined with 1000 bootstrap replicates
and the heuristic search options. Each data set was
also analyzed under the Bayesian inference using
MrBayes (Ronquist & Huelsenbeck, 2003) with the
model estimated with Modeltest version 3.6 (Posada
& Crandall, 1998; Posada & Buckley, 2004). Four
simultaneous chains were run with trees sampled
every 100 generations. The number of generations
needed to reach stationarity was determined by plot-
ting likelihood scores against generations.
1.4 Divergence time estimates
Because there are no fossils for Paranepheliinae
to calibrate the molecular clock, a rate of ITS evolu-
tion of other plant lineages from the literature was
used to estimate divergence times in Paranepheliinae.
We used the nucleotide substitution rate of the Ha-
waiian silversword alliance (Baldwin & Sanderson,
1998), a group of Asteraceae known as a good exam-
ple of insular adaptive radiation. The rate of the
Hawaiian silversword alliance is considered to be
equivalent with that of Espeletia Mutis ex Humb. &
Bonpl., an Andean endemic genus of Asteraceae
(Rauscher, 2002).


SOEJIMA et al.: Speciation and hybridization of Paranephelius in Andes

379


















Fig. 3. Scanning electron micrographs (SEM) of upper leaf surfaces of Paranephelius. A, Smooth and hairless surface (ISV 11397). B, Strongly
bullate and hairless surface (ISV 2778). C, Bullate and dense tomentose surface (ISV 11843). Bar=1 mm.































Fig. 4. Pseudonoseris. A–C, P. discolor. A, Plants growing in rocky soils in full sun (Quipuscoa 3338). B, Basal leaves with thin arachnoid
tomentum (Quipuscoa 3338). C, Capitulum (Quipuscoa 3338). D–G, P. szyszylowiczii. D, Habit along disturbed roadside (ASA 17548). E, Capitulum
(ASA 17548). F, Basal leaf, adaxial surface (ASA 17548). G, Basal leaf, abaxial surface (ASA 17548).
Journal of Systematics and Evolution Vol. 46 No. 3 2008 380
2 Results
2.1 Nuclear ribosomal ITS data
The lengths of the ribosomal DNA regions of the
subtribe Paranepheliinae were 239 or 240 bp for the
ITS1, 205 or 206 bp for the ITS2, and 165 bp for the
5.8S. Additionally, the 5′-end of the 26S region (26
bp) was also included in the analyses. The sequence
boundaries between the two ITS regions and three
coding regions (18S, 5.8S, and 26S) of rDNA were
determined following Baldwin (1992). Three acces-
sions (Kunkel 965, Wood 8345, and Wood 16192)
could not be amplified with the primers for this re-
gion. There were 28 variable sites and two indels
within the subtribe Paranepheliinae, 11 of which were
phylogenetically informative within the subtribe (Fig.
5). After the alignment with the outgroups, the matrix
had 705 nucleotide sites with 99 informative sites.
Bayesian analysis of the ITS data resulted in stabiliza-
tion of the likelihood scores at ca. 350,000 genera-
tions, and 875 trees were discarded as burn-in.
Pairwise distance (HKY85) among the sequences
within Paranephelius ranged from 0.00 to 2.17% with
the maximum between Zapata 06 and Funk 12088.
Between Paranephelius and Pseudonoseris, the
maximum distance was 3.33% (between ISV 2778
and Ps discolor 1i). Comparing Paranepheliinae and
the other genera of tribe Liabeae, the pairwise distance
ranged from 7.78% (between Ps discolor 2d and Erato
DC.) to 21.29% (between ISV 2778 and Chrysactin-
ium Wedd.).
On the ITS tree, Ps. discolor and Paranephelius
formed a clade (BS=PP=100%), and Ps. discolor was
sister to the clade of Paranephelius (BS=61%,
PP<80%). Two groups can be roughly recognized in
Paranephelius, designated here as Group A and
Group B (Figs. 5, 6). These two groups diagnostically
differed in four nucleotide substitutions (the sites 208,
492, 634, and 656). In addition to these diagnostic
sites, there were sequence variations within each
group. In Group A, two subgroups (A1 and A2) were
recognizable, and at least seven subgroups were found
in Group B (B1-7). In some accessions, direct se-
quencing detected multiple ITS sequences as sin-
gle-site additive polymorphisms with the occurrence
of overlapping double peaks on both complementary
strands for these diagnostic sites. Eight accessions
(ISV 8136, Dillon 2884, ISV 12100, Zapata 13A,
Zapata 10, Zapata 12A, Dillon 2843, ASA 16384)
were heterozygous on some of the four sites. These
heterozygosity could be interpreted as combinations of
any of the two haploid groups. In the phylogenetic
analysis of Paranephelius, these eight accessions were
excluded because of these polymorphisms.
Direct sequencing of the ITS regions of some
accessions of Pseudonoseris yielded many double
peaks. We thus sequenced about 20 PCR products of
the ITS regions for each accession by cloning all
accessions of Pseudonoseris. The NJ method was
adopted to show similarities between closely related
sequences. Figure 7 shows an NJ tree based on data
from the direct sequencing of Paranephelius and the
cloned sequences of Pseudonoseris. We found multi-
ple ITS sequences in all accessions of Pseudonoseris.
The sequences of the two accessions of Ps. discolor
formed a clade sister to Paranephelius, while all
sequences of the four accessions of Ps. szyszylowiczii
were nested in the clade of Paranephelius.
2.2 Intergenic region between trnL and trnF
(trnLF) of chloroplast DNA
Based on the ITS results, 19 Paranephelius and
five Pseudonoseris accessions were selected to be
sequenced for the trnLF region. Three taxa, Chiono-
pappus benthamii S. F. Blake, Munnozia annua
(Muschl.) H. Rob. & Brettell, and Philoglossa pur-
pureodisca H. Rob., were used as outgroups. The
lengths of the sequenced trnLF regions of the subtribe
Paranepheliinae were either 869 bp or 863 bp. Bayes-
ian analysis of the trnLF data resulted in stabilization
of the likelihood scores at ca. 200,000 generations,
and 500 trees were discarded as the burn-in. No node
was supported with more than 80% posterior prob-
ability in the trnLF tree. Paranephelius and Pseu-
donoseris formed a clade with relatively low support
(BS=68%, PP=80%). There were only two parsimony
informative sites out of six variable sites within
Paranephelius, and the relationships within
Paranephelius was thus not well resolved (Fig. 8).
However, Ps. discolor was sister to the clade of the
remaining accessions in the strict consensus tree. It
was noted that the four accessions of Ps. szyszylowic-
zii were nested in the clade of Paranephelius, al-
though the bootstrap value of this clade was less than
50%.
2.3 Morphological variation
Species of Paranephelius (Robinson, 1977) have
been differentiated based on a combination of charac-
ters including leaf shape, adaxial leaf surface mor-
phology, and floral characters. The phyllary shape and
pubescence are variable and unreliable for differenti-
ating the two groups revealed in the ITS sequences.
However, the overall leaf shape and lobing are rather
uniform in Group A, which has pinnately dentate
leaves (Fig. 9), generally ovate or oblong phyllaries
(Fig. 2: L), and bullate adaxial leaf surfaces (Fig. 2: D,
SOEJIMA et al.: Speciation and hybridization of Paranephelius in Andes

