全 文 ：Journal of Systematics and Evolution 46 (3): 322–332 (2008) doi: 10.3724/SP.J.1002.2008.08050
(formerly Acta Phytotaxonomica Sinica) http://www.plantsystematics.com
Phylogenetic inferences in Prunus (Rosaceae) using chloroplast
ndhF and nuclear ribosomal ITS sequences
1Jun WEN* 2Scott T. BERGGREN 3Chung-Hee LEE 4Stefanie ICKERT-BOND
5Ting-Shuang YI 6Ki-Oug YOO 7Lei XIE 8Joey SHAW 9Dan POTTER
1(Department of Botany, National Museum of Natural History, MRC 166, Smithsonian Institution, Washington, DC 20013-7012, USA)
2(Department of Biology, Colorado State University, Fort Collins, CO 80523, USA)
3(Korean National Arboretum, 51-7 Jikdongni Soheur-eup Pocheon-si Gyeonggi-do, 487-821, Korea)
4(UA Museum of the North and Department of Biology and Wildlife, University of Alaska Fairbanks, Fairbanks, AK 99775-6960, USA)
5(Key Laboratory of Plant Biodiversity and Biogeography, Kunming Institute of Botany, Chinese Academy of Sciences, Kunming 650204, China)
6(Division of Life Sciences, Kangwon National University, Chuncheon 200-701, Korea)
7(State Key Laboratory of Systematic and Evolutionary Botany, Institute of Botany, Chinese Academy of Sciences, Beijing 100093, China)
8(Department of Biological and Environmental Sciences, University of Tennessee, Chattanooga, TN 37403-2598, USA)
9(Department of Plant Sciences, MS 2, University of California, Davis, CA 95616, USA)
Abstract Sequences of the chloroplast ndhF gene and the nuclear ribosomal ITS regions are employed to recon-
struct the phylogeny of Prunus (Rosaceae), and evaluate the classification schemes of this genus. The two data
sets are congruent in that the genera Prunus s.l. and Maddenia form a monophyletic group, with Maddenia nested
within Prunus. However, the ndhF data set is incongruent with the ITS data supporting two major groups within
Prunus: one consisting of subgenera Laurocerasus (including Pygeum) and Padus as well as the genus Maddenia
and another of subgenera Amygdalus, Cerasus, and Prunus. The ITS data, on the other hand, support a clade
composed of subgenera Amygdalus and Prunus and Prunus sect. Microcerasus in addition to a paraphyletic grade
of subgenera Laurocerasus and Padus (and the genus Maddenia) taxa. In general, the subgeneric classifications of
Prunus s.l. are not supported. The ITS and ndhF phylogenies differ mainly in interspecific relationships and the
relative position of the Padus/Laurocerasus group. Both ITS and ndhF data sets suggest that the formerly recog-
nized genus Pygeum is polyphyletic and that the distinction of the subgenera Padus and Laurocerasus is not
supported. The biogeographic interactions of the temperate and tropical members in the Padus/Laurocera-
sus/Maddenia alliance including Pygeum are shown to be highly dynamic and complex.
Key words ITS, ndhF, phylogeny, Prunus, Rosaceae.
Prunus L. (Rosaceae) consists of approximately
250 species distributed across the Northern Hemi-
sphere and into the sub-tropics and tropics, including a
large number of economically significant species such
as cherries, peaches, plums, apricots, almonds and a
wide variety of ornamentals and timber species (Lee
& Wen, 2001). Many taxa are important fruit crops,
and several have been used as such since prehistoric
times (Komarov, 1971; Schery, 1972; Watkins, 1995).
The genus is usually included in the subfamily
Amygdaloideae Arn., which is also known as the
Prunoideae Focke (see Robertson, 1974). The Amyg-
daloideae has traditionally contained four genera:
Prunus s.l., Prinsepia Royle, Maddenia Hook. f. &
Thoms., and Oemleria Rchb. (=Osmaronia Greene,
Nuttallia Torrey & Gray; see Landon, 1975) and is
distinguished from other rosaceous subfamilies by its
simple leaves, drupaceous fruits, superior ovaries, and
x=8 chromosome number (Robertson, 1974; Ghora &
Panigrahi, 1995). While most treatments follow this
concept of the subfamily (e.g., Rehder, 1940; Robert-
son, 1974; Ghora & Panigrahi, 1995), several other
workers raised the subfamily to family rank, either as
Amygdalaceae or Drupaceae, though this treatment is
less common (e.g., Rydberg, 1900, 1917; Berry, 1930;
Small, 1933; Dahlgren, 1983; Mai, 1984). The genus
Exochorda Lindl., often placed in the subfamily
Spiraeoideae Arn. because of its five carpels that
produce capsular fruits, has sometimes been allied
with the Amygdaloideae, based on several lines of
evidence (Stebbins, 1958; Goldblatt, 1976; Zhang,
1992; Morgan et al., 1994; Lee & Wen, 2001). Other
genera formerly included in this subfamily include
Pygeum Gaertner [which was merged into Prunus
subgen. Laurocerasus Duhamel (Kalkman, 1965)],
Plagiospermum Oliver [which was shown to be
synonymous with the genus Prinsepia (Rehder,
1915)], and the various segregate genera from Prunus
s.l. (Lee & Wen, 2001). Recently, Potter et al. (2007)
proposed a new classification of Rosaceae based on
molecular phylogenetic analyses, in which they
———————————
Received: 11 April 2008 Accepted: 25 April 2008
* E-mail: .
WEN et al.: Prunus phylogeny

323
recognized three subfamilies: Rosoideae, Dryadoideae
and Spiraeoideae. The newly defined Spiraeoideae
includes all genera previously assigned to Amygda-
loideae and Maloideae. Monophyly of the traditional
Amygdaloideae was not supported. Instead, Prunus
s.l., along with Maddenia and Pygeum, is treated in
the tribe Amygdaleae of the subfamily Spiraeoideae,
while Exochorda, Oemleria, and Prinsepia are placed
in tribe Osmaronieae of the same subfamily.
