全 文 ：昆 虫 学 报 Acta Entomologica Sinica，February 2016，59(2):239 － 246 doi:10. 16380 / j. kcxb. 2016. 02. 014
Mechanisms of population regulation in the mutualism
between Breynia vitis-idaea (Euphorbiaceae)and
Epicephala vitisidaea (Lepidoptera，Gracillariidae)
ZHANG Jing#，YIN Yan-Ｒu#，WANG Zhi-Bo，ZHANG Zhen-Guo，
HU Bing-Bing，SHI Fu-Chen* ，LI Hou-Hun*
(College of Life Sciences，Nankai University，Tianjin 300071，China)
Abstract:【Aim】In order to study how Breynia vitis-idaea controls the population of Epicephala vitisidaea．
【Methods】The phenology of B． vitis-idaea and the biology of E． vitisidaea were recorded． The flower-visiting
Epicephala moth species were identified by dissecting their genitalia． For calculating larval survivorship，the
number of larvae inside the fruit，the number of exit holes and oviposition scars on the fruit were counted by
dissecting the stalked and non-stalked fruits in summer and late autumn． Finally，the proportion of stalked and
non-stalked fruits in summer and late autumn was calculated． 【Ｒesults】In Xiamen，Fujian Province，B． vitis-
idaea plants have five peak fruiting stages annually and the flower-visiting Epicephala moths have five
generations each year． The species of moth was identified as E． vitisidaea by contrasting its genitalia． E．
vitisidaea larvae need to consume 2 － 4 seeds to develop into mature larvae． B． vitis-idaea could produce two
types of fruits，stalked and non-stalked． The larval survivorship of E． vitisidaea was obviously higher in non-
stalked fruits than in stalked fruits as well as in late autumn than in summer． Meanwhile，the proportion of non-
stalked fruits of B． vitis-idaea was obviously higher in late autumn (82． 04%)than in summer (31． 53%)．
【Conclusion】This study revealed the mechanism by which mutualistic interactions are more stable． B． vitis-
idaea can control the larval survivorship of E． vitisidaea by producing the stalk in base of fruits． By seasonally
regulating the ratio of stalked and non-stalked fruits，Breynia plants prevent seed overconsumption in summer
and ensure Epicephala survival through the winter． The auto-regulating mechanism of B． vitis-idaea plays an
important role in keeping stability of mutualistic system between B． vitis-idaea and E． vitisidaea．
Key words:Breynia vitis-idaea;Epicephala vitisidaea;fruit stalk;larval survivorship;obligate pollination
mutualism;regulation mechanism
基金项目:国家自然科学基金项目(30930014，31311140254)
作者简介:张晶，女，1984 年 9 月生，辽宁大连人，研究方向为植物与蛾类协同进化，E-mail:2009jingzhang@ 163． com;尹彦入，女，1986 年9
月生，四川内江人，研究方向为环境与资源植物学，E-mail:yinyru@ 126． com
#共同第一作者 Authors with equal contribution
* 通讯作者 Corresponding authors，E-mail:lihouhun@ nankai． edu． cn;fcshi@ nankai． edu． cn
收稿日期 Ｒeceived:2015-11-03;接受日期 Accepted:2016-01-23
1 INTＲODUCTION
Mutualisms are reciprocally beneficial inter-
specific interactions，which are viewed as balanced
antagonistic interactions that nonetheless provide net
benefits to each mutualist (Axelrod and Harnilton，
1981;Bull and Ｒice，1991)． They are found at all
levels of biological organization and are widely
appreciated for their fundamental importance in the
evolution and maintenance of biodiversity (Herre et
al．，1999;Kawakita and Kato，2006)． There are，
however，clear conflicts of interests in mutualisms
between plants and seed-parasitic pollinators． Plants
depend on the pollination of flower-visiting insects
but must limit the seed consumption by the larvae of
these insects． In the mutualism between figs and fig
wasps，for example，figs limit the number of eggs
laid by fig wasps by varying the length of flower
styles (Cook and Ｒasplus， 2003 )． In the
mutualisms between yucca trees and yucca moths，
and between Glochidion trees and Epicephala moths，
plants selectively abscise flowers with high egg loads
