Content uploaded by Richard M. K. Saunders
Author content
All content in this area was uploaded by Richard M. K. Saunders on Jun 30, 2017
Content may be subject to copyright.
Phytotaxa 311 (3): 255–262
http://www.mapress.com/j/pt/
Copyright © 2017 Magnolia Press Article PHYTOTAXA
ISSN 1179-3155 (print edition)
ISSN 1179-3163 (online edition)
Accepted by Li-Bing Zhang: 12 Jun. 2017; published: 30 Jun. 2017
https://doi.org/10.11646/phytotaxa.311.3.5
255
Emended description and resurrection of Kadsura matsudae (Schisandraceae)
KENJI SUETSUGU1, TIAN-CHUAN HSU2, TSUGUTAKA TOMA3, TAKASHI MIYAKE4, RICHARD M. K.
SAUNDERS5
1Department of Biology, Graduate School of Science, Kobe University, 1-1 Rokkodai, Nada-ku, Kobe, 657-8501, Japan;
email: kenji.suetsugu@gmail.com
2Herbarium of Taiwan Forestry Research Institute, No. 53, Nanhai Rd., Taipei 100, Taiwan
3Ginowan City, Okinawa, Japan
4Faculty of Education, Gifu University, Gifu, Japan
5School of Biological Sciences, The University of Hong Kong, Pokfulam Road, Hong Kong, China
Abstract
The taxonomic identity of Kadsura matsudae is reevaluated. This taxon is often treated as a synonym of K. japonica, a
species known from Japan, Korea, and Taiwan. When studying the type materials of K. matsudae, however, we noted that
some of its morphological characters, such as the non-contiguous thecae in adjacent stamens, do not fall within the typical
variation range of K. japonica. Phylogenetic analysis suggests that although K. matsudae is retrieved within the sect. Kad-
sura clade, it is not closely related to K. japonica. We therefore propose that K. matsudae should be recognized as a distinct
species. Because the protologue of K. matsudae lacks sufficient detail with regards to stamen morphology, which is one of
the most important characteristics for identifying the species, we provide an emended description of K. matsudae based on
the holotype specimen and newly collected specimens.
Key words: Kadsura, Schisandraceae, Japan, Ryukyu Islands, Taiwan, taxonomy
Introduction
The genus Kadsura Kaempfer ex Jussieu (1810: 340) comprises 11–16 species of scandent and twining woody vines
(Saunders 1998, Lin 2002). Kadsura possesses many morphological and anatomical features that have been interpreted
as ancestral: although the flowers are unisexual, the gynoecium consists of numerous free carpels that are spirally
arranged around an elongated floral axis or torus (Saunders 1998). Each carpel is ascidiate and poorly differentiated,
possessing a small pseudostyle that is either narrow and subulate, or else slightly enlarged distally to form a subpeltate
pseudostigma (Saunders 1998).
Although the center of diversity lies in southern China, the genus is widely distributed in Asia, extending from
Sri Lanka eastwards to the Philippines, and from Java and the Lesser Sunda Islands northwards to southern Korea and
Japan (Smith 1947, Saunders 1998). So far, only one species, Kadsura japonica (Linnaeus 1753: 536). Dunal (1817:
57), has been reported from Japan (Saunders 1998). Previous studies in Taiwan, however, have recorded two species:
K. japonica from the main island, and K. philippinensis Elmer (1908: 277) from the outlying islets Ludao and Lanyu
(Li & Chaw 1996). The record of K. philippinensis was later verified as K. heteroclita Craib (1925: 28) (Lin 2002),
and a third species, K. oblongifolia Merrill (1923: 241), was more recently recognized from the western and southern
parts of the main island (Lin 2005).