Classification within the genus Prunus s.l. has
been varied. Prunus is distinguished from the other
three genera in the traditionally defined subfamily
Amygdaloideae by the combination of its single carpel
(rarely 2), five sepals (occasionally more), bisexual
flowers (rarely andromonoecious, see Wolfe & Dra-
palik, 1999), and a solid stem pith. Tournefort (1700)
offered the first classification of Prunus s.l. by pro-
posing six genera based on fruit morphology: Amyg-
dalus L., Armeniaca Miller, Cerasus Miller, Lauro-
cerasus, Persica Miller, and Prunus (s.s.). Linnaeus
(1753) reduced these six genera to two by merging
Persica into Amygdalus and putting the rest (including
another genus Padus Miller) into Prunus. Since then,
many other classifications have been proposed for
Prunus s.l., recognizing as many as seven or more
distinct genera (De Candolle, 1825; Hutchinson, 1964;
Browicz, 1969), or one broadly defined genus
(Rehder, 1940). Many classifications treated Prunus
inclusively with several subgenera or sections, fol-
lowing Bentham and Hooker (1865), and Focke
(1894) (e.g., Koehne, 1911; Rehder, 1940; Fernald,
1950; Robertson, 1974; Ghora & Panigrahi, 1995). On
the other hand, several workers divided Prunus into
multiple genera (e.g., Hutchinson, 1964; Browicz,
1969; Komarov, 1971; Yü et al., 1986). Rehder’s
(1940) treatment of Prunus in the inclusive sense with
five subgenera [Amygdalus, Cerasus, Laurocerasus,
Padus, and Prunophora Neck. (=Prunus s.s.)] and
twelve sections is favored by several workers
(Bate-Smith, 1961; Robertson, 1974; Ghora & Pani-
grahi, 1995; Lersten & Horner, 2000). Krüssmann
(1978) recognizes section Microcerasus of subgenus
Cerasus as a distinct subgenus, Lithocerasus Ingram,
resulting in a total of six subgenera and 14 sections
recognized within the genus. For a detailed history of
classification within Prunus s.l., see McVaugh (1951),
Kalkman (1965), Ghora and Panigrahi (1995), and
Lee and Wen (2001).
Several recent phylogenetic studies of Prunus
have been conducted. One of the earliest works was
done by Mowrey and Werner (1990) who examined
isozyme profiles of 34 species from subgenera
Prunus, Amygdalus, Cerasus and Lithocerasus (sec-
tion Microcerasus of subgen. Cerasus sensu Rehder).
They found support for the subgenera Prunus, Amyg-
dalus, and Cerasus. The most noteworthy exception of
their hypothesis, with respect to the classically recog-
nized subgeneric groupings, was that several species
of subgen. Lithocerasus were nested within subgen.
Prunus.
Zhang (1992) sampled wood anatomy in the
Rosaceae, including 83 samples of Prunus s.l. He
found that the genera (subgeneric levels in other
treatments) that comprise Prunus s.l. form a mono-
phyletic group nested within the other amygdaloid
genera. Within Prunus s.l., sect. Armeniaca of subgen.
Prunus (sensu Rehder) and subgen. Amygdalus are
suggested to be the most derived, subgen. Cerasus is
sister to that, then a clade of Padus and some mem-
bers of Laurocerasus (“group A”), then Prunus s.s.,
while the Pygeum group and the remainder of Lauro-
cerasus (“group B”) are considered the least advanced
groups.
Chloroplast DNA restriction sites were used to
construct the phylogeny of eight cultivated members
of Prunus (Badenes & Parfitt, 1995). While too few
species were sampled to study Prunus classification,
this study suggested that subgenera Amygdalus and
Prunus are more closely related to one another than
either is to subgenus Cerasus.
Lersten and Horner (2000) examined leaf crystals
in several members of Prunus s.l., and suggested that
the Prunophora (=Prunus s.s.) and Amygdalus sub-
genera are related and apparently are the most ad-
vanced subgenera. The Cerasus and Laurocerasus
subgenera are intermediate and very diverse, while
Padus is suggested to be the least advanced subgenus
and the furthest from subgenera Prunus and Amygda-
lus.
Lee and Wen (2001) employed nuclear ribosomal
ITS sequences to construct the phylogeny of Prunus.
The ITS data suggest a close relationship between
subgenera Prunus and Amygdalus. They reported that
subgenera Padus and Laurocerasus are closely re-
lated, and form a basally branching paraphyletic
group. They also found that Maddenia is nested within
the Padus/Laurocerasus group, and that subgenera
Cerasus and Padus are both polyphyletic. Bortiri et al.
(2001) used the nuclear ribosomal ITS and the chloro-
plast trnL-F spacer region to construct the phylogeny
of Prunus. Even though the resolution of the trnL-F
tree was relatively low, the combined analysis showed
a congruent phylogeny as seen in Lee and Wen
(2001). To further resolve some deep nodes Bortiri et
Journal of Systematics and Evolution Vol. 46 No. 3 2008 324
al. (2002) employed sequences of the s6pdh (sorbitol
6-phosphate dehydrogenase gene). The resolution
remained low and the authors proposed a rapid radia-
tion in the early history of the genus. Shaw and Small
(2004) focused on Prunus sect. Prunocerasus using
seven noncoding chloroplast DNA regions and also
sampled widely throughout the other subgenera of the
genus. Their data yielded a well-resolved phylogenetic
hypothesis showing support for subgen. Prunus sect.
Prunus, subgen. Prunus sect. Prunocerasus, and
subgen. Amygdalus. Three species of subgen. Cerasus
sect. Microcerasus were nested within the pruno-
amygdaloid clade that was sister to other species of
subgen. Cerasus. All molecular analyses of the above
mentioned studies have only included a few taxa from
tropical regions.
This study aims to provide further insights into
the phylogenetic relationships of Prunus using se-
quences of the chloroplast ndhF gene, and the nuclear
ribosomal ITS regions. We expanded our sampling of
tropical members of Prunus, especially the unsampled
Pygeum group. Our primary goal of the study is to
examine how the previously poorly sampled subgen.
Laurocerasus including the Pygeum group is related
to its putative closest relative subgen. Padus and how
the two putative subgenera are related to the rest of the
genus. Questions to be addressed include: (1) Is the
generic status of Pygeum supported? (2) What are the
phylogenetic relationships among the subgenera
within the genus Prunus s.l.? (3) Are any of the
current classifications of Prunus s.l., supported by
molecular evidence? (4) Is the chloroplast ndhF data
set congruent with the other molecular data sets
described above (nDNA ITS and s6pdh and cpDNA
trnL-F and other regions included in Shaw and Small,
2004)? and (5) What are the relationships between the
temperate and the tropical members of Prunus s.l.?
