Annals of Botany 103: 39 –44, 2009
doi:10.1093/aob/mcn212, available online at www.aob.oxfordjournals.org
A new pollination system: brood-site pollination by flower bugs
in Macaranga (Euphorbiaceae)
Chikako Ishida1, Masumi Kono2 and Shoko Sakai3,*
1
Center for Ecological Research, Kyoto University, Otsu 520-2113, Japan, 2Department of Botany, Graduate School
of Science, Kyoto University, Kyoto 606-8502, Japan and 3Research Institute for Humanity and Nature,
Kamigamo, Kita-ku, Kyoto 603-8047, Japan
Received: 29 June 2008 Returned for revision: 1 September 2008 Accepted: 22 September 2008 Published electronically: 7 November 2008
† Background and Aims Macaranga (Euphorbiaceae) is a large genus of dioecious trees with approx. 260 species.
To date, only one pollination study of the genus has reported brood-site pollination by thrips in M. hullettii. In this
study, the pollination system of Macaranga tanarius is reported.
† Methods The study was conducted on Okinawa and Amami Islands, Japan. Flower visitors on M. tanarius were
collected and their pollen load and behaviour on the flowers examined, as well as inflorescence structure and
reward for the pollinators.
† Key Results The most abundant flower visitors found on the male and female inflorescences were Orius atratus
(Anthocoridae, Hemiptera), followed by Decomioides schneirlai (Miridae, Hemiptera). Pollen load on O. atratus
from flowering pistillate inflorescences was detected as well as from staminate flowers. Orius atratus and
D. schneirlai are likely to use the enclosed chambers formed by floral bracts as breeding sites before and
during flower anthesis, and feed on nectar on the adaxial surface of flower bracts. The extrafloral nectary has
a ball-shaped structure and the contained nectar is not exposed; the hemipterans pierce the ball to suck out
the nectar.
† Conclusions The results indicate that the plant is pollinated by flower bugs breeding on the inflorescences. This
study may be the first report of pollination systems in which flower bugs are the main pollinators. Similarity of
pollination systems between M. hullettii and M. tanarius indicates that the two brood-site pollination systems
have the same origin. The pollinator species belongs to a predacious group, whose major prey includes thrips.
The pollination system might represent a unique example of evolution from predatory flower visitors feeding
on the pollinators (thrips) to the main pollinators.
Key words: Amami, brood-site pollination, dioecy, Euphorbiaceae, extrafloral nectary, Hemiptera,
Macaranga tanarius, Okinawa, Orius atratus
IN T RO DU C T IO N
Flowering plants have evolved amazingly diverse pollination
systems. Although the majority of plants provide nectar and/
or pollen, various other rewards have been adopted, especially
for insect pollinators, reflecting their diverse ecology. Such
rewards include special kinds of oil (e.g. Sérsic and Coccuci,
1999; Coccuci and Vogel, 2001) or resin (e.g. Armbruster,
1997), mating sites (e.g. Gottsberger, 1999) and heat (e.g.
Seymour et al., 2003; Yuval et al., 2006). The type of
reward may be an important factor to determine specificity
of the pollinators to plants and vice versa.
Close relationships between plants and pollinators are often
found in brood-site (or nursery) pollination systems, in which
plants provide breeding sites as a reward for pollination. In
these pollination systems, pollinator larvae grow on the
flowers or inflorescences, feeding on ovules, pollen and other
reproductive organs (reviewed by Sakai, 2002; Dufaÿ and
Anstett, 2003). Well-documented examples of such pollinators
are ovule parasites of plants, including fig wasps pollinating
figs (Moraceae; e.g. Janzen, 1979; Machado et al., 2001),
yucca moths pollinating Yucca (Agavaceae; e.g. Pellmyr,
* For correspondence. E-mail [email protected]
2003) and Epicephala moths pollinating Glochidion spp. and
their allies (Phyllanthaceae; Kato et al., 2003). In these interactions, both plants and pollinators cannot survive without
their obligate partners, and are considered to have tightly
co-evolved.
