ISSN: 1021-5506

Zoological Studies
An International Journal

Volume 51, Number 2

March, 2012

Published by Biodiversity Research Center

Academia Sinica, Taipei, Taiwan
 Zoological Studies
CHIEF EDITOR MANAGING EDITOR

LI, WEN-HSIUNG LEE, SIN-CHE

Biodiversity Research Center, Biodiversity Research Center,
Academia Sinica, Taipei, Taiwan Academia Sinica, Taipei, Taiwan

ADVISORY BOARD
COLEMAN, DAVID C., USA KNOWLTON, NANCY, USA WU, CHUNG-I, USA
,
EDWARDS, JAMES, Denmark O BRIEN, STEPHEN J., USA

EDITORIAL BOARD

AYALA, FRANCISCO J., USA HWANG, PUNG-PUNG, Taiwan TING, CHAU-TI, Taiwan

CHANG, CHING-FONG, Taiwan LEE, LING-LING, Taiwan TSO, I-MIN, Taiwan

CHANG, ERNEST S., USA LOOF, ARNOLD DE, Belgium WU, SHI-KUEI, USA

CHEN, CHAOLUN ALLEN, Taiwan McCULLOUGH, DALE R., USA XIA, XUHUA, Canada

CHIANG, TZEN-YUH, Taiwan MOK, MICHAEL HIN-KIU, Taiwan YEN, SHEN-HORN, Taiwan

DAI, CHANG-FENG, Taiwan RANDALL, JOHN E., USA YU, HON-TSEN, Taiwan

HUANG, RU-CHIH C., USA SHAO, KWANG-TSAO, Taiwan YU, SIMON S.J., USA

ASSISTANT EDITORS

CHEN, CHUN-CHIAO VANESSA, Biodiversity Research Center, WU, CHIA-CHI KIKI, Biodiversity Research Center, Academia
Academia Sinica, Taipei, Taiwan Sinica, Taipei, Taiwan
ISI Journal Citation Reports®Ranking: 2010: 65/145 (Zoology)
Impact Factor: 1.046

The publication of Zoological Studies, a bimonthly journal, is

supported by Biodiversity Research Center, Academia Sinica,
Taipei 115, Taiwan. Phone and Fax No.: 886-2- 27899529,
E-mail:zoolstud@gate.sinica.edu.tw; URL: http://zoolstud.sinica.
edu.tw

This journal has been awarded by the National Science Council, Taiwan. It
can be available from Editorial Office, Biodiversity Research Center, Academia
Sinica, Taipei 115, Taiwan. Printed by Cabin Printing Co., Ltd. 1st Fl., No. 30,
Lane 210, Sec. 2, Fu-Shin S. Rd., Taipei 100, Taiwan.
Monogamous System in the Taiwan Vole Microtus
kikuchii Inferred from Microsatellite DNA and Home
Ranges (photo by Y.C. Chang)
Zoological Studies 51(2): 137-142 (2012)

Determination of the Thermal Tolerance of Symbiodinium Using the

Activation Energy for Inhibiting Photosystem II Activity
Jih-Terng Wang1,*, Pei-Jie Meng2,3, Yi-Yun Chen1, and Chaolun Allen Chen4,5,6
1
Graduate Institute of Biotechnology, Tajen Univ., Pingtung 907, Taiwan
2
National Museum of Marine Biology and Aquarium, Checheng, Pingtung 944, Taiwan
3
Institute of Marine Biodiversity and Evolution, National Dong Hwa Univ., Checheng, Pingtung 944, Taiwan
4
Institute of Oceanography, National Taiwan Univ., Taipei 108, Taiwan
5
Biodiversity Research Center, Academia Sinica, Nangang, Taipei 115, Taiwan
6
Taiwan International Graduate Program (TIGP)- Biodiversity, Academia Sinica, Nangang, Taipei 115, Tawian

(Accepted October 4, 2011)

Jih-Terng Wang, Pei-Jie Meng, Yi-Yun Chen, and Chaolun Allen Chen (2012) Determination of the thermal
tolerance of Symbiodinium using the activation energy for inhibiting photosystem II activity. Zoological Studies
51(2): 137-142. Holobionts with different Symbiodinium clades or subclades display varying levels of thermal
tolerance; however, an index to quantify and standardize this difference has not yet been formulated. In this
study, the potential for the activation energy (Ea) to inhibit photosystem (PS)II being used to represent the heat
tolerance of Symbiodinium was investigated. As the Ea required for PSII heat denaturation increased, the PSII
apparatus in the algae remained stable at higher temperatures; thus, PSII activity was maintained at higher
temperatures. The Ea was determined by fitting the kinetics data of the decrease in the maximum quantum
yield (Fv/Fm) of freshly isolated Symbiodinium (FIS) at an elevated temperature to the Arrhenius equation. The
results indicated that the PSII activity of FIS linearly decreased with an increase in the incubation time under
thermal stress (r 2 > 0.95), and the rate of PSII denaturation significantly fit the Arrhenius equation (r 2 > 0.95)
after a logarithmic transformation. Comparisons between 5 Symbiodinium subclades indicated that D1a, known
as the most heat-tolerant subclade, showed the highest Ea value (348 ± 16 kJ/mole), which was significantly
(p < 0.05) higher than those of B1, C1, C3, and C15 (126-262 kJ/mole). The reliability of the Ea calculation
was confirmed by the low coefficient of variation (< 10%), suggesting that it can reliably be used to quantify the
thermal tolerance of Symbiodinium. http://zoolstud.sinica.edu.tw/Journals/51.2/137.pdf

Key words: Coral bleaching, Activation energy, PSII activity, Symbiodinium.

S ymbiodinium algae, the dinoflagellates
mostly found in symbiosis with corals and sea
anemones, are widely considered to underpin the
ecological success of cnidarian-alga symbioses
in shallow, nutrient-poor waters (Muscatine and
Porter 1977, Falkowski et al. 1993). However,
thermal stress caused by increasing seawater
temperatures results in a breakdown of symbiotic
associations and seriously threatens coral reefs
worldwide (Hoegh-Guldberg et al. 2007, Lesser
2007). Thermal breakdown of coral-Symbiodinium
symbioses causing coral bleaching was found
to be closely related to the thermal inhibition of
photosystem (PS)II activity of symbiotic algae
(Hill et al. 2004, Takahashi et al. 2008). With the
9 currently described clades (A-I) and numerous
subclades within Symbiodinium (see review in
Coffroth and Santos 2005), the algae were also
shown to exhibit different extents of tolerance to
thermal stress in culture or in hospite (Bhagooli
and Hidaka 2003, Rowan 2004, Tchernov et al.
2004, Robinson and Warner 2006, Sampayo et al.

*To whom correspondence and reprint requests should be addressed. Tel: 886-8-7624002. Fax: 886-8-7621645.
E-mail:jtw@mail.tajen.edu.tw

137
138 Wang et al. – Thermal Tolerance Index of Symbiodinium
2008). Selecting thermally-tolerant Symbiodinium
clades or subclades; therefore, was proposed
as a way to promote the survival of corals in the
coming century (Chen et al. 2005a, Berkelmans
and van Oppen 2006). This proposal was based
either on the biogeographic distribution of thermal-
tolerant Symbiodinium in historically warming
regions (Chen et al. 2005a b, LaJeunesse et al.
2010) or on thermal-tolerant experiments under
controlled laboratory conditions (Rowan 2004,
Tchernov et al. 2004, Sampayo et al. 2008).
However, conflict occurs when thermal tolerance is
determined among different Symbiodinium clades
or subclades. For example, both biogeographic
and physiological experiments showed that
Symbiodinium clade D (specifically subclade D1a)
is the most heat-tolerant clade compared to clades
A, B, and C (Rowan 2004, Chen et al. 2005b,
LaJeunesse et al. 2010). However, analysis of the
thylakoid membrane integrity showed that there
are also thermal-tolerant subclades within clades A,
B, and C, suggesting that a priori ribosomal DNA
phylotyping is not diagnostic for thermal sensitivity
of Symbiodinium associations (Tchernov et al.
2004). To resolve this conflict, it is necessary to
develop a quantitative comparison with a single
parameter or index to determine the thermal
tolerance among Symbiodinium clades and
subclades.
The thermal tolerance between different
Symbiodinium clades or subclades has been
compared by estimating the temperature-
dependent performance of the photosynthesis-
irradiation response (Iglesias-Prieto et al. 1992,
Rowan 2004), the degree of decrease in PSII
activity during heat treatment (Bhagooli and Hidaka
2003, Rowan 2004, Robinson and Warner 2006,
Sampayo et al. 2008), or thermal sensitivity to
the induction of stress proteins (or enzymes) and
their related genes (Downs et al. 2000, Brown et
al. 2002, Souter et al. 2011). However, comparing
results between studies has been difficult due to
the experimental designs and conditions used. In
this study, we attempted to develop a universal
index, as indicated by the activation energy (Ea)
for thermally inhibiting PSII activity, to represent the
thermal tolerance of members of Symbiodinium,
since the PSII activity of Symbiodinium is closely
associated with photosynthesis of the alga and
its symbiotic stability with corals (Robinson and
Warner 2006, and references therein). Moreover,
the function of the PSII apparatus is determined by
the natural state of several proteins, such as the
D1 protein, peridinin-chlorophyll-a-binding proteins,
the chlorophyll-a/chlorophyll-c2/peridinin protein
complex, etc. (Takahashi et al. 2008). Thus,
the loss of PSII activity is expected to follow the
process of protein denaturation. If the denaturation
of PSII proteins follows a first-order reaction, then
the Ea for thermally inhibiting PSII activity could
be calculated from the Arrhenius equation (a
kinetic equation for measuring the Ea by linearly
regressing the rate constant on the reaction
temperature in °K). This Ea value could potentially
represent the thermal tolerance of Symbiodinium.
Based on this idea, this study was conducted by
subjecting freshly isolated Symbiodinium (FIS)
to elevated temperatures, measuring the rate of
decline in PSII activity (as indicated by Fv/Fm over
time), and then fitting the rate constants to the
Arrhenius equation to calculate the Ea.
MATERIALS AND METHODS
Symbiodinium isolation and subclade typing
Freshly isolated Symbiodinium (FIS) samples
used in this study were designated Symbiodinium
C1, C3, C15, D1a, and B1 from 4 hard corals
(Stylophora pistillata, Acropora humilis, Porites
lutea, and Galaxea fasicularis) and a sea anemone
(Aiptasia pulchella), respectively, based on a
recent study (Wang et al. 2011). The corals were
collected by scuba diving in Kenting National Park,
Taiwan (21°55'54"N, 120°44'45"E), and the sea
anemone was obtained from a laboratory culture
as described in Wang et al. (2011). Isolation of
Symbiodinium from each replicate of the animal
host was conducted as previously described
(Wang and Douglas 1997, Wang et al. 2011).
Briefly, coral fragments having about 100 cm2 of
live tissue were stripped of tissue using an air
blast, and tentacles of 10 Aiptasia pulchella were
homogenized with a tissue grinder. After mixing
with 2-3 volumes of artificial seawater (Instant
Ocean, Sarrebourg Cedex, France), the resultant
slurry was passed through a 15-μm nylon mesh to
remove debris. Symbiodinium was then isolated
by centrifugation at 860 xg for 3 min and washed
with artificial seawater 3 times. Symbiodinium was
preserved in 80% ethanol before conducting the
phylotype analysis by resolving the polymerase
chain reaction (PCR) product of the ribosomal
internal transcribed spacer (ITS) 2 in denatured
gel gradient electrophoresis (DDGE) developed
by LaJeunesse et al. (2003) and modified as
described in Wang et al. (2011). Since the co-
 Zoological Studies 51(2): 137-142 (2012)
existence of multiple clades or subclades in a
single coral host is well documented (Chen et
al. 2005a b, Berkelmans and van Oppen 2006),
FIS phylotyping of the ITS2 gene by PCR-
DGGE represented the dominant Symbiodinium
population from which the host was isolated. To
obtain Ea data from a single subclade, the kinetics
data were abandoned if the PCR-DGGE suggested
the possibility of a mixture of clades or subclades
of FIS in the preparation (data not shown).
Fluorescence methodology
FIS samples with about (0.5-0.8) × 10 6
cells/ml, counted with an improved Neubauer
hemocytometer (Marienfeld, Germany), were
maintained at 25°C under dim light (< 5 μE/m2/s,
PAR) for 1 h before proceeding with heat treatment.
Measurement of changes in the maximum
quantum yield [Fv/Fm = (Fm - Fo)/Fm] of FIS at the
elevated temperatures began by suspending an
algal pellet, collected from centrifugation of 10 ml
of an algal suspension at 860 xg for 2 min, in the
original volume of artificial seawater which had
been prewarmed to the experimental temperature
(of 31, 33, 35, 37, 39, or 41°C). The value of Fv/Fm
of the FIS suspension was directly measured at
2-min intervals with a DIVING-PAM fluorometer
(Walz, Germany) at the DIVING-PAM setting of
8 for measuring the light and saturating flash of
the actinic light. The F v/F m of treated FIS was
measured under indoor illumination (< 10 μE/m2/s,
PAR), and heat treatment was completed within
12 or 14 min depending on the temperature used.
The F v/F m value of a control FIS that remained
at 25°C under dim light for 4 h was examined to
evaluate the quality of FIS used in the experiment.
Kinetics and statistical analyses
The rate constant, k (1/min), of PSII pro-
tein denaturation at each temperature was
obtained from the slope of the linear regression
of Fv/Fm values against incubation times. Then,
each k value was natural-logarithmically (ln)
transformed to produce an Arrhenius plot with
1/T in °K. The fitness of the kinetics data to the
Arrhenius equation, [ln(k) = ln(A) - (Ea/R)(1/T)],
was examined by a linear regression of ln(k)
against 1/T. Therefore, the Ea of each sample was
calculated from the Arrhenius equation obtained
above with the gas constant, R (= 8.314 J/mol/°K).
The coefficient of variation (CV) was used to
examine the reproducibility between experiments.
139
Comparisons of Ea values between Symbiodinium
subclades were made using a one-way analysis
of variance (ANOVA) following by Fisher ’s
least significance difference (LSD) test, with a
significance level of p < 0.05.
RESULTS
Fv/Fm values of FIS from each preparation,
which ranged 0.618-0.675, were comparable
between Symbiodinium subclades with a 4-h
incubation at 25°C under dim light. When data
from mixed populations of Symbiodinium subclades
were excluded, Fv/Fm values of all subclades tested
(C1, B1, C3, C15, and D1a) linearly decreased
with incubation time at an elevated temperature, as
shown by data for Symbiodinium subclade D1a in
figure 1A. Coefficients of the linear regression (r 2)
of the decrease in Fv/Fm values with incubation time
at each treated temperature were all significant
(p < 0.05), and were 0.982 ± 0.008 (mean ± S.D.,
n = 40) for D1a, 0.979 ± 0.010 (n = 35) for C15,
0.981 ± 0.015 (n = 30) for C3, 0.980 ± 0.019
(n = 30) for B1, and 0.988 ± 0.007 (n = 40) for C1.
When the logarithmically transformed denaturation
rate (k) of PSII was plotted against 1/T, coefficients
of the linear regression were also significant
(p < 0.05), and showed a good fit to the Arrhenius
equation (r 2 = 0.942-0.985), as shown by D1a data
in figure 1B. The regression coefficients obtained
were 0.961 ± 0.019 (n = 8) for D1a, 0.966 ± 0.012
(n = 7) for C15, 0.965 ± 0.020 (n = 6) for C3, 0.965
± 0.017 (n = 6) for B1, and 0.966 ± 0.013 (n = 8)
for C1. The Ea for PSII denaturation was then
calculated from each Arrhenius equation (Table
1). The Ea values significantly differed among the
5 different Symbiodinium subclades (F4,30 = 288.3,
p < 0.001). The post-hoc analysis with Fisher’s
LSD test also indicated that Symbiodinium D1a
displayed the highest Ea, followed in order by C15,
C3, B1, and C1 (Table 1). Ea values for D1a and
C15 were almost 2-fold higher than those of B1
and C1.
DISCUSSION
This study proposes that the activation energy
for inhibiting PSII activity under thermal stress
could be used to represent the thermal tolerance
of Symbiodinium. With such an index, thermal
tolerances among Symbiodinium subclades could
be compared on a universal scale. In order to
140 Wang et al. – Thermal Tolerance Index of Symbiodinium