1 Material and Methods
Our sampling included 59 (ndhF) or 51 (ITS)
accessions of Prunus s.l. The samples covered all five
subgenera of Prunus s.l. (Amygdalus, Cerasus, Lau-
rocerasus, Padus, and Prunus; sensu Rehder, 1940),
and most of the subgeneric sections, as well as the
genus Maddenia and six species formerly classified in
Pygeum. The outgroups included species of tribe
Osmaronieae (Exochorda, Oemleria, and Prinsepia),
which were formerly classified with Prunus in
Amygdaloideae and species of two other genera of
Spiraeoideae, Physocarpus Maxim. and Lyonotham-
nus A. Gray.
Total DNA was extracted from leaf material with
the CTAB method of Doyle and Doyle (1987). DNA
amplifications were performed in 100-μL reactions
following Wen and Zimmer (1996). Most of the ITS
PCR products were purified using millipore columns
(Ultrafree-MC Filter Unit, 30,000 NMWL, Millipore,
Bedford, Massachusetts, USA), while all of the ndhF
PCR products and a portion of the ITS products were
purified using Wizard® purification preps (Cat.
#A7170, Promega, Madison, WI, USA). Sequences
were generated by automated sequencing (ABI
PRISM® 377XL, Perkin-Elmer) and Big Dye chemis-
try. Two primers (C26A and N18L18 in Wen &
Zimmer, 1996) were used to obtain the entire ITS and
5.8S regions from both directions. The ndhF region
was sequenced using several published primers
(ndhF-274R, ndhF-536R, ndhF-803R, ndhF-972R,
ndhF-1318R, ndhF-1318, ndhF-2110R; Olmstead &
Sweere, 1994) and one primer (ndhF-1577pr:
CGTTTATTAGTATTGCTCGKTTTG) that was
designed in this study. All the sequences were depos-
ited in GenBank (see Table 1 for accession numbers).
Phylogenetic analyses were performed using the
maximum parsimony (Swofford et al., 1996) and the
Bayesian inference methods. Parsimony analysis was
performed with tree bisection-reconnection branch
swapping, MulTrees on, and simple taxon addition in
PAUP* version 4.0b10 (Swofford, 2003). Parsimony
bootstrap support for each clade was estimated as
above from 500 heuristic search replicates, with 100
random taxon addition replicates saving all optimal
trees at each step.
The optimal model of molecular evolution was
determined by the Akaike Information Criterion (AIC)
using Modeltest ver. 3.7 (Posada & Crandall, 1998;
Posada & Buckley, 2004). In each case the optimal
model was the General Time Reversible model, with
rate heterogeneity modeled by assuming that some
proportion of sites are invariable and that the rate of
evolution at other sites is modeled using a discrete
approximation to a gamma distribution [GTR+I+Γ].
Bayesian inferences were implemented in MrBayes
version 3.1.2 (Huelsenbeck & Ronquist, 2001) with
the model estimated above and each gene was parti-
tioned. We used one cold and three heated chains, with
random initial trees. Trees were generated for
2,000,000 generations, with sampling every 100
generations. Following a burn-in period of the first
2000 generations, 19,800 trees were sampled from the
posterior distribution to calculate the posterior prob-
abilities (PP).
Congruence among the two different data sets
WEN et al.: Prunus phylogeny

325
Table 1 Taxa of Prunus and outgroups sampled for this study, and GenBank accession numbers (the classification system of Prunus follows Rehder,
1940; and US, CS, and F in parentheses are acronyms of the US National Herbarium, Colorado State University Herbarium, and Field Museum of
Natural History Herbarium, respectively)
Taxon Source and voucher1) GenBank accession (ITS; ndhF)
Subgen. 1. Prunus2)
Sect. 1. Prunus
P. insititia L. 7307 USA, Illinois, cult. Morton Arboretum: Wen 7307 (US) EU669097; EU669166
P. maritima Marsh. 7311 USA, Illinois, cult. Morton Arboretum: Wen 7311 (US) EU669098; EU669168
P. murrayana Palmer 7283 USA, Texas, Brewster Co.: Wen 7283 (US) EU669099; EU669099
P. salicina Lindl. China, Zhejiang Prov.: Wen 3020 (CS) AF179486, AF179487; EU669143
P. spinosa L. 7308 USA, Illinois, cult. Morton Arboretum: Wen 7308 (US) EU669100; EU669167
Prunus sp. 8080 China, Chongqing Shi: Wen 8080 (US) –; EU669171
Sect. 2. Prunocerasus Koehne
P. americana Marsh. 4021 USA, Colorado, Larimer Co.: Lee & Wen 4021 (CS) AF179488; EU669150
P. americana Marsh. 4061 Canada, Alberta: Whitcher s.n. (CS) AF179489; EU669125
P. americana Marsh. 5011 USA, Colorado, Larimer Co.: Lee & Wen 5011 (US) –; EU669137
P. angustifolia Marsh. USA, Florida, Jackson Co.: Gholson s.n. (CS) AF179490; EU669131
P. nigra Ait. USA, Colorado, cult. CS TS88139: Lee & Wen 4024 (CS) AF179491,AF179492; EU669128
P. umbellata Ell. USA, Florida, Jackson Co.: Gholson s.n. (CS) AF179493; EU669120
Sect. 3. Armeniaca (Lam.) Koch.
P. armeniaca L. var. mandshurica
Maxim.
USA, Colorado, cult. CS TS81501: Lee & Wen 4025 (CS) AF179494, AF179495; EU669149
P. mume (Sieb.) Sieb. & Zucc. China, Zhejiang Prov.: Wen 3043 (CS) AF179496, AF179497; EU669141
Subgen. 2. Amygdalus (L.) Focke
P. andersonii Gray USA, Nevada, Douglas Co.: M. Beck s.n.(CS) EU669083; EU669151
P. davidiana (Carr.) Franch. USA, Missouri, cult. MBG 1981-1933-1: H. M. Davis s.n. (CS) EU669084; EU669142
P. dulcis (Mill.) Webb. USA, Missouri, cult. MBG 1983-0585: H. M. Davis s.n. (CS) EU669085; EU669146
P. fasciculata (Torr.) Gray USA, California, Riverside Co.: M. Beck s.n.(CS) EU669086; EU669153
P. fremontii Wats. USA, California, Riverside Co.: M. Beck s.n.(CS) EU669087; EU669152
P. havardii Mason USA, Colorado, cult. CS: Wen s. n. (US) EU669096; EU669165
P. persica (L.) Batsch. China, Zhejiang Prov.: Wen 3017 (CS) AF179562; EU669129
P. tenella Batsch. USA, Colorado, cult. CS TS93054: Lee & Wen 4011 (CS) AF179560, AF179561; EU669119
P. triloba Lindl. USA, Colorado, cult. CS s.n.: S. Berggren s.n. (CS) EU669088; EU669140
Subgen. 3. Cerasus Pers.
