Botanical Studies (2006) 47: 307-318.

*

Corresponding author: e-mail: erecta9283@yahoo.com.tw.

INTRODUCTION

A total ca. 750 Ficus species worldwide are distributed

in tropical to subtropical areas (Corner, 1965; Berg, 1989),

and each species is typically pollinated by females of its

own specific species of fig wasp (Galil, 1973; Wiebes,

1979; Van Noort and Compton, 1996). The relationship

between Ficus species and their pollinating wasps

(Agonidae, Chalcidoidea, Hymenoptera) is considered

to be an extreme instance of plant-animal co-evolution

(Janzen, 1979). Fig trees are defined by the syconium

(syncarp or fig), a unique enclosed inflorescence, which

is lined with several dozen to thousands of tiny, unisexual

flowers (Berg, 1989; Verkerke, 1989; Tzeng et al., 2001).

This is also the arena for interactions with fig wasps.

The morphologically specific fig wasp enters the syconia

through the bract-lined entrance of the ostiole, often losing

their wings and antennae in the process. The pollen-loaded

wasps then pollinate the female flowers. In turn, wasp

reproduction is dependent on the fig, as its larvae feed on

galling flowers (Ramirez, 1970; Galil, 1973; 1977; Janzen,

1979; Van Noort and Compton, 1996).

Fig trees can be classified as either monoecious or

dioecious, with each group comprising roughly 50% of

the species (Corner, 1965; Berg, 1989). In monoecious

fig trees, male and female flowers line the inner wall of

the same syconium. Female flowers, differing in style

and length, comprise the imperfect heterostyle (Verkerke,

1989). Additionally, the ovules of female flowers can

be pollinated for seed-production and oviposited for fig

wasp developing the gall, in which larvae feed (Bronstein,

1988). Several weeks later, adult fig wasps emerge and

mate while still in the syconium. The wingless males can

then cut a tunnel out of the syconia, often dying inside

the natal syconium. The female pollen-carrier emerges

from the now-mature male flowers, either through passive

ducting or packing special pollen pockets and depart to

search for receptive syconia (Galil and Eisikowitch, 1968;

eCOlOgy

Pollinational-mutualism strategy of Ficus erecta var.

beecheyana and Blastophaga nipponica in seasonal

guandaushi Forest ecosystem, Taiwan

Hsy-Yu TZENG

1,

*, Fu-Yuan LU

2

, Chern-Hsiung OU

3

, King-Cherng LU

3

, and Li-Jung TSENG

3

1

Hengchun Center, Taiwan Forestry Research Institute, 203 Kungyuan Rd., Hengchun 94644, Pingtung, TAIWAN

2

Department of Forestry and Natural Resources, National Chiayi University, 300 University Rd., Chiayi 60004, TAIWAN

3

Department of Forestry, National Chunghsing University, 250 Kuokwang Rd., Taichung 40227, TAIWAN

(Received November 23, 2004; Accepted January 26, 2006)

ABSTRACT.

This study investigates pollination mutualism between F. erecta var. beecheyana and its

obligate pollinator Blastophaga nipponica using detailed phenology data from the seasonal Guandaushi Forest

Station. The symbiotic cycle resembled that of the F. carica and B. psenes in southern France. Blastophaga

nipponica emerged from the D-phase syconia of over-winter male crops and entered the receptive male

syconia to lay its eggs during the spring male main crop reproductive period. Several weeks later, dozens of

offspring emerged and pollinated the receptive syconia of the summer female major crop. A few pollinators

entered the receptive male syconia to oviposite, and their offspring wintered inside the male fig as larvae or

pupae. The fig trees could control the developmental period of wasp-producing syconia during pollination.

The peak of B-phase syconia (pollinators pollinate or set eggs) of both genders appeared earlier than the

abundant D-phase syconia (pollinators released) by about two to three weeks. On the other hand, the flowering

syconia of both genders occurred abundantly, staggered about two to four weeks after heavy rainfall, and the

fly-out pollinators would pollinate or lay eggs during this period. This fig flowering phenology accommodates

the shorter life span of the obligate species-special pollen carrier to enter the receptive syconia for effective

oviposition or seed-setting. Ficus erecta var. beecheyana and B. nipponica thus have a successful mutualism

strategy of pollination at the Guandaushi Forest Station.

Keywords: Blastophaga nipponica; Ficus erecta var. beecheyana; Guandaushi Forest Ecosystem; Mutualism;

Pollination.

308

Botanical Studies, Vol. 47, 2006

Bronstein et al., 1990). Because monoecy have syconia,

with ripeness is synchronized within an individual tree

and are typically asynchronous between trees, the female

pollinators flying out of the syconia must search for

another receptive syconium on another tree to oviposite

and pollinate in (Galil et al., 1970; Janzen, 1979; Wiebes,

1979; Bronstein et al., 1990; Bronstein and Patel, 1992;

Thomson et al., 1997).

Dioecious fig species, however, have divided the

production of seeds and the rearing of pollinators into

two inflorescent types on separate plants: syconia male

and female figs. Male and short-style female flowers

(gall flowers, for fig wasp ovipositing), line the male

syconium while long-style female flowers (seed flowers

for seed production) and/or neutral flowers line the female

syconium (Weiblen et al., 1995; Tseng, 1999; Tzeng

et al., 2001). Short- and long-style female flowers are

differentiated by length of style, morphology, structure of

stigma and ovary (Verkerke, 1990; Beck and Lord, 1988a;

Tseng et al., 2000; Tzeng et al., 2001), and by pollination

physiology (Verkerke, 1990; Beck and Lord, 1988b),

which displays the perfect heterostyle (Verkerke, 1989;

1990).