test the hypothesis, 5 Symbiodinium subclades,
for which the tolerance or sensitivity to heat was
compared in the literature (LaJeunesse et al. 2003,
Fabricius et al. 2004, Rowan 2004, Berkelmans
and van Oppen 2006), were selected for testing in
this study.
When a Symbiodinium alga is stressed due
to an elevated temperature, many physiological
responses are evoked, including upregulation of
stress protein synthesis, downregulation of normal
protein synthesis, and an increase in protein
denaturation (as reviewed by Brown et al. 2002).
It would be easy to obtain the correlation between
heat-stress indicators of the tested organism and
temperature, but none of them can be summarized
to a constant value to reflect the heat-stress
response or tolerance of a Symbiodinium alga
without a kinetics analysis. Kinetics studies on the
increase and subsequent collapse in the rate of
respiration or heart beat were used to represent
the thermal tolerance of a snail (Stenseng et
al. 2005), crab (Stillman 2002), and shellfish
(Dahlhoff and Somero 1993). Increases in the
rates of respiration and heart beat usually follow
Q 10 over a wide range of temperatures. With
photosynthetic algae, a decline in PSII activity
during heat treatment was found, in this study,
to be very suitable for a kinetics analysis of the
thermal deterioration of Symbiodinium algae for 4
reasons. First, the PSII activity of Symbiodinium
can be instantly determined in situ; therefore,
the time interval for the kinetic analysis can be
precisely controlled. Second, the PSII activity of
Symbiodinium was proven to be closely related to
the photosynthetic capability of the alga (Robinson
and Warner 2006, and references therein). Third,

(A) (B)
0.8 -2

0.7
-3
0.6

0.5
ln (k) (min-1)

-4
Fv /Fm

0.4
-5
0.3

0.2
-6
0.1

0.0 -7
0 2 4 6 8 10 12 14 16 3.18 3.20 3.22 3.24 3.26 3.28
Time (min) T-1 × 103 (°K)

Fig. 1. Representative data obtained from freshly isolated Symbiodinium subclade D1a. (A) Decrease in the maximum quantum yield
(Fv/Fm) of Symbiodinium when incubated at 33 (●), 35 (○), 37 (▼), 39 (■), and 41°C (□). (B) An Arrhenius plot obtained from data in (A).
The equations and coefficients of the linear regression of Fv/Fm against time are: y = -0.0027x + 0.6876, r 2 = 0.984 (33°C); y = -0.0042x
+ 0.677, r 2 = 0.991 (35°C); y = -0.0131x + 0.6984, r 2 = 0.979 (37°C); y = -0.0438x + 0.7445, r 2 = 0.993 (39°C); and y = -0.0551x + 0.7394,
r 2 = 0.967 (41°C). That for ln(k) on 1/T is y = -41315x + 128.96, r 2 = 0.968.

Table 1. Activation energy (Ea) for inhibiting photosystem II activity of freshly isolated Symbiodinium under
thermal stress. n, number of replicates from different colonies; Ea, activation energy, the data of which
followed by the same superscript letter do not significantly differ at p = 0.05 according to Fisher’s LSD test;
CV, coefficient of variation

Cnidarian host Symbiodinium subclade n Ea (kJ/mole) CV (%)

Stylophora pistillata C1 8 126 ± 10 a 7.6

Aiptasia pulchella B1 6 144 ± 7 b 4.9
Acropora humilis C3 6 214 ± 7 c 3.2
Porites lutea C15 7 262 ± 25 d 9.4
Galaxea fasicularis D1a 8 348 ± 16 e 4.5
 Zoological Studies 51(2): 137-142 (2012)
the stability of the PSII apparatus is determined
by the natural state of a set of proteins, especially
the D1 protein (Waner et al. 1999, Takahashi et al.
2008). Fourth, the kinetics of protein denaturation
under heat treatment were reported to follow a
1st-order reaction and comply with the Arrhenius
equation (Weijers et al. 2003), and the Ea for
the thermal inhibition (or denaturation) of PSII
proteins can be easily obtained from the Arrhenius
equation. Consequently, the data obtained in this
study indicated that Ea values for inhibiting PSII
activity of each Symbiodinium subclade were
consistent with previous studies (LaJeunesse
et al. 2003, Fabricius et al. 2004, Rowan 2004,
Tchernov et al. 2004, Berkelmans and van Oppen
2006, Díaz-Almeyda et al. 2011), i.e., D1a and C15
were the 2 most thermally tolerant Symbiodinium
among the tested subclades. Reproducibility of
the Ea data for each Symbiodinium subclade was
determined to be acceptable by examining values
of the CV which ranged 3.2%-9.4%.
In summary, a high regression coefficient
(r 2 > 0.95) obtained from the kinetics data and
the low CV between replicates (< 10%) indicated
that the calculated Ea values for PSII denaturation
were stable and reliable. This evidence suggests
that the Ea for inhibiting PSII activity during
heat stress can be used to quantify the thermal
tolerance of Symbiodinium; thus, facilitating
ecological, physiological, and evolutionary studies
of coral symbiosis and bleaching biology. Based
on this idea, it is also possible to develop an index
to represent bleaching susceptibility of corals by
selecting proper physiological indicator(s), changes
in which with an increase in heating temperature or
light intensity follow a 1st-order reaction.
Acknowledgments: The authors would like to
thank members of the Coral Reef Evolutionary
Ecology and Genetics (CREEG) Group, Biodi-
versity Research Center, Academia Sinica
(BRCAS) for field support and D.P. Chamberlin
for English editing. This work was supported by
an Academic Sinica Thematic grant (2008-2010)
to JTW and CAC, and a National Science Council
grant (NSC96-2628-B-001-004-MY3) to CAC. This
is CREEG-BRCAS contribution no. 69.
REFERENCES
Berkelmans R, MJH van Oppen. 2006. The role of zoox-
anthellae in the thermal tolerance of corals: a “nugget of
hope” for coral reefs in an era of climate change. Proc. R.
141
Soc. Lond. B 273: 2305-2312.
Bhggooli R, M Hidaka. 2003. Comparison of stress suscep-
tibility of in hospite and isolated zooxanthellae among five
coral species. J. Exp. Mar. Biol. Ecol. 291: 191-197.
Brown BE, CA Downs, RP Dunne, SW Gibb. 2002. Exploring
the basis of thermotolerance in reef coral Goniastrea
aspera. Mar. Ecol. Progr. Ser. 242: 119-129.
Chen CA, JT Wang, LS Fang, YW Yang. 2005a. Fluctuating
algal symbiont communities in Acropora palifera
(Scleractinia: Acroporidae) from Taiwan. Mar. Ecol. Progr.
Ser. 295: 113-121.
Chen CA, YW Yang, NV Wei, WS Tsai, LS Fang. 2005b.
Symbiont diversity in the scleractinian corals from tropical
reefs and non-reefal communities in Taiwan. Coral Reefs
24: 11-22.
Coffroth MA, SR Santos. 2005. Genetic diversity of symbiotic
dinoflagellates in the genus Symbiodinium. Protist 156:
19-34.
Dahlhoff E, GN Somero. 1993. Effects of temperature on
mitochondria from abalone (genus Haliotis): adaptive
plasticity and its limits. J. Exp. Biol. 185: 151-168.
Díaz-Almeyda E, PE Thomé, M El Hafidi, R Iglesias-Prieto.
2011. Differential stability of photosynthetic membranes
and fatty acid composition at elevated temperature in
Symbiodinium. Coral Reefs 30: 217-225.
Downs CA, E Mueller, S Phillips, JE Fauth, CM Woodley.
2000. A biomarker system for assessing the health of
coral (Montastrea faveolata) during heat stress. Mar.
Biotechnol. 2: 533-544.
Fabricius KE, JC Mieog, PL Colin, DH Idip, MJH van Oppen.
2004. Identity and diversity of coral endosymbionts
(Zooxanthellae) from three Palauan reefs with contrasting
bleaching, temperature and shading histories. Mol. Ecol.
13: 2445-2458.
Falkowski PG, Z Dubinsky, L Muscatine, L McCloskey. 1993.
Population control in symbiotic corals. BioScience 43:
606-611.
Hill R, AWD Larkum, G Frankart, M Kuhl, PJ Ralph. 2004.
Loss of functional photosystem II reaction centres in
zooxanthellae of corals exposed to bleaching conditions:
using fluorescence rise kinetics. Photosynth. Res. 82: 59-
72.
Hoegh-Guldberg O, PJ Mumby, AJ Hooten, RS Steneck, P
Greenfield, E Gomez et al. 2007. Coral reefs under rapid
climate change and ocean acidification. Science 318:
1737-1742.
Iglesias-Prieto R, JL Matta, WA Robins, RK Trench. 1992.
Photosynthetic response to elevated temperature in the
symbiotic dinoflagellates Symbiodinium microadriaticum in
the culture. Proc. Natl. Acad. Sci. USA 89: 10302-10305.
LaJeunesse TC, WK Loh, RV Woesik, O Hoegh-Guldberg,
GW Schmidt, WK Fitt. 2003. Low symbiont diversity in
southern Great Barrier Reef corals relative to those of the
Caribbean. Limnol. Oceanogr. 48: 2046-2054.
LaJeunesse TC, R Smith, M Walther, J Pinzón, DT Pettay,
M McGinley et al. 2010. Host-symbiont recombination
versus natural selection in the response of coral-
dinoflagellate symbioses to environmental disturbance.
Proc. R. Soc. Lond. B 277: 2925-2934.
Lesser MP. 2007. Coral reefs bleaching and global climate
change: Can coral survive the next century? Proc. Natl.
Acad. Sci. USA 104: 5250-5260.
Muscatine L, JW Porter. 1977. Reef corals: mutualistic
symbiosis adapted to nutrient poor environment.
142 Wang et al. – Thermal Tolerance Index of Symbiodinium
BioScience 27: 154-157.
Robinson JD, ME Warner. 2006. Differential impacts of
photoacclimation and thermal stress on the photobiology
of four different phylotypes of Symbiodinium (Pyrrhophyta).
J. Phycol. 42: 568-579.
Rowan R. 2004. Thermal adaptation in reef coral symbionts.
Nature 430: 742.
Sampayo EM, T Ridgway, P Bongaerts, O Hoegh-Guldberg.
2008. Bleaching susceptibility and mortality of corals are
determined by fine-scale differences in symbiont type.
Proc. Natl. Acad. Sci. USA 105: 1444-1449.
Souter P, LK Bay, N Anderakis, N Császár, FO Seneca, MJH
van Oppen. 2011. A multilocus, temperature stress-
related gene expression profile assay in Acropora
millepora, a dominant reef-building coral. Mol. Ecol.
Resources 11: 225-421.
Stenseng E, C Braby, GN Somero. 2005. Evolutionary and
acclimation-induced variation in the thermal limits of heart
function in congeneric marine snails (genus Tegula):
implications for vertical zonation. Biol. Bull. 208: 138-144.
Stillman JH. 2002. Causes and consequences of thermal
tolerance limits in rocky intertidal porcelain crabs, genus
Petrolisthes. Integr. Compar. Biol. 42: 790-796.
Takahashi S, S Whitney, S Itoh, T Maruyama, M Badger. 2008.
Heat stress causes inhibition of the de novo synthesis
of antenna proteins and photobleaching in cultured
Symbiodinium. Proc. Natl. Acad. Sci. USA 105: 4203-
4208.
Tchernov D, MY Gorbunov, C de Vargas, SW Yadav, AJ
Milligan, M Häggblom, PG Falkowski. 2004. Membrane
lipids of symbiotic algae are diagnostic of sensitivity to
thermal bleaching in corals. Proc. Natl. Acad. Sci. USA
101: 13531-13535.
Wang JT, AE Douglas. 1997. Nutrients, signals, and photo-
synthate release by symbiotic algae: the impact of taurine
on the dinoflagellate alga Symbiodinium from the sea
anemone Aiptasia pulchella. Plant Physiol. 114: 631-636.
Wang JT, PJ Meng, E Sampayo, SL Tang, CA Chen. 2011.
Photosystem II breakdown induced by reactive oxygen
species (ROS) in the freshly-isolated Symbiodinium of
Montipora (Scleractinia; Acroporidae). Mar. Ecol. Progr.
Ser. 422: 51-62.
Warner MK, WK Fitt, GW Schmidt. 1999. Damage to photo-
system II in symbiotic dinoflagellates: a determinant of
coral bleaching. Proc. Natl. Acad. Sci. USA 96: 8007-
8012.
Weijers M, PA Barneveld, MAC Stuart, RW Visschers. 2003.
Heat-induced denaturation and aggregation of ovalbumin
at neutral pH described by irreversible first-order kinetics.
Protein Sci. 12: 2693-2703.
Zoological Studies 51(2): 143-149 (2012)