381




















































Fig. 5. The ITS informative sites among Paranephelius.
Journal of Systematics and Evolution Vol. 46 No. 3 2008 382


Fig. 6. The 50% majority-rule consensus tree derived from the maximum parsimony analysis of ITS sequences of Paranepheliinae (CI=0.760,
RI=0.773). Bootstrap support (>50%) and Bayesian posterior probabilities (>80%) are shown below the branches (BS/PP), but PP is not shown in
the outgroups. The clades which are not collapsed in the strict consensus tree are indicated with thickened branches. Black box designates a site
change, white box indicates a site change presumed to be a parallel evolution, and the numbers above the boxes indicate the site positions designated
in Fig. 5. The collection locality is cited by the name of the department for the accessions collected in Peru.
SOEJIMA et al.: Speciation and hybridization of Paranephelius in Andes

383
































































Fig. 7. The neighbor-joining tree based on the ITS data including
cloned sequences of Pseudonoseris. The alphabets at the end of the
names of Pseudonoseris indicate each of the cloned sequences.



Fig. 8. The strict consensus tree derived from maximum parsimony
analysis of trnLF sequences of Paranepheliinae (CI=1.000, RI=1.000).
Bootstrap support (>50%) and Bayesian posterior probabilities (≥ 80%)
are shown below the branches (BS/PP). The collection locality is cited
by the name of the department for the accessions collected in Peru.
Asterisks show hybrids presumed from ITS sequences.


F, G; Fig. 3: C) with or without dense pubescence.
The leaves of Group B are variable from unlobed
ovate (Fig. 2: J) to dentately-lobed elliptic in shape
(Fig. 2: O), but predominately ovate (Fig. 10). Its
adaxial leaf surfaces may range from smooth (Fig. 2:
J) to bullate (Fig. 2: O; Fig. 3: B) and glabrous to
pubescent (Fig. 3: A, C).
3 Discussion
3.1 Monophyly of the subtribe Paranepheliinae
Within the tribe Liabeae, Paranephelius and
Pseudonoseris constitute subtribe Paranepheliinae and
they share the specialized morphological characters
such as acaulescent habits, rosulate or basally aggre-
gated leaves, corolla tubes with long discs, and pollen
ultrastructure of pseudocaveate tectum, i.e., not
strictly caveate, but rather with thin basal columellae
under the spines (Feuer & Dillon, 1982; Funk et al.,
Journal of Systematics and Evolution Vol. 46 No. 3 2008 384


Fig. 9. Leaf shapes in Group A.



Fig. 10. Leaf shapes in Group B.
SOEJIMA et al.: Speciation and hybridization of Paranephelius in Andes

