Incompatibility and Pollination of Distylous Psychotria serpens (Rubiaceae) in the Ryukyu Islands, Japan

2016·14.3 MB·English

by  Sugawara, Takashi

The Japanese Society for Plant Systematics
ISSN 1346-7565 Acta Phytotax. Geobot. 67 (1): 37–45 (2016)
Incompatibility and Pollination of Distylous Psychotria serpens
(Rubiaceae) in the Ryukyu Islands, Japan
tAkAshi sugAwArA1,*, kentA wAtAnAbe2, tokushiro tAkAso3,
MitsutAke tAbAtA4 And AkirA shiMizu5
1Makino Herbarium, Graduate School of Science and Engineering, Tokyo Metropolitan University, 1-1 Minami-Oh-
sawa, Hachioji, Tokyo, 192-0397, Japan. *[email protected] (author for correspondence); 2Okinawa National College
of Technology, 95 Henoko, Nago, Okinawa, 905-2192, Japan; 3Tropical Biosphere Research Center, University of the
Ryukyus, 1 Nishihara, Okinawa, 903-0213, Japan; 4Asani-shinmachi, Amami, Kagoshima, 894-0041, Japan;
5Department of Biological Science, Graduate School of Science and Engineering, Tokyo Metropolitan University, 1-1
Minami-Ohsawa, Hachioji, Tokyo, 192-0397, Japan
Psychotria serpens (Rubiaceae) is usually distylous in floral morphology, but uncertainty still remains
as to whether the two morphs of the species have self- and intramorph-incompatibility system and also
on the efficiency of pollinators. We examined self- and intramorph-incompatibility as well as efficiency
in flower visitors to carry out pollination in natural populations. Pollination experiments indicated that
P. serpens is self- and intramorph-incompatible, as in other typical distylous species. Flowers of the two
morphs produced nectar from the base of the corolla tube and attracted various insects, such as short-
tongued wasps, bees, hoverflies, butterflies, etc. Among the flower visitors, the short-tongued wasps,
bees and hoverflies were considered to be efficient pollinators for P. serpens in the Ryukyu Islands.
Key words: Distyly, fruit set, incompatibility, nectar, pollinator, Psychotria serpens, Rubiaceae, wasp
Distyly is a genetically controlled floral di- es in floral polymorphism (Charlesworth &
morphism characterized by long (L)-styled and Charlesworth 1979, Barrett 1992, Lloyd & Webb
short (S)-styled morphs. Distylous flowers usu- 1992).
ally show self- and intramorph-incompatibility Psychotria L. (Rubiaceae) is a woody genus
(Ganders 1979, Barrett 1992, Naiki 2012). How- containing more than 1800 species. It occurs
ever, it is known in some species that the two widely in tropical to subtropical regions world-
morphs are self-compatible (Bawa & Beach 1983, wide (Mabberley 2008). Five species of Psy-
Barrett 1992, Riveros et al. 1995, Rodriges & chotria are in Japan. Among them, Psychotria
Consolaro 2013), or functionally dioecious (Bar- serpens L. is a perennial liana distributed widely
rett & Richards 1990, Sugawara et al. 2011, Wata- from southwestern Honshu and Shikoku through
nabe et al. 2014a), although their flowers are di- the Ryukyu Islands in Japan to Taiwan, south-
stylous in floral morphology. The self- and intra- eastern China and Indochina (Yamazaki 1993,
morph- incompatibility is of basic importance not Yang 1998, Tao & Taylor 2011). In a recent study,
only for maintaining the distylous breeding sys- it was reported that P. serpens was distylous in
tem but also for considering evolutionary chang- floral morphology, and rarely possessed homo-
NII-Electronic Library Service
The Japanese Society for Plant Systematics
38 Acta Phytotax. Geobot. Vol. 67
stylous morphs together with two distylous firming reciprocal pollen transfer are necessary.
morphs within a population (Sugawara et al. In this study, we aimed 1) to confirm the compat-
2013). However, uncertainty still remains as to ibility system in self- and intramorph-pollination
whether the species shows incompatibility in in P. serpens, 2) to identify the effective pollina-
self- and intramorph-pollination. Meanwhile in tors and 3) to compare the reproductive nature
Psychotria. boninensis Nakai, a species endemic with species in adjacent areas.