to prevent excessive seed destruction by moth larvae
(Pellmyr and Huth，1994;Goto et al．，2010)．
Kawakita and Kato (2004) found that an
Epicephala moth obligately pollinates B． vitis-idaea．
They found that single Epicephala larva consumed a
fraction of seeds within developing fruit in B． vitis-
idaea and some of the fruits were left untouched． For
many of these untouched fruits with indication of moth
oviposition，they suggested that egg / larval mortality of
240 昆虫学报 Acta Entomologica Sinica 59 卷
Epicephala moths is an important factor assuring seed
set in these plants． Svensson et al． (2010)showed
that system-specific chemistry is not necessary for
efficient host location by exclusive pollinators in
mutualism between B． vitis-idaea and its host-
specific Epicephala pollinator． Zhang et al．
(2012b)found that different structures in the female
ovipositor and oviposition mode caused different
population of co-pollinators:E． mirivalvata and E．
lativalvaris．
In Fujian Province，we found that B． vitis-idaea
had two types of fruits， stalked and non-stalked
(Fig． 1:A) ，and confirmed that E． vitisidaea moth
was the obligate pollinator of B． vitis-idaea plant．
Here we focused on the two types of fruits of B． vitis-
idaea to investigate the phenology，observed the life
history，oviposition behavior and larval survivorship
of E． vitisidaea． Finally，we answered how the B．
vitis-idaea controlled the population of E． vitisidaea
through bearing stalked and non-stalked fruits．
Fig． 1 Morphology of Breynia vitis-idaea
A:Fruits of two types on one branch:a fruit without a stalk (left) ，and a fruit with a stalk (right) ;B:An E． vitisidaea moth laying an egg;C:Male
flower;D:Female flower． WS:Non-stalked;S:Stalk．
2 MATEＲIALS AND METHODS
2． 1 Studied plant
Breynia vitis-idaea (Burm． f．)Fischer is an
evergreen and perennial species of Phyllanthaceae．
It is a monoecious shrub that can reach 1． 5 － 5． 0 m
in height． It grows in montane scrubs at 150 － 1 000
m in Fujian Province and is widely distributed in
China (Guangdong，Guangxi，Guizhou，Taiwan，
Yunnan) ，across southeastern Asia and the Indian
subcontinent (Li，1994;Li and Gilbert，2008)．
Sepals of both male and female flowers are fused in
an inverted mitriform (Fig． 1:C，D) ，which makes
it hard to be touched by other flower visitors except
Epicephala moths． Stamens and stigmas are
concealed within the calyx，making them difficult to
be touched by common flower visitors． Upon
maturation，the calyxes of male flowers dehisce to
form a small pit at the apex． Fruits are spherical and
red to red-purple in color． According to Li and
Gilbert (2008) ，the plants with stalked fruits belong
to B． officinalis，and the plants with non-stalked
fruits belong to B． vitis-idaea． However，Li (1994)
considered B． officinalis as a synonym of B． vitis-
idaea． In this study，we firstly found that B． vitis-
idaea with two fruit types coexist on the same
individual plant (Fig． 1:A)． Here we call these
fruits“stalked”or“non-stalked”． The stalked fruit
is elevated a short distance above the calyx，whereas
the non-stalked fruit clings to the top of the calyx． So
we regard the presence or absence of stalk as well as
the variation of the leaf shape as intraspecific
variations， and confirm that B． officinalis is a
synonym of B． vitis-idaea． As the host plant，B．
vitis-idaea provides food that the E． vitisidaea larvae
use to grow and develop．
2 期 ZHANG Jing et al．:Mutualism between Breynia vitis-idaea and Epicephala vitisidaea 241
2． 2 Studied insect
Epicephala vitisidaea Li，Wang ＆ Zhang，is a
pollinating moth species of Gracillariidae in the order
Lepidoptera (Zhang et al．，2012a)． Proboscises of
the pollinating adult female have cilia for collecting
and transporting pollen grains． These cilia are absent
from the non-pollinating adult males． E． vitisidaea