Larval Development of Fertilized “Pseudo-Gynodioecious” Eggs

Suggests a Sexual Pattern of Gynodioecy in Galaxea fascicularis
(Scleractinia: Euphyllidae)
Shashank Keshavmurthy1, Chia-Min Hsu1,2, Chao-Yang Kuo1, Vianney Denis1, Julia Ka-Lai
Leung1,3, Silvia Fontana1,4, Hernyi Justin Hsieh5, Wan-Sen Tsai5, Wei-Cheng Su5, and Chaolun
Allen Chen1,2,6,7,*
1
Biodiversity Research Center, Academia Sinica, Nangang, Taipei 115, Taiwan
2
Institute of Oceanography, National Taiwan Univ., Taipei 106, Taiwan
3
Institute of Life Sciences, National Taiwan Normal Univ., Taipei 106, Taiwan
4
Univ. of Milan-Bicocca, Piazza della Scienza 2, Milan 20126, Italy
5
Penghu Marine Biological Research Center, Makong 880, Taiwan
6
Institute of Life Science, National Taitung Univ., Taitung 950, Taiwan
7
Taiwan International Graduate Program (TIGP)- Biodiversity, Academia Sinica, Nangang, Taipei 115, Tawian

(Accepted September 22, 2011)

Shashank Keshavmurthy, Chia-Ming Hsu, Chao-Yang Kuo, Vianney Denis, Julia Ka-Lai Leung, Silvia
Fontana, Hernyi Justin Hsieh, Wan-Sen Tsai, Wei-Cheng Su, and Chaolun Allen Chen (2012) Larval
development of fertilized “pseudo-gynodioecious” eggs suggests a sexual pattern of gynodioecy in Galaxea
fascicularis (Scleractinia: Euphyllidae). Zoological Studies 51(2): 143-149. Galaxea fascicularis possesses
a unique sexual pattern, namely “pseudo-gynodioecy”, among scleractinian corals. Galaxea fascicularis
populations on the Great Barrier Reef, Australia are composed of female colonies that produce red eggs and
hermaphroditic colonies that produce sperm and white eggs. However, white eggs of hermaphroditic colonies
are incapable of being fertilized or undergoing embryogenesis. In this study, the reproductive ecology and
fertilization of G. fascicularis were examined in Chinwan Inner Bay, Penghu, Taiwan in Apr.-June 2011 to
determine the geographic variation of sexual patterns in G. fascicularis. Synchronous spawning of female and
hermaphroditic colonies was observed between 17:30 and 20:00 (1 h after sunset) between 24-28 May 2011 (7-11
nights after the full moon in May), and at same times between 22-24 June 2011 (6-8 nights after the full moon in
June). Red eggs were significantly larger than white eggs, although both types of eggs had a distinct nucleus,
which was located at the edge of the eggs, suggesting that they were in the final stage of maturation and ready
to release gametes. Crossing experiments showed that both white and red eggs could be fertilized in vivo, and
they synchronously developed into swimming larvae, suggesting that instead of being pseudo-gynodioecious,
the sexual pattern of G. fascicularis is gynodioecious. http://zoolstud.sinica.edu.tw/Journals/51.2/143.pdf

Key words: Gynodioecy, Pseudo-gynodioecy, Galaxea fascicularis, Reproductive mode, Synchronous spawning.

S exual patterns and modes of development
are the most important life-history traits in
scleractinian corals, and have been one of the
major research themes over the last 3 decades
(reviewed in Richmond and Hunter 1990, Harrison
and Wallace 1990, Baird et al. 2009, Harrison
2011). Three sexual patterns (hermaphroditic,
gonochronic, and mixed) and 2 modes of
development (broadcast-spawned gametes and
brooded larvae) were identified (Harrison 2011).

*To whom correspondence and reprint requests should be addressed. Shashank Keshavmurthy, Chia-Min Hsu, and Chao-Yang Kuo
contributed equally to this work. Tel: 886-2-27899549. Fax: 886-2-27858059. E-mail:cac@gate.sinica.edu.tw

143
144 Keshavmurthy et al. – Gynodioecy in Scleractinian Corals

Among the 444 species studied, 295 species
are hermaphroditic and 109 are gonochronic (or
dioecious). The remaining species are either
mixed or have contrasting modes of reproduction.
For the mode of development, 354 species spawn
gametes into the water, and 60 species brood
larvae (Harrison 2011).
Galaxea spp. were originally described as
being simultaneous hermaphrodites (Harrison et
al. 1984). However, subsequent research at the
Great Barrier Reef (GBR), Australia demonstrated
that Galaxea species have populations composed
of female colonies that spawn pinkish-red eggs,
and hermaphroditic colonies that produce sperm
and lipid-filled white eggs (Harrison 1989).
Hermaphroditic G. fascicularis colonies produce
functional sperm that can fertilize spawned,
pigmented eggs of female colonies (Fig. 1).
However, white eggs contain unusually large
lipid spheres, cannot undergo fertilization, and
function to lift the sperm bundles up to the water
surface where the buoyant pigmented eggs
accumulate, suggesting that these white eggs
potentially enhance fertilization success (Harrison
1989). Harrison (2011) suggested that the
pseudo-gynodioecious sexual pattern in at least
some Galaxea species is therefore functionally
gonochronic. However, this detailed observation
was only made in the GBR, and the sample sizes
of hermaphroditic and female colonies were
relatively small (n = 2 for each sex). Further
studies outside the GBR with a larger sample size
of colonies are necessary to confirm the pseudo-
gynodioecious sexual pattern of Galaxea spp.
In this study, the reproductive ecology and
fertilization of G. fascicularis were studied in detail
at Chinwan Inner Bay (CIB), Penghu Is., Taiwan.
Galaxea fascicularis is one of the dominant coral
species of the scleractinian community at CIB
(Hsieh 2008, Hsieh et al. 2011). This provided
us with the opportunity to study the reproductive
ecology and reexamine the sexual pattern of G.
fascicularis.

(A) (B)

(C) (D)

Fig. 1. Galaxea fascicularis larval development. (A) White bundle containing white eggs and sperm; (B) red bundle full of red eggs
only; (C) white eggs within a hermaphroditic polyp each with a clear nucleus (arrow), and (D) red eggs within a female polyp each with
a clear nucleus (arrow). Scale bars = 200 μm.
 Zoological Studies 51(2): 143-149 (2012)
MATERIALS AND METHODS
Study site and sample collection
Coral spawning was observed at CIB
(23°31'N, 119°33'E), Penghu Is., Taiwan in Apr.-
June 2011. CIB is a semi-enclosed embayment
where coral communities have developed on top
of volcanic rocks, with 75 species of scleractinian
corals described (Hsieh 2008, Hsieh et al. 2011).
Over 50 colonies of G. fascicularis with a colony
size of > 10 cm in diameter were collected,
deposited in individual buckets, and moved
to tanks with a continuous seawater flow and
aeration system at the joint marine laboratory of
the Biodiversity Research Center, Academia Sinica
(BRCAS)-Penghu Marine Biological Research
Center (PMBRC) at CIB. Acropora muricata
was also collected for reference to compare
developmental stages from fertilized eggs to
elongated planular larvae (Miller and Ball 2002).
Observation of spawning and crossing expe-
riments
Observations of spawning behavior at CIB
began on 12 Apr. 2011, 5 d before the full moon in
Apr., based on previous observations (Chen et al.
unpubl. data). Throughout the spawning period,
seawater flow in the tanks was stopped daily, by
turning the taps off after sunset (ca. 18:30 at CIB).
If no spawning was observed on any particular
day, seawater flow was restored after 22:30. The
time of release of gamete bundles was recorded
once polyps and tentacles were retracted, and
colored bundles, either white or pinkish-red,
were released to the surface of the buckets. On
24 May 2011, 10 colonies of G. fascicularis with
white eggs and 10 colonies with red eggs were
labeled for bundle collection. Gamete bundles
released to the surface of the water in the buckets
were separately scooped up using recycled
plastic cups, and brought back to the laboratory
for crossing experiments. Both white and red
bundles were filtered through a plankton mesh
with a 150-μm-mesh size to separate eggs and
sperm. Aliquots of eggs and sperm were collected
for size measurements and density counts.
Sperm density was diluted to 105-106/ml for the
crossing experiment (Willis et al. 1997 2006).
White and red eggs were mixed and fertilized
with diluted sperm. Developmental stages were
observed every hour and categorized based on
stages described for Acropora by Miller and Ball
145
(2000) using the same terminology. A series of
photographs was taken using an Olympus 5050
camera (Tokyo, Japan) attached to the eyepiece
of an Olympus light microscope to obtain images
of the developmental stages between white and
red eggs until the swimming planular larval stage.
Galaxea fascicularis white and red eggs inside
the coral tissues were photographed under 40x
magnification (objective lens 4x and eyepiece
10x) using an Olympus microscope (model SZ40)
fitted with an Olympus C5050 digital camera. The
gonads were placed in a Petri dish immersed in
seawater without a cover. Images of white and red
egg were photographed under 100x magnification
(object 10x and eyepiece 10x) using an Olympus
microscope (model CX31) fitted with an Olympus
E510 digital camera. The same gonads were
moved to a glass slide and gently put on the slide
cover without any pressure. Time-series photos of
G. fascicularis were taken under 40x magnification
(objective lens 4x and eyepiece 10x) using an
Olympus microscope (model SZ40) fitted with
Olympus SP350 and C5050 digital cameras. The
cameras were fitted directly to the eyepiece of
the microscope to obtain the photos. Time-series
photos of Acropora muricata were taker under 40x
magnification with an Olympus C5050 camera.
The egg size and scale shown in the photos were
obtained by micro-ruler photo of a hemocytometer
obtained at the respective magnifications.
RESULTS
Galaxea fascicularis colonies at CIB,
Penghu Is. were either female (pinkish-red eggs)
or hermaphroditic (white eggs with sperm sacs)
(Fig. 1A, B). No spawning was observed for G.
fascicularis in Apr. 2011 (normal spawning period
in Penghu begins from Apr.). However, on 24
May 2011, 7 nights after the full moon of May,
synchronous spawning of G. fascicularis (> 30
colonies) was first observed at 19:30, 1 h after
sunset at the Penghu Is. with a peak of gamete
bundles released at around 20:00 (Fig. 1A, B).
Continued release of bundles was observed the
following 4 nights with a decrease in the number
of colonies spawned on the 8th night after the full
moon (Table 1). Another synchronous spawning
event of over 30 colonies was observed on 22
June, 6 nights after the full moon of June (Table 1).
Some colonies spawned multiple times either on
different nights in May or continuously in June.
Dissecting gamete bundles suggested
146 Keshavmurthy et al. – Gynodioecy in Scleractinian Corals
that both white and red eggs of G. fascicularis
were mature and had reached the same stage
just before spawning. Some white eggs from
hermaphroditic colonies possessed a clear
nucleus close to the edge of the egg, as seen in
red eggs (Fig. 1C, D). Spawned white eggs had
a significantly (t-test = -72.1769, p < 0.01) smaller
mean diameter (290.20 ± 2.60 μm, n = 171) than
red eggs (438.58 ± 3.13 μm, n = 184) (Fig. 2).
Fertilization experiments showed that both
white and red eggs were mature, and embryo
development was synchronous (Fig. 3). Two-
cell cleavage was observed during the 1st hr
after fertilization (Fig. 3A). The time of the initial
development (cell-cleavage stage) cycle in G.
fascicularis embryos was similar to that of A.
muricata before reaching the prawn-chip stage (Fig.
3C-F). Galaxea fascicularis took 8 hr to reach the
prawn-chip stage, while A. muricata needed at
least 12 hr after fertilization (Miller and Ball 2000).
Also, embryonic development from the donut to the
pear stage in G. fascicularis was significantly faster
than that of A. muricata (Fig. 3I-P). The swimming
ability of planular larvae fertilized from white eggs
did not differ from that of larvae from red eggs.
DISCUSSION
Our study provides several lines of evi-
dence, including final maturation, fertilization,
and embryonic and larvae development, to
demonstrate that the sexual pattern of the G.
fascicularis population at CIB, Penghu Is., Taiwan
is gynodioecious. This is the 1st record of the
Table 1. Month, date, days after the full moon,
time of spawning (hours after sunset), and
numbers of colonies spawned of G. fascicularis in
Chinwan Inner Bay, Penghu, Taiwan in 2011
Month Date No. of days after Time of spawning Number of
a full moon (h after sunset) colonies spawned
from a total of n = 50
May 24 7 1
25 8 1
26 9 1
27 10 1 <5
28 11 1 <5
June 22 6 1
23 7 1
24 8 1
-, not spawning.
gynodioecious sexual pattern in scleractinian
corals.
Egg, embryonic, and larvae development in
Galaxea fascicularis
According to a previous study (Harrison
1989) conducted at the GBR, Australia, the
reproduction mode in G. fascicularis was reported
to be pseudo-gynodioecious suggesting that
white eggs produced by this species cannot be
fertilized. This raises the question as to why white
eggs of G. fascicularis at CIB, Penghu Is., Taiwan
were fertile, but those in the GBR, Australia were
not? Two possible scenarios are proposed to
explain this difference. First, our observed results
may have been due to geographic differentiation
between G. fascicularis populations in the GBR,
Australia and those at CIB, Penghu Is., Taiwan.
In some scleractinian corals, sexual patterns
and reproductive modes can vary in different
geographic regions (reviewed in Harrison 2011).
For example, histological studies on Pocillopora
damicornis colonies in Japan indicated that
brooded planulae develop from eggs, and may
500
Mean diameter of eggs (µm)
450
400
350
300
250
Red White
Egg colour
Fig. 2. Difference in egg sizes between female (red egg color)
> 30 and hermaphroditic (white egg color) colonies of Galaxea
> 30 fascicularis. The egg size data were plotted using software R
<5
to generate a box plot. The upper and lower hinge of the box
indicate 75th and 25th percentile of the data set. The line in the
middle of the box represents median for each data set of egg
> 30
sizes indicating a skewed data set. Vertical dotted lines with
- whiskers at top and bottom represent maximum and minimum
- values. Circles in the figure are outliers with values outside
the 25%-75% interval. The absence of circles for red eggs
indicates that there were no outliers.
 Zoological Studies 51(2): 143-149 (2012) 147