One of the authors (TT) found an unknown population of Kadsura growing on Okinawa Island in the Ryukyu
Islands, Japan. After examining its morphology and comparing it to previously recorded species, we are convinced
that the plant should be treated as K. matsudae Hayata (1920: 4). Kadsura matsudae was first described from Ariko,
southern Taiwan. This taxon is often treated as a synonym of K. japonica, a species found in Japan, Korea, and Taiwan
(Saunders 1998). Although Saunders (1998) was unable to locate the holotype of K. matsudae, we located it at the
herbarium TI: examination of this specimen revealed that some of its morphological characters, such as the non-
contiguous thecae in adjacent stamens, do not fall within the typical variation range of K. japonica (Saunders 1998).
We therefore believe that K. matsudae should be treated as a distinct species.
SUETSUGU ET AL.
256 • Phytotaxa 311 (3) © 2017 Magnolia Press
During our herbarium investigations, we also found several additional localities of K. matsudae in other Ryukyu
Islands, extending its distribution range in Taiwan beyond the previous reports from southern Taiwan. Because the
protologue of K. matsudae lacks sufficient detail with regards to stamen morphology (Hayata 1920), which is one of
the most important characteristics for identifying Kadsura species (Saunders 1998), we provide an emended description
of K. matsudae based on the holotype specimen supplemented with newly collected specimens. The phylogenetic
position of K. matsudae was also investigated based on the partial regions of trnL-F and ITS.
Materials and methods
Morphological observations
In order to compare the morphology of K. matsudae with previously recorded species, we undertook a thorough
literature review and consulted specimens from the following herbaria TI, TNS, KYO, KPM, OSA and TAIF, as well
as online digitized plant specimens such as JSTOR Global Plants (http://plants.jstor.org/).
DNA barcoding
For DNA isolation, leaf materials of K. matsudae newly discovered in Okinawa Prefecture and K. japonica in Gifu
Prefecture were collected and desiccated in the field using silica gel. DNA was extracted using either: (i) Gentra Puregene
Tissue Kit (Qiagen, Germany) after the silica-dried leaves were ground to fine powder with a mortar and pestle; or
(ii) the single-step protocol developed by Thomson and Henry (1995). We amplified the partial regions of trnL-F and
ITS with KOD FX Neo DNA polymerase (Toyobo, Japan) and sequenced following published protocols (Liu et al.
2006).
The resulting sequences were submitted to GenBank (accession numbers LC228067–LC228069 for trnL-F and
LC225771–LC225773 for ITS) and compared with Kadsura japonica and other species of Kadsura in GenBank. The
sequence alignment for each locus was initially performed using MUSCLE, and then manually edited in SeaView ver.
4.4.2 (Gouy et al. 2010). The aligned sequences (ITS: 665 bp; trnL-F: 757 bp) were used to construct phylogenetic
relationships with MEGA7 (Kumar et al. 2016) using maximum likelihood (ML) methods. The best-fitting model for
each dataset was determined using the Akaike Information Criterion (AIC) in the model test implemented in MEGA7:
HKY+G+I for the ITS data, and HKY for the trnL-F data. Gaps were excluded from the analysis, and the reliability
of each branch was assessed using bootstrap analysis (500 replicates). Schisandra bicolor (trnL-F: DQ342264, ITS:
KP689681) was included as the outgroup (Liu et al. 2006; Zhang et al. 2015).
Results and Discussion
Kadsura exhibits considerable variation in androecial structure, with three main morphological types apparent; this
has been used (Smith 1947) as the basis for a sectional classification of the genus, viz. sections Cosbaea (Lemaire)
A. C. Smith (1947: 162), Kadsura (Kaempfer ex Jussieu) A. C. Smith (1947: 163), and Sarcocarpon (Blume) A. C.
Smith (1947: 163). Kadsura sect. Cosbaea, consisting of only one species (K. coccinea (Lemaire) A. C. Smith), is
characterized by free stamens that are spirally arranged around the receptacle and sterile apical appendages, whereas
species in K. sect. Kadsura and K. sect. Sarcocarpon possess compact subglobose androecia derived from the close
appression of short, fleshy stamens (Smith 1947, Saunders 1998, Liu et al. 2006). While species in K. sect. Kadsura
have stamens with lateral thecae and connectives that are broader than they are thick, so that the thecae of adjacent
stamens are contiguous, those in K. sect. Sarcocarpon possess stamens with dorsolateral thecae and the connectives
that are as broad as they are thick, so that the thecae of adjacent stamens are not contiguous (Smith 1947, Saunders
1998).
