ArticlePDF Available

Species circumscription of the Caltha leptosepala polyploid complex (Ranunculaceae) based on molecular and morphological data

Authors:

Abstract and Figures

The Caltha leptosepala species complex (Ranunculaceae) is taxonomically unresolved, with authors of various regional floras recognizing different names and numbers of species. Integrating molecular, morphological, cytological, and geographic data, we describe three species in the complex, restoring two species names, C. biflora and C. chionophila, in addition to recognizing C. leptosepala. Based on chloroplast and nuclear ribosomal phylogenies, we illustrate key morphological synapomorphies for the three Caltha species, assess the usefulness of previously used morphological characters, and provide a dichotomous key for their field identification. A neotype is designated for C. leptosepala var. howellii because the originals were lost or destroyed.
Content may be subject to copyright.
1
This is an uncorrected proof. All tables and figures appear at the end of the manuscript.
The final version appears in:
Phytotaxa 316 (3): 201–223. doi.org/10.11646/phytotaxa.316.3.1
2
Species circumscription of the Caltha leptosepala polyploid complex (Ranunculaceae) based
on molecular and morphological data
KEIR M. WEFFERLING* & SARA B. HOOT
Department of Biological Sciences, PO Box 413, University of Wisconsin-Milwaukee, Milwaukee
WI, 53201, U.S.A.
*author for correspondance. Email: keirwefferling@gmail.com
Abstract
The Caltha leptosepala species complex (Ranunculaceae) is taxonomically unresolved, with
authors of various regional floras recognizing different names and numbers of species.
Integrating molecular, morphological, cytological, and geographic data, we describe three
species in the complex, restoring two species names, C. biflora and C. chionophila, in addition to
recognizing C. leptosepala. Based on chloroplast and nuclear ribosomal phylogenies, we
illustrate key morphological synapomorphies for the three Caltha species, assess the usefulness
of previously used morphological characters, and provide a dichotomous key for their field
identification. A neotype is designated for C. leptosepala var. howellii because the originals were
lost or destroyed.
Key words: allopolyploid, hybrid origin, marshmarigold, multiple origins, neotype, North
America, pollen
Introduction
The subalpine marshmarigold polyploid complex, Caltha leptosepala Candolle (1818: 310)
sensu lato (Ranunculaceae), is a heterogenous group of herbaceous plants growing in
mountainous regions of western North America. A recent cytogeographical study showed that
hexaploids (2n = 6x = 48) are more southerly in distribution, growing mainly in the southern
Rockies, Cascade-Sierra axis, and Coast Ranges (Wefferling et al. 2017). Dodecaploids (2n =
12x = 96) generally occupy the north of the range, from the Northern Rockies and Cascades in
3
the south to the Alaska Range in the north (Wefferling et al. 2017). Known nonaploids (2n = 9x
= 72) are limited to a single population in the Northern Rockies of Idaho, U.S.A., in a contact
zone between hexaploids and dodecaploids (Wefferling et al. 2017). The Pacific Northwest of
North America is an important intersection of three major lineages: two hexaploids and a
morphologically intermediate dodecaploid (Smit & Punt 1969, Wefferling et al. 2017). The
present study reviews the taxonomic history of the complex, examines molecular, biogeographic,
cytological, and morphological evidence for species-level recognition of taxa that are currently
subsumed within Caltha leptosepala, and provides an updated taxonomy and key for the Caltha
leptosepala polyploid complex. A brief history of the species that were, at one time or another,
subsumed in Caltha leptosepala is given.
In 1818, Candolle described two species of Caltha in western North America: C. biflora
Candolle (1818: 310) (Fig. 1) and C. leptosepala Candolle (1818: 310) (Fig. 2). Lawson (1884)
recognized only one of Candolle’s (1818) two species, relegating C. biflora to C. leptosepala
var. biflora (Candolle) Lawson (1884: 69), though with reservations about this placement. Huth
(1892) also recognized only one species: C. leptosepala with two varieties, C. leptosepala var.
rotundifolia Huth (1892: 68) and C. leptosepala var. howellii Huth (1892: 68) (Fig. 3). Huth
(1892) questioned whether Candolle’s original description of C. biflora was an accurate
description of the type specimen (Greene 1899), pointing out that Candolle (1818) referred to the
leaves as having a “sinu latissimo”, or a very broad sinus at the leaf base. Indeed, in the holotype
specimen for C. biflora (Fig. 1), if not prevented by the mounting and drying process, the sinuses
of at least some of the leaves would be closed, with overlapping leaf auricles (Huth 1892, Greene
1899). Regional variation in characters led Greene (1899) to segregate the complex into nine
species: C. biflora, C. malvacea Greene (1899: 75), C. leptosepala, C. macounii Greene (1899:
77), C. chelidonii Greene (1899: 78), C. howellii (Huth) Greene (1899: 79), C. rotundifolia
(Huth) Greene (1899: 80), C. chionophila Greene (1899: 80) (Fig. 4), and C. confinis Greene
(1899: 76; though C. confinis, based on a single incomplete specimen [US 270276, collected by
J. T. White s.n.], is almost certainly C. palustris Linneaus [1753: 558]; Smit & Punt 1969).
Rydberg (1900), working on the Flora of Montana and Yellowstone National Park, recognized C.
rotundifolia, and described C. uniflora Rydberg (1900: 474). Davis (1900) recognized a
combination of species described by Candolle (1818), Greene (1899), and Rydberg (1900), and
additionally described two varieties: C. biflora, C. chionophila, C. leptosepala, C. confinis, C.
4
chelidonii, C. uniflora, C. rotundifolia var. howellii (Huth) Davis (1900: 15), and C. leptosepala
var. macounii (Greene) Davis (1900: 16). Abrams (1944) recognized Caltha biflora subsp.
howellii (Huth) Abrams (1944: 175) and C. leptosepala var. rotundifolia Huth. Hitchcock et al.
(1964) described C. biflora var. rotundifolia (Huth) Hitchcock (1964: 335) and C. leptosepala
var. sulfurea Hitchcock (1964: 337). Smit & Punt (1969) described three populations of C.
leptosepala based on leaf, flower, and pollen morphology, corresponding with three major
geographic regions in western North America: C. leptosepala subsp. biflora (Candolle) Smit in
Smit & Punt (1969: 26) in the Sierran, Cascadian and Klamath ranges, C. leptosepala subsp.
leptosepala in the southern Rocky Mountains, and a morphologically intermediate “Caltha
leptosepala coll.” in the Canadian Rockies and Coastal Mountains. Later, Smit (1973)
reclassified Caltha leptosepala subsp. biflora as C. leptosepala subsp. howellii (Huth) Smit
(1973: 143). Ford (1997) recognized only C. leptosepala with no separation among intraspecific
taxa. Other studies of Caltha, such as Hoffmann (1999), Schuettpelz & Hoot (2004), Cheng &
Xie (2014), and Wefferling et al. (2017) followed Smit’s (1973) classification of two subspecies
of C. leptosepala, while Liu et al. (2016) recognized C. leptosepala and C. howellii.
By integrating molecular, biogeographic, cytological, and morphological data, we aim to
build on earlier work and provide an updated taxonomy and key for the Caltha leptosepala
polyploid complex.
Materials and Methods
Sampling:The sampling of Schuettpelz & Hoot (2004) served as a starting point, and
included nine outgroup species of Caltha. From the Caltha leptosepala polyploid complex, we
included two accessions each of the hexaploid taxa, from either the Cascades of Washington and
Oregon in the United States (identified as “Caltha leptosepala subsp. howellii” in Wefferling et
al. 2017) or from the Rockies of Colorado and Idaho (identified as “C. leptosepala subsp.
leptosepala” in Wefferling et al. 2017), and four accessions of dodecaploid C. leptosepala from
the Coastal Range in southeast Alaska, the North Cascades of Washington, the Blue Mountains
of northeastern Oregon, and the Northern Rockies of Idaho (identified as “Northern Caltha
leptosepala in Wefferling et al. 2017) (Appendix 1).
5
Genomic DNA was extracted from silica-dried or herbarium leaf material using either the
DNeasy Plant Mini Kit (Qiagen, Valencia, California, U.S.A.) or the GeneJET Plant Genomic
DNA Purification Mini Kit (Thermo Fisher Scientific, Waltham, Massachussetts, U.S.A.)
following manufacturers’ protocols after homogenization in a mortar and pestle under liquid
nitrogen or in 2 mL tubes with a tungsten bead and lysed in a TissueLyser II (Qiagen) bead mill
for two 30 sec cycles at 20 Hz.
Molecular data collection:—For PCR amplification and Sanger sequencing, we targeted
both nuclear ribosomal internal transcribed spacer (ITS) and chloroplast (cpDNA) regions in
order to track hybridization events and the direction of crosses. Molecular data included ITS,
cpDNA intergenic spacer rpL32–trnLUAG (rpL32–trnL), and trnLUAA–trnFGAA (trnL–trnF) (Table
1). PCR was performed in 20 µL reactions as follows: 7 µL GoTaq Green Master Mix (Promega
Corporation), 10 µL water, and 1 µL each of 10 µM upstream and downstream primers. Initial
ITS amplification primers were based on those of Nickrent et al. (1994) and then redesigned
based on preliminary sequence data (Table 1). The cpDNA rpL32–trnL primer sequences were
obtained from Shaw et al. (2007) then modified as necessary (Table 1). CpDNA trnL–trnF
primers were from Taberlet et al. (1991) and Azuma et al. (2011) (Table 1). PCR conditions for
all reactions were as follows: 5 min at 94º C, followed by 41 amplification cycles (1 min at 94º C
1 min at 52º C 1 min 30 sec at 72º C), a final extension step of 7 min at 72º C, then cooled to 4º
C. After checking for successful amplification on a 1% TBE agarose gel with ethidium bromide,
7 µL of each PCR amplicon were treated with 4 units of Exonuclease I (Thermo Fisher
Scientific) and 1 unit of FastAP thermosensitive alkaline phosphotase (Thermo Fisher Scientific)
in a total volume of 10 µL at 37º C for 30 min and 80º C for 15 min. Amplicons were then sent
to the University of Chicago comprehensive cancer center DNA sequencing and genotyping
facility (http://cancer-seqbase.uchicago.edu/) for Sanger sequencing of both DNA strands on an
Applied Biosystems (Foster City, California, U.S.A.) 3730XL 96-capillary automated DNA
sequencer using the amplification primers. When direct sequencing of ITS amplicons resulted in
polymorphic chromatograms (putative hybrids with insertions or deletions among ribotypes), the
mixed PCR product was column-purified using the QIAquick Gel Extraction Kit (Qiagen) and
then cloned with the pGEM-T easy vector system and JM-109 competent E. coli cells (Promega
Corporation, Madison, Wisconsin, U.S.A.). Successfully transformed (white) colonies were
picked and re-amplified using the RNA polymerase promoter primers SP6 (5’-
6
TATTTAGGTGACACTATAG-3’) and T7 (5’-TAATACGACTCACTATAGGG-3’) with the
following PCR conditions: 2 min at 94º C, followed by 30 cycles of 15 sec at 94º C 15 sec at 40º
C, 45 sec at 72º C, a final extension step of 5 min at 72º C, then cooled to 4º C. Amplicons were
sequenced in one direction with the SP6 primer. After initial sequences were compared from
hexaploids and dodecaploids, repeat-specific primers (Rauscher et al. 2002) were designed to
separate ribotypes in putative hybrids (Table 1).
Data analysis:—Alignments (Table 2) were partitioned by gene region, codon, and
spacer regions (8 cpDNA partitions, 3 ITS partitions). Partitionfinder 1.1.1 (Lanfear et al. 2012)
determined the optimal partitioning scheme as comprising 3 partitions for cpDNA regions, and
no partitioning of ITS (including ITS1, 5.8S, and ITS2). The nuclear and concatenated
chloroplast datasets (Table 2) were analyzed separately using Bayesian Markov chain Monte
Carlo (MCMC), maximum likelihood (ML), and maximum parsimony (MP) approaches.
Posterior probability (PP) values 95% and bootstrap (BS) values of 70% (Hillis & Bull 1993,
Alfaro et al. 2003) were considered moderate to strong support.