385
2007; Robinson & Marticorena, 1986). The mono-
phyly of the subtribe Paranepheliinae was also sup-
ported by a cladistic analysis using morphological
characters (Funk et al., 1996). In this study, molecular
phylogenetic analyses of both ITS and trnLF se-
quences showed Paranephelius and Pseudonoseris
form a well supported clade (BS=PP=100% in the ITS
tree, BS=68%, PP=80% in the trnLF tree).
3.2 Intergeneric relationship between Paranephe-
lius and Pseudonoseris
Although the phylogenetic resolution within the
subtribe Paranepheliinae is relatively low, some of our
results are consistent with the previous phylogenetic
hypotheses based on morphology.
The majority of Paranephelius accessions used
in this study are from northern Peru with a few from
Bolivia and Argentina, representing the northern and
southern ends of its distribution range. Although the
species boundaries within Paranephelius are still not
well defined, our samples cover the range of morpho-
logical variation within the genus. Pseudonoseris has
three species, with Ps. discolor from Dept. Puno on
the eastern versant and the other two from about 1250
km to the north on the western versant of the Andes
(Ps. striata in Dept. Lambayeque and Ps. szyszylowic-
zii in Dept. Cajamarca). Four accessions of Ps.
szyszylowiczii and two of Ps. discolor were included
in this study.
Morphologically, Pseudonoseris is easily distin-
guished from Paranephelius by containing latex in its
vegetative organs, erect branching inflorescences (Fig.
4), simple low-alveolate surface of the receptacle, and
shortened outer series of the pappus. These distinctive
characters of Pseudonoseris were considered to be a
less specialized condition than those of Paranephe-
lius, and the lower elevation of its distribution is also
considered to indicate ancestral features of Pseu-
donoseris (Robinson, 1983; Funk et al., 1996). Our
results showed that Ps. discolor is sister to the clade of
Paranephelius in both the ITS and trnLF trees.
However, the positions of the Ps. szyszylowiczii
accessions in the trnLF tree require further considera-
tion. In the trnLF tree, although the bootstrap and
posterior provability values are not high, all four Ps.
szyszylowiczii accessions are nested in the clade of
Paranephelius. This pattern may be the result of
chloroplast capture due to hybridization between the
two genera. Actually, the ITS sequences of these
accessions of Ps. szyszylowiczii showed several
double-peaked sites which indicated the existence of
multiple sequences. Several (15–24) cloned rDNA
fragments were thus sequenced for each accession of
Pseudonoseris, and we found up to ten ITS sequences
within an individual slightly different from each other.
When all of them were included in the phylogenetic
analysis, the trees obtained were quite messy. But
some of them had 5.8S sequences with a few site
changes or deletions compared to other 5.8S se-
quences common to Paranephelius. The analysis
excluding those with different 5.8S yielded a tree
shown in Fig. 7. Most of the sequences from an
individual formed a clade except for those of Ps
szyszy 1 and 2 which appear in separated clades. On
the other hand, all cloned sequences of Ps. discolor
formed a clade sister to the clade of Paranephelius
(Fig. 7).
Because Ps. discolor is sister to the Paranephe-
lius clade in both the ITS and the trnLF trees, it is
likely that Paranephelius and Pseudonoseris separated
before the divergence of Paranephelius. Concerning
the positions of Ps. szyszylowiczii, two hypotheses are
available. One is that Ps. szyszylowiczii originated
within Pseudonoseris, and subsequently hybridized
with Paranephelius as the maternal parent. Both the
ITS and trnLF trees indicate that all four accessions
used in this study are not pure Ps. szyszylowiczii but
progenies of hybridization. The ITS sequences from
the Pseudonoseris hybrids might be homogenized to
some degree by concerted evolution. Another hy-
pothesis is that Ps. szyszylowiczii itself originated via
hybridization between Pseudonoseris and Paranephe-
lius. The scattered positions of each accession of Ps.
szyszylowiczii on the trees suggest multiple origins of
these plants. The non-monophyly of the accessions Ps
szyszy 1 and 2 in the ITS tree may even implicate
backcrosses with Paranephelius. Nevertheless, both
hypotheses assume multiple, complicated intergeneric
hybridization. It is noted that Pseudonoseris szyszy-
lowiczii is morphologically intermediate between Ps.
discolor and P. uniflorus. Further studies are neces-
sary to test the hypothesis of this intergeneric hy-
bridization event.
Multiple ITS sequences can also arise via gene
duplication through polyploidization. The only chro-
mosome count available for Pseudonoseris is n=12
for Ps. szyszylowiczii (Dillon & Turner, 1982; Robin-
son et al., 1985), whereas Paranephelius has the
counts of n=9, 14, and 29 (Robinson et al., 1985;
Sundberg & Dillon, 1986). With the lack of chromo-
somal data of the plants we used in this study, we
cannot further consider the possible effects of poly-
ploidy.
Journal of Systematics and Evolution Vol. 46 No. 3 2008 386
3.3 Phylogeny and taxonomic implications in
Paranephelius