While the gross morphology of Kadsura matsudae indicates a close affinity with K. sect. Kadsura, it is easily
distinguished from other species in the section because its adjacent stamens are not strictly contiguous. Its connectives
are nevertheless broader than thick (a characteristic feature of K. sect. Kadsura), and it is furthermore noteworthy that
the apex of K. matsudae androecium is sterile, and that thecae are located on opposing sides of the connective, which
are both features of K. sect. Kadsura that are not observed in K. sect. Sarcocarpon. On this basis, K. matsudae clearly
belongs to K. sect. Kadsura.
KADSURA MATSUDAE (SCHISANDRACEAE) Phytotaxa 311 (3) © 2017 Magnolia Press • 257
In addition to the non-contiguous thecae in adjacent stamens, Kadsura matsudae can be distinguished from other
members of sect. Kadsura by reference to several morphological characters such as apocarp size (5.5–9.0 mm long vs.
10–22 mm long for K. heteroclita), pericarp thickness (very thin vs. thick for K. philippinensis), and peduncle length
(17–33 mm long vs. 30–50 mm long for K. longipedunculata Finet & Gagnepain (1905: 53)).
DNA barcoding based on trnL-F and ITS regions supported a close relationship between Kadsura matsudae and
other species in K. sect. Kadsura. The BLAST search based on the ITS sequence of K. matsudae Toma 93 resulted in
homologies as high as 669/672, 668/672, 671/672 and 670/672 bp with sequences of K. heteroclita, K. japonica, K.
philippinensis, and K. longipedunculata (GenBank accession nos. KP689696, KP689704, KP689690, and JF976708,
respectively) in GenBank. Kadsura sect. Kadsura is supported as monophyletic in the ML phylogenetic trees based
on both the ITS and trnL-F datasets (Fig. 4). The ML reconstructions based on trnL-F sequences also suggest that K.
matsudae is basal to a clade comprising the other species in K. sect. Kadsura and does not appear to be closely related
to K. japonica within sect. Kadsura.
Taxonomic treatment
Kadsura matsudae Hayata (1920: 4, as “K. matsudai”) emend. Suetsugu & T.C.Hsu (Figs. 1–4).
Typ e :—TAIWAN. Pingtung Co.: Ariko, January 1917, E. Matsuda s.n. (holotype TI!).
FIGURE 1. Holotype of Kadsura matsudae deposited in TI.—A. Habit.—B. Male flower.
Emended description:—Woody vines, glabrous. Leaves elliptic or rarely ovate-elliptic, (4–)5.5–11 cm long, (1.5–)2.5–
4.7(–5.5) cm wide, length: width ratio 1.9–2.6(–4.0); papyraceous to subcoriaceous; base acute to cuneate; apex
acute to acuminate; margins shallowly dentate or entire; primary vein variably slightly to highly impressed above,
prominent below; secondary veins 3–6 pairs per leaf, often indistinct due to presence of intersecondary veins, (slightly)
arcuate; petioles 10–26 mm long. Flowers either borne solitary in axils of leaves or in axils of fugaceous bracts,
occasionally cauliflorous; tepals pale yellow to pale green; outermost tepal ovate to triangular, 1.3–2.2 mm long,
2.2–3.0 mm wide, length:width ratio 0.6–0.9, highly reduced to 0.1–0.2 of length of largest; innermost tepal elliptic,
5.3–10.8 mm long, 3.0–5.2 mm wide, length:width ratio 1.8–2.2, reduced to 0.5–0.7 of length of largest; largest tepal
generally ovate, occasionally elliptic, 10.0–16.3 mm long, 7.5–10.9 mm wide, length:width ratio 1.2–1.6. Male flowers
SUETSUGU ET AL.