Bayesian MCMC phylogenetic inference was conducted using MrBayes 3.2.2
(Huelsenbeck & Ronquist 2001, Ronquist & Huelsenbeck 2003, Ronquist et al. 2012), with
datasets partitioned as described above, using reversible jump MCMC (Huelsenbeck et al. 2004).
Four chains (three heated) were run, sampling trees every 500 generations, until reaching a
conservative convergence diagnostic of average standard deviation of split frequencies 0.005
(i.e., much lower than the 0.1 default value in MrBayes 3.2; Ronquist et al. 2012). Additionally,
stationarity and convergence of runs were assessed visually by monitoring trace plots of
parameters using Tracer v1.6 (Rambaut et al. 2014) and checking that effective sample sizes for
all parameters were 200 (Drummond & Bouckaert 2015). After discarding 25% of steps, a
50% majority rule consensus tree was generated. RAxML 7.2.8 (Stamatakis 2006, implemented
through Geneious 7.1.6, Kearse et al. 2012) was used for ML analysis. We used the GTR
GAMMA model of sequence evolution (as recommended for trees with < 50 taxa in the RAxML
manual v8.2.X, Stamatakis 2016) with rapid bootstrapping, a search for the best-scoring ML tree,
and 1,000 BS replicates. PAUP* version 4.0b10 (Swofford 2002) was used for MP analysis. To
assess branch support, BS analyses were performed using a full heuristic search, with 500
replications of 20 random additions each; maxtrees were set to 5,000, and a 50% majority rule
consensus tree was generated.
7
Morphology:—Morphological and ecological data were gleaned from herbarium
specimens and our own collections and observations of Caltha in the field. Herbarium specimens
were borrowed from the following herbaria (using abbreviations as in the Index Herbariorum):
ALA, ASC, CIC, CSU, DAV, HSU, ID, MO, MONTU, NYBG, OSC, RBCM, RM, RSA, SRP,
US, USFS/RM, UWM, V, and WTU. Digitized type specimens of Caltha were examined from
BM, CAN, GH, K, NDG, NY, US, and WTU (the ! symbol is used to denote type specimens that
were seen by the first author, albeit in their digital form; see descriptions in Taxonomic treatment
section below). Morphological characters were examined using a stereomicroscope, measured,
and scored for samples from each of the cytotypes and putative taxa encompassing the entire
geographic range (for a total of 140 specimens). Particular attention was given to the following
characters: ratio of leaf blade length (including the leaf auricles) to width; form of auricles or
basal leaf lobes (closed: large and overlapping to cover leaf sinus; diplophyllous: upturned over
the leaf blade; or open: auricles not covering the sinus); number of flowers per stem (1–4);
filament width relative to anther width (filamentous: much narrower than anthers; intermediate:
approximately the same width as anthers; or broad: wider than anthers); attachment of carpels or
follicles to peduncle (sessile, substipitate, or stipitate); and shape of stylar beak (straight, curved,
or rolled into tight fiddlehead form).
Scanning electron microscopy (SEM) of 48 pollen samples was also performed, and
number and shape of apertures (porate or colpate) were scored. For SEM, anthers were collected
from herbarium specimens, dried overnight in a 50º C oven, then gently tapped over or rolled on
stubs with ultra smooth carbon adhesive tabs (Electron Microscopy Sciences, Hatfield, PA,
U.S.A.), coated with iridium (6 nm deposition at a 90º angle, followed by 5 nm at ~30º angle),
and examined using a Hitachi S-4800 field emission scanning electron microscope at 3.0 kV. We
examined at least 30 pollen grains of each specimen, and for publication we selected pollen
grains that were “typical” of the specimen. Image processing and capture functions were made
through the Hitachi PC-SEM software.
All figures were prepared (brightness and contrast, cropping, etc.) using Adobe Illustrator
CS5 and Adobe Photoshop CS5.
8
Results
The phylogenetic tree topologies derived from different approaches (Bayesian MCMC, ML, or
MP) were all similar or identical when comparing a single dataset (cpDNA or ITS alone; Figs. 5,
6), but were discordant between datasets. Well-supported clades, by all measures (PP, MLBS,
and MPBS) and in both datasets (cpDNA and ITS), included the “Psychrophila group”
(Schuettpelz & Hoot 2004), Caltha chionophila (ID, CO) + C. leptosepala (ID, OR), and C.
biflora (OR, WA) + C. leptosepala (AK, WA) (Figs. 5, 6). In contrast, the C. leptosepala
complex as a whole was either monophyletic (cpDNA; Fig. 5) or paraphyletic (ITS; Fig. 6). Each
C. leptosepala individual yielded two ITS ribotypes, one of which grouped with C. biflora, the
other with C. chionophila, both with strong support (Fig. 6). Additionally, cpDNA from 12x C.
leptosepala specimens either grouped with C. biflora or C. chionophila with strong support (Fig.
5).
Of the morphological characters examined, leaf length to width ratio (Wefferling et al.,
2017), filament width, and pollen morphology (number and shape of pori or colpi) (Table 3)
were most consistent with molecular data (Wefferling & Hoot, unpublished data) and cytological
determination (Wefferling et al. 2017). However, pollen characters were not entirely consistent
with leaf macromorphology or molecular data (Table 3; Figs. 7–10; Wefferling & Hoot,
unpublished data). Pollen from C. biflora ranged from pantoporate to pantocolpate (Fig. 7).
Caltha chionophila was almost always tricolpate (Fig. 8) with some notable exceptions (Fig. 8H,
I). Caltha leptosepala ranged from tricolpate (Fig. 9G) to pantoporate (Fig. 9I). Several
specimens had malformed, variably sized, and apparently inviable pollen, including some
hexaploids (Fig. 8I), dodecaploids (Fig. 9I), hybrid or non-hybrid “aneuploids” (based on flow
cytometry data, Wefferling et al. 2017; Fig. 10A, B), and the single nonaploid specimen (Fig.
10C).
All members of the species complex can be described as follows: fleshy hairless herbs
arising 5 to 40 cm from thick caudices or short rhizomes; simple petiolate leaves basal (to
cauline), with dentate, crenate, or subentire margins; leaf bases cordate to sagittate, or auricles
overlapping (sometimes upturned and covering part of the lamina; i.e., diplophyllous); plants
with 1 to 6 perfect, hypogynous flowers, apetalous with 5 to 12 (to 18) white (maturing to
yellow-white) linear to oblong petaloid sepals (abaxially blue- or green-tinted); many (up to 50)
9
stamens with filiform to broad and flattened filaments, the filaments covered with trichomes; 4 to
12 (to 32) sessile to stipitate carpels with stigma and style straight to slightly curved, carpels
maturing to many-seeded follicles; ovules anatropous, seeds dark, with slight “checkerboard”
texturing on seed coat, endosperm present, and embryo small relative to mature seed (Hitchcock
& Cronquist 1973, Morris 1973, Smit 1973, Ford 1997).
Discussion
The present study includes more specimens of the Caltha leptosepala complex than previous
phylogenetic work on the genus and provides novel insights into relationships among members
of the polyploid complex. In particular, examining the phylogeny in light of ploidy level
variation (Wefferling et al. 2015, 2017) and morphology (Wefferling et al. 2017, present study)
allows for a better understanding of the biological diversity present in the group. Examination of
type specimens (Figs. 1–4) and 140 additional accessions from across the geographical range of
the complex supports the recognition of three species in the polyploid complex: hexaploid C.
biflora, hexaploid C. chionophila, and allododecaploid C. leptosepala.
The discordance between nuclear and chloroplast phylogenies (Figs. 5, 6) is consistent
with some earlier work on the genus (Schuettpelz & Hoot 2004, Cheng & Xie 2015), and the
topologies are similar (Cheng & Xie 2015, Liu et al. 2016) or identical (Schuettpelz & Hoot
2004) to previous work. Notably, we did not combine nuclear and chloroplast data as in previous
studies due to the presence of multiple nuclear ribotypes in the dodecaploid Caltha leptosepala.
The discordant topologies between nuclear and chloroplast datasets could be explained by longer
coalescent times in nuclear genomes due to a larger effective population size than in chloroplast
genomes paired with incomplete lineage sorting (Rautenberg et al. 2010). Alternatively,
introgressive hybridization could explain the discordance (Hardig et al. 2000, Yoo et al. 2002,
Rautenberg et al. 2010), though it is not clear which populations or lineages would be involved
in such crosses.
Through the use of ribotype-specific primers and/or cloning of ITS PCR product, we
were able to demonstrate a likely hybrid origin of the sampled dodecaploids between ancestral
10
Caltha biflora and C. chionophila (Fig. 6). Further, in the cpDNA dataset allododecaploid C.
leptosepala groups with either C. biflora or C. chionophila (Fig. 5), providing evidence for
reciprocal origins of allododecaploid C. leptosepala.
Individual specimens are most easily grouped by leaf blade length to width ratio, filament
width, and shape and number of colpi (Table 3). These characters, along with geography, are
almost always sufficient for discriminating among Caltha biflora, C. chionophila, and C.
leptosepala. However, pollen can be misleading in differentiating among taxa, despite reports of
the informativeness of such characters (Smit & Punt 1969). For example, C. biflora almost
always has pantoporate pollen (Fig. 7), as found by Smit & Punt (1969), but some specimens
have colpate pollen (Fig. 7I). Similarly, C. chionophila displays tricolpate pollen (Fig. 8), but
sometimes diverges from this morphology, with colpi merged (Fig. 8H) or very poorly
developed (8I). Notably, the previous exceptional cases occur in mixed-ploidy populations in the
Pacific Northwest of North America (Wefferling et al. 2017). Pollen from allododecaploid C.
leptosepala is usually 4–8-colpate (Fig. 9), but quite variable, with pantoporate (Fig. 9H),
tricolpate (Fig. 9I), or malformed and apparently inviable (Fig. 9G) grains. Some rare plants,
such as nonaploids and putative aneuploids (Wefferling et al. 2017), showed variable and
malformed (Fig. 10C) or tricolpate pollen (Fig. 10B). One putative aneuploid with the
morphology of C. biflora (Wefferling et al. 2017) is apparently a hybrid (Wefferling & Hoot,
unpublished data) with malformed to pantocolpate pollen (Fig. 10A).
In conclusion, some difficulties remain in discriminating between Caltha chionophila
and C. leptosepala in the Northern Rockies (Wefferling et al. 2017) where the two are
sometimes morphologically very similar. There are also some specimens that morphologically
match C. biflora on the Olympic Peninsula (Washington, U.S.A.) and Vancouver Island (British
Columbia, Canada), but have putatively aneuploid genomes (Wefferling et al. 2017) and hybrid
origins (Wefferling & Hoot, unpublished data). Nevertherless, a combination of leaf and anther
characters should allow for field identification in almost all cases. The few exceptions we have
seen to the overall molecular, cytogenetic, and morphological patterns among members of the
marshamarigold polyploid complex should be addressed through studies sampling a larger
number of individuals from across the range.
11
Taxonomic treatment
Caltha biflora Candolle (1818: 310). C. leptosepala var. biflora (Candolle) Lawson (1884: 69).
C. leptosepala subsp. biflora (Candolle) Smit (1969: 26).
Type:—CANADA. British Columbia: Northwest coast of North America, inland behind Banks
Island (between Haida Gwaii and mainland), 1792, Menzies s.n. (holotype BM!). Fig. 1.
Caltha leptosepala var. howellii Huth (1892: 68). C. howellii (Huth) Greene (1899: 79). C.
rotundifolia (Huth) Greene var. howellii (Huth) Davis (1900: 15). C. biflora subsp.
howellii (Huth) Abrams (1944: 175). C. leptosepala subsp. howellii (Huth) Smit (1973:
143).
Type:—U.S.A. California: near Colby, Butte County, 1896, R.M. Austin s.n. (neotype,
designated here, NDG!). Fig. 3.
Caltha malvacea Greene (1899: 75).
Type:—U.S.A. Washington: Cascade Mountains, 1838–1842, Wilkes Expedition 484 (Lectotype
US!).
Morphology and cytology:—Leaf blades 0.7–1 × as long as wide, reniform to orbicular
(rarely emarginate), often diplophyllous (especially in the southern part of the range), margins
crenate (in the north) to subentire (in the south) (Figs. 1, 3); flowers (1–)2(–3) per stem;
filaments filiform, 0.1–0.2(–0.3) mm wide (Giblin et al. in press), narrower and often longer than
anthers (Figs. 1, 11A); pollen pantoporate to pantocolpate (Fig. 7); follicles with stylar beak 0.1–
0.2 mm in length (Giblin et al. in press); hexaploid (2n = 6x = 48, Wefferling et al. 2015;
holoploid genome size ca. 10–18 pg/2C, Wefferling et al. 2017).