adults have a grayish brown or yellowish brown
forewing (Fig． 1:B;Fig． 2:A)with three parallel
white stripes that extend obliquely outward from the
basal 1 /3，3 /5，and 3 /4 of the costal margin． They
also have a broad creamy white band on the dorsal
margin． The costa is nearly parallel dorsoventrally，
with long dense ventral setae in male genitalia．
Mature larvae are 5． 0 － 6． 5 mm in length，with a
black head， a dark red thorax， and a white
abdomen． The middle part of each abdominal
segment is red or dark red(Fig． 2:B)． The body is
sparsely covered with setae． As the host plant，B．
vitis-idaea provides food that the E． vitisidaea larvae
use to grow and develop． Larvae feed on B． vitis-
idaea seeds． Larvae exit the fruit when they have
developed to maturity and pupate on the leaves．
Cocoons are 6． 0 － 7． 5 mm in length，white，
elliptical in shape，with some white grains attached
to the surface (Fig． 2:C)． Pupae are 4． 0 － 5． 0
mm in length，fusiform，and pale yellow during early
pupal stages (Fig． 2:D)． They gradually become
grayish brown before eclosion．
Fig． 2 Life history of Epicephala vitisidaea
A:Coupling moths;B:A mature larva that has exited from a fruit and is ready to pupate on the host leaf;C:A pupal cocoon woven by a mature larva
on a host leaf;D:A pupa that has been stripped from a cocoon．
2． 3 Determination of mutualism systems
We tracked the development of B． vitis-idaea
flowers and fruits in Xiamen，Fujian Province from
June 2010 to September 2011 for a phenological
study and collected fruits when they had matured． To
ascertain the oviposition behavior of E． vitisidaea，
we dissected 109 pollinated female flowers that were
collected from June 2010 to September 2011 at
Xiamen University，Jinbang Park，Hongshan Park
and Gulang Island in Xiamen，Fujian Province． We
also dissected the genitalia of flower-visiting
Epicephala moths to identify the species． To
determine the impact of the fruit stalk on larval
survivorship，we collected and dissected stalked
fruits of different developmental stages，specifically
asking whether the eggs had yielded viable larvae．
We dissected 532 stalked and 397 non-stalked
mature fruits from 15 individual B． vitis-idaea plants
in summer (June to August) and late autumn
(November)． We counted the number of oviposition
scars on the fruit to confirm the number of eggs and
counted both the number of larvae inside the fruit
and the number of exit holes to calculate larval
survivorship． To calculate emergence rates， we
reared larvae (see below)and counted the number
of pupae and the number of adults that survived． The
242 昆虫学报 Acta Entomologica Sinica 59 卷
emergence rate equaled the number of adults divided
by the number of pupae．
Mature fruits were collected randomly from 15
individual plants every fruiting period (4 in Xiamen
University，3 in Jinbang Park，5 in Hongshan Park，
and 3 in Gulang Island in Xiamen，Fujian Province)
and placed in transparent plastic cylindrical boxes
(8． 5 cm × 12 cm) to calculate the proportion of
stalked and non-stalked fruits (the proportion of
stalked or non-stalked fruits = the number of stalked
or non-stalked fruits / the number of all sampled
fruits)． We reared E． vitisidaea larvae and recorded
how and when the larvae exited the fruit，pupated，
and emerged through field and laboratory
observations． Pupae were kept in transparent vitreous
cylindrical boxes (2． 7 cm × 7． 2 cm) ，which were
put in 40% －60% humidity conditions．
2． 4 Statistical analyses
Statistical analyses were performed with SPSS
20 (SPSS，Chicago)． Different seasons and fruit
types were analyzed using a two-way analysis of
variance (GLM statistical model)． Differences
between the means were compared using Duncan’s
multiple-range tests at P ＜ 0． 05．
3 ＲESULTS