258 • Phytotaxa 311 (3) © 2017 Magnolia Press
FIGURE 2. Kadsura matsudae newly discovered in Okinawa Island.—A. Male flower.—B. Female flower.—C. Male flower with red
androecium (left) and male flower with yellow androecium (right).—D. Fruit.
KADSURA MATSUDAE (SCHISANDRACEAE) Phytotaxa 311 (3) © 2017 Magnolia Press • 259
with 9–12 tepals; androecium of ca. 40 stamens, yellow or red; anthers sessile, spirally arranged in a subglobose to
ellipsoid head, 5.2–6.6 mm in diameter, connectives broader than thick, with lateral thecae, adjacent stamens not
contiguous; peduncle 17–26 mm long. Female flowers with 10–13 tepals; gynoecium of ca. 40 carpels, 6.0–7.5 mm in
diameter; carpels 1.5–2.2 mm long, 1.0–1.6 mm wide, length:width ratio 1.2–1.8; pseudostyle broad with subpeltate
pseudostigma; peduncle 17–33 mm long. Fruit composed of ca. 30 apocarps, ripening red to dark purple, apocarps
5.5–9.0 mm long, 5.5–8.0 mm wide, length:width ratio 0.9–1.2, apocarps sessile, pericarp thin, showing the shape of
the seeds after drying; peduncle slightly elongated, 26–40 mm long. Seeds 1–3 per apocarp, reniform, 4.0–4.4 mm
long, 4.4–5.2 mm wide, length:width ratio 0.8–0.9.
FIGURE 3. Kadsura matsudae newly discovered in Okinawa Island.—A. Androecium.—B. Gynoecium. Scale bar = 2 mm
Additional specimens examined:—JAPAN. Ryukyu. Okinawa Pref., Okinawa Island, Motobu Town, Henaji,
17 October 2016, Toma 93 (OSA!); JAPAN. Ryukyu. Okinawa Pref., Okinawa Island, Kunigami Village, Benoki, 18
October 2016, Toma 98 (OSA!); Ryukyu. Okinawa Pref., Okinawa Island, Motobu Town, Henaji, 9 October 2016,
Toma 101-01 (OSA!); Ryukyu. Okinawa Pref., Okinawa Island, Motobu Town, Henaji, 20 October 2016, Toma 101-
02 (OSA!); JAPAN. Ryukyu. Okinawa Pref., Okinawa Island, Kunigami Village, Jashiki, 18 October 2016, Tom a
103-01 (OSA!); JAPAN. Ryukyu. Okinawa Pref., Okinawa Island, Kunigami Village, Jashiki, 21 October 2016,
Toma 103-02 (OSA!); JAPAN. Ryukyu. Okinawa Pref., Okinawa Island, Kunigami Village, Uka, 21 November 2016,
Toma 110 (OSA!); JAPAN. Ryukyu. Okinawa Pref., Okinawa Island, Kunigami Village, Benoki, 18 October 2016,
Toma 128 (OSA!); JAPAN. Ryukyu. Okinawa Pref., Okinawa Island, Kunigami Village, Uka, 20 October 2016, Tom a
129-1 (OSA!); JAPAN. Ryukyu. Okinawa Pref., Okinawa Island, Kunigami Village, Uka, 28 Nobember 2016, Tom a
129-2 (OSA!); JAPAN. Ryukyu. Okinawa Pref., Okinawa Island, Kunigami Village, Sate, 28 November 2016, Tom a
131 (OSA!); JAPAN. Ryukyu. Okinawa Pref., Okinawa Island, Nago City, Mt. Katsuudake, 21 September 1940, Y.