Sect. 1. Microcerasus Webb.
P. besseyi Bailey USA, Colorado, cult. CS TS85155: Lee & Wen 4023 (CS) AF179498, AF179499; EU669121
P. glandulosa Thunb. USA, Colorado, cult. Ft. Collins: S. Berggren s.n. (CS) EU669089; EU669147
P. tomentosa Thunb. USA, Colorado, cult. CS TS81261: Lee & Wen 4010 (CS) AF179500; EU669122
Sect. 2. Pseudocerasus Koehne
P. campanulata Maxim. USA, Washington DC, cult. USNA 58776: Lee & Wen 4014 (CS) AF179501, AF179502; EU669123
P. incisa Thunb. USA, Washington DC, cult. USNA 58816: Lee & Wen 4071 (CS) AF179504, AF179505; EU669145
P. mahaleb L. USA, Colorado, cult. CS TS83156: Lee & Wen 4015 (CS) AF179523, AF179524; EU669134
P. nipponica Matsum. var. nipponica USA, Washington DC, cult. USNA 45734: Lee & Wen 4077 (CS) AF179507, AF179508; EU669144
P. pensylvanica L. f. 7298 USA, Wisconsin: Wen7298 (US) EU669090; EU669138
P. subhirtella Miq. var. subhirtella USA, Washington DC, cult. USNA 61383: Lee & Wen 4080 (CS) AF179519; EU669135
Sect. 3. Phyllomahaleb Koehne
P. maximowiczii Rupr. 4079 USA, Washington DC, cult. USNA 62ER: Lee & Wen 4079 (CS) AF179526; EU669124
Sect. 4. Lobopetalum (Koehne) T. T.Yü & C. L. Li
P. dielsiana Schneid. 8091 China, Chongqing Shi: Wen 8091 (US) –; EU669172
Subgen. 4. Padus (Moench) Koehne
P. grayana Maxim. USA, Washington DC, cult. USNA 46329: Lee & Wen 4073 (CS) AF179531; EU669136
P. maackii Rupr. 4009 USA, Colorado, cult. CS TS78092: Lee & Wen 4009 (CS) AF179532, AF179533; EU669139

Journal of Systematics and Evolution Vol. 46 No. 3 2008 326
Table 1 (continued)
Taxon Source and voucher1) GenBank accession (ITS; ndhF)
P. napaulensis K. Koch. 6470 China, Yunnan: Wen 6470 (US) EU669106; EU669159
P. padus L. var. commutata Dipp. USA, Colorado, cult. CS TS82097: Lee & Wen 4027 (CS) AF179527; EU669132
P. phaeosticta Maxim. China, Taiwan: H-Y Liu s.n. (F) EU669095; EU669163
P. serotina Ehrh. 7229 USA, Illinois: Wen 7229 (US) EU669104; EU669160
P. vana J. F. Macbr. Ecuador: Talfur & Villacres 123 (F) EU669105; EU669148
P. virginiana L. var. virginiana USA, Colorado, Larimer Co.: Lee & Wen 4022 (CS) AF179536, AF179537; EU669126
P. virginiana var. demissa (Nutt.)
Murr.
USA, Colorado, cult. CS: S. Berggren s.n. (CS) EU669101; EU669127
Subgen. 5. Laurocerasus Koehne
P. caroliniana Aiton USA, Florida, Gadsden Co.: Gholson 9-20-98 (CS) AF179540; EU669130
P. ilicifolia (Nutt.) Walp. USA, California, Santa Barbara: D. A. Young s.n. (CS) AF179543, AF179544; EU669133
P. laurocerasus L. 5001 Cult. AA 889-72-D: Lee & Wen 5001 (CS) AF179545, AF179546; EU669118
P. undulata (D. Don) Roem. 6452 China, Yunnan: Wen 6452 (US) EU669108; EU669155
P. zippeliana Miq. 6030 Vietnam, Lao Cai: Wen 6030 (US) –; EU669170
Prunus sp. 5928 Vietnam, Lao Cai: Wen 5928 (US) –; EU669157
Prunus sp. 6812 Costa Rica, Puntarenas Prov., Monteverde area: Wen 6812 (US) EU669092; EU669161
Prunus sp. 6846 Costa Rica, Puntarenas Prov., Monteverde area: Wen 6846 (US) EU669093; EU669162
Prunus sp. 7042 Costa Rica, San Jose Prov.: Wen 7042 (US) EU669094; EU669164
Prunus sp. 8419 Malaysia, Paso: Wen 8419 (US) –; EU669156
Pygeum group
Pygeum stipulaceum King 8418 Malaysia, Paso: Wen 8418 (US) EU669103; EU669175
Pygeum topengii Merr. 5813 China, Guangdong: Wen 5813 (US) EU669110; EU669154
Prunus africana (Hook. f.) Kalkman
6226
USA, New York, cult Cornell Univ.: Wen 6226 (US) EU669109; EU669158
P. grisea (Blume Ex Müll. Berol.)
Kalkman 8262
Philippines, Los Banos: Wen 8262 (US) EU669102; EU669173
P. arborea (Blume) Kalkman 8431 Malaysia, Paso: Wen 8431 (US) –; EU669174
P. malayana Kalkman 8366 Malaysia, Pahang, Cameron Highlands: Wen 8366 (US) EU669107; EU669176
Exochorda giraldii Hesse var. wilsonii
(Rehder) Rehder
USA, Massachusetts, cult. AA 11626-C: Lee & Wen 5003 (CS) AF179555, AF179556; EU669114
Maddenia hypoleuca Koehne USA, Massachusetts, cult. AA 665-65-A: Lee & Wen 5005 (CS) AF179549, AF179550; EU669117
Oemleria cerasiformis (Hook.& Arn.)