be produced sexually (Diah Permata et al. 2000).
Different reproductive patterns occur in the
eastern Pacific and Gulf of California populations
of P. damicornis, which are characterized by
the production of eggs and sperm and inferred
spawning of mature gametes, but there is no
evidence of brooding or planular production in
those populations (Glynn et al. 1991, Colley et
al. 2006, Chavez-Romo and Reyes-Bonilla 2007,
Glynn and Colley 2009). It was suggested by
Harrison (2011) that variations in reproductive
characteristics and life-history traits recorded
among populations in different regions indicate
that these characteristics are unusually variable in
this species. Alternatively, P. damicornis may be
a species complex containing cryptic species with
different reproductive patterns (Flot et al. 2008,
Souter 2010). Determining whether G. fascicularis
with the fertilization capability of white eggs
from CIB and the GBR is the same species with
unusually variable life-history traits or there are
cryptic species with different reproductive patterns
requires further investigation using a molecular
genetic analysis.
Another explanation is that results may
have been due to the small sample size of G.
fascicularis (2 female and 2 hermaphroditic)
colonies utilized in the fertilization trials at the GBR
(Harrison 1989). The percentage of mature white
eggs in hermaphroditic colonies was relatively low
compared to those large lipid bodies in gamete
bundles released into the water column, thereby
reducing the chance of obtaining fertile eggs for
further observations of embryo development. In
our study, large numbers of gamete bundles were
collected from both female and hermaphroditic G.

Acropora muricata Galaxea fascicularis Acropora muricata Galaxea fascicularis

(A) 2 cell, 2 hr (B) 2 cell, 1 hr (I) dount, 16 hr (J) dount, 9 hr

W
W
R R

(C) 8 cell, 4 hr (D) 8 cell, 3hr (K) fat dount, 19 hr (L) fat dount, 11 hr

R
W
W
R

(E) 32 cell, 6 hr (F) 32 cell, 5 hr (M) Pear, 47 hr (N) Pear, 13 hr

W

W R

(G) prawn-chip, 12 hr (H) prawn-chip, 8 hr (O) spindle planular larvae, 76 hr (P) spindle planular larvae, 66 hr

R
W
R
W

Fig. 3. Embryo stages of Galaxea fascicularis and Acropora muricata. The time of each stage is indicated in hours after fertilization.
(A, B) Two-cell stage; (C, D) 8-cell stage; (E, F) 32-cell stage; (G, H) prawn-chip stage; (I, J) donut stage; (K, L) fat-donut stage; (M, N)
pear stage; (O, P): spindle planular larvae. R and W = Developmental stages form red and white fertilized eggs. Scale bar = 200 μm.
148 Keshavmurthy et al. – Gynodioecy in Scleractinian Corals
fascicularis colonies, and fertilization took place
with large quantities of gametes that increased the
chances of observing serial embryo development
of white eggs from hermaphroditic colonies.
Further investigations of the percentage of fertile
white eggs in gamete bundles of hermaphroditic
colonies and of the survival, settlement, recruitment
success, and growth of derived juvenile corals are
needed to confirm the contribution of white eggs to
G. fascicularis populations.
Overall, results from this study showed that
embryonic development time is much shorter
in G. fascicularis compared to that in Acropora.
The length of embryonic development could
affect dispersal and recruitment among different
spawning corals (Nakamura and Sakai 2010). For
example, among spawning pocilloporid corals,
larvae that develop relatively more rapidly have
higher recruitment at sites where adult coral cover
is high. In contrast, recruitment is not related to
adult coral cover in acroporid and poritid corals,
the embryonic development times of which are
relatively slow (Nakamura and Sakai 2010).
The shorter embryonic development time might
facilitate G. fascicularis settling locally faster, and
helping it become the dominant species after
a series of disturbances and disappearance of
acroporid corals (Acropora and Montipora) after
a cold shock event in 2008 at CIB (Hsieh et al.
2008 2011). The recruitment of acroporid corals
may be slower because sources of larvae are from
neighboring coral communities outside CIB.
Sexual pattern of Galaxea fascicularis
gynodioecy
Completion of embryonic and larval deve-
lopment of white eggs from hermaphroditic colonies
suggests that the sexual pattern of G. fascicularis
is gynodioecious, instead of pseudo-gynodioecious
as proposed by Harrison (1989). Studies on
plant reproductive systems have indicated that
gynodioecy is a transitional step towards dioecy
(gonochorism) from hermaphroditism (reviewed
in Charlesworth 2006). This scenario might be
applicable to the evolution of sexual pattern traits
in Galaxea. Galaxea is the only coral genus that
possesses a sexual pattern of gynodieocy, and
phylogenetic studies have relocated Galaxea from
the family Oculinidae to the Euphyllidae, where it
forms a sister clade to the genus Euphyllia (Fukami
et al. 2008, Dai and Horng 2010). The sexual
patterns of 8 Euphyllia species can be divided into
either dioecious species with spawned gametes
(e.g., E. ancora) or hermaphroditic species with
brooded larvae (e.g., E. glabrescens) (Veron
2000). Gynodieocy in Galaxea might represent a
transitional step of sexual pattern evolution in the
family Euphyllidae. In addition, gynodieocy also
suggests a unique inheritance mode of genetics
in Galaxea compared to true hermaphroditic or
dioecious species. Further work on ancestral
reconstruction of life-history traits and genetic
structuring of populations should provide insights
into the evolutionary novelty of gynodioecy in
Galaxea among scleractinian corals.
Acknowledgments: Many thanks go to the
staff of the Penghu Marine Biological Research
Center (PMBRC), Council of Agriculture for logis-
tical support, and members of the Coral Reef
Evolutionary Ecology and Genetics (CREEG)
l a b o r a t o r y, B i o d i v e r s i t y R e s e a r c h C e n t e r,
Academia Sinica (BRCAS) and 2 anonymous
reviewers for constructive comments. CMH and
JKL are recipients of a PhD fellowship, and SK
is the recipient of a postdoctoral fellowship from
Academia Sinica (2010-2012). VD is the recipient
of a postdoctoral fellowship from the National
Science Council (NSC), Taiwan. This study was
made possible by an Academia Sinica Thematic
Grant (AS-100-TP2-A02) and grants from the
NSC (NSC99-2621-B-001-006-MY3) to CAC and
(NSC98-2313-B-056-001-MY3) HJH. This is
CREEG-BRCAS contribution no. 75, and BRCAS-
PMBRC Joint Marine Laboratory contribution no. 1.
REFERENCES
Baird AH, JR Guest, BL Willis. 2009. Systematic and bio-
geographical patterns in the reproductive biology of
scleractinian corals. Annu. Rev. Ecol. Syst. 40: 551-571.
Charlesworth D. 2006. Evolution of plant breeding systems.
Curr. Biol. 16: R726-R735.
Chavez-Romo HE, H Reyes-Bonilla. 2007. Sexual repro-
duction of the coral Pocillopora damicornis in the southern
Gulf of California, Mexico. Cienc. Mar. 33: 495-501.
Colley SB, PW Glynn, AS Mayt, JL Mate. 2006. Species-
dependent reproductive responses of eastern Pacific
corals to the 1997-1998 ENSO event. In Proceedings
of the 10th International Coral Reef Symposium, vol. 1,
Okinawa, Japan, 2004, pp. 61-70.
Dai CF, S Horng. 2010. In, Scleractinia Fauna of Taiwan 1.
The Complex group. National Taiwan University, Taipei
Taiwan, pp. 175.
Diah Permata W, RA Kinzie III, M Hidaka. 2000. Histological
studies on the origin of planulae of the coral Pocillopora
damicornis. Mar. Ecol. Progr. Ser. 200: 191-200.
Flot JF, H Magalon, C Cruaud, A Couloux, S Tillier. 2008.
Patterns of genetic structure among Hawaiian corals of
 Zoological Studies 51(2): 143-149 (2012)
the genus Pocillopora yield clusters of individuals that are
compatible with morphology. Compt. Rend. Biol. 331:
239-247.
Fukami H, CA Chen, AF Budd, A Collins, C Wallace, YY Chuang
et al. 2008. Mitochondrial and nuclear genes suggest
that stony corals are monophyletic but most families of
stony corals are not (order Scleractinia, class Anthozoa,
phylum Cnidaria). PLoS ONE 3: 1-9.
Glynn PW, SB Colley. 2009. Survival of brooding and
broadcasting reef corals following large-scale
disturbances: Is there any hope for broadcasting
species during global warming? In Proceedings of the
11th International Coral Reef Symposium, vol. 1, Ft.
Lauderdale, FL, USA, 2008, pp. 361-365.
Glynn PW, NJ Gassman, CM Eakin, J Cortés, DB Smith, HM
Guzmán. 1991. Reef coral reproduction in the eastern
Pacific: Costa Rica, Panamá, and Galápagos Islands
(Ecuador). I. Pocilloporidae. Mar. Biol. 109: 355-368.
Harrison PL. 1989. Pseudo-gynodioecy: an unusual breeding
system in the scleractinian coral Galaxea fascicularis.
Proc. 6th. Int. Coral Reef Symp. 2: 699-705.
Harrison PL. 2011. Sexual reproduction of scleractinian
corals. In Z Dubinsky, N Stambler, eds. Coral reefs: an
ecosystem in transition Part 3. USA: Springer, pp. 59-85.
Harrison PL, RC Babcock, GD Bull, JK Oliver, CC Wallace,
BL Willis. 1984. Mass spawning in tropical reef corals.
Science 223: 1187-1188.
Harrison PL, CC Wallace. 1990. Reproduction, dispersal and
recruitment of scleractinian corals. In Z Dubinsky, ed.
Coral reefs. Amsterdam: Elsevier, pp. 133-207.
Hsieh HJ, KS Chen, YL Lin, YA Huang, AH Baird, WS Tsai et
149
al. 2011. Establishment of a no-take area (NTA) could
not guarantee the preservation of coral communities in
Chinwan Inner Bay, Penghu, Taiwan. Zool. Stud. 50: 443-
453.
Hsieh HJ, IL Shen, MS Jeng, WS Tsai, WC Su, CA Chen.
2008. Tropical fish killed by the cold. Coral Reefs. DOI:
10.1007/s00338-008-0378-3.
Miller DJ, EE Ball. 2000. The coral Acropora: what it can
contribute to our knowledge of metazoan evolution and
the evolution of developmental processes. BioEssays 22:
291-296.
Nakamura MS, K Sakai. 2010. Spatiotemporal variability in
recruitment around Iriomote Island, Ryukyu Archipelago,
Japan: implications for dispersal of spawning corals. Mar.
Biol. 157: 801-810.
Richmond RH, CL Hunter. 1990. Reproduction and recruitment
of corals: comparisons among the Caribbean, the tropical
Pacific, and the Red Sea. Mar. Ecol. Progr. Ser. 60: 185-
203.
Souter P. 2010. Hidden genetic diversity in a key model
species of coral. Mar. Biol. 157: 875-885.
Willis BL, RC Babcock, PL Harrison, CC Wallace. 1997.
Experimental hybridization and breeding incompatibilities
within the mating systems of mass spawning reef corals.
Coral Reefs 16: S53-S65.
Willis BL, MJH van Oppen, DJ Mille, SV Vollmer, DJ Ayre.
2006. The role of hybridization in the evolution of reef
corals. Ann. Rev. Ecol. Evol. Syst. 37: 489-517.
Veron JEN. 2000. Corals of the world. Townsville, Australia:
Australian Institute of Marine Science.
Zoological Studies 51(2): 150-159 (2012)