Adult female pollinators emerge from D-phase syconia

and search for male receptive syconia to oviposit or female

receptive syconia to pollinate during their extremely short

lifespan ranging from several hours to two days (Kjellberg

et al., 1988; Tzeng, 1997; Wu, 1996; Tseng, 1999). After

pollination or oviposition, the wasp usually dies inside the

syconium it last visited. Otherwise, unvisited syconia are

then aborted.

Dioecious figs must produce D- and B-phase male

syconia during the same period to maintain a continuous

pollinator life cycle (Anstett et al., 1995; Tzeng, 1997;

Kameyama et al., 1999), and pollination success is

dependent on the continuous cyclical production of

pollinators with pollen-entering receptive syconia (Chen,

1998; Tseng, 1999; Harrison et al., 2000). Syconia

phenology has been shown to be useful in investigating

the influence of climatic fluctuation and biotic interactions

in some studies on dioecious figs in seasonal environments

which show sexual specialization (Valdeyron and

Lloyd, 1979; Spencer et al., 1996; Tzeng et al., 2003;

Chang, 2003). Dioecious figs do not necessarily ensure

outcrossing (Corlett, 1987), but the fig trees must imitate

the morphology, color, and odor of their syconia to keep

the fig wasps from distinguishing male from female

syconia and, thereby, only setting eggs (Grafen and

Godfray, 1991; Ware et al., 1993; Ware and Compton,

1994); they must also be prevented from recognizing the

fig phenology between male and female trees (Kjellberg

et al., 1987; Kjellberg and McKey, 1989; Spencer et al.,

1996; Patel and McKey, 1998; Harrison et al., 2000;

Tzeng et al., 2004).

The short lifespan of a pollinator (Kjellberg et al.,

1988; Wu, 1996; Tseng, 1999) and the receptive syconia

can prevent aborting when pollinators do not pollinate

or oviposite on the female flowers during a 2- to 3-week

period (Khadari et al., 1995; Tzeng, 1997). Kjellberg et al.

(1987) elucidated the stability of the symbiotic relationship

between the dioecious figs F. carica L. and Blastophaga

psenes L. and their pollinators during seasonal

environments. However, no studies have identified the

relationship between the life cycle of the pollinator¡Xwith

their short-lived escape from D-phase syconia¡X and the

receptive syconia for both genders year round. The goal

of the present study is to elucidate the syconia phenology

o f F. erecta Thunb. var. beecheyana (Hook. et Arn.)

King and determine the relationship between flowering-

phase syconia reproduction and the dynamic processes

of the obligate pollinator Blastophaga nipponica Grandi.

under sporadic rainfall. Additionally, this study identifies

the pollination-mutualism ecology between a fig and its

pollinator in the seasonal environment at the Guandaushi

Forest Station.

Studied site

The studied area was located at the Guandaushi Forest

Station of the Hue-Sun Experimental Forest Station (24o4¡¦

N, 12 o80¡¦E), one of the Long Term Ecological Research

(LTER) sites in Central Taiwan (Figure 1). During the

study period of 1996, the average annual rainfall, relative

humidity, and yearly temperature were 2,596.9 mm,

79.1%, and 21.0¢XC, respectively (data from the station).

Most of the rainfall (96.2% of total) occurred during

the rainy and typhoon seasons. The average rainfall

over ten years was 2,683.3 mm (Figure 2, from 1987 to

1996), and 93.2% occurred during the warm-wet season.

According to the classification of Thomthwite, this study

area belongs to AB¡¦wa¡¦, a wet but warmed wintertrochene

klima (Yu, 2001). The vegetation of the site is typical of

a warm forest and is characterized by a Ficus ¡V Machilus

vegetation zone at about 500-800 m (Liu, 1968; Su, 1992;

Lu and Ou, 1994).

F igu re 1. T he s tudied area was located at the Guandaus hi

Forest Station of the Hue-Sun Experimental Forest Station in

Central Taiwan.

TZENG et al. ¡X Pollinational-mutualism strategy of

F. erecta

var.

beecheyana

and

B. nipponic

a in Taiwan

309

Biology of Ficus erecta var. beecheyana

Ficus erecta Thunb. var. beecheyana (Hook. et Arn.)

King (Figure 7a) is a small semi-deciduous to deciduous

tree or big shrub from 2 to 5 m, distributed mainly in

southern China, from Ryukyu, Taiwan, Orchid Is., Hong

Kong, to the Malay Peninsula (Corner, 1965; Liu et al.,

1994; Liao, 1995; Tzeng, 2004). It is a pioneer species,

usually occurring in the lowland to medium altitude of

disturbed forests, second forests, or along roadsides (Liu

et al., 1994; Liao, 1995; Tzeng, 2004). The fig¡¦s obligate

pollinator is Blastophaga nipponica Grandi. (Figure 3a,

b) (Chen and Chou, 1997). The non-pollinating fig wasp,

Sycoscapter inubiae Ishii (Figure 3c, d), which oviposites

through the syconium exterior during the C-phase, and

its offspring may feed on the pollinator ¡¦s larvae inside

the ovary (Wu, 1996; Tzeng, 1997). The special-species

pollinator is also shared with F. erecta Thunb. var. erecta

(Okamoto and Tashiro, 1981; Chen and Chou, 1997).

This fig is morphologically gynodioecious and

functionally dioecious, with populations comprised of

female and hermaphrodite individuals, and this study

refers to the two tree sexes as female and male (Tzeng et

al., 2001). The syconia are axillary but male syconia are

generally born on the older branches while the female

syconia are mostly initiated on new branches (Tzeng et

al., 2003). The classification of the post-budding period

development processes of F. erecta var. beecheyana

is based on Galil and Eisikowitch (1968) and on

modifications by Tzeng et al. (2001).