Landon
USA, Massachusetts, cult: AA 275-85-A: Lee & Wen 5002 (CS) AF179553, AF179554; EU669115
Prinsepia uniflora Batal. USA, Colorado, cult. CS TS81293: Lee & Wen 4086 (CS) AF179559; EU669116
Physocarpus monogynus (Torr.) Coult. USA, Colorado, Larimer Co.: Owens 205 (CS) –; EU669112
Lyonothamnus floribundus Gray USA, California, cult. SBBG 63-048: D. A. Young s.n. (CS) AF179558; EU669111
Holodiscus discolor (Pursh) Maxim.7257 USA, Texas, Jeff Davis Co.: Wen 7257 (US) EU669091; EU669113
1) Abbrevations: AA=Arnold Arboretum, CS=Colorado State University Arboretum, MBG=Missouri Botanical Garden, SBBG=Santa Barbara
Botanical Garden, USNA=United States National Arboretum. 2) Subgeneric and sectional name Prunus was used instead of Prunophora and
Euprunus, respectively, following the International Code of Botanical Nomenclature (McNeill et al., 2006).

was tested using the incongruence length difference
(ILD) test in PAUP* using 1000 data bipartitions and
analyzing a maximum of 10,000 trees for each (Farris
et al., 1995).
2 Results
2.1 Phylogenetic analysis of ndhF data
With gaps treated as missing data, the parsimony
analysis of the ndhF data generated 196200 maximally
parsimonious trees (MPT’s) with a length of 815 steps,
a consistency index (CI) of 0.71, a COI excluding
uninformative characters of 0.56, and a retention
index (RI) of 0.86. The strict consensus tree is shown
in Fig. 1.
The genus Maddenia is nested within Prunus s.l.
and shows a close relationship with some taxa of
subgenera Laurocerasus (including some species of
the Pygeum group) and Padus. The subgenera Padus
and Laurocerasus (along with Maddenia) form a
monophyletic group with Bayesian posterior probabil-
ity (PP) 99%, but bootstrap value (BV) less than 50%
(Fig. 1). This large Laurocerasus-Pygeum-Padus-
Maddenia clade contains two major subclades, each
with taxa of the subgenera Laurocerasus, Pygeum and
Padus. The monophyly of each of those three
WEN et al.: Prunus phylogeny

327


Fig. 1. Strict consensus of 196200 maximally parsimonious trees of the ndhF sequence data (CI=0.71, CI excluding uninformative characters=0.56,
and RI=0.86). Numbers above the lines are bootstrap values, and the numbers below the branches are Bayesian posterior probabilities. The abbre-
viations Amy, Ce, Lau, Pad, and Pr stand for subgenera Amygdalus, Cerasus, Laurocerasus, Padus, and Prunus, respectively; and Pyg represents the
Pygeum group.
Journal of Systematics and Evolution Vol. 46 No. 3 2008 328
subgroups is thus not supported by the ndhF data.
Four Asian species formerly classified in Pygeum
form a clade, but neither the Asian Pygeum topengii
nor the African Prunus africana (also formerly classi-
fied in Pygeum) is included in that clade. Subgenera
Amygdalus, Cerasus, and Prunus form another
strongly supported monophyletic group (PP=100,
BV=81), with Prunus maackii and the core members
of Cerasus forming a subclade (PP=100, BV=68; Fig.
1). The monophyly of subgenus Cerasus is, however,
not supported because Prunus besseyi, P. tomentosa,
and P. glandulosa are unresolved in the Amygdalus-
Prunus-Cerasus clade.
2.2 Phylogenetic analysis of ITS data
Treating gaps as missing data, the parsimony
analysis of the ITS data generated 49200 MPT’s with
a length of 791 steps, a CI of 0.56, a CI excluding
uninformative characters of 0.45, and an RI of 0.70.
The strict consensus tree is shown in Fig. 2.
As in the ndhF analysis, the ITS analysis indi-
cates that Maddenia is nested within Prunus, and
more specifically within a group consisting of mem-
bers of the Padus and Laurocerasus subgenera (Fig.
2). Neither subgenus Padus nor subgenus Laurocera-
sus is monophyletic, and most of the species of Lau-
rocerasus (including Pygeum) form a paraphyletic
grade within which the rest of the genus is nested.
Here, the four sampled Asian species of Pygeum form
a well-supported monophyletic group, but Prunus
africana is not included in that clade. Subgenus
Cerasus is poorly resolved and clearly not mono-
phyletic. Prunus besseyi, P. glandulosa, and P. to-
mentosa of subgenus Cerasus group with taxa of
subgenera Amygdalus and Prunus (Fig. 2). The
Prunus and Amygdalus subgenera are closely related,
and they form a monophyletic group with low support
with the aforementioned three taxa from subgenus
Cerasus (Fig. 2).
A partition homogeneity test (Farris et al., 1995)
performed in PAUP* (Swofford, 2003) resulted in a
p-value of 0.01, indicating significant incongruence
between the two data sets, which therefore were not
combined for further analysis.
3 Discussion
3.1 Maddenia nested within Prunus
Maddenia, a small genus of five species in the
Himalayas and China (Rehder, 1940), was shown to
be nested within Prunus s.l. In each analysis, it was
found embedded in a clade composed of members of
the subgenera Laurocerasus and Padus (Figs. 1, 2).
Maddenia shares several characters with these sub-
genera, including racemose inflorescences and gener-
ally monocarpellate flowers, but has been given
generic status based on its dioecious flowers that have
ten sepals and no petals (as opposed to five in
Prunus). These characters may not clearly delimit
Maddenia from Prunus. Some species in Prunus
subgenus Laurocerasus (specifically, former members
of the genus Pygeum that were merged into subgenus
Laurocerasus) have ten perianth parts that are indis-
tinguishable or only slightly distinguishable as petals
or sepals (see Kalkman, 1965). Furthermore, Sterling
(1964) points out that the occurrence of dioecy among
species of Maddenia is inconsistent, and that the fruits
of Maddenia and the Pygeum group share several
characters. The close alliance of Maddenia with the
Laurocerasus and Padus group seems noteworthy in
that at least one species in subgenus Laurocerasus,
Prunus caroliniana, frequently shows andromonoecy
(Wolfe & Drapalik, 1999), a breeding system believed
to be a precursor of dioecy (Bertin, 1982; Solomon,
1986). If Maddenia is indeed derived from a common
ancestor of some members of the Laurocerasus/Padus
group, this is a specific example of a dioecious species
arising from bisexual and andromonoecious ancestors.
The evolution of breeding systems potentially exem-
plified in the Maddenia-Prunus alliance deserves
further study.