Alternatively, specificity for and reliance on their partners
may be variable or relatively low in other brood-site pollination systems, in which plants offer parts of reproductive
organs other than ovules to their pollinators (Sakai, 2002). In
these systems, shifts from one pollinator group to another
may occur after evolution of brood-site pollination. However,
such examples have rarely been reported, and frequencies of
and conditions for shifts of pollinators are largely unknown.
In this study, the pollination system of Macaranga tanarius
(Euphorbiaceae) is reported. Macaranga is a moderately large
genus of dioecious trees including approx. 260 species. Many
of these species are pioneer trees, and about 30 are myrmecophytes (Davies et al., 2001). Only one pollination study on the
genus has reported brood-site pollination by thrips in
M. hullettii (Moog et al., 2002). Although M. tanarius also
has a brood-site pollination system, it was found that its pollinators were flower bugs (Hemiptera) rather than thrips. As far
as is known, this is the first report of a plant pollinated mainly
# The Author 2008. Published by Oxford University Press on behalf of the Annals of Botany Company. All rights reserved.
For Permissions, please email: [email protected]
40
Ishida et al. — Pollination by flower bugs in Macaranga
by hemipterans. The similarity of the pollination systems of
M. tanarius and M. hullettii indicates that their brood-site pollination systems have the same origin, and a pollinator shift
occurred after brood-site pollination had evolved in
Macaranga.
M AT E R IA L S A ND M E T HO DS
Study sites
Collection and observation in the field were conducted in the
northern part of Okinawa Island (25846’N, 126838’E) in
April 2007 and the eastern part and along the northern coast
of Amami Island (28820’N, 129828’E) in May 2007. Both
islands are in the subtropical region of Japan, about 330 km
from each other. The annual mean temperature in Okinawa
is about 22.4 8C, and the average annual total rainfall is
about 2000 mm. The northern area of Okinawa is mostly
covered with broadleaf evergreens of the genus Castanopsis,
while the south-central section is largely occupied by
cultivated land. The annual mean temperature of Amami is
21.1 8C, and the average annual total rainfall is about
3000 mm. Most parts of the island are covered by secondary
Castanopsis forests or secondary Pinus forests. Cultivated
lands is mainly in the coastal region of the island.
Study species
Macaranga tanarius (Euphorbiaceae) is a dioecious pioneer
tree that grows up to 10 m tall and 15 cm in diameter at breast
height. The trees are usually found in the open or in disturbed
areas such as forest edges, roadsides and secondary forests.
The species has a wide distribution from southern Japan
through Peninsular Malaysia, Borneo, and New Guinea. In
Japan, M. tanarius produces inconspicuous green inflorescences clustered at the end of the shoots in April – June.
A staminate inflorescence is composed of about ten branchlets bearing 15– 30 apetalous flowers along a 15- to
20-cm-long rachis (Fig. 1B). Flowering trees can be recognized from a distance, since top of the branchlets are visible
above the canopy. Each flower has approximately ten
anthers, subtended by a bract approx. 0.5 mm in width and
length. The flower is loosely covered by bracts when flowering
(Fig. 1D). A pistillate inflorescence has branchlets thicker and
shorter than those of a staminate inflorescence. The branchlets
do not elongate above the leaves, thus they are not visible from
outside of the canopy (Fig. 1A). Each branchlet produces five
to ten apetalous flowers in clusters. Each flower has a twolobed stigma and is subtended by a bract 0.5 mm in width
and 1.0 cm in length (Fig. 1C). Pistillate inflorescences
contain fewer flowers than staminate inflorescences. Stigmas
are tightly covered with bracts even during anthesis, and
there are only narrow gaps between overlapping bracts. The
adaxial surface of the bracts of both sexes is covered with trichomes, which are especially dense around the base.