to the Bonin (Ogasawara) Islands, Japan, and
closely related to P. serpens (Watanabe & Suga- Materials and Methods
wara 2015), it was recently reported that its two
floral morphs showed self- and intramorph-in- Study site and species
compatibility, and that the female reproductive Field investigations were carried out at three
potential of the S-styled morph declined func- populations located in southwestern Japan: a pop-
tionally (Kondo et al. 2007, Sugawara et al. 2014). ulation at Akasaki on Amami-Oshima Island
Moreover, P. homalosperma A. Gray, a small tree (28˚24´N, 129˚29´E, 160 m a.s.l.), Kagoshima
endemic to the Bonin Island, shows partial intra- Prefecture; at population at Mount Nago-dake
morph-compatibility within the S-styled morphs (26˚35´N, 128˚00´E; and a population at Mount
(Watanabe et al. 2014b). Therefore, self- and in- Oppa-dake (26˚40´N, 127˚57´E) on Okinawa Is-
tramorph-incompatibility is presumed to some- land, Okinawa Prefecture.
what differ among the species of Psychotria. Psychotria serpens L. is a woody liana, usu-
To understand not only the reproductive na- ally at sunny forest edges and frequently climb-
ture but also the modification of morphological ing trees. At the study sites above, the plants oc-
and mating patterns in populations of distylous curred at the forest edge along roadsides and were
species, it is important to clarify the pollination mostly climbing on trees. Anthesis is usually
system confirming reciprocal pollen transfer be- from May to July at the study sites.
tween individuals of the two morphs (Ganders
1979, Barrett & Richards 1990). Previously, Gan- Pollination experiments
ders (1979) noted that distylous species were pol- Artificial pollination experiments were con-
linated mostly by insects and occasionally by ducted from 3 to 5 July, 2013, to determine the
hummingbirds. Although there have been inves- reproductive potential of L- and S-styled morphs.
tigations of pollination systems of Psychotria in We examined seed set following three treatments
neotropical forests (Castro & Oliveira 2002, Cas- on flowers from three marked individuals of each
tro & Araujo 2004, Ramos 2006, Rodriges & morph in a natural population at Akasaki on
Consolaro 2013), few investigations have been Amami-Oshima: 1) hand self-pollination with
conducted on the species of Psychotria in eastern pollen grains from the same individual; 2) intra-
Asia; detailed information is available only for P. morph-pollination with pollen grains from other
rubra (Lour.) Poir., a functionally dioecious individuals of the same morph; 3) intermorph-
shrub widely distributed in the Ryukyu Islands pollination with pollen grains from a different
and Taiwan (Watanabe et al. 2014a, Watanabe & morph. In this case, individuals of the same
Sugawara 2015). Little information is available morph were more than 10 m apart, while two in-
for P. serpens (Kato 2000). The two species often dividuals of different morphs were 2 to 5 m apart.
occur together in the same area, or in adjacent ar- Pollen for self- and cross-pollination was ran-
eas in the Ryukyu Islands. domly collected from flowers of the three indi-
To understand the reproductive nature of viduals, each of which was used for checking
morphologically distylous Psychotria serpens, fruit production. Details on number of flowers
further investigation on the incompatibility sys- examined in each individual are indicated in Ta-
tem of the two morphs in self- and intramorph- ble 1. In addition, we examined 4) open pollina-
pollination and on the pollination system con- tion of flowers on four individuals, including the
NII-Electronic Library Service
The Japanese Society for Plant Systematics
February 2016 Sugawara &al.–Incompatibility and pollination of Psychotria serpens 39
above three of each morph in the same study site. logical Systematics, Department of Biological
To exclude flower visitors in treatments 1) – Science, Tokyo Metropolitan University.
3), inflorescences were covered with Kulpack
(Matsunami Glass, Ltd.) before anthesis and re- Statistical analyses
covered after the treatments. On 31 July and 1 All statistical tests were performed using R
August 2013, fruit set percentage (number of ver. 3.2.0 software (R Core Team 2015). Fisher’s
fruits / number of flowers examined) was deter- exact test was performed for comparison of fruit
mined. set between L- and S-styled morphs, and Welch’s
t-test for comparison of nectar volume between
Examination of nectar volume the two morphs.