The multiple ITS sequences detected in Pseu-
donoseris have made the phylogenetic analysis within
Paranephelius using this marker difficult. We at-
tempted another analysis using a data set excluding
Ps. szyszylowiczii to infer the infrageneric relation-
ships in Paranephelius. This analysis suggested two
genetically differentiated groups, which can be recog-
nized by four site changes in Paranephelius (Fig. 5).
The two groups are designated as Group A and Group
B (Figs. 6, 9, 10).
The clade of Group A is composed of accessions
from northern Peru with strong support (BS=83%,
PP=100%) and can be distinguished by the four site
changes (Fig. 5). Group B includes the remaining
accessions from Peru, Bolivia, and Argentina. Taxa of
Group A and Group B are sympatric in Dept. Caja-
marca, where the greatest diversity of this genus is
detected (Fig. 1). The accessions of Group B are more
diverse morphologically than those of Group A. The
leaf characters such as shape, texture of surfaces, and
hair density, as well as phyllary shape have been
considered to be important in classifying Paranephe-
lius (Robinson, 1977). The accessions of Group A
have pinnately lobed leaves (Fig. 2: D, F, G; Fig. 9)
and bullate adaxial surfaces with dense tomentose
pubescence (Fig. 3: C). The accessions of Group B are
more variable; but the leaves are usually ovate, with
varying degrees of lobing (Fig. 10) and the adaxial
surfaces are variable from smooth (Fig. 2: J; Fig. 3: A)
to strongly bullate (Fig. 2: O; Fig. 3: B).
Paranephelius is one of the most easily recog-
nizable genera in the high-elevation, alpine communi-
ties of the central Andes (2500–4500 m), but the
species delimitations are notoriously difficult (Robin-
son, 1977). Authors have used a combination of
characters to diagnose species but the variation de-
tected within a population is often so considerable that
it defies recognition of species boundaries. The mor-
phological features of Group B show wide variation
across their geographic ranges and correspond to the
taxonomic descriptions of P. ovatus, P. bullatus, P.
jelskii, and P. wurdackii. The molecular analysis
suggests that P. asperifolius, a species with pinnately
lobed leaves and distributed from Bolivia to north-
western Argentina is included in Group B
(Funk12088, Funk11314). The descriptions of P.
uniflorus and P. ferreyrii, which have pinnately lobed,
bullate surface leaves and dense pubescence, corre-
spond to Group A. Assigning names to herbarium
material was not adopted in this study because a
detailed taxonomic revision of Paranephelius is in
preparation. The described species will be evaluated in
light of the molecular results.
3.4 Evidence of hybridization in Paranephelius
Biparentally inherited ITS may be polymorphic
due to hybridization. It is known that such polymor-
phism can be homogenized rapidly by concerted
evolution (Graur & Li, 1999). However, there are
some examples of additive heterozygosity in ITS,
which could be regarded as results of recent hybridi-
zation between different ITS sequence types (e.g.,
Krigia Schreb., Kim & Jansen, 1994; Miscanthus
Andersson, Hodkinson et al., 2002; Callicarpa L.,
Tsukaya et al., 2003; Mitchella L., Yokoyama et al.,
2003; and Cardamine L., Lihová et al., 2004).
Group A and Group B are distinguished by four
diagnostic sites (Fig. 5). Among the 45 accessions of
Paranephelius, eight (ISV 8136, Dillon 2884, ISV
12100, Zapata 13A, Zapata 10, Zapata 12A, Dillon
2843, ASA 16384) are heterozygous for the diagnostic
sites of the two groups. Direct sequencing of the ITS
region without cloning may not detect all heterozy-
gous sites (Gravendeel et al., 2004). The homozygous
state of certain sites such as site 208 does not neces-
sarily argue against their hybrid origin, and the se-
quences of these eight accessions support that these
accessions are hybrids between Group A and Group
B. Furthermore, morphological characters also support
this hypothesis of hybrid origins. The division of the
leaves is the most discernable character between
Group A and Group B; and the leaves of the eight
accessions are intermediate between simple-unlobed
and pinnately-lobed conditions. The accessions ISV
10335 from western Cajamarca (Prov. Chota) and
Zapata 12A from Calla Calla, Dept. Amazonas in
Group B showed heterozygosity at some sites (Fig. 5).
Their multiple sequences implicate that hybridization
between subgroups of Group B may also have oc-
curred.
Both Group A and Group B are distributed sym-
patrically in northern Peru. For example, in Cumbe-
mayo, Zapata 01 and Zapata 04 of Group A were
collected in the same area as Zapata 02, Zapata 03,
and Zapata 05 of Group B. Pollen exchange between
such morphologically and genetically differentiated
plants may not be difficult. Andean grasslands have
been used as pastureland for an extended period of
time, in addition to environmental disturbance such as
road construction and large-scale mining. Such human
activities might be followed by secondary contact
between the once isolated species, and could have
resulted in hybridization. No putative hybridization
SOEJIMA et al.: Speciation and hybridization of Paranephelius in Andes