3． 1 Life history of E． vitisidaea and phenology
of B． vitis-idaea
By dissecting the genitalia of flower-visiting
Epicephala moths，we determined that the pollinators
of B． vitis-idaea were E． vitisidaea moths． The
studied insect and voucher plant specimens are
deposited in the Insect Collection of Nankai
University (NKUM)． Together these species formed
a one-to-one obligate pollination mutual relationship．
In Fujian，B． vitis-idaea had five peak fruiting
stages annually． The first generation of male and
female flowers began to bud from late March to early
April，reaching the peak flowering stage in late
April． The following peak flowering stages happened
in early June，middle July，late August and early
October． E． vitisidaea had five generations
correspondingly，lasting from the 1st generation in
middle April to the 5th generation in early
November， with the overwintering generation
emerging in mid-April． Mature larvae of the
overwintering generation exited during early January
and overwintered as pupae (Table 1)． The flowers
Table 1 Life history of Epicephala vitisidae in Xiamen，Fujian Province
Generation
Apr． May Jun． Jul． Aug． Sep． Oct． Nov． Dec． Jan．
F M L F M L F M L F M L F M L F M L F M L F M L F M L F M L
1st
● ● ●
－ － － － － －
□ □ □ □ □
+ + + +
2nd
● ● ● ●
－ － － － － － －
□ □ □ □ □
+ + + +
3rd
● ● ● ●
－ － － － － － －
□ □ □ □ □
+ + + +
4th
● ● ● ●
－ － － － － －
□ □ □ □ □
+ + + +
5th
● ● ● ●
－ － － － － － － － －
(□) (□) (□) (□) (□) (□) (□) (□) (□) (□) (□)
+ + +
●:Egg;－:Larva;□:Pupa;+:Adult; (□) :Overwintering pupa;F:First ten days;M:Middle ten days;L:Last ten days．
of B． vitis-idaea needed 10 － 25 d to develop to mature
flowers from buds， and pollinated female flowers
needed 20 － 25 d to develop to mature fruits． The non-
pollinated female flowers could develop without pollen
grains，and then withered 10 d later． All the seeds in
withered flowers were infertile．
3． 2 Biology of E． vitisidaea
E． vitisidaea larvae must consume 2 － 4 seeds
(from stalked or non-stalked fruits)to develop into
mature larvae according to our study． Each mature
larva gnawed a round hole in the fruit wall to exit．
They then produced a cocoon and pupated on the
2 期 ZHANG Jing et al．:Mutualism between Breynia vitis-idaea and Epicephala vitisidaea 243
leaves of the host or a nearby plant． The pupal stage
lasted 5 － 8 d． Emerged adults mated on leaves of
the host or a nearby plant． Adult females (n = 37)
actively collected pollen grains from male flowers
(Fig． 3:A)to pollinate female flowers(Fig． 3:B)
and then oviposited on these flowers． E． vitisidaea
female moths may continuously pollinate 2 － 3 female
flowers from bottom base to top end along the same
branch． The female moths deposited pollens on the
stigma by repeatedly rubbing the outspread tip of the
proboscis against stigma (Fig． 3:B) ，and laid eggs
subsequently (Fig． 1:B)． We also observed some
female moths repeating this oviposition behavior on the
same flower，but they never repeated it more than three
times． E． vitisidaea adults survived for 3 － 5 d． E．
vitisidaea is an obligate pollinator of B． vitis-idaea．
Fig． 3 Behavior of Epicephala vitisidaea moths
A:Female moth collecting pollen grains from a male flower;B:Female moth pollinating a female flower;C:An egg laid on a non-stalked fruit of B．
vitis-idaea;D:An egg laid on a stalked fruit of B． vitis-idaea． S:Stalk．
3． 3 Number of E． vitisidaea eggs deposited on
B． vitis-idaea female flowers
By dissecting female flowers (n = 109) ，we
determined that all pollinated female flowers of B．
vitis-idaea loaded E． vitisidaea eggs． This agreed
with our field observation that all Epicephala
individuals oviposited after pollinating flowers at
night (Fig． 1:B)． Most pollinated female flowers
(80． 73%)only had one egg，and the rest had 2 － 4