Flowering stages
Three flowering stages of trees were distinguished. Stage 1:
in most of the inflorescences, the space between overlapping
bracts is ,1 mm; the size of immature anthers and stigmas
is less than half the size at maturity (anther, ,0.5 mm in
width and length; stigma, ,2 mm in length). Stage 2: the
space is 1 – 3 mm; anthers or stigmas are almost the same
size as those in mature flowers (anther, 0.5 – 1 mm in width
and length; stigma, 2 – 4 mm in length). Stage 3: at least
some of the flowers are in anthesis; in staminate flowers,
anthers dehisce, and pollen is exposed; in pistillate flowers,
the tip of the stigma is split and the surface seems wet
(Fig. 1C).
Collection and observation of flower visitors
Flower visitors hidden inside the enclosed bracts were collected on 18 and 39 trees on Okinawa and Amami, respectively. First, part of the inflorescence was put into a plastic
bag and then a branchlet was cut off to reduce possible loss
of floral visitors during collection. All the collections were
made during daytime (0700 – 1800 h). All insect visitors
were identified to the order, and abundant adult hemipterans
were identified to species level. Hemipteran larvae in
subsamples were also indentified into species. Reference
collections are kept at Kyoto University. In addition,
highly mobile flower visitors were assessed by observing
inflorescences directly (staminate, 5 h; pistillate, 7 h) or
indirectly using a video camera (staminate, 10 h; pistillate,
6 h) both during daytime (0600– 1800 h) and night-time
(1800 – 2300 h).
To compare flower visitors between the sexes and among
the flowering stages, insects were collected on five branchlets
per tree from 22 staminate and 17 pistillate trees on Amami
Island. The number and sex ratio of the main pollinator
species, Orius atratus (see Results), were compared between
the staminate and pistillate trees using the Mann – Whitney
U-test, and among flowering stages of each sex using the
Kruskal – Wallis test. The statistical tests were conducted
using R software (R Development Core Team, 2005).
To assess pollen transfer, adult hemipterans were collected
from 230 branchlets of seven pistillate trees at Stage 3 and
25 branchlets of three staminate trees at Stage 3 on Amami
Island. The insects were examined for pollen load under a
microscope.
In the field, behaviour of flower visitors inside the bracts
was observed by carefully spreading apart the floral bracts to
extend the openings. For observation in the laboratory,
flower visitors collected alive were kept in a Petri dish with
part of the inflorescences with about five flowers. Feeding
behaviour was observed under a binocular microscope.
R E S U LT S
Flower visitors
In total, 1184 and 758 flower visitors were collected from staminate and pistillate inflorescences, respectively, before and
during anthesis (Fig. 2) on Okinawa and Amami (Table 1).
The adult insect visitors were dominated by two species of
Hemiptera, Orius atratus (Anthocoridae) and Decomioides
schneirlai (Miridae) (Table 1). The density of O. atratus did
not differ between staminate (7.6 + 10.0 individuals per 5
Ishida et al. — Pollination by flower bugs in Macaranga
A
B
C
D
E
41
F
*
n
F I G . 1. Macaranga tanarius and pollinator: (A) overview of pistillate inflorescences (scale bar ¼ 2 cm); (B) overview of staminate inflorescences (scale bar ¼
3 cm); (C) flowering pistillate inflorescence in which the outer bracts are extended to show the lobed stigma (scale bar ¼ 5 mm); (D) flowering staminate inflorescences visited by Orius atratus (scale bar ¼ 2 mm); (E) Orius atratus sucking nectar on the adaxial surface of a bract (scale bar ¼ 1 mm); (F) scanning electron
micrograph of nectaries on the adaxial surface of a bract (n, mature nectary before sucking; *, nectary already sucked and deflated; scale bar ¼ 100 mm).
inflorescences, n ¼ 22) and pistillate (4.2 + 3.3, n ¼ 17) trees,
but the proportion of females was significantly higher on
pistillate trees (0.91 + 0.16, N ¼ 14) than on staminate trees
(0.56 + 0.18, n ¼ 16) (P ¼ 0.00011, Mann – Whitney U-test).