Taxonomic notes:—Regarding the original type specimen of Caltha leptosepala var.
howellii, collected by J. Howell in 1882 in the Cascade Mountains and deposited in “HGB”
(HGB was Howell’s abbreviation for “Herbarium generale Berolinense” at Berlin, The Botanic
Garden and Botanical Museum Berlin-Dahlem, current acronym B; B. Hellenthal, Museum of
Biodiversity and Greene-Nieuwland Herbarium, pers. comm.): “…If the specimen has ever been
part of our herbarium [B] then it was probably lost in WW2” (R. Vogt, Botanischer Garten und
Botanisches Museum Berlin-Dahlem, pers. comm.). Another specimen collected by Howell
(NDG17323) would be an ideal neotype, but lacks clear morphology of the carpels, among other
important features. Therefore, NDG17325 (Fig. 3), collected by R.M. Austin near Colby, Butte
12
Co. in 1896 is selected as neotype as (1) it is mentioned by Greene (1899), (2) the leaves bear
resemblance to those drawn in Huth's (1892) treatment (his figure 8), and (3) it is a complete
specimen showing stamens, nearly mature and immature carpels, etc. The diplophyllous nature
of the leaves (Huth 1892) is not so clear in this specimen, but one upturned auricle can be seen in
the leftmost leaf.
Geographic distribution:—Lowland to subalpine in and west of Coast Ranges and
Cascade-Sierra axis; from southeast Alaska south through coastal British Columbia, Washington,
Oregon, California, and western Nevada (Wefferling et al. 2017, Giblin et al. in press).
Additional specimens examined (* pollen sampled for SEM, n = 12; † ploidy level
estimated via FCM or chromosome count [Wefferling et al. 2017], n = 27):—CANADA: British
Columbia: N. end of first firebreak on Croman Rd. Woss Area Vancouver Island, 50.175566, -
126.363954, elevation 390 m, Stevenson S. 518*, 6/5/1975 (V). British Columbia: 14.5 miles on
road from Shawnigan Lake to Port Renfrew, 48.616667, -123.899996, elevation 580 m, Calder,
J.A. and K.T. MacKay 29381, 21-May-1961 (US). British Columbia: Mountain at head of
McClintock Bay, Masset Inlet, Graham Island, 53.640189, -132.574431, elevation 310 m,
Calder, J.A., D.B.O. Savile, and R.L. Taylor 21584*, 18-Jun-1957 (V). British Columbia,
Vancouver Island, Brooks Quadrant, Harris Peak, Brooks Peninsula, 50.225, -127.725, elevation
825 m, Ogilvie, R.T., W.J. Schofield and R.J. Hebda 848912, 9-Aug-1984 (V). British Columbia:
Three Arm creek, San Juan River Valley, 48.587475, -124.124074, elevation 900 m, Hebda, R.
and G. Allen 91-17, 9-Jul-1991 (V). British Columbia: ca. 1/4 mile beyond end of Feona Rd.
(Canadian Forest Products' setting #M-50), ca. 20 miles S of Woss, northern Vancouver Island,
49.930832, -126.563567, elevation 823 m, Bavis, P. 770608-001, 8-Jun-1977 (V). British
Columbia, North Coast, Princess Royal Island, NW end of island, ridge E of Home Bay,
53.274999, -129.05722, elevation 639 m, Marr, K.L. and C. Copley KM6340, 25-Jun-2005 (V).
U.S.A.: Alaska: Thorne Bay, Lyman Anchorage Cove, 55.6, -132.58, elevation 91 m, Koval, V.L.
5*, 19-May-1991 (ALA). Alaska: Deer Mtn., 3.5 miles southeast of Ketchikan, 55.38, -131.57,
elevation 730 m, Jaques 1209, 2-Jul-1972 (OSC). Alaska: Ketchikan, Harriet Hunt Lake road,
55.478649, -131.607508, elevation 230 m, Williams, M. 3094, 5-Jul-1972 (WTU). California:
Kimshew Point quad. High Cascade Range, below the head of Keyser Creek and just below the
dirt road, 0.5 km E of Bald Mountain Lookout, 39.9525, -121.4775, elevation 1695 m, Janeway,
L.P. 7564*, 26-May-2002 (CSU). California: Lassen National Forest (NF), 3.3 miles on rd. 110,
13
off Hwy. A21, 40.502731, -121.129142, elevation 1920 m, Meyer, K.M. and A. Townesmith
145†, 21-Jul-2007 (DAV). California: McKay Camp Meadows, 40.91, -123.04, elevation 1524
m, Mesler 909, 12-Jul-2011 (HSU). California: Dunsmuir Quad. Bog around Cedar Lake, ca. 11
mi. SW of Mt. Shasta City, 41.2109, -122.5045, elevation 1737 m, Taylor, M.S. 2778, 4-Jun-
1980 (MO). California: Kaiser Wilderness, Sierra NF, George Lake drainage, 37.295611, -
119.169917, elevation 2737 m, Wefferling, K.M. and L. Woo 21*, 27-Jun-2012 (UWM).
California: Kaiser Wilderness, Sierra NF, 20 m from Kaiser Pass Rd., 37.300556, -119.104806,
elevation 2743 m, Wefferling, K.M. 25†, 28-Jun-2012 (UWM). California: Shasta-Trinity NF,
meadow above and NW of Upper Gumboot Lake, 41.209722, -122.516306, elevation 1904 m,
Wefferling, K.M. and L. Woo 30*, 3-Jul-2012 (UWM). California: Shasta-Trinity NF, E of
Pacific Crest Trail, above Fawn Meadow, 41.239972, -122.508556, elevation 1997 m,
Wefferling, K.M. and L. Woo 33†, 5-Jul-2012 (UWM). Nevada: Lake Tahoe Basin Mgmt Unit
(USFS), Incline Lakes area, ca. 0.5 mi. SW of Incline Lake, 39.289, -119.934, elevation 2500 m,
Christie, K. 1808, 15-Jun-2010 (ASC). Nevada: Hobart Creek Reservoir, Lake Tahoe State Park,
6 miles E of Carson City, 39.1928, -119.8697, elevation 2347 m, Johnson, J.M. 097*, 1-Jun-
1997 (NYBG). Oregon: Rogue River NF, Oregon hwy 230 to FS road 6510 to 6515, two miles
south of summit of Hershberger Mountain, 43.0197, -122.4519, elevation 1676 m, Baldwin, C.
428, 28-Jun-1994 (ID). Oregon: Mt. Hood, at Government Camp on Hwy 26, 45.3, -121.76,
elevation 1158 m, Chambers, K.L. 1687, 31-May-1961 (NYBG). Oregon: Cascade Range,
Willamette National Forest, Bruno Meadows area, about 5 air miles SE of Idanha, 44.6473, -
122.0011, elevation 1244 m, Halse, R.R. 5869†, 13-Jul-2000 (OSC). Oregon: Cascade Range,
Willamette National Forest, Bruno Meadows, along F.S. Road 2234, about 5 air miles SE of
Idanha, 44.6474, -122.007, elevation 1244 m, Halse, R.R. 4746, 9-Jul-1994 (RSA). Oregon:
Fanno Bog, 44.851718, -123.596, elevation 850 m, Wilson, B.L. 6986, 11-Jun-1994 (SRP).
Oregon: Klamath NF, Mt Ashland, Pacific Crest Trail, 42.075694, -122.726861, elevation 2000
m, Wefferling, K.M. and L. Woo 38†, 6-Jul-2012 (UWM). Oregon: Umpqua NF, Abbott Butte,
just outside Rogue-Umpqua Divide Wilderness, 42.941167, -122.547417, elevation 1790 m,
Wefferling, K.M. and L. Woo 39*†, 7-Jul-2012 (UWM). Oregon: Umpqua NF, side of Silent
Creek near Diamond Lake inlet, 43.126361, -122.161111, elevation 1586 m, Wefferling, K.M.
43†, 8-Jul-2012 (UWM). Oregon: Mt. Hood NF, near trailhead for Elk Meadows and Sahalie
Falls, 45.322361, -121.634111, elevation 1359 m, Wefferling, K.M. and L. Woo 44†, 16-Jul-2012
14
(UWM). Oregon: Mt. Hood Wilderness, Mt. Hood NF, on trail to Paradise Park along PCT,
45.357445, -121.752278, elevation 1686 m, Wefferling, K.M. 61†, 8-Aug-2012 (UWM). Oregon:
Umpqua NF, Umpqua side of Rogue-Umpqua Divide, S end Donegan Prairie, 42.920087, -
122.590767, elevation 1605 m, Wefferling, K.M. and L. Woo 42†, 7-Jul-2012 (UWM). Oregon:
Mt. Hood NF, along creek below Mt. Hood Meadows Drive, 45.326963, -121.660281, elevation
1585 m, Wefferling, K.M. 47s†, 16-Jul-2012 (UWM). Oregon: Mt. Hood Wilderness, NW end of
Elk Meadows, 45.34525, -121.620528, elevation 1570 m, Wefferling, K.M. 68*†, 1-Jul-2013
(UWM). Oregon: Mt. Hood Wilderness, NW end of Elk Meadows, 45.34525, -121.620528,
elevation 1570 m, Wefferling, K.M. 69*†, 1-Jul-2013 (UWM). Washington: Mink Lake, Olympic
NP, 47.949332, -123.868271, elevation 940 m, Harthill, M.P. 1069*, 11-Aug-1972 (RSA).
Washington: Mink Lake, Olympic NP, 47.9479, -123.8677, elevation 950 m, Harthill, M.P. 988,
11-Aug-1972 (RSA). Washington: Gifford Pinchot NF, Indian Heaven Wilderness, outside of
Carson, 46.00317, -121.78873, elevation 1424 m, Meyer A. s. n., 24-Sep-2011 (UWM).
Washington: Alpine Lakes Wilderness Mt. Baker-Snoqualmie NF, near outlet of Rainy Lake,
47.514215, -121.537258, elevation 1116 m, Wefferling, L. 1, 10-Jun-2012 (UWM). Washington:
Mt. Baker-Snoqualmie NF, Heather Lake, 48.071806, -121.783917, elevation 740 m, Wefferling,
K.M. 48†, 21-Jul-2012 (UWM). Washington: Alpine Lakes Wilderness, Mt. Baker-Snoqualmie
NF, between Mason and Kulla Kulla Lakes, 47.426806, -121.54625, elevation 1341 m,
Wefferling, K.M. 50*†, 22-Jul-2012 (UWM). Washington: Alpine Lakes Wilderness, Mt. Baker-
Snoqualmie NF, outlet of Nimue Lake, 47.539556, -121.397167, elevation 1138 m, Wefferling,
K.M. and L. Wefferling 52, 23-Jul-2012 (UWM). Washington: Alpine Lakes Wilderness, Mt.
Baker-Snoqualmie NF, below Denny Mountain, 47.436972, -121.456333, elevation 1114 m,
Wefferling, K.M. and B. Wefferling 54†, 30-Jul-2012 (UWM). Washington: Gifford-Pinchot NF,
below road W of Takhlakh Lake, 46.274691, -121.606949, elevation 1265 m, Wefferling, K.M.
55†, 1-Aug-2012 (UWM). Washington: Gifford-Pinchot NF, Takhlakh Meadow, 46.270444, -
121.58875, elevation 1382 m, Wefferling, K.M. 57†, 1-Aug-2012 (UWM). Washington: Gifford-
Pinchot NF, Takhlakh Meadow, 46.269707, -121.588677, elevation 1400 m, Wefferling, K.M.
58†, 1-Aug-2012 (UWM). Washington: Gifford-Pinchot NF, Takhlakh Meadow, 46.269707, -
121.588677, elevation 1400 m, Wefferling, K.M. 59s†, 1-Aug-2012 (UWM). Washington: Sauk
Mountain, 48.523121, -121.598597, elevation 1550 m, Shrum, J. SM5†, 22-Jul-2015 (UWM).