eggs (Fig． 4)． E． vitisidaea laid eggs between the
ovary and the calyx of B． vitis-idaea female flowers
(Fig． 1:B)． The eggs could be deposited on the
upper (Fig． 3:C) ，middle，or lower part of the
ovary． The female flower buds did not have stalks
and the stalks developed before the eggs hatched
(Fig． 3:D)．
3． 4 Statistics of E． vitisidaea eggs and
larval survivorship
We calculated the number of eggs associated
Fig． 4 Percent of Epicephala vitisidaea eggs per
pollinated female flower of Breynia vitis-idaea
with each fruit by adding the number of larvae and
the number of exit holes． Larval survival rates in
stalked fruits (27． 31%， n = 495 in summer;
29． 73%，n = 37 in late autumn)were lower than
244 昆虫学报 Acta Entomologica Sinica 59 卷
those in non-stalked fruits (63． 30%，n = 228 in
summer;72． 69%，n = 169 in late autumn) (Table
2)． Moth larvae were less likely to survive on
stalked fruit， so this type of fruit limited seed
consumption． During the summer，the proportion of
stalked fruit was high，which promoted seed survival
during these critical months． Non-stalked fruit
predominated during the late autumn， which
increased larval survivorship and reduced the impact
of low emergence rates on the size of the
overwintering E． vitisidaea population．
Table 2 Number of Epicephala vitisidaea eggs and larval survivorship in two fruit types of
Breynia vitis-idaea in summer and in late autumn
Season Fruit type Examined fruits
Number of larvae per fruit
0 1 2 3
Number of
eggs
Number of
living larvae
Larval survivorship
(%)
Summer
(Jul． － Aug．)
Non-stalked fruits 228 98 91 39 0 267 169 63． 30
Stalked fruits 495 378 92 25 0 520 142 27． 31
Sum 723 476 183 64 0 787 311 39． 52
Late autumn
(Nov．)
Non-stalked fruits 169 59 66 41 3 216 157 72． 69
Stalked fruits 37 26 11 0 0 37 11 29． 73
Sum 206 85 77 41 3 253 168 66． 40
Larval survivorship =(the number of larvae + the number of exiting holes)/ the number of eggs in fruits．
During the summer (June to August)we reared
80 E． vitisidaea larvae (57 from stalked fruits and 23
from non-stalked fruits) that developed into pupae．
Forty-four of these pupae (31 from stalked fruits and
13 from non-stalked fruits) emerged as adults
(emergence rate = 55%)． E． vitisidaea overwinter as
pupae for 4 months (Table 1)． We observed 30
overwintering mature larvae (4 from stalked fruits and
26 from non-stalked fruits) ，only seven mature larvae
(1 from stalked fruits and 6 from non-stalked fruits)
pupated and emerged as adults in the spring
(emergence rate = 23． 3%)． As such，emergence
rates are lower for overwintering pupae． Larval
survivorship and proportion of stalked fruits were
significantly affected by an interaction of seasons and
fruit types (Table 3)． These studies have revealed a
positive correlation between the proportion of non-
stalked fruits and larval survivorship (Fig． 5)． The
high percentage of stalked fruits was related with the
low larval survivorship in summer and the high
percentage of non-stalked fruits was related with the
high larval survivorship in late autumn．
Table 3 Two-way ANOVA for larval survivorship of
Epicephala vitisidaea and fruit type of Breynia vitis-idaea
in summer and late autumn
df
Larval survivorship Proportion of non-stalked fruits
F P F P
Season (Se) 1 0． 007 0． 936 0． 006 0． 939
Stalked (St) 1 267． 1 ＜0． 01 141． 6 ＜0． 01
Se × St 1 10． 99 ＜0． 01 1372 ＜0． 01
Significant P-values are in bold．
4 DISCUSSION AND CONCLUSION
Explaining the evolutionary stability of
interspecific mutualism remains one of the major
challenges in evolutionary biology (Sachs et al．，2004;
West et al．，2007;Kiers and Denison，2008)． Some
figs limit the number of eggs laid by fig wasps by
varying the length of the style，thereby ensuring seed
production (Cook and Ｒasplus，2003;Wang et al．，