Both O. atratus and D. shneirlai were observed moving
quickly on inflorescences. Among the visitors, 22.1 –66.3 %
were hemipteran larvae. The majority of the larvae on both
staminate and pistillate inflorescences (98.6 % of 69 and
96.8 % of 63, respectively) were those of O. atratus, and the
rest were those of D. shneirlai. During the observation
periods for mobile visitors, no visitors other than the two
hemipteran species were observed. Significant differences in
the numbers and sex ratio of hemipterans among flowering
stages were not observed.
Eighteen per cent (n ¼ 50) and 9.3% (n ¼ 107) of male and
female O. atratus, respectively, collected from flowering
pistillate inflorescences (Stage 3) had pollen loads. However,
pollen was not found on D. schneirlai (n ¼ 28; Table 2). In
staminate inflorescences of Stage 3, 100% of O. atratus and
D. schneirlai had pollen on their bodies (Table 2).
Nectary and feeding behaviour of the pollinators
Ball-like structures filled with liquid were scattered on the
adaxial base of the bracts covered with trichomes both on
staminate and pistillate inflorescences (Fig. 1F). The cuticle
on the nectary seemed to be relatively tough since it did not
break when squeezed with the tips of tweezers. Hemipteran
42
Ishida et al. — Pollination by flower bugs in Macaranga
(A)
Adult Orius
Hemipteran larvae
Adult Decomioides
(B)
Staminate
Number of individuals
Number of individuals
30
20
10
0
1 (5)
2 (8)
3 (9)
Pistillate
1 (5)
Flowering stage
2 (5)
3 (9)
Flowering stage
F I G . 2. Number of hemipterans collected from five branchlets on staminate (A) and pistillate (B) inflorescences. Columns show an adult Orius atratus, an adult
Decomioides schneirlai and hemipteran larvae as indicated. Vertical bars represent s.e.
TA B L E 1. Summary of flower visitors collected on staminate and pistillate trees on Okinawa and Amami Islands
No. of sites
Okinawa (15)
Amami (16)
Sex
No. of trees
No. of branchlets
No. of insects
O. atratus (%)
D. schneirlai (%)
Hemipteran larvae (%)
Others (%)
Staminate
Pistillate
Staminate
Pistillate
13
5
22
17
66
41
130
310
440
86
744
672
42.5
65.1
22.6
32.3
20.0
12.8
10.1
7.3
37.3
22.1
66.3
59.7
0.2
0.0
1.1
0.7
‘Others’ include thrips and hemipteran insects.
TA B L E 2. Proportion of hemipteran insects with body pollen
collected from flowering pistillate and staminate inflorescences
Family, species
Anthocoridae, Orius atratus
Miridae, Decomioides
schneirlai
Sex
Pistillate tree
(%)
Staminate tree
(%)
Male
Female
Male
18.0 (50)
9.3 (107)
0.0 (15)
100.0 (16)
100.0 (10)
100.0 (2)
Female
0.0 (13)
100.0 (2)
Sample sizes are given in parentheses.
adults and larvae were observed to probe the adaxial base of
the bract (Fig. 1E). They inserted their mouthparts and
sucked the contents at the rate of 1– 2 balls s21. Once the
content was sucked out, the ball became deflated (Fig. 1F).
No signs of feeding damage were observed on bracts and
other floral tissues, or of flower bugs feeding on pollen.
DISCUSSION
This study suggests that Macaranga tanarius is pollinated by
flower bugs on Amami and Okinawa Islands. Other than hemipteran larvae, the flower bugs Orius atratus were the predominant visitors to both staminate and pistillate inflorescences of
M. tanarius, on both Okinawa and Amami Islands. The presence of body pollen on O. atratus collected from pistillate
inflorescences indicated that the bugs moved from staminate
to pistillate trees, thus contributing to pollination. They were
observed to search and suck nectar very actively on the inner
surface of the bract. It is likely that they deposit pollen on
stigmas when they crawl into the chamber formed by floral
bracts, and when they are feeding on nectar on the inner
surface of the bract. The contribution to pollination of the
second most abundant species, Decomioides schneirlai, was
uncertain, since it was not possible to find body pollen on individuals collected from pistillate inflorescences. Wind pollination is quite unlikely considering that pistillate flowers are
tightly enclosed in the floral bracts during the whole flowering
period, and that pistillate inflorescences are not exposed
outside of the canopy. Thus we consider that O. atratus is
the most important pollinator of M. tanarius, although the efficiency of the insects as pollinators still remains to be evaluated
by measuring frequency of inter-tree movement and pollen
deposition efficiency. Dominance of hemipterans (Miridae)
among flower visitors on the plants has also been reported
from west Malaysia (Moog, 2002).