Diverse Interactions between Corals and the Coral-Killing Sponge,

Terpios hoshinota (Suberitidae: Hadromerida)
Jih-Terng Wang1,*, Yi-Yun Chen1, Pei-Jie Meng2,3, Yu-Hsuan Sune3, Chia-Min Hsu4, Kuo-Yen Wei1,
and Chaolun Allen Chen4,5,6
1
Graduate Institute of Biotechnology, Tajen Univ., Pingtung 907, Taiwan
2
National Museum of Marine Biology and Aquarium, Pingtung 944, Taiwan
3
Institute of Marine Biodiversity and Evolution, National Dong Hwa Univ., Checheng, Pingtung 944, Taiwan
4
Institute of Oceanography, National Taiwan Univ., Taipei 108, Taiwan
5
Biodiversity Research Center, Academia Sinica, Nangang, Taipei 115, Taiwan
6
Taiwan International Graduate Program (TIGP)- Biodiversity, Academia Sinica, Nangang, Taipei 115, Tawian

(Accepted September 15, 2011)

Jih-Terng Wang, Yi-Yun Chen, Pei-Jie Meng, Yu-Hsuan Sune, Chia-Min Hsu, Kuo-Yen Wei, and Chaolun
Allen Chen (2012) Diverse interactions between corals and the coral-killing sponge, Terpios hoshinota
(Suberitidae: Hadromerida). Zoological Studies 51(2): 150-159. Terpios hoshinota is an encrusting sponge
which can kill corals by overgrowing them. However, little is known about interactions between sponges and
corals. Using visual observations and scanning electron microscopy (SEM), 4 features, including hairy tips,
thick tissue threads, compact edges, and disintegrated tissues, displayed at the coral-facing front of Terpios
were summarized from examining 20 species of corals. Hairy tips, found on 13 species of coral victims, were
occupied by cyanobacteria, sponge tissues, and spicules. Thick tissue threads, found on only 7 coral species,
were obviously an extension of Terpios tissues. Twelve coral species displayed a compact edge at the Terpios-
coral border, in which some Terpios fronts had extruding spicules. Disintegrated tissue was only found on the
coral side in 5 species of coral, but that on the sponge side was only found on 1 coral species. Only a few
disintegrated tissues being found at the Terpios-coral border suggests that allelochemicals are not the major
player in Terpios-coral interactions. The interactions also did not display species specificity, except in the case
of Terpios having been retrogressively grown over by a coral, which was only found in Millepora exaesa. Under
SEM examination, coral nematocysts were usually found on the surface of the invading Terpios, but they did not
seem to retard the growth of the sponge. In summary, exploitation of the substratum by T. hoshinota on coral
does not move forward in a consistent manner. The performance of Terpios, such as when overgrowing a coral,
building a clear border, or being retrogressively overgrown by a coral, may rely on the viability status of both
organisms. http://zoolstud.sinica.edu.tw/Journals/51.2/150.pdf

Key words: Terpios, Cyanobacteria, Coral-killing sponge, Substrata competition.

I
n coral reefs, sponges are a well-known
space competitor (Suchanek et al. 1983, Rützler
2002), but few are recognized as real threats to
the survival of corals. Terpios hoshinota Rützler
and Muzik, 1993, a cyanobacteriosponge, is an
exception, as its high growth rate can encrust
almost every type of hermatypic coral encountered.
Its widespread infection was first reported in Guam
(Bryan 1973), and subsequently in the Ryukyus,
Japan (Rützler and Muzik 1993, Reimer et al. 2011)
and Green I. (Lyudao), Taiwan (Liao et al. 2007).
Terpios hoshinota was also found in Truk Lagoon
in American Samoa, Cebu I. in the Philippines,
Thailand (Plucer-Rosario 1987), and even on the

*To whom correspondence and reprint requests should be addressed. Tel: 886-8-7624002. Fax: 886-8-7621645.
E-mail:jtw@mail.tajen.edu.tw

150
 Wang et al. – Coral-Terpios Interactions
Great Barrier Reef (Fujii et al. 2011). Damage
caused by an invasion of Terpios caused nearly
30% loss of coral coverage on some reefs in
Guam (Plucer-Rosario 1987). At Green I., Taiwan,
an unprecedented overgrowth by Terpios on corals
was found in 2006, which also caused almost 30%
coral coverage loss along a 100-m transect belt
(Liao et al. 2007). The complete recovery from
a Terpios encrustation, e.g., at Anae I. in Guam,
took more than 10 yr, when the disturbance level
decreased (Plucer-Rosario 1987). Therefore, once
a Terpios outbreak occurs, there will be long-term
impacts on a coral reef ecosystem and on activities
that rely on a healthy condition of the reefs.
Ecologically, T. hoshinota is distributed above
the limit of the euphotic zone, probably due to
the presence of endosymbiotic photosynthetic
cyanobacteria (Bryan 1973, Plucer-Rosario
1987, Rützler and Muzik 1993). A histological
examination of T. hoshinota indicated that the
sponge contained a high percentage (> 50%)
of intercellular cyanobacteria and 5%-18% of
cells were in the dividing stage (Rützler and
Muzik 1993, Hirose and Murakami 2011). High
abundances and activities of cyanobacteria
contained in T. hoshinota suggest that a potential
source of the sponge’s nutrients is derived from
photosynthetic bacteria (Rützler and Muzik
1993). It was hypothesized by Bryan (1973)
that Terpios probably kills coral for nutrients
with toxic chemicals, but comparisons between
tissue-depleted and healthy coral suggested
that the sponge might just overgrow the coral
surface to occupy more space (Plucer-Rosario
1987). During growth, Terpios moves forward by
lateral propagation, extending short, fine tendrils
across crevices to new substrate (Rützler and
Muzik 1993). Terpios hoshinota can also develop
tissue threads, instead of whole sheets of tissue,
to move over a shaded area and establish new
territory (Soong et al. 2009). However, Terpios
occasionally exhibits retrogression (i.e., negative
growth) and can even be overgrown by some
corals (e.g., Montipora and Porites) or red
calcareous algae (Plucer-Rosario 1987). Thus,
Terpios does not always win during its advance.
Interactions between Terpios and corals were
examined from the viewpoint of changes in the
bacterial community. Tang et al. (2011) indicated
that invasion by Terpios onto corals initiates a
shift in the coral bacterial community from one on
healthy corals to that found on corals with black-
band disease. Their results suggested that harmful
bacteria weakening the coral might favor Terpios
151
outcompeting the coral for substratum (Tang et al.
2011).
As yet, only limited information briefly des-
cribing how Terpios invades victimized corals
is available (Bryan 1973, Plucer-Rosario 1987,
Rützler and Muzik 1993, Soong et al. 2009, Tang
et al. 2011), and it is not clear how different coral
species respond to an invasion by Terpios at the
coral-sponge border under a fine scale. Therefore,
the aim of this study was to examine the border
between these 2 antagonists with scanning
electron microscopy (SEM). Our findings provide
insights into interactions between an aggressively
invading sponge and its coral victims.
MATERIALS AND METHODS
Sample collection and maintenance
Field observations of coral-Terpios inter-
actions were conducted at Gon-Guam and Chai-
Ko, Green I., Taiwan (22°39'N, 121°29'E) from Aug.
2008 to July 2010. Due to the strong northeasterly
monsoon in winter, observations were made more
intensively during summer (May-Sept.). During
the investigation, interactions between coral and
Terpios were recorded with an underwater camera.
To further examine the interaction border between
corals and Terpios, 19 species of scleractinian
coral and 1 hydrozoan coral with T. hoshinota
invasion were collected by scuba diving from 3-5 m
in depth at Gon-Guam and Chai-Ko on 28 July
2010 and examined by SEM. The 19 scleractinian
corals included Isopora palifera, Montipora
aequituberculata, Mon. peltiformis, Hydnophora
rigida, Favia stelligera, Psammocora digitata,
Echinopora lamellose, Echinophyllia aspera,
Goniastrea edwardsi, G. aspera, Pocilliopora
verrucosa, Acropora digitifera, Stylophora pistillata,
Platygyra ryukyuensis, Leptoria phrygia, Favites
chinensis, Cyphastrea microphthalma, Porites
lutea, and Por. cylindrical. The hydrozoan coral
examined was Millepora exaesa. Every species
was duplicated by collecting a sample from 2
different colonies, and the interactions were
photographed before collection. Terpios-coral
specimens were sealed in a plastic bag underwater
when collected and preserved in fixative once the
diver had left the water.
Terpios hoshinota on I. palifera was also
collected and maintained in an aquarium (60 × 45
× 45 cm) equipped with illumination (12-h:12-h light
dark regime and 70-90 μE/m2/s photosynthetically
152 Zoological Studies 51(2): 150-159 (2012)

active radiation, temperature control (25°C),
filtration (EHEIM, Deizisau, Germany), and a
protein skimmer. The seawater level in the tank
was kept at only 20 cm deep, and 2 underwater
pumps were used to create flow above the sponge.
Terpios hoshinota could grow along the cut edge of
the original coral substrata and also onto the shell
debris at the bottom of the tank. Newly growing
sponge on the shell debris was also examined
by SEM for comparison of Terpios on a non-coral
substratum.
SEM method
Freshly collected Terpios hoshinota was
persevered in fixative (2.5% glutaraldehyde,
2% paraformaldehyde, and 5% sucrose in
0.1 M phosphate buffer at pH 7.0) overnight at
4°C. Subsequently, specimens were washed in
phosphate buffer and post-fixed in 2% OsO4/0.1 M
phosphate buffer (pH 7.3) overnight. Standard
procedures were used to prepare coated samples
for SEM observations. Coral-Terpios samples
were dried in a critical-point dryer (Hitachi HCP-2,
Tokyo, Japan), and coated with platinum in an ion
sputter (Hitachi E1010). SEM observations were
made on an SEM (Hitachi S-3500N) at a voltage of
5 kV.
RESULTS
After examining interactions of T. hoshinota
with 19 species of scleractinian coral and 1
hydrozoan coral, the results indicated that there
were 4 types of interactions between them, i.e.,
hairy tips, thick tissue threads, compact edges,
and disintegrated tissues (Table 1). The 2 most
common features at the coral-facing growth front
of Terpios were hairy tips and compact edges,
which were respectively found in 13 (totally 23
specimens) and 12 (totally 23 specimens) species
of victim corals. Thick tissue threads were less
often found at the coral-facing growth front of
the sponge, and were found in only 7 (totally 8
specimens) species of coral victims, and only that
on Por. cylindrical was found in both specimens
examined. Disintegrated tissues were more rarely

Table 1. Morphological characterization of interactions between T. hoshinota and corals. “+” and “-” signs
respectively represent the presence and absence of a character in the 2 replicates

Sponge-coral border

Coral specimen Terpios hoshinota Coral

Hairy tips Thick tissue threads Compact edges Disintegrated tissues Disintegrate tissues