3.2 Relationships within Prunus s.l.
Subgenera Laurocerasus (including Pygeum) and
Padus and the genus Maddenia form a monophyletic
group that is sister to the rest of Prunus s.l. in the
ndhF tree (Fig. 1). These relationships are also sup-
ported by cpDNA trnL-trnL-trnF+trnS-trnG-trnG+
psbA-trnH sequences (JS, JW and DP unpublished
study). These same taxa form a paraphyletic group in
the ITS tree (Fig. 2). Lersten and Horner’s (2000) leaf
crystal data suggests that subgenus Padus probably is
a less advanced group within Prunus s.l., with subge-
nus Laurocerasus next (see also Kalkman, 1965), but
the data presented here do not support this view as
members of each subgenus are intermixed among two
sister clades. In any case, the ndhF parsimony data
(Fig. 1) presented here suggest a Padus/Laurocera-
sus/Maddenia alliance (PP=99, BV<50).
The subgenera Laurocerasus (including Pygeum)
and Padus share racemose inflorescences, small
flowers, and small floral bracts, but have been delim-
ited because members of subgenus Laurocerasus
generally have evergreen leaves, naked peduncles, and
axillary inflorescences (flowers are in new terminal
racemes within subgenus Padus) (see Rehder, 1940;
WEN et al.: Prunus phylogeny

329


Fig. 2. Strict consensus of 49200 maximally parsimonious trees of the nuclear ribosomal ITS sequence data (CI=0.56, CI excluding uninformative
characters=0.45, and RI=0.70). Numbers above the lines are bootstrap values, and the numbers below the branches are Bayesian posterior probabili-
ties. The abbreviations Amy, Ce, Lau, Pad, and Pr stand for subgenera Amygdalus, Cerasus, Laurocerasus, Padus, and Prunus, respectively; and Pyg
represents the Pygeum group.


Hutchinson, 1964). Since this study suggests that
members of the two subgenera are intermixed; thus,
the evolution of the delimiting characters of these two
subgenera (particularly evergreen leaves) may have
occurred several times.
Taxa sampled from subgenus Cerasus form a
monophyletic group in the ndhF tree (Fig. 1) with two
exceptions: the inclusion of Prunus maackii, tradi-
tionally placed in subgenus Padus, and the absence of
the representatives of section Microcerasus (sensu
Rehder, 1940). In the ndhF data, the core Cerasus
group is shown to be a monophyletic lineage branch-
ing from the Amygdalus/Prunus alliance, while in the
ITS data it is poorly resolved (Figs. 1, 2).
Prunus maackii, a species from Asia that has
been placed in subgenus Padus (Rehder, 1940),
should probably be included with members of subge-
nus Cerasus, as some workers have chosen to do
(Krüssmann, 1978). Prunus maackii has a moderately
sized raceme that is more akin to subgenus Padus
(5–7 cm long, 10–20+flowers per raceme), but its
flowers have campanulate hypanthia and are compara-
tively larger, a trait more like some members of
subgenus Cerasus. Subgenus Padus typically has
Journal of Systematics and Evolution Vol. 46 No. 3 2008 330
small flowers with cup-shaped hypanthia (Rehder,
1940). Prunus maackii, which Rehder (1940) chose to
put in subgenus Padus, forms natural hybrids with
Prunus maximowiczii, a species he placed in subgenus
Cerasus that has a short, 5–6 flowered raceme.
Subgenus Cerasus section Microcerasus of
Prunus s.l. was sampled with three species (P. bes-
seyi, P. glandulosa, and P. tomentosa), which were
nested within groups composed of subgenera Amyg-
dalus and Prunus in the analyses (Figs. 1, 2). This
corroborates the data seen in Lee and Wen (2001),
Bortiri et al. (2001) and Shaw and Small (2004).
Section Microcerasus shows axillary buds in threes
along with short pedicellate flowers, a trait more like
members of subgenus Amygdalus than subgenus
Cerasus, but they do lack the typical bloom or pubes-
cence found on the fruits of species in subgenera
Prunus and Amygdalus. Lersten and Horner (2000)
noticed that the leaf crystals in several Microcerasus
species showed similarities with subgenera Prunus (=
Prunophora) and Amygdalus, and isozyme studies
revealed several members of Microcerasus that
grouped with members of subgenera Prunus and
Amygdalus (Mowrey & Werner, 1990). It also has
been shown that Microcerasus species can form
hybrids with cultivated members of subgenera Prunus
and Amygdalus, while other members of subgenus
Cerasus do not (Watkins, 1995).
The Microcerasus group has generally been
treated as part of subgenus Cerasus (Rehder, 1940;
Ghora & Panigrahi, 1995), but Focke (1894) chose to
raise the group to subgeneric status (=Untergattung)
equal to the other traditional subgenera (like Cerasus).
Krüssmann (1978) followed a similar approach by
treating this taxon as subgenus Lithocerasus with
three sections. This subgeneric treatment circum-
scribes subgenus Cerasus section Microcerasus sensu
Rehder. Our analysis does not suggest that the Micro-
cerasus group deserves subgeneric status equal to the
other group within Prunus s.l., since Microcerasus is
found in several places within the Prunus/Amygdalus
alliance, but the workers who treated this group as a
separate subgenus recognize that some of these spe-
cies do not show a close relationship with subgenus
Cerasus. Both data sets from this study echo that
assertion. Mowrey and Werner (1990) suggest that
Lithocerasus (=section Microcerasus sensu Rehder) is
not supported as a natural group, since members are
found nested in several different other groups in their
data (this is also supported by unpublished cpDNA
trnL-trnL-trnF+trnS-trnG-trnG+psbA-trnH sequences
of JS, JW and DP). More samples need to be exam-
ined to determine the phylogenetic position of other
members of section Microcerasus.
Members of these two subgenera (Prunus and
Amygdalus) are intermixed in the ITS trees (Fig. 2),
and unresolved in the ndhF tree (Fig. 1), the former
supporting the conclusion neither subgenus is mono-
phyletic and the later not refuting this claim (also see
Bortiri et al. 2001, 2002; Lee & Wen, 2001; Shaw &
Small, 2004). Kalkman (1965) points out that all the
groups (subgenera) within Prunus s.l. “are not very
sharply delimited” morphologically and the molecular
data from this study seem to support that assertion in
the Padus/Laurocerasus group. Furthermore, Watkins
(1995) also notes that hybridization is common within
Prunus s.l., even between the traditional subgenera
(Rehder, 1940). The Prunus/Amygdalus groups show
close ties with each other, including morphological
similarities, such as sulcate, bloomy fruits, furrowed
and/or rough-pitted stones, and flowers in fascicles or
umbels.