As far as is known, this study is the first report of pollination
systems in which flower bugs are the main pollinators. Only a
few studies have reported hemipterans or flower bugs as
pollinators, despite the fact that they are frequently found
in flowers, feeding on flower visitors such as thrips
(Yasunaga, 1997). Fahn and Shimony (2001) reported that
hemipteran insects are the main flower visitors in Ecballium
elaterium (Curcurbitaceae), yet their contribution as pollinators are uncertain. Another report showed that hemipteran
larvae feeding on pollen facilitated selfing in Roridula
(Roridulaceae; Anderson et al., 2003).
Ishida et al. — Pollination by flower bugs in Macaranga
The presence of large numbers of hemipteran larvae in both
staminate and pistillate inflorescences at all flowering stages
suggests that the hemipterans use the inflorescences as breeding sites before and during flower anthesis. Orius spp. are
known to lay their eggs within the plant tissue (Tommasini,
2003). Their larvae do not have functional wings, thus their
mobility is quite limited. Enclosed chambers formed by
floral bracts may be ideal sites for reproduction, since they
provide a shelter from natural enemies as well as a food
source. One remarkable character of this hemipteran pollination is that the main food for the pollinators provided by the
plant is nectar secreted by the nectaries scattered on the
adaxial surface of bracts (Fig. 1F). In most flowering plants,
nectar is secreted from secretory tissues and exposed on the
epidermis (reviewed by Pacini et al., 2003). However, the
extrafloral nectary for pollinators of M. tanarius is a ballshaped structure, and the nectar inside is not exposed unless
the cuticle is torn open. This type of nectary can provide a
food source only for insects with mouthparts suitable for
inserting and sucking, such as hemipterans, and may contribute to eliminate nectar robbers. A similar structure of pollinator rewards is known for the elaiophore of oil-producing
flowers of Sisyrinchium species (Iridaceae; Cucucci and
Vogel, 2001). These flowers are pollinated by oil-collecting
bees with special scrapers on their forelegs, with which they
disrupt the cuticle.
Although O. atratus is likely to be the most important pollinator of M. tanarius, the importance of M. tanarius for the
pollinator is unclear. The genus Orius comprises a group of
flower bugs known to be predators of minute insects, such as
thrips and mites, which are found on plants, while they also
feed on pollen and other plant tissues. Species such as Orius
tristicolor, O. insidiosus, O. sauteri and O. minutus are commonly used for biological pest control (e.g. Kawai, 1995;
Kohno and Kashio, 1998; Tommasini, 2003). These species
can also be reared on pollen, but their growth and/or survival
rates are often significantly lower (Funao and Yoshiyasu, 1995;
Richards and Schmidt, 1996; Maeda et al., 2002). Thus nectar
composition or dependence of O. atratus on nectar for nutrient
during the flowering period may be an interesting subject for
further studies, considering that nitrogen-rich food seems to
be essential for omnivorous Orius spp. (Coll and Guershon,
2002). Orius atratus was described about 10 years ago, and
its ecology is mostly unknown. The species has been recorded
from flowers of different trees, such as Mallotus japonicus
(Euphorbiaceae), Schima wallichi (Theaceae) and Trema
orientalis (Ulmaceae; Yasunaga, 1997). Since O. atratus has
been collected out of flowering season of M. tanarius
(Yasunaga, 1997), it may survive feeding on herbivorous
insects when the nectar of M. tanarius is not available.