Scleractinian coral
Acropora digitifera ++ -- ++ -- +-
Isopora palifera ++ -- -- -- --
Montipora aequituberculata ++ -- -- -- --
Montipora peltiformis ++ -- -- -- --
Porites cylindrical +- ++ ++ -- --
Porites lutea -- +- ++ -- --
Pocilliopora verrucosa ++ -- -- -- --
Psammocora digitata ++ -- -- -- --
Stylophora pistillata -- +- ++ -- +-
Cyphastrea microphthalma +- -- ++ -- --
Echinopora lamellosa ++ -- ++ -- --
Favia stelligera ++ -- ++ -- --
Favites chinensis -- -- ++ -- --
Goniastrea aspera ++ -- -- -- --
Goniastrea edwardsi -- -- ++ -- ++
Hydnophora rigida +- +- -- -- +-
Leptoria phrygia -- +- ++ -- --
Platygyra ryukyuensis -- +- ++ -- --
Echinophyllia aspera ++ -- -- -- --
Hydrozoan coral
Millepora exaesa -- +- +- +- +-
 Wang et al. – Coral-Terpios Interactions 153

observed at the coral-facing growth front of the
sponge, which was only found in 1 specimen
of Mil. exaesa. On the coral side, disintegrated
tissues were also rarely observed; only 5 (totally 6
specimens) coral species displayed disintegrated
tissues at the coral-Terpios border. Eighty percent
of specimens of the 19 species of coral displayed
comparable color morphs and Symbiodinium
densities with nearby corals in the same colony
which had not been attacked by Terpios (see
example photos in Figs. 1A, 2A, 3B). Table 1 also
indicates that on I. palifera, Mon. aequituberculata,
Mon. peltiformis, Poc. verrucosa, Psa. digitata, and
Eph. aspera, the sponge displayed only 1 feature,
hairy tips, at the coral-facing growth front; but the
sponge on the other 14 species of coral displayed
more than 1 feature.
Typical examples of detailed interactions
between corals and Terpios are shown in figures
1-6. Figure 1 shows lots of hairy tips along
the Terpios growth front on I. palifera. Hairy
tips of Terpios touch the coral surface when
it moves forward. As shown in figure 1A, the
coral surface at the boundary next to Terpios
displayed no significant changes in the color
morph or Symbiodinium density. Under SEM exa-
mination, hairy tips were found to be occupied
by cyanobacteria, sponge tissues, and spicules
(Fig. 1B-D). Nematocysts obviously from the
victim coral were also found on the surface of the
hairy tips (Fig. 1D). Direct exposure of internal
cyanobacteria and spicules from the hairy tips
(Fig. 1C, D) was caused by the loss of the fragile
pinacoderm during SEM processing.
In figure 2, the compact edge and thick
tissue threads at the Terpios growth front on Pla.
ryukyuensis are shown. The thick tissue threads
were obviously an extension of sponge tissues.
However, the microscopic image of the compact
edge of Terpios revealed only spicules but no
sponge tissue or cyanobacteria extruding from
the sponge front (Fig. 2B, C). Figure 2B and 2C
also indicate that there was no direct contact
between the sponge front and coral tissues, and
no obvious disintegration was found on the coral
surface. Sometimes, a clear border was also
found at the coral-Terpios interface. As shown in
figure 3A, the sponge seemed to hold its growth

(A) (B)

SE WD36.5 mm 5.00 kV x90 500 μm

(C) (D)

SE WD36.5 mm 5.00 kV x450 100 μm SE WD38.5 mm 5.00 kV x900 50 μm

Fig. 1. Terpios hoshinota displaying hairy tips at the growth front in an interaction with coral. (A) An example from Terpios invading
Isopora palifera; (B-D) SEM examination of the hairy tips found in (A) at different magnifications. The white arrowhead indicates the
location of hairy tips. C, cyanobacteria; N, nematocyst; S, spicules.
154 Zoological Studies 51(2): 150-159 (2012)

by showing a compact edge at the boundary
with disintegrated tissues of S. pistillata. This
interaction was not prevalent throughout the entire
Stylophora colony, because thick tissue threads
derived from the sponge were also found to
interact with the coral on different branches of the
same colony (Fig. 3B). Under SEM examination,
there was a 200-500-μm wide border without coral
or sponge tissues between the 2 antagonists (Fig.
3C). When zooming into the Terpios front at the
coral-interacting side, as shown in figure 3, many
stinging nematocysts were found at the leading
edge of Terpios tissues. Opposition against
Terpios growth was also found in an extreme case
during our field survey: 1 A. digitifera colony had
maintained the same boundary with Terpios for
more than a year with no advance or regression.
Disintegration of coral tissues at the coral-
Terpios border, even though very seldom, was
also found. Taking G. edwardsi as an example,
the coral tissue displayed disintegration and had
disintegrated into filamentous residues along
the border contacting Terpios (Fig. 4A). When
examined at high magnification, the growth front
of Terpios seemed to move forward by penetrating
underneath the disintegrated coral tissue (Fig.

(A)

(B) (C)

TF

TF

CS CS

SE WD5.7 mm 5.00 kV x120 250 μm SE WD5.8 mm 5.00 kV x1.0 k 50 μm

Fig. 2. Terpios hoshinota displaying a compact edge and thick tissue threads at the growth front in an interaction with coral. (A) An
example of Terpios infecting Platygyra ryukyuensis; (B, C) SEM examination of the growth front with a compact edge of the sponge
found in (A), indicated by a white arrowhead, at different magnifications. Thick tissue threads from the sponge are marked by white
arrows. CS, coral surface; S, spicules; TF, Terpios front.
 Wang et al. – Coral-Terpios Interactions 155

(A) (B)

(C) (D) TS

TS N

DB

CS
SE WD33.1 mm 5.00 kV x30 1 mm SE WD34.1 mm 5.00 kV x300 100 μm

Fig. 3. Coral displaying both disintegrated and comparatively normal tissues along the growth front of T. hoshinota on different
branches of the same colony. (A, B) Example from Terpios infecting Stylophora pistillata which displays (A) disintegrated coral tissues
and (B) comparatively normal coral tissues at the coral-sponge interface. (C, D) SEM examination of the coral-sponge interface with
disintegrated coral tissues at different magnifications. Disintegrated coral tissue is marked by a white arrowhead, and thick tissue
threads from the sponge are marked by white arrows. CS, coral surface; DB, dead coral boundary; N, nematocyst; S, spicules; TS,
Terpios surface.

(A) (B)
CS

NC

TS
TS
NC

S
SE WD20.8 mm 5.00 kV x35 1 mm SE WD20.2 mm 5.00 kV x300 100 μm

Fig. 4. SEM examination of the disintegration of Goniastrea edwardsi tissues at the growth front of T. hoshinota. (A, B) The
same specimen at different magnifications; the white arrowhead indicates the growth direction of Terpios. CS, coral surface; NC,
disintegrated coral tissue; S, spicules; TS, Terpios surface.
156 Zoological Studies 51(2): 150-159 (2012)

4B). Disintegration was also found on the sponge Terpios.

side at the interaction of Mil. exaesa and Terpios.
On Mil. exaesa, Terpios showed a retreat or
curving-back of the sponge tissue at the coral-
sponge front (Fig. 5A). However, at another part
of the border between Mil. exaesa and Terpios,
thick tissue threads of the growth front of Terpios
had crossed over the coral tissue and touched
down a certain distance behind the border (Fig.
5B). When examining the interaction found in
figure 5A by SEM, there was a 100-300-μm wide
border with disintegrated Terpios tissues in which
spicules were exposed (Fig. 5C). At the coral
front interacting with Terpios, many spirocysts
were found to be protruding from the coral surface
(Fig. 5D). Along a disintegrated Terpios tissue
zone (Fig. 6A), a piece of Mil. exaesa tissue at
the coral-Terpios border also had spicules that
had penetrated through the coral surface (Fig.
6B), suggesting retrogressive growth of coral on
In order to compare the growth of the Terpios
on coral with a non-coral substratum, the sponge
moving onto fragments of bivalve shells was
also examined by SEM. As shown in figure 7A,
Terpios had expanded its tissue onto the shell
fragments by the 7th d of maintenance in the
aquarium. The SEM examination indicated that
rather than showing no direct contact between
the coral and Terpios growth front, the sponge
tissue had firmly attached to the surface of the
shell fragments, and a group of protruding spicules
was nearly completely free of sponge tissues and
cyanobacteria (Fig. 7B).
DISCUSSION
By examining 20 species of coral with
Terpios hoshinota invasion, it was found that

(A) (B)

TS
CS

TS
CS

(C) (D)