387
has been found outside northern Peru.
3.5 Evolutionary history of Paranepheliinae in
relation to Andean geologic history
The diversity of ITS sequences in Paranephelii-
nae is compared with that of other plants in previous
reports. The maximum pairwise sequence difference
within Paranephelius is 1.7% (11 sites of the 636
aligned characters) found between ASA 17554 and
ISV 2778. In a previous study on the evolutionary rate
of ITS sequences, Baldwin and Sanderson (1998)
estimated the rate of diversification of the Hawaiian
silverswords (Asteraceae: Heliantheae) using paleo-
climatic and fossil data. According to their result, the
maximum within-group divergence is about 4.5% for
the group with the divergence with its most recent
common ancestor estimated to be about 5 mya.
Rauscher (2002) shows that the maximum diversity in
the Espeletia complex, an Andean highland group of
Asteraceae (Heliantheae), is about 3.9%. The Espe-
letia complex is distributed in the upper montane
forest and the páramo of the northern Andes which are
thought to have existed only for the last 2–4 million
years. Therefore, Rauscher (2002) considers that the
evolutionary rates of ITS sequences of the Espeletia
complex and silversword are similar. Compared with
these results, the diversity of Paranephelius is 1.7%,
less than half of the other two reports. This level of
divergence may suggest that modern Paranephelius
may have diverged much later, perhaps in the early
Pleistocene or possibly the late Pliocene. Outside the
Asteraceae, the intraspecific genetic divergence of
Saxifraga oppositifolia L. is comparable to that of
Paranephelius. Holderegger and Abbott (2003) esti-
mated the genetic divergence of S. oppositifolia as
1.5%. Although they did not determine the divergence
time, the two major lineages of S. oppositifolia are
thought to have become isolated during the Pleisto-
cene. Our result estimated the divergence time of
Paranepheliinae as in the early Pleistocene is consis-
tent with that of the circumpolar species of Saxifraga
L.
The subtribe Paranepheliinae is morphologically
quite distinctive in the Liabeae (Feuer & Dillon, 1982;
Robinson & Marticorena, 1986; Funk et al., 1996).
The present molecular analyses confirm the mono-
phyly of Paranepheliinae, but do not resolve its close
sister group. Microliabum Cabrera, a small genus in
Argentina, is presumed to be the sister to the subtribe
(V. Funk, pers. comm.), however, the sequences of
Microliabum were not available to test this hypothesis.
Although the low genetic diversity of Paranepheliinae
found in this study suggests a relatively recent radia-
tion, the morphological distinctiveness of
Paranepheliinae within the tribe Liabeae seems to
indicate its ancient origin. In comparison with the
other two subtribes in the Liabeae (Liabinae, 9 genera
and ca. 90 spp.; and Munnoziinae, 4 genera and ca. 50
spp.), the taxonomic diversity of Paranepheliinae is
relatively small. It seems that the ancestor of
Paranepheliinae originated in the alpine region of the
Andes and has remained poorly diversified for a long
time unless extinctions may have eliminated many
members of the group. The average of pairwise diver-
gence of ITS sequences between Paranepheliinae and
the outgroup is 11.8%. Using the nucleotide substitu-
tion rate in the silverswords, the origin of
Paranepheliinae is estimated to be about 13 mya
during the middle Miocene. During this period, the
climate was deteriorating and the central Andes were
actively uplifting. The distribution of Paranepheliinae
is from Peru through Bolivia and into northwestern
Argentina, and most diverse in the Department of
Cajamarca, Peru, near the northern limit of their
distribution. The Department of Cajamarca, the region
directly south of the Huancabamba Depression is one
of the centers of biodiversity in the Andes (Young et
al., 2002; Sánchez-Baracaldo, 2004; Sánchez &
Dillon, 2006). The Andes is older in the south, and
younger in the north. The ancestor of Paranepheliinae
may have originated in the southern mountainous area
of the Andes. The uplift of the Andes accompanied by
climatic changes of getting drier and colder may have
led to the expansion of distribution toward the north.
In general, the Andean alpine environments are wetter
in the north (páramo) and more xeric in the south
(puna). The expansion perhaps stopped at the Huan-
cabamba Depression (HD in Fig. 1), a major bio-
geographic barrier for many plants (Ayers, 1999;
Weigend, 2002). The wet and fragmented environ-
ment there may have stimulated the adaptive radiation
and diversification of Paranepheliinae.
For the alpine region of the central Andes, Wei-
gend et al. (2004) also showed little genetic differen-
tiation in Nasa ser. Grandiflorae (Loasaceae). In their
analysis of the trnL intron, the seven species of the
series Grandiflorae constituted a clade, but most of
them formed a polytomy within the clade. Their
results along with our observations on Paranepheliinae
seem to support a model of recent speciation and
geographic differentiation in the central Andes.
Journal of Systematics and Evolution Vol. 46 No. 3 2008 388
4 Conclusions
1. Paranephelius and Pseudonoseris comprise a
monophyletic group endemic to the central Andean