Nevertheless, some circumstantial evidence indicates the
importance of M. tanarius for O. atratus. For example, the
northern limit of distribution of both M. tanarius and
O. atratus, Amami Island, coincides (Ohba, 1989; Yasunaga,
1997). Moreover, O. atratus is most abundant in the flowering
season of M. tanarius in Japan, judging from collection
records of specimens (Ohba, 1989; Yasunaga, 2001).
Several other aspects of the pollination system remain to be
revealed. For example, the advantage of breeding of pollinators on female plants is unclear. Dispersal of pollinators
43
from female trees without pollen load does not contribute to
reproductive success, while breeding of pollinators on staminate inflorescences well before anthesis may facilitate pollen
dispersal by increasing the number of pollinators. Most dioecious plants with a brood-site pollination system attract pollinators by deceit to female plants without rewards (Dufaÿ and
Anstett, 2003). Presence of many female conspecifics on pistillate inflorescences may promote visits of male pollinators,
which carry more pollen than do females. The role of
Decomioides schneirlai in this pollination system is also
unknown; is it a co-pollinator of O. atratus, or competitor?
Besides, it is not known what triggers movements of the sedentary pollinators between trees. There is no explanation either
for why the sex ratio of O. atratus differs between staminate
and pistillate trees.
Only one published study on pollination in the genus reported
brood-site pollination by thrips in M. hullettii (Moog et al.,
2002). Abundant thrips have also been observed on inflorescences of close relatives of the species (Ishida, 2008). Thrips
are tiny insects with piercing/sucking mouthparts and are
found on many different plants, although host specificity is
largely unknown (Williams et al., 2001; Mound, 2005).
Pollination by thrips has been reported in several plant
groups, in most of which pollen is the major reward for the pollinators (e.g. Popowia, Momose et al., 1998; Macrozamia,
Mound and Terry, 2001; Castilla, Sakai, 2001; Antiaropsis,
Zerega et al., 2004). Among thrip-pollinated plants,
Macaranga is unique in that its reward for pollinators is not
pollen but nectar secreted by trichomes on the adaxial surface
of bracts (Moog et al., 2002). In addition, thrips begin inhabiting and breeding in flowers well before anthesis.
The similarity of the pollination systems between
M. hullettii and M. tanarius indicates that brood site pollination by thrips and flower bugs has the same origin. The
thrips and hemipteran pollination systems found in
Macaranga are common in that the pollinators use enclosed
bract chambers as breeding and feeding sites before and
during anthesis, both in staminate and pistillate trees. In
addition, the plants provide nectar rather than pollen for the
pollinators and their larvae, and the nectaries are located on
the adaxial surface of bracts. Therefore, we consider that pollinator shifts between insects breeding on flowers have
occurred after brood-site pollination.
One possible scenario might be that hemipteran visitors,
which had been originally predators of thrips pollinators,
became the main pollinators when plants evolved a ball-like
nectary, which requires a strong piercing mouthpart to suck
out the nectar. Both O. atratus and D. schneirlai belongs to
a predacious group, whose major prey includes thrips.
Predacious hemipteran insects are occasionally observed
together with thrips on the inflorescence of Macaranga and
Mallotus (S. Sakai, unpubl. res.), a sister genus of
Macaranga (Kulju et al., 2007). Future studies might reveal
that the pollination system represents a unique example of
evolution from predatory flower visitors feeding on the pollinators (thrips) to main pollinators. Evolution and specificity
of plant – pollinator interactions is still largely unknown in
brood-site pollination systems other than pollination by ovule
parasites. Macaranga may provide an interesting material for
further studies.
44
Ishida et al. — Pollination by flower bugs in Macaranga
AC KN OW LED GEMEN T S
We thank Dr Makoto Kato for comments on an earlier version
of the manuscript and identification of hemipteran insects; Dr
Katsuyuki Kohno for information on hemipterans; Dr Atsushi
Kawakita for advice on the field study; Dr Hiroshi Tobe for
permission to use research equipment; and two anonymous
reviewers for useful comments on the earlier version of the
manuscript. This study was financially supported in part by
Grants-in-Aid for Scientific Research (no. 20405009) by the
Ministry of Education, Culture, Sports, Science and
Technology of Japan.