Kimura and Y. Hurusawa s.n. (TI!); JAPAN. Ryukyu. Okinawa Pref., Iriomote Island, Taketomi Town Shirahama,
30 September 2003, K. Yasuda 2048 (KYO!); JAPAN. Ryukyu. Okinawa Pref., Ishigaki Island, Kabira, 2 November
1959, S. Hatsushima 206061 (KAG!); JAPAN. Ryukyu. Okinawa Pref., Ishigaki Island, Mt. Suri, 21 August 1919,
S. Kawagoe 4036 (KAG!); JAPAN. Ryukyu. Okinawa Pref., Iriomote, Shirahama, 19 August 2002, J. C. Wang et
al. 11806 (TAIF!); JAPAN. Ryukyu. Okinawa Pref., Kumejima Island, Shimajiri, near Mt. Aradake, 30 September
2005, G. Kokubugata 7031 (TNS!). TAIWAN. Hsinchu Co.: Hsinkuang, 11 August 2013, P. F. Lu 25977 (TAIF!);
Litungshan, 19 September 2000, S. W. Chung 3063 (TAIF!); Simakusi Giant Trees, 27 August 2011, P. F. Lu 22720
(TAIF!); Szmakusz, 20 July 2002, P. F. Lu 4459 (TAIF!). Hualien Co.: Kiraikei-Asahi, 8 August 1919, E. Matuda 210
(TI!); Kiraikei-Asahi, 8 August 1919, E. Matuda 275 (TI!); Chilai, 4 August 1919, E. Matsuda s.n. (TAIF!); Hopping
logging trail, 28 July 1993, J. C. Wang et al. 8638 (TAIF!). Ilan Co.: Ssuchi Village, 6 August 2014, C. F. Chen 5315
SUETSUGU ET AL.
260 • Phytotaxa 311 (3) © 2017 Magnolia Press
FIGURE 4. Molecular maximum likelihood phylogenetic analyses of Kadsura species. A. ITS data (highest log likelihood -1310.5853).
B. trnL-F data (highest log likelihood -1234.9454). Numbers above the branches represent bootstrap values ≥50%.
KADSURA MATSUDAE (SCHISANDRACEAE) Phytotaxa 311 (3) © 2017 Magnolia Press • 261
(TAIF!); Nanshan-Kiretei, 20 August 1969, Yamazaki et al. 555 (KAG!). New Taipei City: Urai, 15 July 1931, T.
Suzuki 4716 (TI!). Taipei City: Sendyosan, T. Suzuki 5396 (TI!); Mt. Tsaikungkeng, 14 July 1999, S. C. Wu et al. 875
(TAIF!). Taitung Co.: Pinan, 26 November 1896, T. Makino s.n (TI!); Kotosho, July 1912, T. Kawakami & S. Sasaki
s.n. (TI!); Botel Tobago, 23 August 1962, C. E. Chang 3177 (KYO1); Botel Tobago, 28 Aug 1985, C. E. Chang 17129
(TI!); Lanyu, 4–17 August 1968, G. Ikeda 2329 (KAG!); Mt. Hungtou, 19 July 2013, C. F. Chen 4668 (TAIF!).
Distribution and phenology:—Our field surveys and herbarium investigation showed that K. matsudae is
distributed in many islands in central and southern Ryukyu and Taiwan. In addition, our field and herbarium surveys
failed to find any K. japonica flowering individuals in some Ryukyu Islands, such as Okinawa Island. This suggests
that K. matsudae may possibly have been mistaken for the more widespread species K. japonica, which has a similar
gross morphology. Extensive surveys during the flowering season are urgently required in order to reveal the precise
distribution of K. matsudae in the Ryukyu Islands and Taiwan. Flowering was observed from late-July to mid-
November, and fruiting from mid-October to mid-February.
GenBank accession no:—Toma 9 3 : LC228067 (trnL-F), LC225771 (ITS). Toma 128: LC228068 (trnL-F),
LC225772 (ITS).