Cordillera, and Pseudonoseris discolor is sister to
Paranephelius.
2. Complicated intergeneric hybridization be-
tween Paranephelius and Pseudonoseris may explain
that Ps. szyszylowiczii is nested within the Paranephe-
lius clade.
3. Two genetically distinguishable groups were
recognized in Paranephelius. Multiple ITS sequences
suggest that hybridization in Paranephelius has
occurred in northern Peru and may be responsible for
the taxonomic confusion and difficulties.
4. The low genetic diversity in Paranephelius
(1.7%) suggests their recent radiation and speciation
in the high Andes.
Acknowledgements We thank the curators of US,
NY, CPUN, F, HAO, and HUT for loaning the her-
barium specimens and NY and US for granting the
permission to sample the collections. AS and JW
acknowledge the Pritzker Laboratory for Molecular
Systematics and Evolution of the Field Museum, and
the Laboratory of Analytical Biology of the National
Museum of Natural History, the Smithsonian Institu-
tion. MOD and JW acknowledge the support of the
US National Science Foundation (DEB-0071506 and
DEB-0415573). MZ thanks the Universidad Privada
Antenor Orrego (HAO) for support of field studies.
Abundio Sagástegui Alva, Isidoro Sánchez Vega,
Segundo Leiva Gonzáles, Erick Rodriguez, and Pedro
Lezama participated in the field expeditions. We are
most grateful to Victor Quipuscoa for his efforts of
collecting Pseudonoseris.
References
Ayers TJ. 1999. Biogeography of Lysipomia (Campanulaceae),
a high elevation endemic: an illustration of species
richness at the Huancabamba Depression, Peru. Arnaldoa
6: 13–27.
Baldwin BG. 1992. Phylogenetic utility of the internal
transcribed spacers of nuclear ribosomal DNA in plants:
an example from the Compositae. Molecular
Phylogenetics and Evolution 1: 3–6.
Baldwin BG, Sanderson MJ. 1998. Age and rate of
diversification of the Hawaiian silversword alliance
(Compositae). Proceedings of the National Academy of
Science USA 95: 9402–9406.
Bazán-Zurita H, Sánchez-V I, Cabanillas-M S, Miranda-A L.
1998. La Jalca de Oro. Minera Yanacocha. Mundo
Grafico, S. A., Lima, Perú. 128.
Cuatrecasas J. 1968. Páramo vegetation and its life forms. In:
Colloquium Geographicum 9 ed. Geoecology of the
mountainous regions of the tropical Americas. Bonn: C.
Troll. 163–186.
Dillon MO. 2005. Familia Asteraceae. In: Sklenár P, Luteyn JL,
Ulloa UC, Jørgensen PM, Dillon MO eds. Flora Genérica
de los Páramos: Guía Ilustrada de las Plantas Vasculares.
Memoirs of The New York Botanical Garden 92: 22–82.
Dillon MO, Turner BL. 1982. Chromosome numbers of
Peruvian Compositae. Rhodora 84: 131–137.
Doyle JJ, Doyle JL. 1987. A rapid DNA procedure for small
quantities of fresh leaf tissue. Phytochemical Bulletin 19:
11–15.
Fehrer J, Gemeinholzer B, Chrtek J Jr, Bräutigam S. 2007.
Incongruent plastid and nuclear DNA phylogenies reveal
ancient intergeneric hybridization in Pilosella hawkweeds
(Hieracium, Cichorieae, Asteraceae). Molecular
Phylogenetics and Evolution 42: 347–361.
Feuer SM, Dillon MO. 1982. Pollen morphology and
ultrastructure of the Liabeae (Asteraceae). Botanical
Society of America Miscellaneous Publication 162: 93.
Felsenstein J. 1985. Confidence intervals on phylogenies: an
approach using the bootstrap. Evolution 39: 783–791.
Funk VA, Robinson H, Dillon MO. 1996. Liabeae: taxonomy,
phylogeny and biogeography. In: Hind DJN, Beentje HJ
eds. Compositae: Systematics. Proceedings of the
International Compositae Conference, Kew, 1994. Vol. 1.
Kew: Royal Botanic Gardens. 545–467.
Funk VA, Robinson H, Dillon MO. 2007. Liabeae (Asteraceae).
In: Kubitzki K, Jeffrey JW eds. The families and genera of
vascular plants. Asterales. Vol. 8. Berlin: Springer-Verlag.
175–180.
Funk VA, Robinson H, McKee GS, Pruski JF. 1995.
Neotropical montane Compositae with an emphasis on the
Andes. In: Churchill SP, Balslev H, Forero E, Luteyn JL
eds. Proceedings of Neotropical montane forest
biodiversity and conservation symposium. Bronx, New
York: The New York Botanical Garden Press. 451–471.
Graur D, Li W-H. 1999. Fundamentals of molecular evolution.
Sunderland, MA: Sinauer Associates.
Gravendeel B, Eurlings MCM, van den Berg C, Cribb PJ. 2004.
Phylogeny of Pleione (Orchidaceae) and parentage
analysis of its wild hybrids based on plastid and nuclear
ribosomal ITS sequences and morphological data.
Systematic Botany 29: 50–63.
Hegarty MJ, Hiscock SJ. 2005. Hybrid speciation in plants:
new insights from molecular studies. New Phytologist
165: 411–423.
Hodkinson TR, Chase MW, Takahashi C, Leitch IJ, Bennett
MD, Renvoize SA. 2002. The use of DNA sequencing
(ITS and trnL-F), AFLP, and fluorescent in situ
hybridization to study allopolyploid Miscanthus
(Poaceae). American Journal of Botany 89: 279–286.
Holderegger R, Abbott RJ. 2003. Phylogeography of the
Arctic-Alpine Saxifraga oppositifolia (Saxifragaceae) and
some related taxa based on cpDNA and ITS sequence
variation. American Journal of Botany 90: 931–936.
Kim H-G, Funk VA, Vlasak A, Zimmer EA. 2003. A phylogeny
of the Munnoziinae (Asteraceae, Liabeae): circumscri-
ption of Munnozia and a new placement of M. perfoliata.
Plant Systematics and Evolution 239: 171–185.
Kim K-J, Jansen RK. 1994. Comparisons of phylogenetic
SOEJIMA et al.: Speciation and hybridization of Paranephelius in Andes