L I T E R AT U R E CI T E D
Anderson B, Midgley JJ, Stewart BA. 2003. Facilitated selfing offers reproductive assurance: a mutualism between a hemipteran and a carnivorous
plant. American Journal of Botany 90: 1009– 1015.
Armbruster WS. 1997. Exaptations link evolution of plant-herbivore and
plant-pollinator interactions: a phylogenetic inquiry. Ecology 78:
1661–1672.
Cocucci AA, Vogel S. 2001. Oil-producing flowers of Sisyrinchium species
(Iridaceae) and their pollinators in southern South America. Flora 196:
26– 46.
Coll M, Guershon M. 2002. Omnivory in terrestrial arthropods: mixing plant
and prey diets. Annual Review of Entomology 47: 267–297.
Davies SJ, Lum SKY, Chan R, Wang LK. 2001. Evolution of myrmecophytism in western Malesian Macaranga (Euphorbiaceae). Evolution 55:
1542–1559.
Dufaÿ M, Anstett M-C. 2003. Conflicts between plants and pollinators that
reproduce within inflorescences: evolutionary variations on a theme.
Oikos 100: 3 –14.
Fahn A, Shimony C. 2001. Nectary structure of unisexual flowers of
Ecballium elaterium (L.) A. Rich (Cucurbitaceae) and their presumptive
pollinators. Annals of Botany 87: 27–33.
Funao T, Yoshiyasu Y. 1995. Development and fecundity of Orius sauteri
(Poppius) (Hemiptera: Anthocoridae) reared on Aphis gossypii Glover
and corn pollen. Japanese Journal of Applied Entomology and Zoology
29: 84–85.
Gottsberger G. 1999. Pollination and evolution in neotropical Annonaceae.
Plant Species Biology 14: 143– 152.
Ishida C. 2008. Pollination system of Macaranga (Euphorbiaceae); plants providing breeding sites as reward for pollination. Master Thesis, Kyoto
University, Japan [in Japanese].
Janzen DH. 1979. How to be a fig. Annual Review of Ecology and Systematics
10: 13–51.
Kato M, Takimura A, Kawakita A. 2003. An obligate pollination mutualism
and reciprocal diversification in the tree genus Glochidion
(Euphorbiaceae). Proceedings of the National Academy of Sciences of
the USA 100: 5264– 5267.
Kawai A. 1995. Control of Thrips palmi Karny (Thysanoptera: Thripidae) by
Orius spp. (Heteroptera: Anthocoridae) on greenhouse eggplant. Applied
Entomology and Zoology 30: 1– 7.
Kulju KKM, Sierra SEC, Draisma SGA, Samuel R, van Welzen PC. 2007.
Molecular phylogeny of Macaranga, Mallotus, and related Genera
(Euphorbiaceae S.S.): insights from plastid and nuclear DNA sequence
data. American Journal of Botany 94: 1726–1743.
Kohno K, Kashio T. 1998. Development and prey consumption of Orius
sauteri (Poppius) and O. minutus (L.) (Heteroptera: Anthocoridae) fed
on Frankliniella occidentalis (Pergande) (Thysanoptera: Thripidae).
Applied Entomology and Zoology 33: 227–230.
Machado CA, Jousellin E, Kjellberg F, Compton SG, Herre EA. 2001. The
evolution of fig pollinating wasps: phylogenetic relationships, character
evolution, and historical biogeography. Proceedings of the Royal
Society, Series B 268: 685–694.
Maeda S, Nakashima Y, Tagashira E, Takagi M. 2002. Effects of sucrose on
the longevity and fecundity of Orius sauteri (Poppius) (Hemiptera:
Anthocoridae). Japanese Journal of Applied Entomology and Zoology
46: 169 –173 [in Japanese].