Washington: Olympic National Park, Boulder Creek, lower site, 47.976209, -123.697533,
15
elevation 677 m, Hunter, G. BL4†, 15-Jul-2015 (UWM). Washington: Olympic National Park,
Little River, 48.04704, -123.504876, elevation 330 m, Hunter, G. LR1†, 15-Jul-2015 (UWM).
Washington: Gifford-Pinchot NF, Takhlakh Meadow, 46.268447, -121.586248, elevation 1400
m, Wefferling, K.M. CR1*†, 1-Jan-2014 (UWM). Washington: Gifford-Pinchot NF, Babyshoe
Pass, 46.268139, -121.604472, elevation 1320 m, Wefferling, K.M. and L. Wefferling 72s, 15-
Jul-2014 (UWM). Washington: Gifford-Pinchot NF, Takhlakh Meadow, 46.268447, -
121.586248, elevation 1400 m, Wefferling, K.M. and L. Wefferling 94s†, 15-Jul-2014 (UWM).
Washington: Gifford-Pinchot NF, Babyshoe Pass, 46.268139, -121.604472, elevation 1320 m,
Wefferling, K.M. and L. Wefferling 74s, 15-Jul-2014 (UWM). Washington: Olympic Peninsula,
Rd. #303 between Rugged Ridge and Pine Mt., 47.948646, -123.961192, elevation 884 m,
Buckingham, N. 02023, 29-Jun-1979 (WTU). Washington: Cascade Mountains of Western
Washington, Mt. Gardner, collected at the intersection of roads 155, 155.5, and 155.6, 47.3667, -
121.5514, elevation 1100 m, Stout, T. SAR4†, 23-Jun-2001 (WTU).
Caltha chionophila Greene (1899: 80).
Type:—U.S.A. Colorado: near Pagosa Peak, 11,000 ft., 1899; Baker s.n. (isolectotypes BM! K!
NDG!). Fig. 4.
Caltha uniflora Rydberg (1900: 474).
Type:U.S.A. Montana: Haystack Peak, 30003300 m., August 1899, Koch s.n. (holotype NY!).
Morphology and cytology:—Leaf blades 1.2–2.5(–3) × as long as wide, +/- sagittate or
auriculate, the generally short auricles little overlapping or not at all, rarely diplophyllous
(upturned over the leaf blade), margins mostly sinuate-dentate to subentire (Fig. 4); flowers 1(–
3) per stem; filaments strongly flattened, 0.5–1(–1.2) mm wide (Giblin et al. in press), wider and
generally shorter than anthers (Figs. 4, 11B); pollen tricolpate (rarely misshapen) (Fig. 8);
follicles with ± curved stylar beak 0.5–1.2 mm in length (Giblin et al. in press); hexaploid (2n =
6x = 48, Wefferling et al. 2015; holoploid genome size ca. 7–10 pg/2C, Wefferling et al. 2017).
Taxonomic notes:—The specimen pictured in Figure 4 (NDG17369) is at the Greene-
Nieuwland Herbarium, determined by E.L. Greene and labeled with his own hand (B. Hellenthal,
pers. comm.). Smit (1973) designated Baker 322 as lectotype, but Greene did not indicate a
16
collection number; the collector, collection locality, and date match his 1899 description of
Caltha chionophila.
Geographic distribution:—Subalpine and alpine habitats in the Rockies and Basin and
Range: central Idaho, western Montana, southeastern Oregon, northeastern Nevada, Wyoming,
Colorado, Utah, Arizona and New Mexico (Wefferling et al. 2017, Giblin et al. in press).
Additional specimens examined (*pollen sampled for SEM, n = 14; † ploidy level
estimated via FCM or chromosome count [Wefferling et al. 2017], n = 8):—U.S.A.: Arizona:
Apache-Sitgreaves National Forest, Clear Cut Spring, White Mtns., 33.903383, -109.484811,
elevation 2835 m, Rink, G. and L. Stevens s.n.*, 27-Jun-2010 (ASC). Colorado: Wet Mountains,
Wet Mountain Valley, Sangre Cristo Range and vicinity, Sawatch Range, San Isabel NF,
Monarch Park, ca. 1.8 road mi. on County Road 231, 38.5153, -106.3234, elevation 3200 m,
Hall, C., B. Jacobs, and A. Morgan 1565*†, 29-Jun-1998 (NYBG). Colorado: Roosevelt NF,
Niwot Ridge at the LTER site, 35 km W of Boulder, 40.0757, -105.593, elevation 3000 m,
Schuettpelz, E.J. 001, 13-Jul-2000 (UWM). Colorado: Roosevelt NF, Ward Quad., Shoreline of
Brainard Lake, 40.0769, -105.57571, elevation 3158 m, Majack, M. 2-1*†, 1-Jan-2014 (UWM).
Idaho: Challis NF, Merriam Lake Basin, series of hummocky alpine marshy meadows at the base
of rock talus along the edge of Merriam Lake, 44.12, -113.75, elevation 2930 m, Bursik, R.
1240*, 7-Jul-1988 (ID). Idaho: Salmon NF, lakeside meadow of upper lake at head of Middle
Fork Little Timber Creek, 44.549619, -113.531627, elevation 2801 m, Brunsfeld, S.J. 1737*, 14-
Jul-1981 (ID). Idaho: Head of Big Eightmile Creek, 44.574, -113.602, elevation 2800 m,
Henderson, D. 3365, 14-Jul-1976 (SRP). Idaho: Spring and stringer at head of Jordan Creek
southwest of Jordan Peak, ca 18 airmiles NNE of Stanley, 44.466, -114.776, elevation 2780 m,
Errter, B., B. Corbin, C. Scott, J. Irwin, and W. Irwin 20192*, 17-Jul-2010 (SRP). Idaho:
Sawtooth NRA, S end Decker Flat, S of junction of FR 315 and FR 037, 44.019044, -
114.858893, elevation 2090 m, Wefferling, K.M. 276h*, 23-Jul-2014 (UWM). Idaho: Summit
Creek, Pioneer Mtns., Challis NF, at pass between Summit Creek and R. Fork Kane Creek,
43.80247, -114.203175, elevation 2890 m, Wefferling, K.M. 299 lepto†, 1-Jan-2014 (UWM).
Idaho: Summit Creek, Pioneer Mtns., Challis NF, at pass between Summit Creek and R. Fork
Kane Creek, 43.801472, -114.201444, elevation 2900 m, Wefferling, K.M. 291h, 24-Jul-2014
(UWM). Idaho: E side Sawtooth Range, Sawtooth NRA, Elk Meadows (NE side, edge of
meadow/forest), 44.266698, -115.094952, elevation 2060 m, Wefferling, K.M. 212 lepto*†, 1-
17
Jan-2014 (UWM). Idaho: White Cloud Peaks, Sawtooth NRA, 4th of July Lake Creek,
44.050251, -114.649835, elevation 2724 m, Wefferling, K.M. 242h, 23-Jul-2014 (UWM). Idaho:
White Cloud Peaks, Sawtooth NRA, Inlet to 4th of July Lake, 44.044306, -114.631222, elevation
2860 m, Wefferling, K.M. 258h, 23-Jul-2014 (UWM). Idaho: Cirque at head of Rock Creek,
below N face of Borah peak, Lost River Range, Challis NF, ca 17 miles NW of Mackay,
44.13717, -113.801859, elevation 3018 m, Moseley, B. 1211*, 23-Aug-1987 (ID). Idaho: E side
Sawtooth Range, Sawtooth NRA, Elk Meadows (NE side, edge of meadow/forest), 44.266698, -
115.094952, elevation 2060 m, Wefferling, K.M. 212lh, 22-Jul-2014 (UWM). Idaho: Kane Creek
cirque, ca. 0.5 mi. E of Kane Lake, Pioneer Mtns., Challis NF, ca. 13 mi. NE of Ketchum,
43.786011, -114.14451, elevation 3170 m, Moseley. B. 1181, 22-Aug-1987 (ID). Montana: La
Marche Lake Meadows, 46.012634, -113.300911, elevation 2522 m, Lackschewitz, K.H. 3880,
17-Jul-1972 (MONTU). Montana: Scapegoat Mtn., Continental Divide (General Summit, east
side), 24 mi SW of Augusta, 47.31, -112.8, elevation 2590 m, McDonald, C.H. 2091, 14-Jul-
1966 (MONTU). Montana: Camp Pass in Lincoln-Scapegoat area above Camp Lake, 47.185, -
113.051, elevation 1830 m, Mooar, M. 9527, 16-Jul-1968 (MONTU). Montana: Beartooth Mtns,
Stillwater Plateau 5 mi. N of Mt. Wood, Custer NF, 7 mi. S of Nye, 45.331487, -109.825172,
elevation 3109 m, Evert, E. s.n., 26-Jul-1992 (RM). Montana: Yellow-flowered plants abundant
in wet soil along a small stream at the head of Nicholia Creek, 44.384363, -112.856781,
elevation 2700 m, Lesica, P. s.n., 2-Jul-2014 (UWM). New Mexico: Sangre Cristo Mtns., west
slope of Wheeler Peak, above Williams Lake, 36.559, -105.4177, elevation 3658 m, Holmgren,
N.H. and P.K. Holmgren 7330*, 21-Aug-1973 (NYBG). New Mexico: Sante Fé Basin, Ski Slide
Area, 35.8, -105.7, elevation 3170 m, Throne, A.L. 12538, 26-Jul-1960 (UWM). Nevada: Ruby
Mountains, Island Lake on west side of Lamoille Canyon, 40.6108, -115.3828, elevation 2987 m,
Tiehm, A. and M. Williams 9675*, 14-Jun-1985 (ID). Nevada: Ruby Mountains, Lamoille
Canyon, 40.648, -115.385, elevation 2900 m, Throne, A.L. 13776, 25-Jul-1949 (UWM). Oregon:
in seep by small creek half way down from Steens summit ridge to Wildhorse Lake by trail,
42.63373, -118.582351, elevation 2670 m, Mansfield, D. 94-50, 26-Jul-1994 (CIC). Oregon:
Steens Mountain, 42.674949, -118.588356, elevation 2700 m, Johanson, J. 07-10*, 21-Jul-2007
(WTU). Utah: Uinta Mtns., along State Route 150, 2.4km (1.5mi) N of the Duchesne County
line, 40.743, -110.871, elevation 3030 m, Holmgren, N.H. and P.K. Holmgren 15424*†, 10-Jun-
2006 (NYBG). Utah: Wasatch Plateau, Manti Creek drainage in the South Fork of Manti
18
Canyon, 2.6 km road distance below Skyline Drive, 39.219167, -111.502222, elevation 3040 m,
Holmgren, N.H. and P.K. Holmgren 16196†, 15-Aug-2009 (NYBG). Utah: North Slope Uinta
Mountains, Wasatch NF, Gunsight Pass, ca. 33 air mi. S of Mountain View, 40.811651, -
110.362809, elevation 3625 m, Refsdal, C.H. 6799 with L. Refsdal, 8-Aug-1995 (RM).
Wyoming: West Slope Wind River Range: vicinity of Meeks Lake and Iron Creek Meadows, ca.
1.25 air mi N of Big Sandy Campground, ca. 30 air mi ESE of Pinedale, 42.699878, -
109.266609, elevation 2805 m, Fertig, W. 7481, 11-Jun-1991 (RM). Wyoming: Medicine Bow
Mountains, N of Headquarters Park off FR 200, 1 air mile N of Wyo Hwy 130, ca. 24 air miles
SE of Saratoga, 41.3386, -106.3726, elevation 3050 m, Lukas, L.E. 2286†, 13-Jul-2007
(USFS/RM). Wyoming: Medicine Bow NF, alpine meadows above Lake Marie, 41.343889, -
106.333667, elevation 3496 m, Wefferling, K.M. 19*†, 21-Jun-2012 (UWM).
Caltha leptosepala Candolle (1818: 310).
Type:—U.S.A. Alaska: Prince William Sound, 1792, Menzies s.n. (holotype BM!). Fig. 2.
Caltha leptosepala var. rotundifolia Huth (1892: 68). C. rotundifolia (Huth) Greene (1899: 80).
C. biflora var. rotundifolia (Huth) Hitchcock (1964: 335).
Type:— U.S.A. Rocky Mountains, 1872, A. Gray s.n. (isotype GH!).
Caltha chelidonii Greene (1899: 78).
Type:—CANADA. Alberta: Yellowhead Pass, Rocky Mountains on Alberta-British Columbia
boundary, Jasper National Park, 13 July 1898, Spreadborough 19250 (holotype CAN!).