2009 )． In the mutualism between Glochidion
acuminatum plant and Epicephala moth，the proportion
of aborted flowers progressively increased with higher
egg load and thus increased the ovule damage (Goto et
al．，2010)． Differing from the mechanism in the above
mutualism，our study concerning the survivorship of E．
vitisidaea larvae in stalked and non-stalked B． vitis-
idaea fruits showed that the host plays a critical role in
regulating the Epicephala population size by varying the
stalk types that can prevent seed overconsumption．
Survivorship rates of E． vitisidaea larvae that
parasitized stalked fruits were lower than those that
parasitized non-stalked fruits (Table 2)． The stalk
began to develop after the female moth laid the eggs，
and E． vitisidaea individuals could not know whether
the female flowers would develop into stalked fruits． As
such，they could not lay their eggs selectively on non-
stalked fruits to improve larval survivorship． The stalk
developed before the E． vitisidaea eggs hatched (Fig．
3:D) ，lengthening the distance between the ovary and
the egg． Larvae that hatched from eggs laid on the
lower or middle part of the ovary of a stalked fruit had
to climb up the stalk to access the ovaries． This climb
consumed a great deal of energy and make the larvae
more prone to be predated，thus increasing costs to
individual larvae，making the feeding process more
difficult，reducing larval survivorship，and decreasing
seed consumption． Larvae hatched from eggs (Fig． 3:
C)laid on the upper part of the ovary of stalked fruits
or from eggs laid on non-stalked fruits did not have to
climb a stalk and could directly feed on seeds． This
decreased energy consumption and increased larval
survivorship．
2 期 ZHANG Jing et al．:Mutualism between Breynia vitis-idaea and Epicephala vitisidaea 245
Fig． 5 Larval survivorship of Epicephala vitisidaea (A)and proportion of Breynia vitis-idaea fruits (B)in summer and late autumn
Bars are means ± SE． Significant differences (P ＜ 0． 05)for each variable are indicated by different lowercase letters． Significant differences for each
variable were compared using Duncan’s multiple-range tests．
According to our field study，the proportion of
non-stalked fruits decreased in summer and the larvae
survivorship also decreased in summer (Fig． 5)
(Table 2)． The decreasing proportion of non-stalked
fruits might attribute to the increasing moth population
in summer， which resulted in more intensive
competition of the common resources (flowers)． This
might lead to overexploitation by more eggs from the
Epicephala moths，and might stimulate the Breynia
plants to develop more stalk fruits to prevent seed
overconsumption． The emergence rates of E． vitisidaea
were lower in late autumn (23． 3%)than in summer
(55%)． In addition，the proportion of non-stalked fruits
of B． vitis-idaea was higher in late autumn (82． 04%)
than in summer (31． 53%) (Table 2)． Accordingly，
larval survivorship was higher in late autumn
(66． 40%)than in summer (39． 52%) (Table 2)．
The high proportion of non-stalked fruits in late autumn
reduced the overwintering danger to E． vitisidaea，
thereby decreasing the impact of low wintertime
emergence rates on the E． vitisidaea population size
and ensuring that enough E． vitisidaea adults emerged
to pollinate B． vitis-idaea in the coming year．
Emergence rates of E． vitisidaea increased in summer
along with the proportion of stalked fruits． This
decreased larval survivorship (39． 52%)and prevented
seed overconsumption by larvae．
We show here that B． vitis-idaea plants actively
regulate the population size of E． vitisidaea moths in a
season-specific manner， thereby creating a stable
mutualistic relationship． This represents a new
mechanism for facilitating mutualism．
ACKNOWLEDGMENTS We are grateful to LI Zhen-Ji，