Nectar volume was determined in flowers
from three individuals of each morph in the Results
Nago-dake population on Okinawa Island from
19 to 21 July 2014. The flowers open in the morn- Pollination experiments
ing and last only a single daytime; the corolla Self-pollinated flowers of both morphs never
tube usually falls before night. Our preliminary set fruit (Table 1). In artificial cross-pollination
observations showed that the nectar secreted experiments between the same morphs (intra-
from the base of corolla tube attained maximum morph-pollination), most of the S-styled morph
volume in the evening. Before anthesis, the inflo- flowers did not set fruit, and only one flower set
rescence was bagged with Kulpack (Matsunami fruit (6.3%). Likewise, none of the L-styled
Glass Ind., Ltd.) to exclude flower visitors. When morph flowers set fruit (Table 1).
the flowers opened during fine weather, nectar In cross-pollination experiments between the
was removed from the flower with 1-µL micro- different morphs (inter-morph pollination), both
capillary tubes (EM minicaps, Hirschmann Lab- the S- and L-styled morph flowers produced
oratory, Germany) during the period from 17:00 fruits at a rate of more than 78% (Table 1). Be-
to 18:00. Volume was determined based on the tween the two morphs, no significant differences
length of the liquid in the tube. were found in fruit set percentage (Fisher’s exact
test, P > 0.05; Table 1).
Observation of flower visitors In open pollination, both L- and S-styled
As stated above, the flowers of Psychotria flowers set fruit at a rate of more than 40%. Al-
serpens open in the morning and last only a sin- though fruit set percentage tended to be greater in
gle day. We observed flower visitors intermittent- the L-styled flowers than in the S-styled ones,
ly for about 5 hours from 7:30 to 17:00 on 23, 26, there were no significant differences between the
27 June 2007 and 21 June 2010 in the Akasaki two flowers (Fisher’s exact test, P > 0.05; Table 1).
population; on 3, 4 July 2010 and 18, 28 June 2011
in the Nago-dake population; and on 23, 30 June Nectar volume
2011 in the Oppa-dake population. The insects Flowers of Psychotria serpens open in the
visiting the flowers were recorded and some were early morning and flowering lasts for one day.
caught for identification. Their feeding behavior The flowers secrete nectar from the base of the
on the flowers was also observed carefully. The corolla tube. Nectar volume varies considerably
frequency of insect visits at each site was tenta- among flowers and even between morphs, but no
tively classified into three categories based on the significant differences were found between the
number of insects during the total observation two morphs in nectar volume (Welch’s t-test, P >
times: frequent visits (+++, more than 10 times), 0.05; Table 2).
occasional visits (++, 4 to 9 times) and rare visits
(+, 1 to 3 times). The insects collected at the study Flower visitors
sites will be deposited in the Laboratory of Zoo- The insects visiting the flowers of the two dif-
NII-Electronic Library Service
The Japanese Society for Plant Systematics
40 Acta Phytotax. Geobot. Vol. 67
tAble 1. Pollination experiments performed on the L- and S-styled flowers of Psychotria serpens on Amami-oshima Island in
2013.
Treatment No. of individuals No. of flowers No. of flowers Fruit set %
Morph examined examined fruited
Self-pollination
L1 2 0
L2 3 0
L3 1 0
Total L 3 6 0 0
S1 6 0
S2 7 0
S3 5 0
Total S 3 18 0 0
Probability* 1
Intramorph cross-pollination (2009)
L1 3 0
L2 8 0
L3 4 0
Total L 3 15 0 0
S1 4 0
S2 10 1
S3 2 0
Total S 3 16 1 6.3
Probability* 0.516
Intermorph cross-pollination (2009)
L1 3 3
L2 11 7
L3 5 5
Total L 3 19 15 78.9
S1 2 1
S2 8 7
S3 10 9
Total S 3 20 17 85.0
Probability* 0.695
Open pollination
L1 207 116
L2 196 65
L3 220 86
L4 141 88
Total L 4 764 355 46.5
S1 192 85
S2 170 52
S3 273 129
S4 163 65
Total S 4 798 331 41.5
Probability* 0.053
*The statistic difference between the two floral morphs in fruit set percentage was examined with Fisher’s exact test.