389
hypotheses among different data sets in dwarf dandelions
(Krigia, Asteraceae): additional information from internal
transcribed spacer sequences of nuclear ribosomal DNA.
Plant Systematics and Evolution 190: 157–185.
Lihová J, Aguilar JF, Marhold K, Feliner GN. 2004. Origin of
the disjunct tetraploid Cardamine amporitana
(Brassicaceae) assessed with nuclear and chloroplast DNA
sequence data. American Journal of Botany 91:
1231–1242.
Luteyn JL. 1999. Páramos: a checklist of plant diversity,
geographical distribution, and botanical literature. Bronx,
New York: The New York Botanical Garden Press. 278.
Mallet J. 2007. Hybrid speciation. Nature 446: 279–283.
Posada D, Buckley TR. 2004. Model selection and model
averaging in phylogenetics: advantages of the AIC and
Bayesian approaches over likelihood ratio tests.
Systematic Biology 53: 793–808.
Posada D, Crandall KA, 1998. MODELTEST: testing the
model of DNA substitution. Bioinformatics 14: 817–818.
Rauscher JT. 2002. Molecular phylogenetics of the Espeletia
complex (Asteraceae): evidence from nrDNA ITS
sequences on the closest relatives of an Andean adaptive
radiation. American Journal of Botany 89: 1074–1084.
Robinson H. 1977. Studies in the Liabeae (Asteraceae). X.
Additions to the genus Paranephelius. Phytologia 35:
233–240.
Robinson H. 1983. A generic review of the tribe Liabeae
(Asteraceae). Smithsonian Contributions to Botany 54:
1–69.
Robinson H, Marticorena C. 1986. A palynological study of the
Liabeae (Asteraceae). Smithsonian Contributions to
Botany 64: 1–50.
Robinson H, Powell AM, King RM, Weedin JF. 1985.
Chromosome numbers in Compositae, XV: Liabeae.
Annals of Missouri Botanical Garden 72: 469–479.
Ronquist F, Huelsenbeck JP. 2003. MrBayes 3: Bayesian
phylogenetic inference under mixed models.
Bioinformatics 19: 1572–1574.
Saitou N, Nei M. 1987. The neighbor-joining method: a new
method for reconstructing phylogenetic trees. Molecular
Biology and Evolution 4: 406–425.
Sánchez-Baracaldo P. 2004. Phylogenetics and biogeography of
the neotropical fern genera Jamesonia and Eriosorus
(Pteridaceae). American Journal of Botany 91: 274–284.
Sánchez VI. 1976. Contribución al Estudio Ecológico de la
Jalca en el Departamento de Cajamarca, Perú. Ph.D.
Dissertation. Trujillo, Perú: Universidad Nacional de
Trujillo. 29.
Sánchez VI, Dillon MO. 2006. Jalcas. In: Moraes RM, Øllgaard
B, Kvist LP, Borchsenius F, Balslev H eds. Botánica
Económica de los Andes Centrales. La Paz: Herbario
Nacional de Bolivia, Instituto de Ecología, Universidad
Mayor de San Andrés. 81–94.
Simpson BB, Todzia CA. 1990. Patterns and processes in the
development of the high Andean flora. American Journal
of Botany 77: 1419–1432.
Smissen RD, Breitwieser I, Ward JM. 2007. Genetic
characterization of hybridization between the New
Zealand everlastings Helichrysum lanceolatum and
Anaphalioides bellidioides (Asteraceae: Gnaphalieae).
Botanical Journal of the Linnean Society 154: 89–98.
Smith SW, Baum DA. 2006. Phylogenetics of the florally
diverse Andean clade Iochrominae (Solanaceae).
American Journal of Botany 93: 1140–1153.
Sundberg S, Dillon MO. 1986. Chromosome numbers. Taxon
35: 409–410.
Swofford DL. 2003. PAUP*: phylogenetic analysis using
parsimony (*and other methods), version 4.0b10.
Sunderland, MA: Sinauer Associates.
Taberlet P, Gielly L, Pautou G, Bouvet J. 1991. Universal
primers for amplification of three non-coding regions of
chloroplast DNA. Plant Molecular Biology 17:
1105–1109.
Tsukaya H, Fukuda T, Yokoyama J. 2003. Hybridization and
introgression between Callicarpa japonica and C. mollis
(Verbenaceae) in central Japan, as inferred from nuclear
and chloroplast DNA sequences. Molecular Ecology 12:
3003–3011.
Weberbauer A. 1936. Phytogeography of the Peruvian Andes.
Field Museum of Natural History, Botany Series 15:
1–587.
Weberbauer A. 1945. El mundo vegetal de los Andes peruanos.
Lima: Ministerio de Agricultura, Dirección de Agricultura,
Estación Experimental Agrícola de La Molina. 776.
Wen J, Zimmer EA, 1996. Phylogeny and biogeography of
Panax L. (the ginseng genus, Araliaceae): Inferences from
ITS sequences of nuclear ribosomal DNA. Molecular
Phylogenetics and Evolution 6: 167–77.
White TL, Bruns T, Lee S, Taylor J. 1991. Amplification and
direct sequencing of fungal ribosomal RNA genes for
phylogenetics. In: Innis MA, Gelfand DH, Sninsky JJ,
White TJ eds. PCR protocols. San Diego, California:
Academic Press. 315–322.
Weigend M. 2002. Observations on the Biogeography of the
Amotape-Huancabamba Zone in Northern Peru. Botanical
Review 68: 38–54.
Weigend M, Gottschling M, Hoot S, Ackermann M. 2004. A
preliminary phylogeny of Loasaceae subfam. Loasoideae
(Angiospermae: Cornales) based on trnL (UAA) sequence
data, with consequences for systematics and historical bio-
geography. Organisms, Diversity and Evolution 4: 73–90.
Yokoyama J, Fukuda T, Tsukaya H. 2003. Morphological and
molecular variation in Mitchella undulata, with special
reference to the systematic treatment of the dwarf form
from Yakushima. Journal of Plant Research 116: 309–315.
Young KR, Ulloa CU, Luteyn JL, Knapp S. 2002. Plant
evolution and endemism in Andean South America: an
introduction. Botanical Review 68: 4–21.