TS

NT Sp

CS
SE WD33.9 mm 5.00 kV x45 1 mm SE WD33.6 mm 5.00 kV x700 50 μm

Fig. 5. Millepora exaesa fighting back against invasion by T. hoshinota. (A, B) The same specimen at different locations; the black
arrowhead indicates a curving back of Terpios tissue; the black arrow indicates a thick tissue thread extending from the growth front
of Terpios; and the white circle indicates the site from which (D) is amplified. CS, coral surface; NT, disintegrated Terpios tissue with
exposed spicules; S, spicules; Sp, spirocyst; TS, Terpios surface.
 Wang et al. – Coral-Terpios Interactions
features of the border between the 2 antagonists
were not uniform among coral species or even
within the same colony. These observations
indicate that interactions between corals and
Terpios are dynamic and also not species-
specific as described in other coral-sponge
interactions (Averts 1998 2000, McLean and
Yoshioka 2008). Averts (2000) also indicated
that the direction of overgrowth by the sponge
might be attributed to the level of compactness
of the coral, suggesting that the health status of
the coral might also be a determining factor in
Terpios infections. Overgrowth by the sponge
when invading a coral was described as occurring
by elevating the sponge’s growing edge (McLean
and Yoshioka 2007), but this was not the only
feature found at the interface of coral-Terpios
interactions. On the coral side, the growing edge
of Terpios often displayed hairy tips, i.e., short, fine
tendrils described by Rützler and Muzik (1993),
which are full of sponge tissues, spicules, and
cyanobacteria as found in the arm-like structure
(A)
CS
NT
SE WD33.7 mm 5.00 kV x35 1 mm
(B)
S
S
SE WD33.6 mm 5.00 kV x300 100 μm
Fig. 6. Millepora exaesa growing over T. hoshinota. (A, B)
The same specimen at different magnifications; the white
circle indicates the site from where (B) is amplified. CS, coral
surface; NT, disintegrated Terpios tissue with spicules exposed;
S, spicules penetrating out of the coral surface.
157
(ALS) of Tang et al. (2011). But ALSs of Terpios
were less often found in the field. We usually
found ALSs when the growing edge of the sponge
ran out of substratum of an invaded coral and
tried to climb over an adjacent coral colony or
perhaps faced strong defense by the coral victim
(e.g., Mil. exaesa as seen in the inset of Fig. 1D).
Another feature at the coral-Terpios border was
the smooth and compact growing edge of Terpios,
which is similar to the interaction in other crustose
sponges, such as Cliona caribbaea and Cli. lampa,
advancing on coral (Rützler 2002).
The observation of a several-millimeter-wide
band of dead zooids paralleling the growing edge
of the sponge usually indicates that allelochemical
interactions are important in spatial competition
(A)
(B)
TS
SE SF WD11.3 mm 5.00 kV x180 250 μm
Fig. 7. Growth of T. hoshinota on shell debris in an aquarium.
(A) Terpios on Isopora palifera maintained in a laboratory
aquarium; the white arrowhead indicates shell debris used for
examination. (B) SEM examination of the shell debris with
Terpios. S, spicules; SF, shell fragment; TS, Terpios surface.
158 Zoological Studies 51(2): 150-159 (2012)
on coral reefs (Jackson and Buss 1975). An
allelochemical effect was also considered one
of the mechanisms by which some aggressive
sponges invaded corals (Jackson and Buss
1975, Porter and Targett 1988). In coral-Terpios
interactions, both specimens from G. edwardsi
and one of 2 specimens from 4 other coral
species displayed disintegration at the coral-
sponge border. Most of the coral specimens
(80%) showed comparable color morphs and
Symbiodinium densities between tissues closely
contacting the sponge and those remote from the
coral-Terpios border. Therefore, even though T.
hoshinota was reported to produce chemicals,
such as nakiterpiosin and nakiterpiosinone,
with potent cytotoxicity (Teruya et al. 2004),
allelochemicals might not be the major mechanism
by which Terpios kills coral during its competition
for substratum. It was more evident that Terpios
kills corals by overgrowing them.
Scleractinian corals exhibit a wide variety
of offensive and defensive mechanisms for
acquiring and maintaining a living space (Connell
1973, Wahle 1980, McCook et al. 2001). One of
the mechanisms is to use stinging warfare such
as nematocysts or spirocysts. The ‘stinging’
mechanism includes the effects of polyps, sweeper
tentacles, and mesenterial filaments. Several
processes were documented in interspecific
competition among corals (reviewed by Lang
and Chornesky 1990) and also between corals
and a range of other animals such as zoanthids
and gorgonians (Karlson 1980, Chornesky 1983,
Chadwick 1987). During overgrowth by Terpios,
some but not all victimized corals were observed
to have ejected nematocysts at the contacting
border. However, the defenses, including poten-
tial effectors not observable by SEM, such as
chemicals, did not seem to be very effective in
most cases. One successful case was found in
the interaction between S. pistillata and Terpios,
in which nematocysts at the surface of the
growing edge of the sponge seemed to deter its
advance. The most effective defense by coral’s
stinging warfare was found in Mil. exaesa, in which
the coral not only caused disintegration of the
Terpios growing edge but was also overgrowing
the sponge in the reverse direction. Of course,
chemical effects derived from Mil. exaesa cannot
be excluded because it is a notorious toxin
producer (Wittle et al. 1971, Shiomi et al. 1989,
Radwan and Aboul-Dahab 2004, Iguchi et al.
2008). Millepora exaesa was further found to
retrogressively grow on invading Terpios according
to findings of spicules protruding from the coral
surface next to the coral-sponge border. This
occurred only when the coral grew on un-degraded
spicules left behind by disintegrated tissues of
Terpios. When the soft tissues of coral moved
over lobed tyrostyle spicules, the spicules were
lifted up and penetrated through the coral tissue as
shown in figure 6B.
Overgrowth by Terpios has caused sub-
stantial losses of coral coverage in Guam (Plucer-
Rosario 1987), the Ryukyus, Japan (Rützler and
Muzik 1993), and Green I., Taiwan (Liao et al.
2007). According to the experience in Guam, the
coral coverage in Guam might recover (Plucer-
Rosario 1987). A similar observation was also
made in Japan (Reimer et al. 2011). Fortunately,
the coverage of Terpios has not further expanded
at Green I. since it was first noted in 2006 (Liao et
al. 2007). The static situation of Terpios coverage
might partly be due to a seasonal typhoon effect,
but the dynamic interactive mode between the 2
antagonists, as revealed in this study, might be
a crucial factor promoting the survival of invaded
corals. If disturbance decreases, Connell (1978)
suggested that coral recolonization is possible
within a period of time. Therefore, it is our hope
that we will see coral recovery from the Terpios
invasion, if the environmental conditions of coral
reefs can be protected from anthropogenic and
natural disturbances.
Acknowledgments: We would like to thank
members of the Coral Reef Evolutionary Ecology
and Genetics (CREEG) Group, Biodiversity
Research Center, Academia Sinica (BRCAS) for
field support, and Profs. K. Soong and E. Hirose
for their valuable comments. This work was made
possible by grants from the National Science
Council, Taiwan (NSC98-2321-B-127-001-MY3) to
JTW and (NSC98-2321-B-001-024-MY3) CAC and
a BRCAS grant to CAC. This is CREEG-BRCAS
contribution no. 72.
REFERENCES
Averts LAM. 1998. Sponge/coral interactions in Caribbean
reefs: analysis of overgrowth patterns in relation to
species identity and cover. Mar. Ecol. Progr. Ser. 175:
241-249.
Averts LAM. 2000. Dynamics behind standoff interactions
in three reef sponge species and the coral Montastrea
cavernosa. Mar. Ecol. 21: 191-204.
Bryan PG. 1973. Growth rate, toxicity and distribution of the
encrusting sponge Terpios sp. (Hadromerida: Suberitidae)
 Wang et al. – Coral-Terpios Interactions
in Guam, Mariana Islands. Micronesica 9: 237-242.
Chadwick NE. 1987. Interspecific aggressive behavior of
the corallimorpharian Corynactis californica (Cnidaria:
Anthozoa): effects on sympatric corals and sea ane-
mones. Biol. Bull. 173: 110-125.
Chornesky EA. 1983. Induced development of sweeper
tentacles on the reef coral Agaricia agaricites: a response
to direct competition. Biol. Bull. 165: 569-581.
Connell JH. 1973. Population ecology of reef-building corals.
In OA Jones, R Endean, eds. Biology and geology of
coral reefs, Vol. II: Biology 1. New York: Academic Press,
pp. 205-245.
Connell JH. 1978. Diversity in tropical rain forests and coral
reefs. Science 199: 1302-1310.
Fujii T, S Keshavmurthy, W Zhou, E Hirose, CA Chen, JD
Reimer. 2011. Coral-killing cyanobacteriosponge (Terpios
hoshinota) on the Great Barrier Reef. Coral Reefs 30:
483.
Hirose E, A Murakami. 2011. Microscopic anatomy and
pigment characterization of coral-encrusting black sponge
with cyanobacterial symbiont, Terpios hoshinota. Zool.
Sci. 28: 199-205.
Iguchi A, S Iwanaga, H Nagai. 2008. Isolation and charac-
terization of a novel protein toxin from fire coral. Biochem.
Biophys. Res. Comm. 365: 107-112.
Jackson JBC, L Buss. 1975. Allelopathy and spatial com-
petition among coral reef invertebrates. Proc. Nat. Acad.
Sci. USA 72: 5160-5163.
Karlson RH. 1980. Alternative competitive strategies in a
periodically disturbed habitat. Bull. Mar. Sci. 30: 894-900.
Lang JC, EA Chornesky. 1990. Competition between scle-
ractinian reef corals - a review of mechanisms and effects.
In Z Dubinsky, ed. Ecosystems of the world, Vol. 25.
Coral reefs. Amsterdam: Elsevier, pp. 209-252.
Liao MH, SL Tang, GM Hsu, KC Wen, H Wu, WM Chen et al.
2007. The “black disease” of reef-building corals at Green
Island, Taiwan - outbreak of a cyanobacteriosponge,
Terpios hoshinota (Suberitidae: Hadromerida). Zool.
Stud. 46: 520.
McCook LJ, J Jompa, G Diaz-Pulido. 2001. Competition
between corals and algae on coral reefs: a review of
evidence and mechanisms. Coral Reefs 19: 400-417.
McLean EL, PM Yoshioka. 2007. Associations and interactions
between gorgonians and sponges. In MR Custódio, G
Lôbo-Hajdu, E Hajdu, GP Muricy, eds. Porifera research:
biochemistry, innovation and sustainability. Série Livros
28. Rio de Janeiro: Museu Nacional, pp. 443-448.
McLean EL, PM Yoshioka. 2008. Substratum effects on the
growth and survivorship of the sponge Desmapsamma
159
anchorata. Carib. J. Sci. 44: 83-89.
Plucer-Rosario G. 1987. The effect of substratum on the
growth of Terpios, an encrusting sponge which kills corals.
Coral Reefs 5: 197-200.
Porter JW, NM Targett. 1988. Allelochemical interactions
between sponges and corals. Biol. Bull. 175: 230-239.
Radwan FY, HM Aboul-Dahab. 2004. Milleporin-1, a new
phospholipase A2 active protein from the fire coral
Millepora platyphylla nematocysts. Comp. Biochem.
Physiol. Part C Toxicol. Pharmacol. 139: 267-272.
Reimer JD, Y Nozawa, E Hirose. 2011. Domination and
disappearance of the black sponge: a quarter century
after the initial Terpios outbreak in southern Japan. Zool.
Stud. 50: 395.
Rützler K. 2002. Impact of crustose Clionid sponges on
Caribbean reef corals. Acta Geol. Hispan. 37: 61-72.
Rützler K, K Muzic. 1993. Terpios hoshinota, a new cyano-
bacteriosponge threatening Pacific reefs. Sci. Mar. 57:
395-403.
Shiomi K, M Hosaka, N Yanaike, H Yamanaka, T Kikuchi. 1989.
Partial characterization of venoms from two species of
fire corals Millepora platyphylla and Millepora dichotoma.
Nippon Suisan Gakk 55: 357-362.
Soong K, SL Yang, CA Chen. 2009. A novel dispersal
mechanism of a coral-threatening sponge, Terpios
hoshinota (Suberitidae, Porifera). Zool. Stud. 45: 596.
Suchanek TH, RC Carpenter, JD Witman, CD Harvell.
1983. Sponge as important space competitors in deep
Caribbean coral reef communities. In ML Reaka, ed. The
ecology of deep and shallow coral reefs. Symposia Series
for Undersea Research, NOAA Undersea Research
Program 1(1). Washington DC: National Oceanic and
Atmospheric Administration (NOAA), pp. 55-60.
Tang SL, MJ Hong, MH Liao, WN Jane, PW Chiang, CB Chen,
CA Chen. 2011. Bacteria associated with an encru-
sting sponge (Terpios hoshinota) and the corals partially
covered by the sponge. Environ. Microbiol. 13: 1179-
1191.
Teruya T, S Nakagawa, T Koyama, T Arimoto, M Kita, D
Uemura. 2004. Nakiterpiosin and nakiterpiosinone,
novel cytotoxic C-nor-D-homosteroids from the Okinawan
sponge Terpios hoshinota. Tetrahedron 60: 6989-6993.
Wahle CM. 1980. Detection, pursuit, and overgrowth of
tropical gorgonians by milleporid hydrocorals: Perseus
and Medusa revisited. Science 209: 689-691.
Wittle LW, RE Middlebrook, CE Lane. 1971. Isolation and
partial purification of toxin from Millepora alcicornis.
Toxicon 9: 329-331.
Zoological Studies 51(2): 160-174 (2012)

Biodiversity of Planktonic Copepods in the Lanyang River (Northeastern

Taiwan), a Typical Watershed of Oceania
Hans-Uwe Dahms1, Li-Chun Tseng2, Shih-Hui Hsiao2, Qing-Chao Chen3, Bong-Rae Kim4, and
Jiang-Shiou Hwang2,*
1
Green Life Science Department, College of Convergence, Sangmyung Univ., 7 Hongij-dong, Jongno-gu, Seoul110-743, South Korea
2
Institute of Marine Biology, National Taiwan Ocean University, 2 Pei-Ning Road, Keelung 202, Taiwan
3
South China Sea Institute of Oceanology, Chinese Academy of Science, Guangzhou 510301, China
4
National Fisheries Research and Development Institute, Inland Fisheries Research Institute, Kyunggi-do 114-3, South Korea

(Accepted September 21, 2011)

Hans-Uwe Dahms, Li-Chun Tseng, Shih-Hui Hsiao, Qing-Chao Chen, Bong-Rae Kim, and Jiang-Shiou
Hwang (2012) Biodiversity of planktonic copepods in the Lanyang River (northeastern Taiwan), a typical
watershed of Oceania. Zoological Studies 51(2): 160-174. To evaluate the environmental status of a typical
Oceania watershed in Taiwan, zooplankton samples were collected bimonthly along the Lanyang River (NE
Taiwan) at 9 different stations including 1 estuarine and 8 freshwater stations during 10 sampling campaigns
from June 2004 to Dec. 2005. Upstream stations showed lower chlorophyll a and temperature values than
downstream stations; the highest chlorophyll a concentration was found in the estuary at all times. We
identified 21 copepod species, belonging to 4 orders, 12 families, and 20 genera in total. Eleven species
were recorded only once among all samples. The Calanoida was restricted to samples from the estuary. The
Poecilostomatoida was only recorded from the estuary and the Lanyang Bridge station. The Harpacticoida was
only recorded from the estuary, Lanyang Bridge, and Tsu-Keng River stations. At 2 mid-section stations, no
copepods were found. The upstream station showed lower abundance, species number, species richness, and
evenness and diversity indices than the downstream and estuarine stations. The estuarine station provided the
highest copepod abundance (3410.05 individuals/m3) and species number (12 species/station) in Aug. 2004
when the waters showed the highest salinities (37 psu), indicating the marine origin of the diverse biota. Among
all samples, there were no significant differences in the abundance, number of species, or indices of richness,
evenness, and diversity among sampling months. In contrast, our analysis clearly showed a succession in
abundance and species composition among sampling months. At the estuarine station, copepod abundances
were significantly positive correlated with salinity (r = 0.880, p = 0.001). Numbers of species were significantly
positive correlated with chlorophyll a (r = 0.790, p = 0.007), salinity (r = 0.780, p = 0.008), and copepod
abundance (r = 0.785, p = 0.007). Copepod abundances were mainly affected by intruding seawater, but there
was no interaction with the month of sampling. http://zoolstud.sinica.edu.tw/Journals/51.2/160.pdf

Key words: Riverine zooplankton, River ecology, Estuary, Copepod mesozooplankton, Plankton communities.

W hen compared to the open ocean, coas-
tal and estuarine ecosystems may be smaller,
in terms of area and volume, but the amount
of organic carbon exported to the deep ocean
through the coastal fringe (1.7 × 1015 tons C/yr)
can reach nearly that of the entire oceanic realm
(13.2 × 10 15 tons C/yr) (Bienfang and Ziemann
1992, Carlsson et al. 1995). Coastal tropical
environments are of particular importance in this
respect (Nittrouer et al. 1995). Increasing attention
has been given to small mountainous rivers with
drainage areas of < 10,000 km2. These smaller
but numerous rivers could collectively be very
important in transporting sediments and particulate

*To whom correspondence and reprint requests should be addressed. Hans-Uwe Dahms and Li-Chun Tseng contributed equally to this
work. Tel: 886-935289642. Fax: 886-2-24629464. E-mail:jshwang@mail.ntou.edu.tw