HOU Xue-Liang， HUANG Zhi-Jun and WU Wei-Huan of
Xiamen University and ＲEN An-Zhi of Nankai University for
their assistance in the field and statistical analyses． Special
thanks are given to three reviewers for reviewing the manuscript
and giving insightful comments and suggestions．
Ｒeferences
Axelrod Ｒ，Harnilton WD，1981． The evolution of co-operation． Science，
211:1390 － 1396．
Bull JJ，Ｒice WＲ，1991． Distinguishing mechanisms for the evolution of
co-operation． Journal of Theoretical Biology，149(1) :63 － 74．
Cook JM，Ｒasplus J-Y，2003． Mutualists with attitude:coevolving fig
wasps and figs． Trends in Ecology ＆ Evolution，18(5) :241 － 248．
Goto Ｒ，Okamoto T，Kiers ET，Kawakita A，Kato M，2010． Selective
flower abortion maintains moth cooperation in a newly discovered
pollination mutualism． Ecology Letters，13(3) :321 － 329．
Herre EA，Knowlton N，Mueller UG，Ｒehner SA，1999． The evolution
of mutualisms:exploring the paths between conflict and cooperation．
Trends in Ecology ＆ Evolution，14(2) :49 － 53．
Kawakita A，Kato M，2004． Obligate pollination mutualism in Breynia
(Phyllanthaceae) :further documentation of pollination mutualism
involving Epicephala moths (Gracillariidae)． American Journal of
Botany，91(9) :1319 － 1325．
Kawakita A，Kato M，2006． Assessment of the diversity and species
specificity of the mutualistic association between Epicephala moths
and Glochidion trees． Molecular Ecology，15(12) :3567 － 3581．
Kiers ET，Denison ＲF，2008． Sanctions，cooperation，and the stability
of plant-rhizosphere mutualisms． Annual Ｒeview of Ecology，
Evolution，and Systematics，39(1) :215 － 236．
Li BT，1994． Flora of China，Vol． 44． Dicotyledoneae，Euphorbiaceae，
Science Press，Beijing． 178 － 184．
Li BT，Gilbert MG，2008． Flora of China． Vol． 11． Dicotyledoneae，
Euphorbiaceae． Missouri Botanical Garden Press，USA． 207 － 209．
Pellmyr O，Huth CJ，1994． Evolutionary stability of mutualism between
yuccas and yucca moths． Nature，372(6503) :257 － 260．
Sachs JL，Mueller UG，Wilcox TP，Bull JJ，2004． The evolution of
cooperation． Quarterly Ｒeview of Biology，79(2) :135 － 160．
Svensson GP，Okamoto T，Kawakita A，Goto Ｒ，Kato M，2010．
Chemical ecology of obligate pollination mutualisms: testing the
‘private channel’hypothesis in the Breynia-Epicephala association．
New Phytologist，186:995 － 1004．
Wang ＲW，Ｒidley J，Sun BF，Zheng Q，Dunn DW，Cook J，Shi L，
Zhang YP，Yu DW，2009． Interference competition and high
temperatures reduce the virulence of fig wasps and stabilize a fig-
wasp mutualism． PLoS ONE，4(11) :e7802．
West SA，Griffin AS，Gardner A，2007． Evolutionary explanations for
cooperation． Current Biology，17(16) :Ｒ661 － Ｒ672．
246 昆虫学报 Acta Entomologica Sinica 59 卷
Zhang J，Hu BB，Wang SX，Li HH，2012a． Six new species of
Epicephala Meyrick，1880 (Lepidoptera:Gracillariidae)associated
with Phyllanthaceae plants． Zootaxa，3275:43 － 54．
Zhang J，Wang SX，Li HH，Hu BB，Yang XF，Wang ZB，2012b． Diffuse
coevolution between two Epicephala species (Gracillariidae)and two
Breynia species (Phyllanthaceae)． PLoS ONE，7(7) :e41657．
小叶黑面神与小叶头细蛾互惠共生
系统的种群调控机制
张 晶#，尹彦入#，王志博，张振国，胡冰冰，石福臣* ，李后魂*
(南开大学生命科学学院，天津 300371)
摘要:【目的】探索小叶黑面神 Breynia vitis-idaea对小叶头细蛾 Epicephala vitisidaea种群数量的调
控机制。【方法】跟踪记录小叶黑面神物候及头细蛾的生物学。解剖在小叶黑面神上访花头细蛾
的外生殖器，鉴定头细蛾种类。对不同时期小叶黑面神有梗和无梗的果实进行解剖，统计果实内幼
虫数量、果实表面孔的数量以及果实表面产卵疤数量，计算头细蛾幼虫存活率。统计不同时期小叶
黑面神有梗和无梗的果实的比例。【结果】在福建厦门小叶黑面神每年有 5 个花果期，相应地，为
小叶黑面神传粉的头细蛾每年有 5 个生活世代。通过解剖，该种头细蛾被鉴定为小叶头细蛾。一
头小叶头细蛾幼虫需要消耗 2 ～ 4 粒种子才能发育成熟。小叶黑面神有两种不同形态的果实:有梗
和无梗。头细蛾幼虫在无梗果实内的存活率明显高于有梗果实，并且晚秋时期头细蛾幼虫的存活
率要高于夏季。小叶黑面神无梗果实的比例在晚秋(82． 04%)要高于夏季(31． 53%)。【结论】本
研究揭示了维持互利共生体系稳定的机制。小叶黑面神能够通过果实基部果梗的有无来调节小叶
头细蛾幼虫的存活率。小叶黑面神通过季节性的调节有梗果实的比率，既有效避免了夏季种子被
过度消耗的风险，又提高了头细蛾在冬季的存活率。小叶黑面神这种自身调控机制对维持小叶黑
面神与小叶头细蛾互惠共生系统的稳定性起到了至关重要的作用。
关键词:小叶黑面神;小叶头细蛾;果梗;幼虫存活率;专性传粉互利共生;调控机制
中图分类号:Q968 文献标识码:A 文章编号:0454-6296(2016)02-239-08
(责任编辑:赵利辉)