A fig tree can produce one to four crops each year for

both genders within the tree, and the syconia productions

were seasonally at population level, which responded to

the fluctuation of climate and showed sexual difference

(Tzeng et al., 2003; Tzeng et al., 2004). The specific

pollinator must fly out the D-phase syconium from its

natal tree to find another B-phase syconium to set eggs or

pollinate in during its short-life span of several hours to

two days (Okamoto and Tashiro, 1981; Wu, 1996; Tzeng,

1997). The phenology of this fig indicated that the syconia

production of each gender was adapted to the environment,

and it would be helpful for seed production, germination,

and maintaining the population of the obligate pollinator,

Bl. nipponica, respectively (Tzeng et al., 2004).

MATeRIAlS AND MeTHODS

Samples selection

A total of 71 trees of F. erecta var. beecheyana were

marked (40 male and 31 female mature trees), from

which 30 mature trees (17 male and 13 female) were

selected for pollination examination. The figs trees were

somewhat closely distributed along the roadside or under

a 36-year-old China fir (Cunninghamia lanceolata) in an

artificial forest at the study site. Two to three branches

roughly 50 cm long from one tree or a whole, small

tree approximately 2 to 3 m high were selected for each

census. The figs, in the form of a shrub to small tree, were

2-5 m high with a diameter at breast height (DBH) of 2-10

cm.

Field observations

Syconia censuses of F. erecta var. beecheyana were

conducted at 5-9 day intervals from November 1995 to

April 1997 at the Guandaushi Forest Station. The numbers

of all syconia in the developmental stage were counted

for each gender during each census. The classification

of syconia development was based on the following

characteristics: (Table 1, Tzeng et al., 2001) male figs

do not have an E-phase, and female figs do not have a

D-phase. Although the syconium acts as both flower and

Figure 2. The ecological climate diagram of Hue-Sun

Experimental Forest Station; (P ) re prese nts perhumid; (H)

represents relative humid; (D) represents relative drought.

Figu re 3. The pollinator of Blastophaga nipponica Grandi.

and non-pollinating fig wasp of Sycoscapter inubiae Ishii. a, c,

indicate male wasps; b, d, indicate female wasp; bars indicate 1

mm.

310

Botanical Studies, Vol. 47, 2006

fruit, the true flowering phases of syconia are actively

B- and D-phases, together forming the critical phase of

syconia development in the mutualistic system. Data were

analyzed using SPSS 8.0. Kendall¡¦s rank correlation was

used to assess between each phenological census and

climate at the study site.

ReSUlTS

Relationship between pollinator life-span and

the flowering-phases of syconia

The B-phase male syconia germinated abundantly

from the male spring crop during March 1996, matching

primarily the mostly D-phase syconia of the 1995 over-

winter syconium crop (Figures 4, 5). The peak of the

D-phase syconium was delayed by roughly three weeks,

after the B-phase syconia of the male trees. The obligate

pollinator, Blastophaga nipponica, emerged from the

D-phase syconia of the 1995 over-winter syconia crop to

search for the B-phase of the male syconia to set eggs.

The pollinator was observed searching for the B-phase of

the male syconia on the same tree or flying away from the

host tree to find the B-phase male syconia on other trees.

The pollinator ¡¦s larvae developed in the 1996 main

male crop in spring between March and June. Several

weeks later the male syconia had successfully developed

from C- to D-phase. D-phase syconia were located in the

bloom, and the pollinator offspring (second generation)

emerged in large numbers primarily from D-phase male

syconia while the B-phase syconia were abundant in the

main female crop in summer. The female pollinators flew

out of their host syconia to search for another B-phase

syconia to set eggs. However, most pollinators joined the

Table 1. Syconia developmental stage, size, and maturity of F. erecta var. beecheyana for both genders.

Phase

Syconium dimater

Syconium maturity

A Pre-female phase ¡ð: 2 ¡V 6 mm

¡ñ: 2 ¡V 21 mm

The young syconia prior to the opening of the ostiole and with tightly closed

ostiolar bracts. Both male and female syconium are orange, red or light

green with dark red spots.

B Female or receptive

phase

¡ð: 6 ¡V 8 mm

¡ñ: 9 ¡V 21 mm

Ostiolar bracts loose. Both seed (female syconia) (Figure 7c) and gall (male

syconia) female flowers (Figure 7d) are ripe. Pollinators, B. nipponica

penetrate into the young syconium (Figures 7b, 8a) and lay eggs into the

ovaries of gall flowers (Figures 8b-d) or pollinate with the seed flowers.

Both male and female syconia are in red or green color with dark red spot.

C Inter-floral phase ¡ð: 6 ¡V 8 mm

¡ñ: 9 ¡V 22 mm

Wasp larvae develop within the occupied ovaries, which are transformed into

galls in male syconium.

D Male phase

¡ñ: 15 ¡V 30 mm Male flowers mature, wasps reach the eclosion stage, and the wingless male

wasp mates with the female wasp which is inside the gall in the syconium

(Figure 7g, h). The female wasps leave the male syconium via the opening

ostiole (Figure 7f) with pollen. The male syconia become soft with dark red

or purple.

E Mature phase

¡ð: 9 ¡V 21 mm Female syconia ripen, becoming dark-purple and soft, sometimes with sweet

liquid outside syconium (Figure 7e).

Figure 4. The relationships between rainfall, temperature and

flowering syconia production of both genders of F. erecta var.

beecheyana at Guandaushi Forest Station from October 1995 to

February 1997.