Acknowledgements
We thank the curators of KYO, OSA, TAIF, TI, and TNS for herbarium access. We also express our appreciation to
Eizi Suzuki, Shigeko Fukumoto, Kumi Hamasaki, Wataru Ohnishi and Masashi Yokogawa for their help with the
examination of herbarium specimens. This study was partly supported by a grant-in-aid by the Japan Society for the
Promotion of Science (15K18470).
References
Craib, W.G. (1925) Flora Siamensis enumeratio. Vol. 1. Siam Society, Bangkok.
Dunal, M.F. (1817) Monographie de la famille des Anonacées. Treuttel & Wiirtz, Paris.
Elmer, A.D.E. (1908) A century of new plants. Leaflets of Philippine Botany 1: 272–359.
Finet, A. & Gagnepain, F. (1905) Contributions à la flore de l’Asie orientale. Bulletin de la Société Botanique de France, Mémoires 52,
Mem. 4: 1–170 + pi. 1–20.
Gouy, M., Guindon, S. & Gascuel, O. (2010) SeaView version 4: a multiplatform graphical user interface for sequence alignment and
phylogenetic tree building. Molecular Biology and Evolution 27: 221–224.
https://doi.org/10.1093/molbev/msp259
de Jussieu, A.L. (1810) Suite des observations sur quelques genres de plantes de Loureiro, accompagnées de notes sur ceux qui composent
la famille des Anonées. Annales du Muséum National d’Histoire Naturelle, Paris 16: 338–340.
Hayata, B. (1920) Icones plantarum formosanarum. Vol. 9. Government of Formosa, Taihoku.
Kumar, S., Stecher, G. & Tamura, K. (2016) MEGA7: Molecular Evolutionary Genetics Analysis version 7.0 for bigger datasets. Molecular
Biology and Evolution 33: 1870–1874.
https://doi.org/10.1093/molbev/msw054
Li, H.L. & Chaw, S.M. (1996) Schisandraceae. In: Huang, T.C. (Ed.) Flora of Taiwan. Vol. 2. 2nd edition. Department of Botany, National
Taiwan University, Taipei, pp. 423–426.
Lin, Q. (2002) Taxonomic notes on some species of Kadsura (Schisandraceae). Bulletin of Botanical Research 22: 399–411.
Lin, Q., Duan, L.D. & Yao, B.F. (2005) Notes on three species of the genus Kadsura Juss. Acta Phytotaxonomica Sinica 43: 567–570.
https://doi.org/10.1360/aps030102
Linnaeus, C. (1753) Species plantarum. Vol. 1. Impensis Laurentii Salvii, Stockholm.
Liu, Z., Hao, G., Luo, Y.B., Thien, L.B., Rosso, S.W., Lu, A.M. & Chen, Z.D. (2006) Phylogeny and androecial evolution in Schisandraceae,
inferred from sequences of nuclear ribosomal DNA ITS and chloroplast DNA trn LF regions. International Journal of Plant Sciences
167: 539–550.
https://doi.org/10.1086/501476
Merrill, E.D. (1923) Diagnoses of Hainan plants, II. The Philippine Journal of Science 23: 237–268.
Saunders, R.M.K. (1998) Monograph of Kadsura (Schisandraceae). Systematic Botany Monographs 54: 1–106.
https://doi.org/10.2307/25096646
SUETSUGU ET AL.
262 • Phytotaxa 311 (3) © 2017 Magnolia Press
Smith, A.C. (1947) The families Illiciaceae and Schisandraceae. Sargentia 7: 1–224.
Thomson, D. & Henry, R. (1995) Single-step protocol for preparation of plant tissue for analysis by PCR. Biotechniques 19: 394–397.
Zhang, J., Chen, M., Dong, X., Lin, R., Fan, J. & Chen, Z. (2015) Evaluation of four commonly used DNA barcoding loci for Chinese
medicinal plants of the family Schisandraceae. PLoS ONE 10: e0125574.
https://doi.org/10.1371/journal.pone.0125574