Momose K, Nagamitsu T, Inoue T. 1998. Thrips cross-pollination of
Popowia pisocarpa (Annonaceae) in a lowland dipterocarp forest in
Sarawak. Biotropica 30: 444–448.
Moog U. 2002. Die Reproduktion von Macaranga (Euphorbiaceae) in
Südostasien: Bestäubung durch Thripse und Kastration durch
Pflanzenameisen. PhD Thesis, Johann Wolfgang von GoetheUniversität, Germany.
Moog U, Fiala B, Federle W, Maschwitz U. 2002. Thrips pollination of the
dioecious ant plant Macaranga hullettii (Euphorbiaceae) in Southeast
Asia. American Journal of Botany 89: 50–59.
Mound LA. 2005. Thysanoptera: diversity and interactions. Annual Review of
Entomology 50: 247–269.
Mound LA, Terry I. 2001. Thrips pollination of the Central Australian cycad,
Macrozamia macdonnellii (Cycadales). International Journal of Plant
Science 162: 147– 154.
Ohba H. 1989. Euphorbiaceae. In: Satake Y, Hara H, Watari S, Tominari T,
eds. Wild flowers of Japan: woody plants, Vol. I. Tokyo: Heibonsha,
269– 321.
Pacini E, Nepi M, Vesprini JL. 2003. Nectar biodiversity: a short review.
Plant Systematics and Evolution 238: 7– 21.
Pellmyr O. 2003. Yuccas, yucca moths and coevolution: a review. Annals of
the Missouri Botanical Garden 90: 35–55.
R Development Core Team. 2005. R: a language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria.
ISBN 3-900051-07-0, URL http://www.R-project.org (9 Sept 2008).
Richards PC, Schmidt JM. 1996. The effects of selected dietary supplements
on survival and reproduction of Orius insidiosus (Say) (Hemiptera:
Anthocoridae). Canadian Entomologists 128: 171– 176.
Sakai S. 2001. Thrips pollination of androdioecious Castilla elastica
(Moraceae) in a seasonal tropical forest. American Journal of Botany
88: 1527–1534.
Sakai S. 2002. A review of brood-site pollination mutualism: plants providing
breeding sites for their pollinators. Journal of Plant Research 115:
161– 168.
Sérsic AN, Cocucci AA. 1999. An unusual kind of nectary in the oil flowers of
Monttea: its structure and function. Flora 194: 393– 404.
Seymour RS, White CR, Gibernau M. 2003. Heat reward for insect pollinators. Nature 426: 243 –244.
Tommasini MG. 2003. Evaluation of Orius species for biological control of
Frankliniella occidentalis (Pergande) (Thysanoptera, Thripidae). PhD
Thesis, Wageningen University, The Netherlands.
Williams GA, Adam P, Mound LA. 2001. Thrips (Thysanoptera) pollination
in Australian subtropical rainforests, with particular reference to pollination of Wilkiea huegeliana (Monimiaceae). Journal of Natural History
35: 1 –21.
Yasunaga T. 1997. The flower bug genus Orius Wolff (Heteroptera:
Anthocoridae) from Japan and Taiwan, part I. Applied Entomology and
Zoology 32: 355– 364.
Yasunaga T. 2001. Family Anthocoridae Fieber, 1836: flower bugs, minute
pirate bugs. In: Yasunaga T, Takai M, Kawasawa T, eds. A field guide
to Japanese bugs. II. Terrestrial Heteropterans. Tokyo: Zenkoku Noson
Kyoiku Kyokai, 278– 294.
Yuval S, Shmida A, Ne’Eman G. 2006. Morning floral heat as a reward to the
pollinators of the Oncocyclus irises. Oecologia 147: 53– 59.
Zerega NJC, Mound LA, Weibren GD. 2004. Pollination in the New Guinea
endemic Antiaropsis decipiens (Moraceae) is mediated by a new species
of thrips, Thrips antiaropsidis sp. nov. (Thysanoptera: Thripidae).
International Journal of Plant Sciences 165: 1017–1026.