TZENG et al. ¡X Pollinational-mutualism strategy of

F. erecta

var.

beecheyana

and

B. nipponic

a in Taiwan

311

Figure 5. The relationships of flowering syconia production between D- and B-phase syconia of both genders for each sample tree of

F. erecta var. beecheyana at Guandaushi Forest Station from October 1995 to February 1997. (Ĉ) indicates female pollinators flying

out from D-phase male syconia. (ċ) indicate female pollinators entering B-phase syconia.

mass of B-phase female syconia from the main female

crop in summer for pollination (Figures 4, 5). A few

female pollinators found the B-phase male syconia of the

sub-male syconia crop for ovipositing. Some offspring

of the third or four generations of pollinators developed

in the sub-male syconia crop from September 1996 to

February 1997 (Figures 4, 5). The pollinational-symbiotic

ecological model is shown in Figure 6.

The incidence of D-phase syconia was well matched

with the abundant B-phase male and female syconia.

The largest number of B-phase syconia of both genders

developed 2 to 3 weeks before the peak of the D-phase

syconia (Figure 4). During the first peak of B-phase

male syconia, approximately 80% of male trees bore

B-phase syconia, and roughly 50% of male trees bore

syconia during the second crop (Figure 4). Roughly

80% of female trees bore B-phase syconia during the

summer main crop, and more than half trees bore B-phase

syconia during the fall sub-main crop (Figures 4, 5). The

frequency distribution of flowering syconia production

of D- and B-phase syconia for both genders were similar

to the proportional distributions of female and male trees

with flowering-phases syconia in each census, a very

significant correlation (Table 2). Each male tree produced

1 to 3 (4) generations of pollinator each year (Tzeng et al.,

2003) with the B-phase and D-phase male syconia present

continuously year round and matching each other (Figures

4, 5, 6).

The relationship between syconia development

and weather

The abundant D- and B-phase syconium reproduction

of both genders was staggered with heavy rainfall by

roughly 2 to 3 weeks although data analysis did not

identify a significant correlation between rainfall and

reproduction (Table 3). The abundance of D-phase syconia

of the 1995 over-winter crop and the B-phase syconium

reproduction in the main male crop in spring occurred

prior to the heavy rainfall in the rainy season (Figure

4). The main male crop in spring developed to the full-

bloom D-phase from June to July between the heavy

rainfall of the rainy and typhoon season. Female fig trees

bore two peaks of abundant B-phase syconia from June

to September (Figure 4). The first peak of the B-phase

female syconia occurred between the heavy rainfall of the

rainy and typhoon season. The second peak of the B-phase

female syconia occurred after the heavy rainfall in the

typhoon season.

The receptive syconia were pollinated or oviposited for

both male and female trees and developed into C-phase.

The abundance of the C-phase syconia of female trees was

significantly correlated with temperature and negatively

correlated with temperature for the males (Table 3).

However, C-phase syconia production was not correlated

with rainfall for either gender (Table 3); C-phase syconia

were plentiful during periods of heavy rainfall in the rainy

and typhoon seasons.

312

Botanical Studies, Vol. 47, 2006

DISCUSSION

Symbiotic relationship between fig and fig

wasp

In the Guandaushi Forest Station, the B- and D-phase

syconia in the flowering stage were numerous during the

specified period. The relationship between F. erecta var.

beecheyana and its obligate pollinator, B. nipponica, is a

symbiotic-pollination relationship, somewhat similar to

the model of F. carica and B. psenes in southern France

(Kjellberg et al., 1987). Ficus erecta var. beecheyana,

which shares affinity with F. carica, belongs to the

subgenus Ficus, section Ficus (Corner, 1965; Tzeng,

2004), and might have processes similar to F. carica.

In a seasonal environment, F. erecta var. beecheyana

produced seasonal and asynchronous crops between its

male and female populations, with the males producing

syconia earlier than the females by roughly 2 to 3 months

(Wu, 1996; Tzeng et al., 2003). At the Guandaushi Forest

Station, fig wasps emerged from the D-phase syconia of

the 1995 over-winter crop between March and April when

the receptive syconia of the main male crop in spring were

Table 2. Kendall rank correlations (£n) between rainfall, temperature, and developmental syconia production for each observation

date of Ficus erecta var. beecheyana (males/females = 17/13, observation times = 74) at Guandaushi Forest Station from October

1995 to February 1997. Each significance test involves a separate risk of a type I error. NS indicates that means in a given column

do not signigicantly differ at the 5% level.

¡ñ D-phase syconia ¡ñ D-phase tree (%) ¡ñ B-phase tree (%) ¡ð B-phase tree (%)

¡ñ D-phase syconia

£n

1.000

0.079

-0.066

£l

0.000

NS

NS

¡ñ B-phase syconia

£n

0.121

0.121

0.834

£l

NS

NS

0.000

¡ð B-phase syconia

£n

-0.066

-0.066

1.000

£l

NS

NS

0.000

Table 3. Kendall rank correlations (£n) between rainfall, temperature, and developmental syconia production for each observation

date of Ficus erecta var. beecheyana (males/females = 17/13, observation times = 74) at Guandaushi Forest Station from October

1995 to February 1997. Each significance test involves a separate risk of a type I error. NS indicates that means in a given column

do not signigicantly differ at the 5% level.

¡ñ B-phase syconia ¡ñ C-phase syconia ¡ñ D-phase syconia ¡ð B-phase syconia ¡ð C-phase syconia

Rainfall

£n

-0.037

-0.076

0.088

0.120

-0.010

£l

NS

NS

NS

NS

NS

Temperature £n

-0.053

-0.185

-0.047

0.521

0.480

£l

NS

0.020

NS

0.000

0.000

Figure 6. Pollinational-symbiotic ecological modal of F. erecta

var. beecheyana and its obviously pollinator, B. nipponica at

Guandaushi Forest Station from October 1995 to February 1997.

Narrows indicate that female pollinators are flying-out from the

D-phase syconium to female syconium.