Within the Amygdalus/Prunus group, both data
sets strongly support the close relationship of Prunus
andersonii, the “desert peach” (Wilken, 1993) and P.
fremontii, the “desert apricot” (Figs. 1, 2; also see
Bortiri et al., 2001, 2002). Both are thorny shrubs (P.
fremontii can be a small tree) from dry areas in the
western United States that share puberulent, dry fruits
and small hypanthia (Munz, 1959; Wilken, 1993).
3.3 Glimpse into the diversification of tropical
members of Prunus s.l.
Subgenus Laurocerasus has been delimited by
characters generally found in tropical habitats, such as
evergreen leaves and bractless racemes (Rehder,
1940). The data from this study suggest the history of
subgenus Laurocerasus is probably more complex,
and grouping of these species into a distinct subgenus
is not supported. In our analysis, subgenus Lauro-
cerasus including Pygeum is intermixed with mem-
bers of the primarily temperate subgenus Padus (Figs.
1, 2). Subgenus Laurocerasus is disjunct across
several tropical and subtropical areas of the world
including Africa, southeast Asia, Central America,
and South America; and separate evolutionary events
may have given rise to members of this group. More
robust phylogenies based on more extensive sampling
of both taxa and characters will be required to thor-
oughly test this hypothesis. Zhang (1992) noted that
subgenus Laurocerasus is a diverse group (even with
the exclusion of Pygeum; cf. Kalkman, 1965), and he
found two disparate clades based on wood anatomy.
Lersten and Horner (2000) noted that leaf crystals in
subgenus Laurocerasus are very diverse, unlike other
WEN et al.: Prunus phylogeny

331
subgenera studied. Lee and Wen (2001) also found
several paraphyletic branches of which members of
subgenera Laurocerasus and Padus were found.
Subgenus Laurocerasus is likely a complex group of
species that does not represent a single, natural rela-
tionship. This group needs further work to understand
the diversifications among these species.
4 Conclusions
Our analysis suggests that (1) Pygeum is poly-
phyletic and its generic status is not supported; (2) the
subgeneric status of Padus and Laurocerasus is not
supported; (3) the relationships among subgenera
Prunus, Amygdalus, and Cerasus are unresolved
because of lack of information in the data sets; (4) we
need more data to more adequately test all of the
proposed or current classification schemes of Prunus;
(5) the ndhF tree is congruent with other published
cpDNA phylogenies, however, it is NOT congruent
with the ITS data presented here and published previ-
ously (Lee & Wen, 2001; Bortiri et al., 2001); and (6)
the biogeographic interactions among tropical and
temperate members in the Padus/Laurocerasus/
Maddenia alliance including Pygeum are shown to be
dynamic and complex. The diversification of Prunus
may have involved reticulate evolution, polyploidy,
and other molecular processes. Future analyses require
more chloroplast as well as single- or low-copy nu-
clear markers.
Acknowledgements This study was supported by
funds from the National Science Foundation (DEB
0515431), the John D. and Catherine T. MacArthur
Foundation, Colorado State University, the Field
Museum, and the Laboratories of Analytical Biology
of the Smithsonian Institution’s National Museum of
Natural History. Jun Wen thanks the organizers of the
Tree of Life symposium for organizing the important
meeting and for the support to attend the symposium.
References
Badenes ML, Parfitt DE. 1995. Phylogenetic relationships of
cultivated Prunus species from an analysis of chloroplast
DNA variation. Theoretical and Applied Genetics 90:
1035–1041.
Bate-Smith EC. 1961. Chromatography and taxonomy in the
Rosaceae, with special reference to Potentilla and Prunus.
Botanical Journal of the Linnean Society 58: 39–54.
Bentham G, Hooker JD. 1865. Genera plantarum. Vol. 1.
London: Reeve & Co.
Berry EW. 1930. Revision of the lower Eocene Wilcox flora of
the southeastern states, with descriptions of new species,
chiefly from Tennessee and Kentucky. U.S. Geological
Survey Professional Paper 156: 1–196.
Bertin RI. 1982. The evolution and maintenance of
andromonoecy. Evolutionary Theory 6: 25–32.
Bortiri ES, Oh H, Gao F-Y, Potter D. 2002. The phylogenetic
utility of nucleotide sequences of sorbitol 6-phosphate
dehydrogenase in Prunus (Rosaceae). American Journal of
Botany 89: 1697–1708.
Bortiri ES, Oh H, Jiang JG, Baggett S, Granger A, Weeks C,
Buckingham M, Potter D, Parfitt DE. 2001. Phylogeny
and systematics of Prunus (Rosaceae) as determined by
sequence analysis of ITS and the chloroplast trnL-trnF
spacer DNA. Systematic Botany 26: 797–807.
Browicz K. 1969. Rosaceae. In: Rechinger KH ed. Flora
Iranica. Graz: Akademische Druck und Verlagsanstalt. 66:
161–203.
Dahlgren R. 1983. General aspects of angiosperm evolution and
macrosystematics. Nordic Journal of Botany 3: 119–149.
De Candolle AP. 1825. Rosaceae. In: Prodromus Systematais
Naturalis Regni Vegetabilis. Paris: Treuttel & Würtz. 2:
525–639.
Doyle JJ, Doyle JL. 1987. A rapid DNA isolation procedure for
small quantities of fresh leaf tissue. Phytochemical
Bulletin 19: 11–15.
Farris JS, Källersjö M, Kluge AG, Bult C. 1995. Testing
significance of incongruence. Cladistics 10: 315–319.
Fernald ML. 1950. Gray’s Manual of Botany. 8th ed. New
York: American Book Co..
Focke WO. 1894. Rosaceae. In: Engler A, Prantl K eds. Die
Natürlichen Pflanzenfamilien. Leipzig: Engelmann. 3:
1–61.
Ghora C, Panigrahi G. 1995. The family Rosaceae in India. Vol.
2. Dehra Dun: Bishen Singh Mahendra Pal Singh.
Goldblatt P. 1976. Cytotaxonomic studies in the tribe
Quillajeae (Rosaceae). Annals of the Missouri Botanical
Garden 63: 200–206.