ferent morphs of Psychotria serpens are summa- hoverflies (Ligyra and Milesia), butterflies
rized in Table 3. Various insects, including short- (Nacadula, Idenopsis, Papilio, etc.) and beetles
tongued wasps (Scolia, Batozonellus, Sphex, (Chlorophorus and Protaetia) visited both S-and
etc.), bees (Lasioglossum, Amegilla, Apis, etc.), L-styled flowers in all three populations on Oki-
NII-Electronic Library Service
The Japanese Society for Plant Systematics
February 2016 Sugawara &al.–Incompatibility and pollination of Psychotria serpens 41
tAble 2. Nectar volume of the L- and S-styled flowers of Psychotria serpens in the Nago-dake population on Okinawa Island
in 2014.
Morph No. of individuals No. of flowers Nectar volume (µL)
examined examined Range Mean ± SD
L-styled 3 26 0.7 - 3.7 2.10 ± 0.74
S-styled 3 43 0.5 - 3.4 1.88 ± 0.57
Probability* 0.192
*The statistic difference between the two floral morphs in nectar volume was examined with Welch’s t-test.
nawa and Amami-Oshima Islands, and they usu- set fruit. Compatibility among morphs exhibited
ally fed on nectar secreted from the corolla tube by the species in our study are common to other
(Table 3). We were unable to find major differ- distylous species of Psychotria in neotropical for-
ences in flower visitors among the populations, ests (Bawa & Beach 1983, Castro & Oliveira
although the number of wasps and bees recorded 2002). It was reported that P. homalosperma, en-
was remarkably smaller in the Oppa-dake popu- demic to the Bonin Islands, was partially capable
lation, perhaps due to differences in the study of fruiting after intramorph pollination of the S-
site. styled morph (Watanabe et al. 2014b). Such intra-
Among the flower visitors, short-tongued morph compatibility is also known in P. nuda in
wasps and small bees were the most frequent vis- Brazil (Castro & Araujo 2004), but seems to be
itors. They actively sucked nectar from the co- rare in distylous plants (Pailler & Thompson
rolla tube. The head of these insects often con- 1997).
tacted the anthers or stigma while they were Distylous breeding systems are generally
sucking nectar (Fig. 1A, B, C), and usually bore known to promote intermorph-pollination (Gan-
pollen on their hairy body. Hoverflies also sucked ders 1979, Barrett 1992). In open pollination of
nectar on the obliquely oriented hairs around the Psychotria serpens, L-styled flowers tend to pro-
entrance of the corolla tube and occasionally duce more fruit than S-styled ones. Greater fruit
their head came in contact with the stigma and set percentage in L-styled flower appears to indi-
anthers. Some butterflies with long legs and a cate asymmetrical pollen transfer, that is, pollen
slender proboscis occasionally visited the flowers removal from the high-level anthers of the S-
to suck nectar, but their head usually did not con- styled morph and transfer to the high-level stigma
tact the anthers (Fig. 1D). Beetles rarely visited, of the L-styled morph is increased while con-
but as they moved about on an inflorescence, trasting transfer in the opposite direction is re-
their bodies came in contact with the anthers and duced. This asymmetrical pollen transfer is com-
stigmas. mon among distylous species, as suggested by
Ganders (1979), Dulberger (1992), and Stone
Discussion (1995).
The volume of nectar secreted by the disty-
Our observations indicate that Psychotria lous flowers of Psychotria serpens is about 2 µL.
serpens exhibits self- and intramorph-incompati- It is not so much compared with about 12 µL in
bility similar to other typical distylous species some species of Psychotria including P. nuda,
(Ganders 1979, Barrett & Richards 1990, Barrett which is pollinated by hummingbirds (Castro &
1992). Besides being preferentially self- and in- Araujo 2004) In our survey, short-tongued wasps,
tramorph-incompatible, one S-styled flower pro- small bees, hoverflies, and butterflies frequently
duced fruit after intramorph cross-pollination. It visited the flowers of P. serpens to obtain nectar.
is unclear as to the reason for fruiting in one flow- Previously, Kato (2000) also reported visits by
er after intramorph cross-pollination, since other small bees (Lasioglossum subopacum) on the
flowers of the same individual observed did not flowers of P. serpens. Among the flower visitors,
NII-Electronic Library Service
The Japanese Society for Plant Systematics
42 Acta Phytotax. Geobot. Vol. 67
tAble 3. Flower visitors observed on the L- and S-styled morphs in three populations of Psychotria serpens on Okinawa and
Amami-Oshima Islands in 2007, 2010 and 2011.