TZENG et al. ¡X Pollinational-mutualism strategy of

F. erecta

var.

beecheyana

and

B. nipponic

a in Taiwan

313

Figure 7. a, Syconia on the branch of Ficus erecta var. beecheyana; b, Blastophaga nipponica inside the B-phase syconium; c,

B-phase seed flowers of section syconium; d, stigma of B-phase gall flowers; e, E-phase female syconium; f, Opening ostiole of

D-phase syconium; g, Male B. nipponica mates with female, which is inside the gall; h, Female B. nipponica leaves the gall. Bars

indicate 10 cm for Figure 7a, 5 mm for Figure 7b, 1 mm for Figure 7c, d, g, h, and 1 cm for Figure 7e, f.

314

Botanical Studies, Vol. 47, 2006

plentiful. Most male fig trees initiated new syconia during

the main male crop but bore asynchronous syconia during

other periods within the population (Tzeng et al., 2003).

Male syconia developed with a similar growth ratio from

A-phase to the B-phase when the 1996 over-winter male

syconia matured and released countless of fig wasps.

If a female founder enters a B-phase male syconium

successfully during the bagged experiment, 100 to 200

offspring could be born during the D-phase syconium

several weeks later (Tzeng, 1997). This would extend

the pollinator population by an average of 152 times for

one founder. In contrast, the average number of progeny

(average = 327) inside one D-phase syconium during the

1996 main male crop in a natural state was significantly

higher than the one founder ¡¦s (T=4.209, £l=0.000), despite

the larvae of the non-pollinator S. inubiae¡Xa parasite¡X

which fed on the larvae of the pollinator inside the ovule

(mean number of pollinators=255; T=3.156, £l=0.003) and

had a negative relationship with the pollinator (Tzeng,

1997). Conversely, the fig wasp population released from

the 1995 over-winter crops was substantially larger than

the number of receptive syconia of the 1996 main male

crop in spring. Due to creating a good match between D-

and B-phase of the male syconia, the pollinator of the

1996 main male crop in spring was extended.

In dioecious figs, wasps can only breed in the gall

flowers of male syconium; consequently, pollination of

female syconia is always lethal (Kjellberg et al., 1987).

Therefore, selection must favor wasps that avoid female

trees. The entry of fig wasps into female syconia to extend

their population in the short-term offers no apparent

benefit (Kjellberg et al., 1987). Hence, to maintain

symbiosis dioecious figs must prevent fig wasps from

developing the ability to distinguish between male and

female syconia. Conversely, syconia display sexual

differences at specific times (Valdeyron and Lloyd, 1979;

Kjellberg et al., 1987; Kjellberg and Maurice, 1989;

Spencer et al., 1996, Tzeng et al., 2003), and fig wasps

have acquired the ability to lay eggs in male syconia

only. However, male and female syconia have modified

their structure, phenology and physiology to prevent the

fig wasp from distinguishing between them (Grafen and

Godfray, 1991; Wu, 1996). Although there is no evidence

that the degree of attraction to female- and male-receptive

syconia of F. erecta var. beecheyana is the same for fig

wasps, it has been demonstrated that fig wasps do not

distinguish between male and female syconia through odor

(Ware and Compton, 1994; Hossaert-McKey et al., 1994).

Pollen carriers are likely to be attracted by receptive

syconia of female trees to pollinate, and surplus pollinators

enter B-phase male syconia to lay eggs to further the

survival of the pollinator population. In brief, fig trees

integrate syconia production phenology and modified

characteristics like odor between male and female syconia

to maintain their mutualism.

Few pollinators entered by chance the male receptive

syconia of the 1996 winter crop, and it was maintained

Figure 8. a, Female B. nipponica often lose their wings in the

ostiole when they penetrate into the B phase syconium; b, c, d,

Female B. nipponica oviposites on the gall flower. Bars indicate

500 £gm for Figure 8a, b, c and 100 £gm for Figure 8d.

TZENG et al. ¡X Pollinational-mutualism strategy of

F. erecta

var.

beecheyana

and

B. nipponic

a in Taiwan

315

Tseng, 1999), F. irisana (Subg. Sycidium, Chen, 1998),

and F. ampelas (Subg. Sycidium, Chang, 2003).

Kjellberg et al. demonstrated that a temporal gap was

a selective pressure favoring the fig wasp¡¦s pollinational-

symbiotic system, concluding that it is characteristic of

F. carica (Kjellberg et al., 1987). However, the receptive

syconia were not visited and were aborted while dozens

of wasps emerged before the development of receptive

syconia in F. carica in Southern France. This might be

a result of long artificial culture and small size samples

(four males and three females) of F. carica employed in

southern France (Kjellberg et al., 1987).

Was the phenomenon of D-phase syconia, which

were in abundance after the B-phase syconia, stable in

the symbiotic system. Because the D-phase syconia

only lasted for a few days (Tzeng et al., 2001), emerging

pollinators had to search for another receptive syconia in

which to lay their eggs or pollinate during their short life-

spans of several hours to 2 days (Ramirez, 1970; Janzen,

1979; Baijnath and Ramcharun, 1983; Kjellberg et al.,

1988; Chen, 1994; Wu, 1996; Chen, 1998; Tseng, 1999).

Observational results indicated that B. nipponica survived

1 to 2 days, and generally less than 1 day during the cold

winter (Wu, 1996; Tzeng, 1997). There were no receptive

syconia to enter for ovipositing or pollinating if the

D-phase syconia developed earlier than B-phase syconia.

Therefore, it is not beneficial for pollinators to emerge

before receptive syconia develop.