Caltha macounii Greene (1899: 77). C. leptosepala var. macounii (Greene) Davis (1900: 16).
Type:— CANADA. British Columbia: Mount Queest, 28 July 1889, Macoun 1255 (syntype
US!).
Caltha leptosepala var. sulfurea Hitchcock (1964: 337).
Type:—U.S.A. Idaho: Custer County, Mount Borah, Rock Creek, 12 August 1944, Hitchcock &
Muhlick 10942 (isotypes US! WTU!).
Morphology and cytology:—Leaf blades 1–1.4(–1.7) × as long as wide, +/- sagittate to
cordate to obovate, auricles variable (diplophyllous or not, sinus open or closed), margins crenate
or dentate (more so near base, tending toward entire near the apex) (Fig. 2); filaments broadly
filiform to flattened, 0.2–0.7(–0.9) mm wide (Giblin et al. in press), slightly narrower to as wide
19
as anthers (Figs. 2, 11C); pollen (3–)4–8(–12) colpate (rarely -porate) or misshapen (Fig. 9);
follicles with curved stylar beak ~0.1 mm in length (Giblin et al. in press); allododecaploid (2n =
12x = 96, Wefferling et al. 2015; holoploid genome size ca. 18–25.5 pg/2C, Wefferling et al.
2017), rarely allononaploid (2n = 9x = 72, Wefferling et al. 2015; holoploid genome size ca.
13.8–16.8 pg/2C, Wefferling et al. 2017); apparently formed (bidirectionally) through
hybridization of Caltha biflora and C. chionophila.
Taxonomic notes:—Candolle (1818) designated as holotype the Menzies collections
from Prince William Sound. This part of the complex’s range is well outside the ranges of
Caltha biflora or C. chionophila; all specimens that we have seen from north of Prince of Wales
Island, Alaska (near Haida Gwaii, British Columbia) are apparently allododecaploids. In 1970, P.
G. Smit annotated the holotype (Fig. 2), indicating that the pollen was colpate.
Geographic distribution:—Montane to subalpine in central Alaska and Yukon south
through British Columbia to Olympic Mountain Range and Cascades of Washington and
Oregon, Sierran California, east to western Alberta, central Idaho, northeastern and southeastern
Oregon, and western Montana (Wefferling et al. 2017, Giblin et al. in press).
Additional specimens examined (*pollen sampled for SEM; n = 15; † ploidy level
estimated via FCM or chromosome count [Wefferling et al. 2017], n = 33):—CANADA:
Alberta: Sunshine Lodge, 51.131817, -115.565006, elevation 2286 m, Ogilvie, R.T. s.n., 19-Jul-
1959 (V). British Columbia: Mount Revelstoke National Park: Along trail to Millar Lake,
51.066667, -118.1, elevation 1890 m, Soper, J.H. 12,754 with M.J. Shchepanek, 31-Jul-1970
(RBCM). British Columbia: Pan Creek, Ilgachuz Range, upper Pan Creek valley, 52.725, -
125.225, elevation 1737 m, Hebda, R. 87-32, 27-Jul-1987 (RBCM). British Columbia:
Tweedsmuir Provincial Park, Tweedsmuir Peak, 53.656667, -126.501667, elevation 1717 m,
Hebda, R and K. Marr KM4086, 9-Aug-2002 (RBCM). British Columbia: Green Mountain,
Vancouver Island, 49.051134, -124.34007, elevation 1300 m, Marr, K., R. Hebda, L. Kennedy,
and White 02-48*, 5-Jul-2002 (RBCM). British Columbia: Muskwa Ranges, Needham Creek,
headwaters of creek, 56.4083, -123.4989, elevation 1750 m, Hebda, R., K. Marr, and W.
MacKenzie KM4777†, 6-Aug-2003 (RBCM). British Columbia: Swannel Ranges, Chase
Mountain, in valley bottom downstream from small lake 1 km S of summit, 56.558569, -
125.255536, elevation 1700 m, Hebda, R. and R. Forsyth KM4857m†, 7-Aug-2003 (RBCM).
British Columbia: Rocky Mtn. foothills, Quintette Mtn., Roman Mtn. summit, 5 km WNW of
20
Quintette Mtn, S of Babcock Creek, 54.888, -120.941, elevation 1650 m, Hebda, R., K.L. Marr,
and R. Forsyth KM5613†, 11-Aug-2003 (RBCM). British Columbia: Skeena Mountains,
Klappan Mountain, small mountain 2 km W of Klappan Mtn. summit near road, 57.238333, -
128.910001, elevation 1751 m, Marr, K.L., R. Hebda, and S. Berger 05-0652†, 6-May-2006
(RBCM). British Columbia: Coast Mountains, Hanna Ridge, S end Hanna Ridge, N of Meziadin
Lake, 56.223056, -129.436944, elevation 1752 m, Marr., K.L., R. Hebda, and W. MacKenzie 06-
0013*†, 13-Jul-2006 (RBCM). British Columbia: Coast Mountains, Copper Mtn., mountain
summit east of town of Terrace, 54.5117, -128.4536, elevation 1199 m, Marr., K.L., R. Hebda,
and W. MacKenzie 06-0523†, 18-Jul-2006 (RBCM). British Columbia: Stikine Ranges, Blue
Sheep Lake, south of Little Blue Sheep Lake, 58.717113, -128.258892, elevation 1685 m, Marr,
K.L., R. Hebda, and W MacKenzie 07-1648†, 12-Aug-2007 (RBCM). U.S.A.: Alaska: Hatcher
Pass, 61.783331, -149.266661, elevation 1200 m, Hawkins, L.L. s.n., 26-Jun-1976 (ALA).
Alaska: Denali National Park and Preserve, Wildhorse Creek Valley, 5.9 km SW of toe of
Kanikula Glacier, 7.3 km upstream of confluence of Wildhorse Creek and Tokositna River,
Alaska Range, AK, 62.653, -150.961, elevation 838 m, Larsen, A. and M.B. Cook 01-0092†, 2-
Jul-2001 (ALA). Alaska: Fairweather Range, mountain E of Brady Glacier, N of Taylor Bay,
58.463567, -136.470102, elevation 360 m, Stratford, L. and J. Grunblatt LS01-7, 20-Jul-2001
(ALA). Alaska: Guyot Hills, 60.168667, -141.405991, elevation 933 m, Loomis, P. and A.
Larsen 1256†, 27-Jul-2003 (ALA). Alaska: Tongass NF, Alexander Archipelago, Baranof
Island, near lake above Lake Diana, 56.8908, -135.045301, elevation 655 m, Calhoun, K. and B.
Krieckhaus 37, 1-Jul-2004 (ALA). Alaska: Afognak Island, Kodiak National Wildlife Refuge,
Blue Fox Bay, head of bay in vicinity of public use cabin, 58.413056, -152.696111, elevation 15
m, Parker, C.L. and S. Studebaker 17293†, 2-Jul-2008 (ALA). Alaska: Kodiak Island, SE region
of island, Big Creek Valley, 11 km NNE of Old Harbor and E side of valley, 57.29537, -
153.29202, elevation 685 m, Parker, C.L. 17964, 5-Aug-2010 (ALA). Alaska: Steep Creek
Basin, near Juneau, 58.393812, -134.515948, elevation 585 m, Taylor, S.G. 55, 1-Jul-1968
(ALA). Alaska: Thompson Pass, Richardson Highway, 61.133256, -145.750058, elevation 900
m, Parker, C.L. 2368, 21-Jul-1990 (RSA). Alaska: St. Elias Mtns, Chilkat R. headwaters, 5km S
of Klukwah Mt., 59.53, -135.83, elevation 910 m, Parker, C.L., A.R. Batten, and D. Blank
9523*, 16-Jul-2000 (UWM). Alaska: Chugach NF, Lost Lake trail from Seward side, 60.202889,
-149.427667, elevation 390 m, Wefferling, K.M. 64†, 15-Jun-2013 (UWM). Alaska: Chugach
21
State Park, Williwaw Lakes trail, 61.112389, -149.660278, elevation 620 m, Wefferling, K.M.
65†, 17-Jun-2013 (UWM). Alaska: Chugach NF, Falls Creek Trail, 61.003389, -149.572778,
elevation 629 m, Wefferling, K.M. 63*†, 13-Jun-2013 (UWM). California: Lily Lake, W side of
the Warner Mountains, Sierran montane forest, Fort Bidwell 15' Quad., 41.976368, -120.202733,
elevation 2130 m, Bartolomew, B. 4327 and B. Anderson*, 24-Jun-1988 (NYBG). Idaho: Bear
Valley Road (FS579), Boise National Forest, old burned forest along small creek south of Bruce
Meadows, 44.367061, -115.27463, elevation 2184 m, Smith, J.F. s.n.†, 27-Apr-2015 (SRP).
Idaho: Salmon-Challis NF, Salmon River Mtns. Beaver Creek drainage, ca. 10 air mi. NE from
Cape Horn Guard Station, 44.446347, -115.035753, elevation 2225 m, Tanaka, T. and C.
Richardson 82*†, 10-Jul-1996 (ID). Idaho: Moses Butte area, near intersection of FR 457, FR
220 and FR 363, in meadow just to south and west of road, 47.0091, -115.8415, elevation 1875
m, Parks, M., L. Stratford, and R. McNeill 620*†, 15-Jul-2004 (ID). Idaho: Salmon River Mtns,
Boise NF, Lowman RD, Canyon Creek at pullout along ID 21, 44.288372, -115.226992,
elevation 2060 m, Wefferling, K.M. 180h†, 21-Jul-2014 (UWM). Idaho: Boise National Forest,
Summit Lake, 44.649947, -115.583245, elevation 2280 m, Smith, J.F. 2962*, 26-Jun-1994
(SRP). Oregon: Mt. Hood Wilderness, NW end of Elk Meadows, 45.34525, -121.620528,
elevation 1570 m, Wefferling, K.M. CR5*†, 1-Jan-2015 (UWM). Montana: Lost Trail Bog, Lost
Trail Pass, near continental divide, 45.693303, -113.953169, elevation 2146 m, Mantas, M. 585*,
5-Jul-1992 (ID). Oregon: Wallowa Whitman National Forest, Anthony Lakes area, streamside
near trail from Anthony Lake to Hoffer Lake, 44.954083, -118.233472, elevation 2212 m,
Wefferling, K.M. and L. Woo 62*†, 12-Aug-2012 (UWM). Oregon: Jefferson Park in the
Cascade Mountains on the border of Marion county and Jefferson County, 44.712148, -
121.797064, elevation 1790 m, Anderson, J. JNS2-1*†, 1-Jan-2015 (UWM). Oregon: Mt. Hood
Wilderness, NW end of Elk Meadows, 45.34525, -121.620528, elevation 1570 m, Wefferling,
K.M. 67*†, 1-Jul-2013 (UWM). Oregon: Mt. Hood Wilderness, NW end of Elk Meadows,
45.34525, -121.620528, elevation 1570 m, Wefferling, K.M. 70†, 1-Jul-2013 (UWM). Oregon:
Fremont NF, three miles N of Dead Horse Lake, at junction of Bald Butte Road #450 and Road
#3411, 42.601, -120.781, elevation 2073 m, Legler, B., S. Gage, W. Gibble, R. Goff, S. Birks, and
K. Davis 1890†, 2-Jul-2004 (WTU). Oregon: McCoy Creek on Steens Mountain, 42.725724, -
118.597244, elevation 2344 m, Lowry II, P.P. 469*, 7-Jul-1976 (OSC). Washington: Mount
Baker Nat'l Forest, Harts Pass, 48.72, -120.67, elevation 2000 m, Muenscher, W.C. 10065, 23-
22
Jun-1939 (UWM). Washington: Gifford-Pinchot NF, Takhlakh Meadow, 46.269707, -
121.588677, elevation 1400 m, Wefferling, K.M. 60†, 1-Aug-2012 (UWM). Washington: Mount
Rainier National Park. Spray Park, 46.944372, -121.750034, elevation 1892 m, Rochefort, R.
SP4†, 22-Jul-2015 (UWM). Washington: Wenatchee Mountains, 47.42, -120.94, elevation 1187
m, collector unknown [Washington Native Plant Society WNPS-2], 1-Jan-2014 (UWM).