Appendix I Classification and voucher data on geographic distribution and origin of voucher accessions of the ingroup,
Paranephelius, Pseudonoseris, hybrids, and outgroups. The application of species names follows the discussion within the manu-
script; no formal synonymy changes are implied. Sequences were deposited in DDBJ (DNA Data Bank of Japan). Sagástegui=ASA,
Sánchez=ISV.
Ingroup: Paranephelius—Argentina. Prov. Salta, Funk 12088 (US), AB355482 (ITS), AB355569 (trnLF). Bolivia. Prov. Chu-
quisaca, Wood 8345 (US), AB355570 (trnLF); Wood 16192 (US), AB355570 (trnLF). Prov. La Paz, Funk 11314 (US), AB355483
Journal of Systematics and Evolution Vol. 46 No. 3 2008 390
(ITS), AB355572 (trnLF). Peru, Dept. Amazonas, Chachapoyas, Wurdack1240 (type: P. wurdackii, US), AB355484 (ITS), AB355573
(trnLF); Zapata 11A (F) AB359077. Dept. Cajamarca, Prov. Cajamarca, Dillon 4600 (F), AB355485 (ITS); Dillon 6472 (F),
AB355486 (ITS); Sánchez 2778 (F), AB355487 (ITS); Sánchez 6904 (F), AB355488 (ITS), AB355574 (trnLF); Sánchez 10414 (F),
AB355489 (ITS); Sánchez 10493 (F), AB355490 (ITS), AB355575 (trnLF); Sánchez 11397 (F), AB355491 (ITS); Sánchez 10487
(F), AB355492 (ITS); Zapata 02 (F), AB355493 (ITS), AB355576 (trnLF); Zapata 03 (F), AB355494 (ITS); Zapata 05 (F),
AB355495 (ITS), Sánchez 4052 (F), AB355500 (ITS); Sánchez 5577 (F), AB355501 (ITS); Zapata 01 (F), AB355502 (ITS); Zapata
04 (F), AB355503 (ITS); Zapata 06 (F), AB355504 (ITS); Zapata 07 (F), AB355505 (ITS). Prov. Chota, Sagástegui 17563 (F),
AB355506 (ITS), AB355580 (trnLF). Prov. Contumazá, Sagástegui 16162 (F), AB355507 (ITS); Sagástegui 17557A (F), AB355508
(ITS), AB355581 (trnLF). Dept. Junin, Prov. Junin, Kunkel 965 (US), AB355577 (trnLF). Dept. La Libertad, Prov. Gran Chimú,
Sagástegui 17554 (F), AB355509 (ITS). Prov. Otuzco, Sagástegui 17177 (F), AB355496 (ITS), Sagástegui 17573A, AB355497
(ITS), AB355578 (trnLF). Prov. Sánchez Carrión, Sagástegui 17340 (F), AB355498 (ITS), Sagástegui 16455 (F), AB355510 (ITS),
Sagástegui 17575A (F), AB355511 (ITS), AB355582 (trnLF), Sagástegui 17578A (F), AB355512 (ITS), AB355583 (trnLF). Prov.
Santiago de Chuco, Sagástegui 17572A (F), AB355499 (ITS), AB355579 (trnLF), Sánchez 11843 (F), AB355513 (ITS).
Paranephelius hybrids—Peru. Dept. Amazonas, Prov. Chachapoyas, Sánchez 8136 (F), AB355514 (ITS), AB355584 (trnLF);
Zapata 10 (F), AB355515 (ITS); Zapata 12A (F), AB355516 (ITS). Dept. Cajamarca, Prov. Cajamarca, Dillon 2884 (F), AB355517
(ITS). Sánchez 12100 (F), AB355518 (ITS), AB355585 (trnLF). Prov. Celendín, Zapata 13A (F), AB355519 (ITS), AB355586
(trnLF). Prov. Chota, Sánchez 10335 (F), AB355520 (ITS), AB355587 (trnLF). Dept. La Libertad, Prov. Pataz, Sagástegui 16384 (F),
AB355521 (ITS). Prov. Sánchez Carrión, Dillon 2843 (F), AB355522 (ITS).
Pseudonoseris discolor (Muschl.) H. Rob. & Brettell—Peru. Dept. Puno, Prov. Sandia, Quipuscoa 3338A (F), AB355529
(ITS-1a), AB355530 (ITS-1d), AB355531 (ITS-1e), AB355532 (ITS-1h), AB355533 (ITS-1i), 3338C (F), AB355523 (ITS),
AB355588 (trnLF), AB355534 (ITS-2b), AB355535 (ITS-2d), AB355536 (ITS-2e), AB355537 (ITS-2f).
Pseudonoseris szyszylowiczii (Hieron.) H. Rob. & Brettell—Peru. Dept. Amazonas, Prov. Chachapoyas, Wurdack 467 (US,
szyszy4), AB355554 (ITS-4e), AB355555 (ITS-4g), AB355556 (ITS-4h), AB355557 (ITS-4i), AB355592 (trnLF). Dept. Cajamarca,
Prov. Celendín, Sánchez 3822 (F, szyszy1), AB355538 (ITS-1a), AB355539 (ITS-1c), AB355540 (ITS-1d), AB355541 (ITS-1e),
AB355542 (ITS-1f), AB355543 (ITS-1h), AB355544 (ITS-1i), AB355589 (trnLF); Sagástegui 17548 (F, szyszy2), AB355545
(ITS-2a), AB355546 (ITS-2c), AB355547 (ITS-2d), AB355548 (ITS-2e), AB355549 (ITS-2g), AB355590 (trnLF); Zapata 20 (F,
szyszy3), AB355550 (ITS-3b), AB355551 (ITS-3 c), AB355552 (ITS-3d), AB355553 (ITS-3e), AB355591 (trnLF).
Outgroup: Chionopappus benthamii S. F. Blake—Peru, Dept. La Libertad, Prov. Gran Chimú, Sagástegui 17543 (F),
AB355524 (ITS), AB355593 (trnLF). Chrysactinium acaule (Kunth) Wedd.—Peru. Dept. La Libertad, Prov. Pataz, Sagástegui
16381 (F), AB355525 (ITS). Erato vulcanica (Klatt) H. Rob.—Costa Rica. Prov. Cartago, Wilbur 30775 (F), AB355526 (ITS).
Munnozia annua (Muschl.) H.Rob. & Brett.—Peru. Dept. La Libertad, Prov. Gran Chimú, Sagástegui 17550 (F), AB355527 (ITS),
AB355594 (trnLF). Philoglossa purpureodisca H. Rob.—Peru. Dept. Cajamarca, Prov. Contumazá, Dillon 4512 (F), AB355528
(ITS), AB355595 (trnLF).

Appendix II Accession numbers for the sequence data obtained from the DNA Data Bank of Japan (DDBJ)
Taxon Accession No. Reference
Liabum bourgeaui Hieron. AF539922 Kim et al., 2002
Dillandia perfoliata (S. F. Blake) V. A. Funk & H. Rob. AF539937 Kim et al., 2002
Sinclairia angustissima (A. Gray) B. L. Turner AF539953 Kim et al., 2002
Chrysactinium acaule (Kunth) Wedd. AF539939 Kim et al., 2002
Munnozia campii H. Rob. AF539927 Kim et al., 2002
Philoglossa mimuloides (Hieron.) H. Rob. & Cuatrec. f. AF539950 Kim et al., 2002
Erato polymnioides DC. AF539946 Kim et al., 2002