B-phase syconia of both tree genders remain on the tree

for roughly 2 to 3 weeks longer than D-phase syconia. In

the monoecious species F. aurea (Bouchaib et al., 1995),

female syconia can survive on the branch for 2 to 5 weeks,

as long as for dioecious species (Bouchaib et al., 1995;

Ware and Compton, 1994; Chen, 1998; Tseng, 1999). The

extended receptive period of the syconia may be a result

of the developmental asynchrony of female flowers inside

the syconia (Tzeng et al., 2001); however, temperature is

the principal factor (Bouchaib et al., 1995).

Although, extending the receptive period wastes the

resources of the trees, a prolonged B-phase provides

increased likelihood for the uncertain flowering period,

low-density population, and seasonal-efficiency of Ficus

for pollination or oviposition (Bouchaib et al., 1995). It is

a critical factor allowing fig and fig wasp to maintain their

pollinational-symbiotic ecological system in a required

minimum number of trees (Bronstein et al., 1990; Anstett

et al., 1995; Bouchaib et al., 1995; Kameyama et al.,

1999). To ensure the future of this relationship, the short-

lived pollinator must enter the B-phase syconia before

they fade. The fig wasps will die before they enter a

receptive syconia to oviposite or pollinate if the receptive

syconia have developed prior to the fig wasp emergence.

That the abundance of receptive syconia occurred

before the numerous of pollinator eclosion (D-phase

syconia) is a flowering characteristic of F. erecta var.

beecheyana. Comparing the cost of a prolonged B-phase

with the successful pollination/oviposition and delayed

by larvae or pupa during the C-phase male syconia in the

winter. For tiny pollinators with a short life span, cold and

rainy weather is fatal (Hill, 1967). However, asynchronous

syconia production can maintain and extend the pollinator

population under such conditions (Ramirez, 1970). In

some male trees, pollinators which escaped from the over-

winter syconia emerge from male syconia and quickly

begin searching for another B-phase male syconia on the

same tree. This can reduce the death ratio and increase the

likelihood of the wasp entering a B-phase male syconia

during the searching period to oviposite.

Ficus erecta var. beecheyana phenologically controls

its gender, preventing the species special pollinator, B.

nipponica, from entering only male syconia. The features

of syconia production within a tree or between trees are

a form of sexual specialization to maintain the fig wasp

populations and attract frugivores for seed dispersal

(Tzeng et al., 2003). Researchers have suggested that it is

the fig tree, not the wasp, which controls the development

time of wasp-producing syconia after pollination occurs

(Kjkllberg et al., 1987).

Observation results showed that B- and D-phase male

syconia were continually present on almost all of the

17 male trees during the study period from Oct. 1995

to Feb. 1997. There were 1 to 3 (4) crops in the male

population (Tzeng et al., 2003) and 1 to 3 (4) generations

of pollinator success in the study population year round.

Although, there was no information on the dynamics of

the pollinator population, less than 20 male trees might

be the smallest survival population of B. nipponica, the

obvious pollinator, in the pollination-symbiotic ecological

system at the Guandaushi Forest Station.

A "temporal gap" between D- and B-phase

syconia production

This study has identified a temporal gap between the

mass production of B- and D-phase syconia in F. erecta

var. beecheyana at the Guandaushi Forest Station. This

phenomenon was somewhat similar to that of F. carica

(Kjellberg et al., 1987) in southern France. However, the

temporal gap of F. erecta var. beecheyana differed from

F. carica in the number of receptive syconia since both

genders were developed earlier than the highest number

of D-phase syconia while female pollinators emerged,

i.e., the emergence of fig wasps from the delayed syconia

(D-phase) occurred earlier than the maturation of the

undelayed syconia (B-phase) in which the fig wasp

oviposited in F. carica (Kjellberg et al., 1987).

The highest D-phase syconia occurred roughly 2 to 3

weeks after the peak of B-phase male and female syconia

in F. erecta var. beecheyana; i.e., the B. nipponica were

released from the D-phase male syconia later than female

syconia of both genders. During the delay period of

D-phase syconia, a few female syconia were entered by

pollinators, released from male syconia. This phenomenon

has been observed in other dioecious figs, such as F. erecta

var. beecheyana (Wu, 1996), F. formosana (Subg. Ficus,

316

Botanical Studies, Vol. 47, 2006

peak production of D-phase, it showed that dozens

of wasps which emerged after the peak of receptive

syconia would benefit evolution between the short life-

span of B. nipponica and F. erecta var. beecheyana

in seasonal Guandaushi Forest Ecosystem. In one big

enough population, it might present another strategy

on the pollination-symbiosis ecological system in the

seasonal environment for other fig species and its obvious

pollinator. And it has to more research in pollination-

symbiosis to understand the co-evolution of fig and fig

wasp.

Acknowledgement. The authors would like to thank

the National Science Council of the Republic of China,

Taiwan (Long Term Ecological Research: Guandaushi

Forest Ecosystem-Pollination Ecology) for financially

supporting this research under Contract No. NSC

87-2613-B-005-085 A07. Hue-Sun Forest Station and

its officials are appreciated for kindly providing research

permits. Prof. Chen Ming-Yih and Prof. Chou Lien-Siang

are appreciated for their valuable discussions. Thanks to

Dr. Chou Liang-Yi for identifying the fig wasp species.

lITeRATURe CITeD

Anstett, M.C., G. Michaloud, and F. Kjellberg. 1995. Critical

populati on s ize for fig/ wa sp mut ualis m in a s ea sonal

environment: effect and evolution of the duration of female

receptivity. Oecolgia 103(4): 453-61.

Baijnath, H. and S. Ramcharun. 1983. Aspects of pollination and

floral development in Ficus capensis Thunb. (Moraceae).

Bothalia 14: 883-888.

Beck, N.G. and E.M. Lord. 1988a. Breeding system in the Ficus

carica, the common fig. I. Floral divers ity. Am. J . Bot.

75(12): 1904-1912.