Washington: Gifford-Pinchot NF, Takhlakh Meadow, 46.268447, -121.586248, elevation 1400
m, Wefferling, K.M. and L. Wefferling 87s†, 15-Jul-2014 (UWM). Washington: Gifford-Pinchot
NF, Chain of Lakes, 46.29321, -121.596996, elevation 1340 m, Wefferling, K.M. and L.
Wefferling 118bh†, 16-Jul-2014 (UWM). Washington: Horseshoe Basin (HB00711), Arnold
Peak, SW face, on bench with steep rocky springs, and open coniferous forest, 48.985377, -
119.924039, elevation 2317 m, Wooten, G. #GW01566, 28-Jun-1992 (WTU). Washington: North
Cascades National Park, Pelton Basin east of Cascade Pass, 48.458683, -121.0433, elevation
1407 m, Knoke, D. 267†, 10-Aug-2002 (WTU). Washington: Okanogan NF, Washington Pass
Viewpoint turnoff, wetland at Hwy. 20 entrance, 48.525081, -120.658038, elevation 1649 m,
Yen, A. C. 03-066 with R. Robohm, S. Bagshaw, L. Van Volkenburgh, and T. Ohlson, 25-Jul-
2003 (WTU). Washington: Okanogan NF, Meadows Campground, meadows S of Hart's Pass,
48.711, -120.676, elevation 1903 m, Rodman, S., D. Tank, C. Spurgeon, and K. Ardern 909*, 25-
Jul-2003 (WTU). Washington: Gifford-Pinchot NF, Takhlakh Meadow, 46.270167, -
121.588028, elevation 1402 m, Wefferling, K.M. 56†, 1-Aug-2012 (UWM). 9x: U.S.A., Idaho: E
side Sawtooth Range, Sawtooth NRA, Elk Meadows (NE side, edge of meadow/forest),
44.266698, -115.094952, elevation 2060 m, Wefferling, K.M. 212rh†, 22-Jul-2014 (UWM).
Key to Caltha leptosepala polyploid complex
1. Leaves as broad as leaf length or broader; filaments filiform (i.e., narrower than anthers) and generally longer
than carpels; in Coastal Ranges (British Columbia and Alaska), Cascades (Washington, Oregon, California),
Klamath - Siskiyous (Oregon, California), or Sierra Nevada (California, western Nevada) … C. biflora
- Leaves longer than broad …………………………………………………………………………………. 2
2. Filaments broader and generally shorter than carpels; in U.S.A. Rockies or Basin and Range (Steens, Ruby Mtns.)
…………………………………………………………………………. C. chionophila
- Filaments no broader than anthers (broadly filiform to about as wide as anthers) and generally longer than carpels;
in all regions except southern Rockies ……………………………………………… C. leptosepala
23
Acknowledgements
The authors thank A.B. Kirkpatrick (Ohio State University) for help with scanning electron
microscopy and for helpful comments on an earlier version of this manuscript; H.A. Owen
(UW-Milwaukee, UW-M) for help with light and scanning electron microscopy; M. Barkworth
(Utah State University), B. Hellenthal (Museum of Biodiversity and Greene-Nieuwland
Herbarium), B. Legler (University of Washington, Burke Museum), and C. Tyrell (Milwaukee
Public Museum) for helpful taxonomic advice; M. Bivin (United States Forest Service, USFS),
T. Humphries (USFS), G. Hunter (Olympic National Park), D. Keeler (USFS), P. Lesica
(University of Montana), M. Majack (Colorado Natural Areas Program), K. Mohatt (USFS), R.
Rochefort (USFS), J. Shrum (USFS), J. Smith (Boise State University), B. Wefferling and L.
Woo for plant collections; L. Wefferling for plant and seed collections and photographs; and P.
Engevold (UW-M) for help with seed germination and propagation. This work would not have
been possible without the work of all the plant collectors and herbarium curators (of ALA, ASC,
BM, CAN, CIC, CSU, DAV, GH, HSU, ID, K, MO, MONTU, NDG, NYBG, OSC, RBCM,
RM, RSA, SRP, US, USFS/RM, UWM, V, and WTU) who made specimens (physical or
digital) available for study. Fieldwork was supported through an International Association for
Plant Taxonomy Research Improvement Grant and by the UW-M Biological Sciences
Department (both to K.M.W.). Finally, the authors thank two anonymous reviewers for helpful
comments, and Q.-e. Yang (editor) for careful editing and helpful input during the submission
process.
References
Abrams, L.R. (1944) Illustrated Flora of the Pacific States, vol. II. Standford University Press, Stanford, 174 pp.
Alfaro, M.E., Zoller, S. & Lutzoni, F. (2003) Bayes or bootstrap? A simulation study comparing the performance of
Bayesian Markov Chain Monte Carlo sampling and bootstrapping in assessing phylogenetic confidence.
Molecular Biology and Evolution 20: 255266. https://doi.org/10.1093/molbev/msg028
Azuma, H., Figlar, R.B., Tredici, P.D., Camelbeke, K., Palmarola-Bejerano, A. & Romanov, M.S. (2011)
Intraspecific sequence variation of cpDNA shows two distinct groups within Magnolia virginiana L. of
Eastern North America and Cuba. Castanea 76: 118123. http://dx.doi.org/10.2179/10-018.1
24
Candolle, A.P. de (1818) [1817] Regni Vegetabilis Systema Naturale, sive Ordines, Genera et Species Plantarum
secundum Methodi Naturalis Normas Digestarum et Descriptarum, vol. 1. Sumptibus sociorum Treuttel et
Würtz, Paris, etc., 564 pp.
Cheng, J. & Xie, L. (2014) Molecular phylogeny and historical biogeography of Caltha (Ranunculaceae) based on
analyses of multiple nuclear and plastid sequences. Journal of Systematics and Evolution 52: 5167.
http://dx.doi.org/10.1111/jse.12051
Davis, K.C. (1900) A Taxonomic Study of North American Ranunculaceae: a Treatment of the Members of this
Natural Order, Whether Found in Gardens or Native in North America (including Mexico). Cornell
University, Ithaca, 173 pp.
Drummond, A. J. & Bouckaret, R. R. (2015) Bayesian Evolutionary Analysis With BEAST. Cambridge University
Press, Cambridge, 249 pp.
Ford, B.A. (1997) Caltha. In: Flora of North America Editorial Committee (Eds.), Flora of North America North of
Mexico, vol. 3: Magnoliophyta: Magnoliidae and Hamamelidae. Oxford University Press, New York, pp.
185, 187189.
Giblin, D.E., Legler, B.S., Zika, P.F. & Olmstead, R.G. (in press) Flora of the Pacific Northwest: An Illustrated
Manual, 2nd Edition. University of Washington Press, Seattle.
Greene, E.L. (1899) Segregates of Caltha leptosepala DC. Pittonia 4: 7381.
Hardig, T.M., Soltis, P.M. & Soltis, D.E. (2000) Diversification of the North American shrub genus Ceanothus
(Rhamnaceae): Conflicting phylogenies from nuclear ribosomal DNA and chloroplast DNA. American
Journal of Botany 87: 108123. http://www.jstor.org/stable/2656689
Hillis, D.M. & Bull, J. J. (1993) An empirical test of bootstrapping as a method for assessing confidence in
phylogenetic analysis. Systematic Biology 42: 182192. https://doi.org/10.1093/sysbio/42.2.182
Hitchcock, C.L. & Cronquist, A. (1973) Flora of the Pacific Northwest. University of Washington Press, Seattle,
730 pp.
Hitchcock, C.L., Cronquist, A., Ownbey, M. & Thompson, J.W. (1964) Vascular Plants of the Pacific Northwest.
Part 2: Salicaceae to Saxifragaceae. University of Washington Press, Seattle, 597 pp.
Hoffmann, M.H. (1999) Biogeographical and evolutionary patterns in the genus Caltha L. (Ranunculaceae).
Botanische Jahrbücher für Systematik, Pflanzengeschichte und Pflanzengeographie 121: 403421.
Huelsenbeck, J.P. & Ronquist, F. (2001) MrBayes: Bayesian inference of phylogenetic trees. Bioinformatics 17:
754755. https://doi.org/10.1093/bioinformatics/17.8.754
Huelsenbeck, J.P., Larget, B. & Alfaro, M.E. (2004) Bayesian phylogenetic model selection using reversible jump
Markov chain Monte Carlo. Molecular Biology and Evolution 21: 11231133.
https://doi.org/10.1093/molbev/msh123
Huth, E. (1892) Monographie der Gattung Caltha. Helios 9: 55103.
Kearse, M., Moir, R., Wilson, A., Stones-Havas, S., Cheung, M., Sturrock, S., Buxton, S., Cooper, A., Markowitz,
S., Duran, C., Thierer, T., Ashton, B., Meintjes, P. & Drummond, A. (2012) Geneious Basic: An
25
integrated and extendable desktop software platform for the organization and analysis of sequence data.
Bioinformatics 28: 16471649. http://dx.doi.org/10.1093/bioinformatics/bts199
Lanfear, R., Calcott, B., Ho, S.Y.W.,& Guindon, S. (2012) PartitionFinder: Combined selection of partitioning
schemes and substitution models for phylogenetic analyses. Molecular Biology and Evolution 29: 1695
1701. https://doi.org/10.1093/molbev/mss020
Lawson, G. (1884) Revision of the Canadian Ranunculaceae. Transactions of the Royal Society of Canada 2: 15
90. http://dx.doi.org/10.5962/bhl.title.39874
Linnaeus, C. (1753) Species Plantarum, vols. 12. Impensis Laurentii Salvii, Holmiæ [Stockholm], 1200 pp.
Liu, B., Hao, Y.-Q., Yang, Y., Sun, T.-T., Lai, Y.-J., Zhang, T. & He, F. (2016) A new species of the
Ranunculaceae from southwestern China based on morphological and molecular data. Phytotaxa 283: 43
53. https://dx.doi.org/10.11646/phytotaxa.283.1.2
Morris, M.I. (1973) A biosystematic analysis of the Caltha leptosepala (Ranunculaceae) complex in the Rocky
Mountains. III. Variability in seed and gross morphological characteristics. Canadian Journal of Botany
51: 22592267.
Nickrent, D.L., Schuette, K.P. & Starr, E.M. (1994) A molecular phylogeny of Arceuthobium (Viscaceae) based on
nuclear ribosomal DNA internal transcribed spacer sequences. American Journal of Botany 81: 1149
1160. http://www.jstor.org/stable/2445477
Rambaut, A., Suchard, M.A., Xie, D. & Drummond, A.J. (2014) Tracer v1.6. Available from
http://beast.bio.ed.ac.uk/Tracer.
Rauscher, J.T., Doyle, J.J. & Brown, A.H.D. (2002) Internal transcribed spacer repeat-specific primers and the
analysis of hybridization in the Glycine tomentella (Leguminosae) polyploid complex. Molecular Ecology
11: 26912702. http://dx.doi.org/10.1046/j.1365-294X.2002.01640.x
Rautenberg, A., Hathaway, L., Oxelman, B. & Prentice, H.C. (2010) Geographic and phylogenetic patterns in
Silene section Melandrium (Caryophyllaceae) as inferred from chloroplast and nuclear DNA sequences.
Molecular Phylogenetics and Evolution 57: 978991. http://dx.doi.org/10.1016/j.ympev.2010.08.003
Ronquist, F. & Huelsenbeck, J.P. (2003) MrBayes 3: Bayesian phylogenetic inference under mixed models.
Bioinformatics 19: 15721574. http://dx.doi.org/10.1093/bioinformatics/btg180
Ronquist, F., Teslenko, M., van der Mark, P., Ayres, D.L., Darling, A., Hohna, S., Larget, B., Liu, L., Suchard,
M.A. & Huelsenbeck, J.P. (2012) MrBayes 3.2: Efficient Bayesian phylogenetic inference and model
choice across a large model space. Systematic Biology 61: 53942.
Rydberg, P.A. (1900) Catalogue of the Flora of Montana and the Yellowstone National Park. Memoirs of the New
York Botanical Garden 1: 1492.