Flower visitors Population
Order Family Species Akasaki Nago-dake Oppa-dake
(2007, 2010) (2010, 2011) (2011)
Hymenoptera
(Wasps)
Scoliidae Scolia kuroiwae +++ +++ Nectar
Scolia fascinate ++ Nectar
Megacampsomeris mojiensis ++ ++ Nectar
Pompilidae Hemipepsis amamiensis ++ Nectar
Batozonellus maculifrons +++ +++ Nectar
Eumenidae Okinawepipona kogimai ++ Nectar
Anterhynchium flavomarginatum ++ ++ Nectar
Stenodynerus kusigematii ++ Nectar
Vespidae Vespula shidai ++ Nectar
Sphecidae Sphex diabolicus +++ +++ Nectar
Isodontia nigella ++ ++ Nectar
Philanthidae Cerceris okumurai +++ Nectar
(Bees)
Colletidae Hylaeus insularum ++ ++ Nectar
Andrenidae Panurginus crawfordi + Nectar
Halictidae Lasioglossum subopacum ++ Nectar
Lasioglossum sp. ++ ++ + Nectar
Anthophoridae Amegilla senahai ++ ++ Nectar
Amegilla florea ++ Nectar
Ceratina okinawana ++ Nectar
Thyreus takaonis ++ Nectar
Apidae Apis cerana ++ Nectar/Pollen
Apis mellifera + +++ Nectar/Pollen
Diptera (Hoverflies)
Bombyliidae Ligyra tantalus ++ ++ Nectar
Syrphidae Milesia oshimaensis ++ Nectar
Lepidoptera(Butterflies)
Lycaenidae Nacaduba kurava ++ + ++ Nectar
Euchrysops cnejus ++ Nectar
Pieridae Eurema hecabe ++ + Nectar
Appias paulina + Nectar
Nymphalidae Argyreus hyperbius + + Nectar
Ypthima riukiuana + + Nectar
Libytheridae Libythea celtis + Nectar
Danaidae Parantica sita ++ Nectar
Ideopsis similis ++ ++ Nectar
Papilionidae Papilio bianor + Nectar
Geometridae Papilio bianor ++ Nectar
Coleoptera (Beetles) Milionia zonea
Cerambycidae Chlorophorus quinquefasciatus + Nectar
Scarabaeidae Protaetia exasperata ++ + Nectar/Pollen
+++. Frequently visited; ++, occasionally visited; +, rarely visited.
NII-Electronic Library Service
The Japanese Society for Plant Systematics
February 2016 Sugawara &al.–Incompatibility and pollination of Psychotria serpens 43
A B
C D
Fig. 1. Flower visitors in the Akasaki (A, C, D) and Nago-dake (B) populations of Psychotria serpens. A. Batozonellus macu-
lifrons (Pompilidae), sucking nectar from flower. B. Isodontia nigella (Sphecidae), sucking nectar. C. Lasioglossum sp.
(Halictidae), sucking nectar. D. Ideopsis similis (Danaidae), sucking nectar.
short-tongued wasps and bees appear to be effi- tropical forests, bees have been regarded as the
cient pollinators of Psychotria serpens, since most common pollinators, together with other in-
their tongues are generally shorter (Proctor et al. sects such as butterflies (Castro & Oliveira 2002,
1996) and thus their mouthparts come in contact Ramos 2006), although hummingbirds (Castro &
with the anthers and stigmas. Hoverflies some- Araujo 2004) and flies (Rodrigues & Consolaro
times visit the flowers to obtain nectar and in do- 2013) also pollinate a few species. It is likely that
ing so occasionally contact the stigma and an- variation in the type of pollinators occurs among
thers. They may also be effective pollinators of P. congeneric plant species. In the present study, the
serpens. Some butterflies occasionally visit the flowers of P. serpens were mainly pollinated by
flowers for sucking, but their head usually does short-tongued wasps, bees, and hoverflies, among
not contact the anthers (Fig. 1D). They may there- which pollination by short-tongued wasps may be
fore be inefficient pollinators. rare in the genus, especially in neotropical for-
In the species of Psychotria examined in neo- ests.