Huelsenbeck JP, Ronquist R. 2001. MrBAYES: Bayesian infer-
ence of Phylogenetic trees. Bioinformatics 17: 754–755.
Hutchinson J. 1964. The genera of flowering plants. Vol. 1.
Oxford: Clarendon Press.
Kalkman C. 1965. The Old World species of Prunus subgen.
Laurocerasus including those formerly referred to
Pygeum. Blumea 13: 1–115.
Koehne BAE. 1911. Die Gliederung von Prunus subgen.
Padus. Berlin: Verhandlungen des Botanischen Vereins
der Provinz Brandenburg. 52: 101–108.
Komarov VL. 1971. Rosaceae—Rosoideae, Prunoideae. In:
Flora of the U.S.S.R. Vol. 10 (English translation).
Washington, D.C.: Smithsonian Institution.
Krüssmann G. 1978. Manual of cultivated broad-leaved Trees
and Shrubs. Vol 3. (Pru-Z. English translation 1986)
Portland: Timber Press.
Landon JW. 1975. A new name for Osmaronia cerasiformis
(Rosaceae). Taxon 24: 200.
Lee S, Wen J. 2001. A phylogenetic analysis of Prunus and the
Amygdaloideae (Rosaceae) using ITS sequences of
nuclear ribosomal DNA. American Journal of Botany 88:
150–160.
Lersten NR, Horner HT. 2000. Calcium oxalate crystal types
and trends in their distribution patterns in leaves of Prunus
Journal of Systematics and Evolution Vol. 46 No. 3 2008 332
(Rosaceae: Prunoideae). Plant Systematics and Evolution
224: 83–96.
Linnaeus C. 1753. Species plantarum. Sweden, Stockholm.
Mai DH. 1984. Karpologische Untersuchungen der Steinkerne
fossiler und rezenter Amygdalaceae (Rosales). Feddes
Repertorium 95: 299–329.
McNeill J, Barrie FR, Burdet HM, Demoulin V, Hawksworth
DL, Marhold K, Nicolson DH, Prado J, Silva PC, Skog
JE, Wiersema JH, Turland NJ. 2006. International code of
botanical nomenclature (Vienna Code). Regnum Vegetabile
146. A.R.G. Gantner Verlag KG.
McVaugh R. 1951. A revision of the North American black
cherries (Prunus serotina Ehrh., and relatives). Brittonia 7:
279–315.
Morgan DR, Soltis DE, Robertson KR. 1994. Systematic and
evolutionary implications of rbcL sequence variation in
Rosaceae. American Journal of Botany 81: 890–903.
Mowrey BD, Werner DJ. 1990. Phylogenetic relationships
among species of Prunus as inferred by isozyme markers.
Theoretical and Applied Genetics 80: 129–133.
Munz PA. 1959. A California Flora. Los Angeles: University of
California Press.
Olmstead RG, Sweere JA. 1994. Combining data in
phylogenetic systematics: an empirical approach using
three molecular data sets in the Solanaceae. Systematic
Biology 43: 467–481.
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.
Potter DF, Eriksson T, Evans RC, Oh S, Smedmark JEE,
Morgan DR, Kerr M, Robertson KR, Arsenault M,
Dickinson TA, Campbell CS. 2007. Phylogeny and
classification of Rosaceae. Plant Systematics and
Evolution 266: 5–43.
Rehder A. 1915. Synopsis of the Chinese species of Pyrus.
Proceedings of the American Academy of Arts and
Sciences 50: 226–239.
Rehder A. 1940. Manual of cultivated trees and shrubs hardy in
North America exclusive of the subtropical and warmer
temperate regions. 2nd ed. New York: MacMillan.
Robertson KR. 1974. The genera of Rosaceae in the
southeastern United States. Journal of the Arnold
Arboretum 55: 303–332, 344–401, 611–662.
Rydberg PA. 1900. Catalogue of the flora of Montana and the
Yellowstone National Park. 1: 1–492.
Rydberg PA. 1917. Flora of the Rocky Mountains and adjacent
plains. New York: Steinman and Foltz.
Schery RW. 1972. Plants for man. 2nd ed. New Jersey: Prentice
Hall.
Small JK. 1933. Manual of the Southeastern Flora. Chapel Hill:
University of North Carolina Press.
Shaw J, Small RL. 2004. Addressing the “hardest puzzle in
American pomology”: phylogeny of Prunus sect.
Prunocerasus (Rosaceae) based on seven noncoding
chloroplast DNA sequences. American Journal of Botany
91: 985–996.
Solomon BP. 1986. Sexual allocation and andromonoecy:
resource investment in male and hermaphrodite flowers of
Solanum carolinense (Solanaceae). American Journal of
Botany 73: 1215–1221.
Stebbins GL. 1958. On the hybrid origin of the angiosperms.
Evolution 12: 267–270.
Sterling C. 1964. Comparative morphology of the carpel in the
Rosaceae. II. Amygdaloideae: Maddenia, Pygeum,
Osmaronia. American Journal of Botany 51: 354–360.
Swofford DL. 2003. PAUP*: Phylogenetic analysis using
parsimony. Version 4.0. Sunderland, MA: Sinauer
Associates.
Swofford DL, Olsen GJ, Waddell PJ, Hillis DM. 1996.
Phylogenetic inference. In: Hillis DM, Moritz C, Mable
BK eds. Molecular Systematics. 2nd ed. Sunderland, MA:
Sinauer Associates. 407–514.
de Tournefort JP. 1700. Institutiones Rei Herbariae. Paris,
France.
Watkins R. 1995. Cherry, plum, peach, apricot and almond. In:
Smartt J, Simmonds NW eds. Evolution of crop plants.
2nd ed. Burnt Mill: Longman Scientific and Technical.
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: 166–177.
Wilken D. 1993. Prunus. In: Hickman JC ed. The Jepson
manual, higher plants of California. Berkeley: University
of California Press.
Wolfe LM, Drapalik DJ. 1999. Variation in the degree of andro-
monoecy in Prunus caroliniana. Castanea 64: 259–262.
Yü T-T, Lu L-T, Ku T-C, Li C-L, Chen S-X. 1986. Rosaceae
(3), Prunoideae. In: Flora Reipublicae Popularis Sinicae.
Beijing: Science Press. 38: 1–133.
Zhang S-Y. 1992. Systematic wood anatomy of the Rosaceae.
Blumea 37: 81–158.