Schuettpelz, E. & Hoot, S.B. (2004) Phylogeny and biogeography of Caltha (Ranunculaceae) based on chloroplast
and nuclear DNA sequences. American Journal of Botany 91: 247253.
http://dx.doi.org/10.3732/ajb.91.2.247
26
Shaw, J., Lickey, E.B., Schilling, E.E. & Small, R.L. (2007) Comparison of whole chloroplast genome sequences to
choose noncoding regions for phylogenetic studies in angiosperms: the tortoise and the hare III. American
Journal of Botany 94: 275288. http://dx.doi.org/10.3732/ajb.94.3.275
Smit, P.G. (1973) A revision of Caltha (Ranunculaceae). Blumea 21: 119150.
Smit, P.G. & Punt, W. (1969) Taxonomy and pollen morphology of the Caltha leptosepala complex. Proceedings
of the Koninklijke Nederlandse Akademie van Wetenschappen , Ser. C. 72: 1627.
Stamatakis, A. (2006) RAxML-VI-HPC: Maximum likelihood-based phylogenetic analyses with thousands of taxa
and mixed models. Bioinformatics 22: 26882690. http://dx.doi.org/10.1093/bioinformatics/btl446.
Stamatakis, A. (2016) The RAxML v8.2.X Manual. Available at https://sco.h-
its.org/exelixis/resource/download/NewManual.pdf
Swofford, D.L. (2002) PAUP* Phylogenetic analysis using parsimony (*and other methods), version 4b10. Sinauer
Associates, Sunderland.
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: 11051109.
Wefferling, K.M., Owen, H.A. & Hoot, S.B. (2015) Caltha leptosepala DC. In: Marhold, K. & Breitwieser, I.
(Eds.), IAPT/IOPB chromosome data 20. Taxon 64: 1350, E36E39. https://doi.org/10.12705/646.42
Wefferling, K.M., Castro, S., Loureiro, J., Castro, M., Tavares, D. & Hoot, S.B. (2017) Cytogeography of the
subalpine marshmarigold polyploid complex (Caltha leptosepala s.l.). American Journal of Botany 104:
271285.
Yoo, K.-O., Lowry II, P.P. & Wen, J. (2002) Discordance of chloroplast and nuclear ribosomal DNA data in
Osmorhiza (Apiaceae). American Journal of Botany 89: 966971. https://doi.org/10.3732/ajb.89.6.966
27
Table 1. Primers used in this study. Nucleotide sequences read 5’ to 3’. Polymorphic nucleotide sites are designated using
IUPAC ambiguity codes. * ribotype specific primers, designed to amplify a single subgenome/ribotype.
Nuclear ribosomal internal transcribed spacer (ITS)
1830F (Nickrent et al. 1994) forward
AACAAGGTTTCCGTAGGTGA
25R (Nickrent et al. 1994) reverse
TATGCTTAAAYTCAGCGGGT
364F (Hoot lab primer) internal forward
ATCGATGAAGAACGTAGCG
390R (Hoot lab primer) internal reverse
CAATTCACACCAAGTATCGC
*Wr13 (this study) ribotype specific reverse
CTGGGGTCGCAGG
*Er14 (this study) ribotype specific reverse
CCTGGGGTCGCAAT
*Ef19_2 (this study) ribotype specific forward
TTGTTGGGATGTGGAATCT
*Wf21_1 (this study) ribotype specific forward
GCAAGATAGGGTACAACAAGC
rpL32trnLUAG
rpL32FSh (Shaw et al. 2007) forward
CAGTTCCAAAAAAACGTACTTC
rpL32F (this study) forward
CRGTTCCKAAAAAACGTACTTC
trnLuag (Shaw et al. 2007) reverse
CTGCTTCCTAAGAGCAGCGT
rpL32R2 (this study) internal reverse
TCGAGGTTGGTATTAAAATTGG
trnLB (this study) internal forward
TTGAACTGTAAGATCGATCAAG
trnLUAAtrnFGAA
trnLF 1 (Azuma et al. 2011) forward
CGTAGCGTCTACCGATTTCG
trnLF A50272 (Tablerlet et al. 1991) reverse
ATTTGAACTGGTGACACGAG
trnLF B49873 (Tablerlet et al. 1991) internal forward
GGTTCAAGTCCCTCTATCCC
trnLF A49855 (Tablerlet et al. 1991) internal reverse
GGGGATAGAGGGACTTGAAC
Table 2. Statistics for gene regions used in molecular dataset.
Gene region
Aligned
length
Taxa/homeologues
% taxon
coverage
ITS
646
15
100
trnL-trnF
837
15
100
rpL32-trnL
952
8
53.3
28
Table 3. Key morphological characters for discriminating among Caltha biflora, C. chionophila, and C. leptosepala.
Taxon
ploidy
(inferred
or
measured)
leaf
length:width
leaf auricles
number of
flowers/stem
filament:
anther width
pollen
Caltha
biflora
2n = 6x =
48
0.7–1
large, often
overlapping
or
diplophyllous
1–3
< 1
9–12-porate (rarely -colpate
with short colpi)
Caltha
chionophila
2n = 6x =
48
1.2–3
small, rarely
overlapping,
often
diplophyllous
1–3
> 1
usually 3-colpate; rarely
malformed
Caltha
leptosepala
2n = 12x =
96 (rarely
9x = 72)
1–1.7
variable;
sometimes
overlapping,
sometimes
diplophyllous
1–3
1
(3-)48(12)-colpate (rarely -
porate)
Appendix 1. List of specimens used in the phylogeny: species name, inferred ploidy level (for Caltha leptosepala sensu lato
specimens only; † indicates specimens with flow cytometry genome size estimates and/or chromosome counts), voucher
information (for new sequences only), GenBank accession number. Gene regions are listed in the following order: internal
transcribed spacer region (ITS; two GenBank numbers given for allododecaploids: biflora/chionophila ribotypes), trnL–F,
rpL32trnL. indicates missing sequence data. Numbers with prefix AY- are from Schuettpelz and Hoot (2004).
Ingroup: Caltha leptosepala DC. (12x) (1), St. Elias Mountains, Chilkat River headwaters, Alaska, U.S.A., Parker, Batten, &
Blank 9523 (UWM63842): MF168897/MF168847, MF168834, MF168804. Caltha leptosepala DC. (12x) (2), Okanogan
National Forest, Washington, U.S.A., Rodman, Tank, Spurgeon, & Ardern 909 (WTU362988): MF168897/identical to
MF168847, MF168838, identical to MF168804. Caltha leptosepala DC. (12x†) (3), Moses Butte area, Idaho, U.S.A., Parks,
Stratford, & McNeill 620 (ID039909): MF168902/MF168857, MF168839, MF168811. Caltha leptosepala DC. (12x†) (4),
Wallowa Whitman National Forest, Anthony Lakes area, Oregon, U.S.A., Wefferling & Woo 62 (UWM65375):
MF168915/MF168862, identical to AY365370, MF168817. Caltha biflora DC. (6x†) (1), Cascade Range, Willamette National
Forest, Bruno Meadows area, Oregon, U.S.A., Halse 5869 (UWM63863): AY365395, AY365369, MF168803. Caltha biflora
DC. (6x) (2), North Cascades near Mt. Gardner, Washington, U.S.A., Stout SAR4 (WTU351856): MF168899, MF168836,
MF168809. Caltha chionophila Greene (6x) (1), Roosevelt National Forest, Niwot Ridge at the LTER site, Colorado, U.S.A.,
Schuettpelz 00-1 (UWM63862): AY365394, AY365370, MF168802. Caltha chionophila Greene (6x†) (2), Pioneer Mtns, Idaho,
U.S.A., Wefferling 299 (UWM65615): MF168882, identical to AY365370, identical to MF168811.
Outgroup: Caltha appendiculata Pers., AY365385, AY365366, —. Caltha dionaeifolia Hook. ƒ., Punta Arenas, Chile,
Holmgren & Wantorp 553 (NY): AY365389, AY365367, MF168799. Caltha introloba F. v. M., AY365387, AY365368, —.
Caltha natans Pallas, AY365398, AY365371, MF168796. Caltha novae-zelandiae Hook. ƒ., Fiordland, New Zealand, Garnock-
Jones 1876 (UWM64078): AY365388, AY365372, MF168800. Caltha obtusa Cheeseman, AY365386, AY365373, —. Caltha
palustris L., Michigan, U.S.A., Hoot 299 (MICH): AY365382 AY365376, MF168797. Caltha sagittata Cav., Chile, Chase 571
(K): AY365399 AY365378, MF168801. Caltha scaposa Hook ƒ. & Thomson, Qinghai, China, Ho, Bartholomew, & Gilbert 569
(MO): AY365396 AY365379, MF168798.
29
Figure 1. Holotype of Caltha biflora, collected by A. Menzies, 1792. “Northwest coast of America, inland behind Banks
Island.” (between Haida Gwaii and mainland British Columbia, Canada). Housed at The Natural History Museum, London,
England (BM565604). Inset photo shows a single narrow, filiform, filament; magnified 5×.
30
Figure 2. Holotype of Caltha leptosepala, collected by A. Menzies, 1792. “Northwest Coast of America, Prince William
Sound.” (coastal Alaska, U.S.A.). Housed at The Natural History Museum, London, England (BM565602). Inset photo shows
narrow filaments; magnified 5×.
31
Figure 3. Neotype of Caltha leptosepala var. howellii, collected by R. M. Austin, July 1896. “Colby, Butte County, Northern
(U.S.A.). California. Housed at the Greene-Nieuwland Herbarium, Notre Dame, U.S.A. (NDG17325). Inset photo shows
narrow filaments; magnified 5×.
32
Figure 4. Isolectotype of Caltha chionophila, collected by C. F. Baker, 1899. “Near Pagosa Peak, 11,000 ft., Colorado”
(U.S.A.). Housed at the Greene-Nieuwland Herbarium, Notre Dame, U.S.A. (NDG17369). Inset photo shows broad filaments;
magnified 5×.
33
Figure 5. Bayesian MCMC phylogram of Caltha species based on concatenated cpDNA data (rpL32trnL and trnL-trnF).
Posterior probability and bootstrap (ML and MP) support is given for each node. indicates branch was not found. Dashed
branches indicate less than moderate support for at least one approach (PP 0.95, BS 70). Psychrophila group and Caltha
leptosepala complex indicated with vertical bars to right. Ploidy level (x = 8) and collection site is given for each sample. AK =
Alaska, CO = Colorado, ID = Idaho, OR = Oregon, WA = Washington (all U.S.A.).
34
Figure 6. Bayesian MCMC phylogram of Caltha species based on nuclear ribosomal DNA (ITS1, 5.8S, and ITS2). Posterior
probability and bootstrap (ML and MP) support is given for each node. indicates branch was not found. Dashed branches
indicate less than moderate support for at least one approach (PP 0.95, BS 70). Dashed lines connect ribotypes from a single
allododecaploid individual. Psychrophila group and Caltha leptosepala complex indicated with vertical bars to right. Ploidy
level (x = 8) and collection site is given for each sample. AK = Alaska, CO = Colorado, ID = Idaho, OR = Oregon, WA =
Washington (all U.S.A.).
35
Figure 7. Pollen of Caltha biflora, determined by morphology and molecular data (Wefferling & Hoot, unpublished data).
††ploidy level determined by chromosome counts (Wefferling et al. 2015); ploidy level estimated by flow cytometry
(Wefferling et al. 2017). A. U.S.A.: Alaska, Koval, V. L. 5 (ALA). B. U.S.A.: Nevada, Johnson, J. M. 097 (NYBG). C. U.S.A.:
California, Janeway, L. P. 7564 (CSU). D. U.S.A.: California, Wefferling, K.M. and L. Woo 21 (UWM). E. U.S.A.: California,
Wefferling, K.M. and L. Woo 30 (UWM). F. U.S.A.: Washington, Wefferling, K.M. 50 (UWM). G. Canada: British Columbia,
Calder, J. A., D. B. O. Savile, and R. L. Taylor 21584 (V). H. U.S.A.: Oregon, Wefferling, K.M. 69 (UWM). I. U.S.A.:
Washington, Wefferling, K.M. CR1†† (UWM). Size bars = 6 µm.
36
Figure 8. Pollen of Caltha chionophila, determined by morphology and molecular data (Wefferling & Hoot, unpublished data).
ploidy level estimated by flow cytometry (Wefferling et al. 2017). A. U.S.A.: New Mexico, Holmgren, N. H., and P. K.