NII-Electronic Library Service
The Japanese Society for Plant Systematics
44 Acta Phytotax. Geobot. Vol. 67
In the Ryukyu Islands, Psychotria serpens Dulberger, R. 1992. Floral dimorphism and their func-
usually occurs near P. rubra, a shrub with a dioe- tional significance in the heterostylous syndrome. In:
cious breeding system derived from distyly Barrett, S. C. H. (ed.), Evolution and Function of Het-
erostyly, pp. 41–84. Springer, Berlin.
(Watanabe et al. 2014a). The two species are sim-
Ganders, F. R. 1979. The biology of heterostyly. New Zea-
ilar in flower size and structure as well as in their
land J. Bot. 17: 607–635.
flowering period. P. rubra is much less frequently Kato, M. 2000. Anthophilous insect community and
visited by pollinators than P. serpens (K. Wata- plant-pollinator interactions on Amami Islands in the
nabe per. observation). Moreover, P. rubra is Ryukyu Archipelago, Japan. Contr. Biol. Lab. Kyoto
Univ. 29: 157–252.
mainly pollinated by flies and short-tongued
Kondo, Y., M. Nishide, K. Watanabe & T. Sugawara.
wasps (Watanabe et al. 2014a), whereas P. ser-
2007. Floral dimorphism in Psychotria boninensis
pens is visited and pollinated by a greater variety Nakai (Rubiaceae) endemic to the Bonin (Ogasawara)
of insects, such as short-tongued wasps, bees, and Islands. J. Jap. Bot. 82: 251–258.
hoverflies. The differences appear to be due to Lloyd, D. G. & C. J. Webb. 1992. The evolution of hetero-
styly. In: Barrett, S. C. H. (ed.), Evolution and Func-
differences in growth habitat between the two
tion of Heterostyly, pp. 41–84. Springer, Berlin.
species; P. rubra usually occurs in darker forests
Mabberley, D. J. 2008. Mabberley’s Plant-Book, Third
and P. serpens is primarily on the sunny edges of Ed. Cambridge University Press, Cambridge.
forests. Since it is still unclear as to the relation- Naiki, A. 2012. Heterostyly and the possibility of its
ship of breeding systems (i.e., heterostyly or dio- breakdown by polyploidization. Pl. Spec. Biol. 27:
3–29.
ecy) with different pollinators between these two
Pailler, T. & J. D. Thompson. 1997. Distyly and variation
species, further studies are necessary to charac-
in heteromorphic incompatibility in Gaertnera vagi-
terize the breeding systems of these species. nata (Rubiaceae) endemic to La Reunion Island.
Amer. J. Bot. 84: 315–327.
We thank Mr. M. Nakaji, Ms. N. Goto, and Dr. E. Oguri Proctor, M., P. Yeo & A. Lack. 1996. The Natural History
for their help in the field, and Drs. A. Iguchi and K. Hata of Pollination. Timber Press, Portland.
for their support in statistical analysis. This study was R Core Team. 2015. R: A language and environment for
partly supported by the Collaborative Research of the statistical computing. R Foundation for Statistical
Tropical Biosphere Research Center, University of the Computing, Vienna. <http://www.R-project.org/>
Ryukyus, to T. Sugawara and by JSPS KAKEKNHI [accessed xx,2015]
Grant No. 26840130 to K. Watanabe. Ramos, F. N. 2006. Floral visitors and pollination of Psy-
chotria tenuinervis (Rubiaceae): distance from the
anthropogenic and nature edges of an Atlantic forest
fragment. Biotropica 38: 383–389.
References
Riveros, G. M., O.R. Barría & A. M. P. Humaña. 1995.