Beck, N.G. and E.M. Lord. 1988b. Breeding system in the Ficus

carica, the common fig. II. Pollination events. Am. J. Bot.

75(12): 1913-1922.

Berg, C.C. 1989. Cla ss ifi cation and distribution of Ficus.

Experientia 45: 605-611.

Bouchaib, K., M. Gibernau, M.-C. Anstett, F. Kjellberg, and M.

Hossaert-Mckey. 1995. When figs wait for pollinators: The

length of fig receptivity. Am. J. Bot. 82: 992-999.

Brons tein, J .L. 1988. Mutualism , antagonism, and the fig-

pollinator interaction. Ecology 69: 1298-1302.

Bronstein, J.L. 1989. A mutualis m at the edge of its range.

Experientia 45: 622-637.

Bronstein, J.L. and A. Patel. 1992. Causes and consequences of

within-tree phenological patterns in the Florida strangling

fig, Ficus aurea (Moraceae). Am. J. Bot. 79: 41-48.

Bronstein, J.L., P. Gouyon, C. Gliddon, F. Kjellberg, and G.

Michaloud. 1990. The ecological consequences of flowering

asynchrony in monoecious figs: a simulation study. Ecology

71: 2145-2156.

Chang, W.C. 2003. Floral phenology and pollination ecology

of Ficus ampelas Burm. f. at Chiayi. MS Thesis, Graduate

Institute of F ores try, College of Agric ult ure, National

Chiayi University, Chiayi, Taiwan, 128 pp. [in Chinese with

English summary].

Chen, C.H. and L.Y. Chou. 1997. The Blastophagini of Taiwan

(Hymenoptera: Agaonidae: Agaoninae). J. Taiwan Mus.

50(2): 113-154.

Chen, Y.L. 1998. Studies of phenology and interaction between

Fic us ir isana El m. (Morace ae) and its fig was ps. MS

Thesis, Department of Entomology, National Chung-Hsing

University, Taichung, Taiwan, 71 pp. [in Chinese with

English summary].

Chen, Y.R. 1994. Phenology and interaction of fig wasps and

Ficus microcarpa L. MS Thesis, Graduate Institute of Plant

Pathology and Entomology, National Taiwan University,

Taipei, Taiwan, 86 pp. [in Chinese with English summary].

Corner, E.J.H. 1965. Check-list of Ficus in Asia and Australasia

with key to identification. Gard. Bull. Sing. 21: 1-186.

Corlett, R.T. 1987. The phenology of Ficus fis tu los a i n

Singapore. Biotropica 19: 122-124.

Galil, J . 1973. Pollinati on in dioecious fi gs pollination of

Ficus fistulosa by Ceratosolen hewitti. Gard. Bull. 16(2):

303-311.

Galil, J. 1977. Fig biology. Endeavour 1: 52-56.

Galil, J. and D. Eisikowitch. 1968. Flowering cycles and fruit

types of Ficus sycomorus in Israel. New Phytol. 67:

745-758.

Galil, J., R. Dulberger, and D. Rosen. 1970. The effects of Sy-

cophaga sycomori L. on the structure and development

of the syconia in Ficus sycomorus L. New Phytol. 69:

103-111.

Grafen, A. and H. C. J. Godfray. 1991. Vicarious selection ex-

plains some paradoxes in dioecious fig-pollinator systems.

Proc. R. Soc. Lond. Ser. Biol. Sci. 245: 73-75.

Harrison, R. D., N. Yamamura, and T. Inoue. 2000. Phenology

of a common roadside fig in Sarawak. Ecol. Res. 15: 47-61.

Hill, D.S. 1967. F igs (Ficus sp.) of Hong Kong. Hong Kong:

Hong Kong Univ. Press, 130 pp.

Hossaert-McKey, M.H., M. Gigernau, and J.E. Frey. 1994. Che-

mosensory attraction of fig wasps to substances produced

by receptive fig. Entomol. Exp. Appl. 70: 185-191.

Janzen, D.H. 1979. How to be a fig. Ann. Rev. Ecol. Syst. 10:

13-51.

Kameyama, T., R. Harrison, and N. Yamamura. 1999. Rersis -

tence of a fig wasp popuation and evolution of dioecy in

figs: a simulation study. Res. Popul. Ecol. 41: 243-52.

Khadari, B., M. Gibernau, M.C. Anstett, F. Kjellberg, and M.

Hossaert-McKey. 1995. When figs wait for pollinators: the

length of fig receptivity. Am. J. Bot. 82: 992-99.

Kjellberg, F., B. Doumesche, and J.L. Bronstein. 1988. Longev-

ity of a fig wasp (Blastophaga psenes). Proc. K. Ned. Akad.

van Wet., Ser. C. Biol. Med. Sci. 91(2): 117-122.

Kjellberg, F., P.H. Gouyon, M. Ibrahim, M. Raymond, and G.

Valdeyron. 1987. The stability of the symbiosis between

TZENG et al. ¡X Pollinational-mutualism strategy of

F. erecta

var.

beecheyana

and

B. nipponic

a in Taiwan

317

dioecious figs and their pollinators: a study of Ficus carica

L. and Blastophaga psenes L. Evolution 41(4): 693-704.

Kjellberg, F. and S. Maurice. 1989. Seasonality in the reproduc-

tive phenology of Ficus: Its evolution and consequences.

Experientia 45: 653-660.

Liao, J.C. 1995. The taxonomic revisions of the family Mora-

ceae in Taiwan (Ed II). Taipei, Taiwan: Department of For-

estry, College of Agriculture, National Taiwan Univ., 202

pp.

Li u, T. 1968. S tu dies on th e cl as si fica tio n of the c lim ax

vegetation communities of Taiwan. 1. Classification of

the climax formations of the vegetation of Taiwan. Bull.

Taiwan For. Res. Inst. No. 166, 25 pp.