Holmgren 7330 (NYBG). B. U.S.A.: Colorado, Hall, C., B. Jacobs, and A. Morgan 1565 (NYBG). C. U.S.A.: Nevada, Tiehm,
A. and M. Williams 9675 (ID). D. U.S.A.: Utah, Holmgren, N. H., and P. K. Holmgren 15424 (NYBG). E. U.S.A.: Idaho,
Errter, B., B. Corbin, C. Scott, J. Irwin, and W. Irwin 20192 (SRP). F. U.S.A.: Wyoming, Wefferling, K.M. 19 (UWM). G.
U.S.A.: Oregon, Johanson, J. 07-10 (WTU). H. U.S.A.: Arizona, Rink, G. and L. Stevens s.n. (ASC). I. U.S.A.: Idaho,
Wefferling, K.M. 212lepto (UWM). Size bars = 6 µm.
37
Figure 9. Pollen of 12x Caltha leptosepala, determined by morphology and molecular data (i.e., two ribotypes were retrieved
from each specimen; Wefferling & Hoot, unpublished data). ††ploidy level determined by chromosome counts (Wefferling et
al. 2015); ploidy level estimated by flow cytometry (Wefferling et al. 2017). A. U.S.A.: Washington, Rodman, S., D. Tank, C.
Spurgeon, and K. Ardern 909 (WTU). B. Canada: British Columbia, Marr, K., R. Hebda, and W. MacKenzie 06-0013
(RBCM). C. U.S.A.: Alaska, Parker, C.L., A.R. Batten, and D. Blank 9523 (UWM). D. U.S.A.: Alaska, Wefferling, K.M. 63
(UWM). E. U.S.A.: Oregon, Wefferling, K.M. CR5†† (UWM). F. U.S.A.: California, Bartolomew, B. 4327 and B. Anderson
(NYBG). G. U.S.A.: Idaho, Parks, M., L. Stratford, and R. McNeill 620 (ID). H. U.S.A.: Oregon, Wefferling, K.M. and L. Woo
62 (UWM). I. Canada: British Columbia, Marr, K., R. Hebda, L. Kennedy, and White 02-48 (RBCM). Size bars = 6 µm.
38
Figure 10. Pollen of putative aneuploid and nonaploid Caltha, determined by morphology and molecular data (Wefferling &
Hoot, unpublished data). ††ploidy level determined by chromosome counts (Wefferling et al. 2015); ploidy level estimated by
flow cytometry (Wefferling et al. 2017). A. Hybrid with morphology of C. biflora, with larger genome size (aneuploid?) than
any other sampled C. biflora (Wefferling et al. 2017); U.S.A.: Washington, Hunter, G. LR5 (UWM). B. Non-hybrid with
morphology of C. chionophila, with larger genome size (aneuploid?) than any other C. chionophila (Wefferling et al. 2017);
U.S.A.: Colorado, Townesmith, A., G. Gust, and L. Nye 202 (UWM). C. Allononaploid (9x) C. leptosepala; U.S.A.: Idaho,
Wefferling, K.M. 212rotA†† (UWM).
Figure 11. Close-up photographs of Caltha flowers. A. C. biflora. U.S.A.: Washington, Alpine Lakes Wilderness. Photo by L.
Wefferling. B. C. chionophila. U.S.A.: Wyoming, Medicine Bow National Forest. C. C. leptosepala. U.S.A.: Alaska, Chugach
National Forest.
Article
Aim In order to understand how a montane polyploid species complex responded to Cenozoic mountain uplift and climate change, we reconstructed the biogeographic history of the subalpine marshmarigold polyploid complex, including Caltha biflora, Caltha chionophila and Caltha leptosepala. Phylogenies at multiple taxonomic levels were used to estimate the timing of species divergence, allopolyploid formation and migration patterns, and to identify Last Glacial Maximum (LGM) refugia and recolonization routes. Location Western North America. Methods A fossil‐calibrated chronogram was estimated for the eudicot order Ranunculales that was in turn used to set age priors on genus‐level Caltha phylogenies. Nuclear ribosomal and chloroplast DNA sequence data were collected from subalpine marshmarigolds, including 161 specimens from across the geographical range of the complex. The datasets were analysed under a strict or relaxed molecular clock and a structured coalescent model of evolution estimated under a Bayesian framework. Results Hexaploids C. biflora and Caltha chionophila diverged in the Upper Miocene to Lower Pleistocene (chloroplast and nuclear dataset estimates overlapping at the Miocene–Pliocene boundary), and upon secondary contact formed allododecaploid C. leptosepala bidirectionally and at least three times. The hexaploids persisted to the south of LGM icesheets and recolonized LGM glaciated regions in multiple waves, mainly from the C. biflora clade in the Cascades and Coast Ranges. Main conclusions This study of a widespread western North American plant lineage shows a complex response to past geological and climatic changes, with multiple refugia, allopolyploidization events, and migrations during the Pleistocene. Bidirectional allododecaploid formation has resulted in cryptic lineages, some of which (C. leptosepala in the Cascades and Coast Ranges) have been more successful at reclaiming glaciated regions than others (C. leptosepala in the Northern Rockies).
Article
Full-text available
Ribosomal DNA (rDNA) internal transcribed spacer (ITS) and 5.8S rDNA sequences were obtained from 22 species of dwarf mistletoes (Arceuthobium — Viscaceae) to test phylogenetic relationships. Interspecific distances ranged from 0 to 21.4% between New World species, values two to five times higher than those measures for the ITS region in other plants. One Old World species (A. oxycedri) and one New World species (A. abietis-religiosae) were remarkably similar to each other but exhibited up to 41% sequence divergence from the remaining species. Minimum length trees support the concept of a verticillately branched subgenus Arceuthobium; however, interspecific distances indicate this group is extremely heterogeneous. Subgenus Vaginata, Section Vaginata, is centered in Mexico and encompasses all the taxa previously placed in this group but is expanded to include several species previously classified in Section Campylopoda (e.g., A. divaricatum, A. rubrum, and A. strictum). The sister group relationship between A. divaricatum and A. douglasii, first seen following isozyme analysis, is supported by ITS sequence data. Section Campylopoda s. s. is now composed of 13 mainly U.S. species that show a high degree of morphological and genetic similarity. The eastern dwarf mistletoe, A. pusillum, is not closely related to A. douglasii but rather with A. bicarinatum from Hispaniola, which suggests that these taxa represent highly modified relicts that shared an ancestor in the early Tertiary. Two endemic species from Mexico and Central America (A. guatemalense and A. pendens) formed a sister group and have been placed in a new Section (Penda). Rapid molecular evolution in Arceuthobium may be associated with the adaptive radiation of this genus on numerous conifer hosts.
Article
Full-text available
Premise of the study: Unrecognized variation in ploidy level can lead to an underestimation of species richness and a misleading delineation of geographic range. Caltha leptosepala (Ranunculaceae) comprises a complex of hexaploids (6x), rare nonaploids (9x), and dodecaploids (12x), all with unknown distributions. We delineate the geographic distribution and contact zones of the cytotypes, investigate morphologies of cytotypes and subspecies, and discuss the biogeography and evolutionary history of the polyploid complex. Methods: Using cytologically determined specimens as reference, propidium iodide flow cytometry was performed on silica-dried samples and herbarium specimens from across the range of C. leptosepala s.l. Genome size estimates from flow cytometry were used to infer cytotypes. A key morphological character, leaf length-to-width ratio, was measured to evaluate whether these dimensions are informative for taxon and/or cytotype delimitation. Key results: Dodecaploids were more northerly in distribution than hexaploids, and a single midlatitude population in the Northern Rockies yielded nonaploids. Genome size estimates were significantly different between all cytotypes and between hexaploid subspecies. Leaf length-to-width ratios were significantly different between subspecies and some cytotypes. Conclusions: Caltha leptosepala presents clear patterns of cytotype distribution at the large scale. Marked differences in morphology, range, and genome size were detected between the hexaploid subspecies, C. leptosepala subsp. howellii in the Cascade-Sierra axis and C. leptosepala subsp. leptosepala in the Rockies. Sympatry between cytotypes in the Cascades and a parapatric distribution in the Northern Rockies suggest unique origins and separate lineages in the respective contact zones.
Article
Full-text available
Based on field investigations, morphological and molecular systematic studies, a new species, Caltha dysosmoides (Ranunculaceae) from southwestern China is described. It differs from all other known congeneric species by the densely dentate leaf margin, the pendulous pedicels, the scarlet flowers, the compressed filaments, and the triangular connective. To better understand taxonomy of this new species, phylogenetic analyses were conducted using a combined dataset from nrITS, plastid trnL-trnF region, and atpB-rbcL spacer. The results indicate that Caltha is monophyletic and C. dysosmoides is sister to C. sinogracilis. The conservation status of the new species is categorized as CR based on IUCN criteria.
Article
Full-text available
Bootstrapping is a common method for assessing confidence in phylogenetic analyses. Although bootstrapping was first applied in phylogenetics to assess the repeatability of a given result, bootstrap results are commonly interpreted as a measure of the probability that a phylogenetic estimate represents the true phylogeny. Here we use computer simulations and a laboratory-generated phylogeny to test bootstrapping results of parsimony analyses, both as measures of repeatability (i.e., the probability of repeating a result given a new sample of characters) and accuracy (i.e., the probability that a result represents the true phylogeny). Our results indicate that any given bootstrap proportion provides an unbiased but highly imprecise measure of repeatability, unless the actual probability of replicating the relevant result is nearly one. The imprecision of the estimate is great enough to render the estimate virtually useless as a measure of repeatability. Under conditions thought to be typical of most phylogenetic analyses, however, bootstrap proportions in majority-rule consensus trees provide biased but highly conservative estimates of the probability of correctly inferring the corresponding clades. Specifically, under conditions of equal rates of change, symmetric phylogenies, and internodal change of less-than-or-equal-to 20% of the characters, bootstrap proportions of greater-than-or-equal-to 70% usually correspond to a probability of greater-than-or-equal-to 95% that the corresponding clade is real. However, under conditions of very high rates of internodal change (approaching randomization of the characters among taxa) or highly unequal rates of change among taxa, bootstrap proportions >50% are overestimates of accuracy.
Book
— We studied sequence variation in 16S rDNA in 204 individuals from 37 populations of the land snail Candidula unifasciata (Poiret 1801) across the core species range in France, Switzerland, and Germany. Phylogeographic, nested clade, and coalescence analyses were used to elucidate the species evolutionary history. The study revealed the presence of two major evolutionary lineages that evolved in separate refuges in southeast France as result of previous fragmentation during the Pleistocene. Applying a recent extension of the nested clade analysis (Templeton 2001), we inferred that range expansions along river valleys in independent corridors to the north led eventually to a secondary contact zone of the major clades around the Geneva Basin. There is evidence supporting the idea that the formation of the secondary contact zone and the colonization of Germany might be postglacial events. The phylogeographic history inferred for C. unifasciata differs from general biogeographic patterns of postglacial colonization previously identified for other taxa, and it might represent a common model for species with restricted dispersal.
Article
Caltha is a widely distributed genus in the buttercup family (Ranunculaceae) showing interesting distribution patterns in both hemispheres. Evolutionary history of Caltha was examined by means of phylogenetic, molecular dating, and historical biogeographic analyses with a more comprehensive sampling than previous studies. The internal transcribed spacer from the nuclear genome and trnL-F and atpB-rbcL regions from the plastid genome were used and analyzed using parsimony and Bayesian methods. Divergence time was estimated using Bayesian dating analyses with multiple fossil calibrations. Historical biogeography was inferred using the Bayes-DIVA method implemented in RASP. We obtained a well-resolved and well-supported phylogeny within the Caltha lineage. Caltha natans Pall. diverged first from the genus and the other species grouped into two clades. Our expanded sampling scheme revealed a complicated evolutionary pattern in the C. palustris complex. Caltha sinogracilis W. T. Wang was resolved to be a member of the C. palustris complex, rather than closely related to C. scaposa Hook. f. & Thomson. Caltha rubriflora B. L. Burtt & Lauener was also revealed to be not just a red-flower form of C. sinogracilis. The diversification of the genus began at 50.5 mya (95% high posterior density: 37.1–63.9 mya), and its ancestral range was very probably in the Northern Hemisphere. The South American species may derive from western North American ancestors that dispersed along the western American Cordillera during the Cenozoic era. The vicariance model of the Southern Hemisphere species proposed by a previous study was rejected in this study.