Self-compatibility in distylous Hedyotis salzmanii
Barrett, S. C. H. 1992. Heterostylous geneic polymor- (Rubiaceae). Plant Syst. Evol. 194: 1–8.
phism: model systems for evolutionary analysis. In: Rodriges, E. B. & H. Consolaro. 2013. Atypical distyly in
Barrett, S. C. H. (ed.), Evolution and Function of Het- Psychotria goyazensis Mull. Arg. (Rubiaceae), an in-
erostyly, pp. 1–29. Springer, Berlin. tramorph self-compatible species. Acta Bot. Bras. 27:
Barrett, S. C. H. & J. H. Richards. 1990. Heterostyly in 155–161.
tropical plants. Mem. New York Bot. Gard. 55: 35–61. Stone, J. L. 1995. Pollen donation patterns in a tropical
Bawa, K. S. & J. H. Beach. 1983. Self-incompatibility distylous shrub (Psychotria suerrensis; Rubiaceae).
systems in the Rubiaceae of a tropical lowland wet Amer. J. Bot. 82: 1390–1398.
forest. Am. J. Bot. 70: 1281–1288. Sugawara, T., N. Tanaka & J. Murata. 2011. Dioecy de-
Castro, C. C. & P. E. Oliveira. 2002. Pollination biology rived from distyly in Morinda villosa Hook. f. (Ru-
of distylous Rubiaceae in the Atlantic rain forest, SE biaceae) occurring in Hukaung Valley, Kachin State,
Brazil. Plant Biol. 4: 640–646. Myanmar. J. Jap. Bot. 86: 9–14.
Castro, C. C. & A. C. Araujo. 2004. Distyly and sequen- Sugawara, T., K. Watanabe & M. Tabata. 2013. Distyly in
tial pollinators of Psychotria nuda (Rubiaceae) in the Psychotria serpens (Rubiaceae) occurring in the
Atlantic rain forest, Brazil. Pl. Syst. Evol. 244: 131– Ryukyu Islands, Japan. Acta Phytotax. Geobot.
139. 64:127–136.
Charlesworth, D. & B. Charlesworth. 1979. A model for Sugawara, T., M. Yumoto, S. Tsuneki & K. Watanabe.
the evolution of distyly. Amer. Nat. 114: 467–498. 2014. Incompatibility and reproductive output in di-
NII-Electronic Library Service
The Japanese Society for Plant Systematics
February 2016 Sugawara &al.–Incompatibility and pollination of Psychotria serpens 45
stylous Psychotria boninensis (Rubiaceae), endemic ceae), an endemic plant of the oceanic Bonin (Ogas-
to the Bonin (Ogasawara) Islands, Japan. J. Jap. Bot. awara) Islands. Flora 209: 641–648.
89: 22–26. Watanabe, K. & T. Sugawara. 2015. Is heterostyly rare on
Tao, C. & C. M. Taylor. 2011. Psychotria L. In: Wu L. Z., oceanic islands? AoB PLANTS doi: 10.1093/aobpla/
P. H. Raven & D. Y. Hong (eds.), Flora of China, vol. plv087
19, pp. 294–301. Science Press, Beijing, and Missouri Yamazaki, T. 1993. Psychotria L. In: Iwatsuki K., T.
Botanical Garden Press, St. Louis. Yamazaki, D. E. Boufford & H. Ohba (eds.), Flora of
Watanabe, K., A. Shimizu & T. Sugawara. 2014a. Dioecy Japan IIIa: 225–227. Kodansha, Tokyo.
derived from distyly and pollination in Psychotria Yang, T. Y. A. 1998. Psychotria L. In: Editorial Commit-
rubra (Rubiaceae) occurring in the Ryukyu Islands, tee of the Flora of Taiwan (eds.), Flora of Taiwan IV,
Japan. Pl. Spec. Biol. 29: 81–191. pp. 315–318, 2nd ed., National Taiwan University,
Watanabe, K., H. Kato & T. Sugawara. 2014b. Distyly and Taipei.
incompatibility in Psychotria homalosperma (Rubia-
Received March 10, 2015; accepted August 10, 2015
NII-Electronic Library Service