Liu, Y.C., F.Y. L u, and C.H. Ou. 1994. Tree s of Tai wan.

Taichung, Taiwan: College of Agriculture, National Chung-

Shing University, pp. 329-348. [in Chinese].

Lu, K.C. and C.H. Ou. 1994. A preliminary Study on the Forest

Vegetation of Guandaushi Forest Ecosystem. Q. J. For. Res.

18(1): 77-108.

Okamoto, M. and M. Tashiro. 1981. Seasonal fluctuation of

the quantity of syconia on male and female trees of Ficus

erecta. Bull. Osaka Mus. Nat. Hist. 35: 43-53.

Patel, A. and D. McKey. 1998. Sexual specialization in two

tropical dioecious figs. Oecologia 115: 391-400.

Ramirez, B.W. 1970. Host specificity of fig wasps (Agaonidae).

Evolution 24: 680-691.

S pencer, H., G. Weiblen, and B. F lick. 1996. Phenology of

Ficus variegata in a seasonal wet tropical forest at Cape

Tribulation, Australia. J. Biogeogr. 23: 467-475.

Su, H.J. 1992. Vegetation of Taiwan: altitudinal vegetation zones

and geographical climatic regions. In C.-I Peng (ed.), The

Biological Resources of Taiwan: a Status Report. Taipei:

Institute of Botany, Academia Sinica Monography Series:

11, pp. 39-53. [in Chinese with English summary].

Thomson, J.D, S. Dent-Acousta, P. Escobar-Raramo, and J.D.

Na son. 1997. Within-crown flowwe ring s ynchrony in

strangler figs, and its relationship to allofusion. Biotropica

29(3): 291-297.

Ts eng, L.J. 1999. F lowering phe nology and pollingnation

ecology of Ficus formosana Maxim. at Hue-Sun Forest

Sta tion. MS Thes is , Depa rt nent of Fores try, National

Chung-Hs ing University, Taichung, Taiwan, 86 pp. [in

Chinese with English summary].

Tseng, L.J., C.H. Ou, F.Y. Lu, and H.Y. Tzeng. 2000. Studies

of the development and morphology of Syconium of Ficus

formosana. Q. J. For. Res. 22(3): 55-68.

Tzeng, H.Y. 1997. The symbios is between Ficus erecta var.

beecheyana and Blastophaga nipponica at Hue-Sun Forest

Station. MS Thesis , Taichung, Taiwan: De partme nt of

Forestry, National Chung-Hsing University, Taipei, Taiwan,

104 pp. [in Chinese with English summary].

Tzeng, H.Y. 2004. Taxonomic S tudy of the Genus Ficus in

Taiwan. Ph. D. Thesis, Taichung, Taiwan: Department of

Fores try, National Chung-Hs ing University, Taichung,,

Taiwan, 396 pp. [in Chinese with English summary].

Tzeng, H.Y., C.H. Ou, and F.Y. Lu. 2001. Morphological study

on the syconia of Ficus erecta var. beecheyana. Taiwan

J. Forest. S ci. 16(4): 295-306. [in Chinese with English

summary].

Tzeng, H.Y., C.H. Ou, and F.Y. Lu. 2003. Syconium phenology

of Ficus erecta var. beecheyana at the Hue-S un F orest

Station. Taiwan J. For. Sci. 18(4): 273-282. [in Chinese

with English summary].

Tze ng, H.Y., C.H. Ou, and F.Y. Lu. 2004. P henology and

sexual differentiation of Ficus erecta var. beecheyana at

Guandaushi Forest Sation, Taiwan. Q. J. For. Res. 26(2):

61-78. [in Chinese with English summary].

Va ldeyron, G. a nd D.G. L loyd. 1979. S ex diffe re nc es and

flowering phenology in the common fig, Ficus carica L.

Evolution 55(2): 673-685.

Van Noort, S. and S.G. Compton. 1996. Convergent evolution

of agaonine and sycoecine (Agaonidae, Chalcidoidea) head

shape in response to the constraints of host fig morphology.

J. Biogeogr. 32: 415-424.

Verkerke, W. 1989. Structure and function of the fig. Experientia

45: 612-622.

Verkerke, W. 1990. Fig anatomy reproduction biology of African

Fic us s pecie s (Morace ae). Mitt. Inst . Al lg. Bot. 231:

427-431.

Ware, A.B. and S.G. Compton. 1994. Responses of fig wasps to

host plant volatile cues. J. Chem. Ecol. 20: 785-802.

Ware, A.B., P.T. Ka ye, S .G. Co mpton, a nd S. va n Noort .

1993. Fig volatiles: their role in attracting polinators and

maintaining pollinator s pec ificity. Pl. S ys t. Ecol. 186:

147-156.

Weiblen, G., H. Spencer, and B. Flick. 1995. Seed set and wasp

predation in dioecius Ficus variegata from an Australian

wet tropical forest. Biotropica 27: 391-394.

Wi eb es , J .T. 1979 . Co- evol ut ion o f fig an d the ir ins e ct

pollinators. Ann. Rev. Ecol. Syst. 10: 1-12.

Wu, H.F. 1996. The symbiosis between Ficus erecta Thunb.

var. beecheyana and Blastophaga nipponica at Yang-Ming

Shan. MS Thesis, Graduate Institute of Zoology, National

Taiwan University, Taipei, Taiwan, 54 pp. [in Chinese with

English summary].

Yu, F. C. 2001. Meteorology. In M.Y. Chen, B.H. Sheu, and

S.H. Wu (eds .), Ecology of Guandaushi Forest S tation.

Taihcung, Taiwan: Experiment Forest of National Chung

Hsing University, pp. 22-35. [in Chinese].

318

Botanical Studies, Vol. 47, 2006