160
 Dahms et al. – Lanyang River Copepods
organic carbon to the ocean (Hsu et al. 1998).
Many of these rivers are present on islands of the
western Pacific, collectively called Oceania (Kao
and Liu 1996 1997). On Oceania islands, such
as Taiwan, high precipitation, steep slopes, small
basin areas, and frequent flood events can induce
high erosion rates (Carry et al. 2002). These
natural characteristics make watersheds much
more vulnerable to anthropogenic perturbations
(Cearreta et al. 2000) such as exacerbation of
erosion induced by human perturbations in the
Lanyang River (Kao and Liu 2002).
Rivers provide a unique gradation of environ-
ments: from pristine waters to a mix of riverine
and seawater in their estuaries. Considering the
high resilience of the estuarine portion of rivers,
analyses of zooplankton community assemblages
along riverine, estuarine, and marine sections of
a river mouth are warranted to understand the
main determinants of zooplankton communities
in estuaries (Thor et al. 2005, Hwang et al. 2000
2006 2009b 2010).
The Lanyang River is a typical watershed on
an Oceania island and is used as an example in
the present study. Shiah et al. (1996) differentiated
3 types of waters in the estuary of the Lanyang
River: river-mouth water, marine seawater, and
mixed water. Amounts of precipitation in the
drainage basin and estuary of the Lanyang River
are influenced by a shift in seasonal currents.
These are driven by the northward flow of the
Kuroshio Current along eastern Taiwan year round
and during winter by the northeasterly monsoon
(Jan et al. 2002, Lee and Chao 2003, Liang et al.
2003, Liu et al. 2003, Hwang et al. 2006, Tseng
et al. 2008b, Hsieh et al. 2011). Although being
the largest tidal river in northeastern Taiwan,
comparatively few biological and hydrological
investigations have been undertaken in the
Lanyang River and its estuary. For example, there
is no information available about zooplankton in
general or copepod community structures, and
particularly about their dynamics. To determine
the ecological health of a river, such as the
Lanyang River, an assessment study needs
to sample a variety of physicochemical and
biological parameters. Studies at higher levels
of organization are important to understand
environmental stressors on ecologically relevant
endpoints such as community diversity. Thus,
organism-level responses are important in
assessing the health of aquatic systems and their
recovery after a disturbance. Establishment of
relationships between stressors and biological
161
responses serve as the basis of management
decisions and environmental remediation practices.
Copepods are claimed to be numerically
the most abundant metazoans (Schminke 2007,
Chang et al. 2010, Hwang et al. 2004 2010, Kâ
and Hwang 2011) and play a central role in the
transfer of carbon from producers to higher trophic
levels in most aquatic ecosystems (Jerling and
Wooldridge 1995). Copepods are the primary
consumers of phytoplankton and are the main prey
items of larval and juvenile fishes that link pelagic
food webs (Tseng et al. 2008a 2009, Vandromme
et al. 2010, Wu et al. 2010). Copepods are used
as indicator species for waters of different qualities
and origins (Bonnet and Frid 2004, Hwang and
Wong 2005, Thor et al. 2005, Hwang et al. 2006
2009a 2010). Understanding the copepod fraction
of the mesozooplankton is thus meaningful to
fundamental ecology and applied environmental
monitoring (Chullasorn et al. 2009 2011, Hwang et
al. 2009b). This also holds for the management
and protection of biological resources of other
riverine watersheds worldwide and in Oceania.
In the present paper, we investigated plank-
tonic copepod assemblages in the freshwater
and estuarine portions of the Lanyang River in
order to understand the major determinants such
as temperature, salinity, chlorophyll (Chl) a, and
copepod abundances and distributions in a typical
smaller watershed of a subtropical Oceania island.
MATERIALS AND METHODS
Study site
The Lanyang River is a comparatively small
watershed in Taiwan but the largest in northeastern
Taiwan (Fig. 1). It originates at an elevation of
3535 m and runs for only 73 km. It is on average
0.5 km wide with a comparatively small drainage
basin area of 820 km 2 (Kao 1996). The main
channel flowing northeasterly, is affected by high
precipitation, and has a steep slope (with a mean
gradient of 1: 21) (Kao 1996). The river mouth
is shallow (< 2 m deep) and narrow. The annual
precipitation in this watershed ranged 2000-
5000 mm for the past 50 yr, with an average of
about 3000 mm (Kao and Liu 2002). This amount
is high as an average value on a global scale,
but typical for Oceania islands. The lithology
and climatic conditions are homogeneous in the
watershed. The basement rock of the Lanyang
River watershed is composed mainly of Tertiary
162 Zoological Studies 51(2): 160-174 (2012)

argillite-slate and metasandstone (Ho 1975).
There are 2 gauge stations along the main channel
of the river. Gauge 1 is located above the tidal
zone at the river mouth. Gauge 2 is located in
the upper part at an elevation of 450 m. The
drainage areas above G1 and G2 are 820 and
273 km 2, respectively. There were 2 massive
road construction events in the study area in the
past 50 yr which were the major anthropogenic
disturbances in the watershed and which also
allowed farming disturbances on otherwise steep
slopes. Vegetable plantations were developed
along the riverbeds and banks of the main stem
as high as 1250 m in elevation. There is little
domestic effluent in the upper and midstream
sections, but heavy discharges in the estuary of
the Lanyang River. Most disturbances are due
to agricultural activities in upstream areas and a
quarry in the midstream portion.
Zooplankton and water sample collection
Based on preliminary surveys of salinity and
copepod distributions, 9 stations were set up at
24°27'41"-24°43'01"N and 121°24'12"-121°49'31"E
along the Lanyang River in order to cover most
of the range of environmental conditions. These
stations are grouped in 3 areas and include fresh
waters and brackish waters in the estuarine river
mouth (Fig. 1, Table 1). The largest distance
between the Gah-Siang (GS) Bridge and the coast
is about 73.00 km. There is a drastic decrease

N Tsu-Keng River Lanyang Bridge

Sung-Luo River Estuary

Yi-Lan County

Niour-Douh Bridge

26° CHINA
Pengjiayu
East China Sea
Ga-Yuan Bridge
Taiwan Strait

Ji-Kwan Bridge 24°

TAIWAN
Gah-Siang
Bridge Sh-Gu-Fuh River Pacific Ocean

22°
120° 122°

Fig. 1. Sampling stations along the Lanyang River in northeastern Taiwan.

Table 1. Station name, abbreviation, code, elevation, and distance to the coast

Location Abbreviation Code for station Elevation (m) Distance to coast (km)

Gah-Siang Bridge GS Bridge A 1534 73.00

Sh-Gu-Fuh River SGF River B 948 67.43
Ji-Kwan Bridge JK Bridge C 808 65.08
Ga-Yuan Bridge GY Bridge D 376 47.10
Niour-Douh Bridge ND Bridge E 209 35.11
Sung-Luo River SL River F 189 28.69
Tsu-Keng River TK River G 90 17.55
Lanyang Bridge LY Bridge H 2 6.21
Estuary Estuary I 1 0.10
 Dahms et al. – Lanyang River Copepods
in elevation from 1534 to 90 m between the GS
Bridge and Tsu-Keng (TK) River stations. The
Sung-Luo (SL) River and TK River stations are
located where these streams merge with the
Lanyang River. Whereas 7 upper stations belong
to the freshwater zone, the 2 stations of Lanyang
(LY) Bridge and the estuary are influenced by
seawater. Waters at the LY Bridge station are less
affected by seawater as they are somewhat farther
from the coast.
Stations were sampled every 2nd month
during 10 sampling periods from June 2004 to
Dec. 2005. Tows were performed from a boat
at the estuary station and by foot at the river
stations. Since the channel of the Lanyang River
is shallow and narrow, boats cannot navigate in
the middle and upstream sections. Zooplankton
samples were obtained by towing a modified North
Pacific (NORPAC) zooplankton net (with a mouth
diameter of 45 cm, a mesh size of 100 µm, and
a length of 180 cm with a Hydrobios flow meter
(Germany) mounted at the center of the net mouth)
horizontally for 10 min at the surface at all stations
(Hwang et al. 2007). Zooplankton were preserved
in a buffered 5% formalin-seawater solution for
later sorting, identification, and counting in the
laboratory.
Water temperature and salinity were mea-
sured with a mercury thermometer and refrac-
tometer (S-100, Tanaka Sanjiro Co.,Ltd., Japan),
respectively. Water samples at 1 m in depth were
collected with Niskin bottles to determine Chl a
concentrations using the fluorometric method
of Parsons et al. (1984). Other parameters like
precipitation, nutrient and organic loadings were
not measured but were taken from the literature
(Kao and Liu 1996 1997).
Copepod enumeration and identification
In the laboratory, zooplankton samples were
subsampled with a Folsom splitter. Procedures
for species identification and counting were similar
to those described by Hwang et al. (1998 2006).
Adult copepods in the subsamples were identified
and counted under a stereomicroscope. Species
were identified according to keys and references
by Chen and Zhang (1965), Chen et al. (1974),
Shih and Young (1995), and Chihara and Murano
(1997). Freshwater copepods were identified
according to Dussart and Defaye (2006) and
Einsle (1996) if not indicated otherwise.
163
Data analysis
Copepod community structures were
analyzed using the Plymouth Routine In
Multivariate Ecology Research (PRIMER)
computer package (Version IV; Clarke and
Warwick 1994). In order to reduce the higher
heteroscedasticity observed in the abundance data
for major taxa, a transformation power (λ = 0.983)
was generated by regression coefficients, that
were simultaneously estimated using a method of
maximizing the log-likelihood function (Box and
Cox 1964). Copepod abundance data were log
(X + 1)-transformed before clustering, using the
matrix of abundances composed of samples and
species. Similarity coefficients between samples
were computed using the Bray-Curtis similarity
and clustering strategy of flexible links. Three
stations (Ji-Kuan (JK) ridge, Ga-Yuan (GY) Bridge,
and Niour-Douh (ND) Bridge) were not considered
in the cluster analysis since they contained no
copepods. For correlations between abiotic
factors and zooplankton abundances, Pearson’s
product moment correlation coefficients were
calculated with the SPSS computer package
(Chicago, IL, USA). The Mann-Whitney U-test
was applied to compare spatial and seasonal
differences in surface-water temperatures. A one-
way analysis of variance (ANOVA) was applied
to reveal differences in abundances, numbers of
species, and indices of richness, evenness, and
diversity among sampling months. The Shannon-
Wiener diversity index (Weaver and Shannon
1949) together with the richness index and Pielou’s
evenness index (Pielou 1966) were applied to
estimate the copepod community composition.
RESULTS
Overview of the Lanyang River
Hydrographic parameters, Chl a, and salinity
We recorded high surface temperatures
from June to Dec. in both years 2004 and 2005.
Stations along the upper stream showed lower
Chl a (0.84 ± 0.25 μg/L) and temperature (19.4
± 4.45°C) values than the downstream stations
(2.35 ± 1.90 μg/L for Chl a and 25.5 ± 4.11°C for
temperature) (Fig. 2). The estuarine station always
had higher concentrations of Chl a (Fig. 2A).
Spatial differences in surface water tempe-
rature were not significant, but seasonal differences
164 Zoological Studies 51(2): 160-174 (2012)

Gah-Siang Bridge Sung-Luo River in temperature were significant (p < 0.05; Mann-
Sh-Gu-Fuh River Tsu-Keng River Whitney U-test), with the lowest temperature of
Ji-Kwan Bridge Lanyang Bridge 9.1°C recorded in Feb. 2005 (GS Bridge station)
Ga-Yuan Bridge Estuary
and the highest temperature of 31.5°C in Aug.
2004 (ND Bridge and LY Bridge stations) (Fig. 2B).
Niour-Douh Bridge
All sampling stations, except the estuarine one,
(A)
7 had freshwater conditions throughout the study
period, with salinities of 0 psu at all times (Fig. 2C).
6 A plume-ward progressive increase in salinity was
obvious in the estuary. The value recorded off the
5 estuary station showed remarkable changes from
0 (i.e., fresh water) to 37 psu, with an average
Chlorophyll a (μg/L)

4
of 9.58 psu, indicating a variable influence of
3 riverine freshwater outflow near the surface from
upstream and of near-bottom intrusion of saline
2 water from the ocean. The salinity was 37.0 psu
when seawater intruded the estuary, and it was 0
1 psu when fresh water flushed the estuary after a
heavy rainfall or during an ebb tide (Fig. 2C). All 3
0
hydrographic parameters showed maximum values
(B)
35 at the estuarine station (Fig. 2A-C).

30 Copepod abundance and diversity

25
In total, 28 species and 21 genera, be-
longing to 12 families and 4 copepod orders
Temperature (°C)

20
were identified from the marine, estuarine, and
riverine portions of the Lanyang River (Table
2). The Poecilostomatoida was only recorded
15
at the estuarine and LY Bridge stations. At the
sampling station closest to the Lanyang River
10
mouth, Apocyclops borneoensis showed the
highest occurrence rate (6.67%). In contrast, 11
5
copepod species were recorded only once among
(C) all samples (with an occurrence ratio (OR) of
40
1.11%). No copepods were found at the following
3 stations: JK Bridge, GY Bridge, and ND Bridge,
located in the lower section of the upper 1/2 of the
30
river (Table 2).
In terms of both diversity and abundance,
Salinity (PSU)

copepods were the most dominant component

20
of the zooplankton in all samplings. Among all
zooplankton, cyclopoid copepods were most
prominent, in terms of both diversity and density at
10
most sampling stations throughout the investigation
period. The number of calanoids was highest in
the estuary. The Poecilostomatoida was only found
0
in the estuary. From all samples, results showed
June Aug. Oct. Dec. Feb. Apr. June Aug. Oct. Dec.
that copepod abundances at the estuarine station
2004 2005 were significantly affected by the intrusion of
Sampling month
seawater. Freshwater copepods were represented
Fig. 2. Chlorophyll a (A), water temperature (B), and salinity by 1 calanoid species of Mongolodiaptomus birulai
(C) in each sampling month at each sampling station. Only the and 9 cyclopoid species of Acanthocyclops sp.,
estuary station exhibited salinities of > 0 psu. Apocyclops borneoensis, Cyclopina sp., Cyclops
 Dahms et al. – Lanyang River Copepods 165

Table 2. Abundance (mean ± S.D. individuals (ind.)/m3), relative abundance (RA, %), and occurrence rate
(OR, %) of copepods (ind./m3) at recorded stations. A, GS Bridge; B, SGF River; C, JK River; D, GY Bridge;
E, ND Bridge; F, SL River; G, TK River; H, LY Bridge; I, estuary

Recorded Mean ± S.D.

Species RA (%) OR (%)
station (ind./m3)

Calanoida
Acartiidae
Acartia (Odontacartia) erythraea Giesbrecht 1889 I 184 3.808 1.11
Acartia (Plantacartia) negligens Dana 1849 I 2.48 ± 2.15 0.103 2.22
Centropagidae
Sinocalanus sp. I 0.4 0.008 1.11
Sinocalanus tenellus (Kikuchi) 1928 I 0.68 0.014 1.11
Diaptomidae
Mongolodiaptomus birulai (Rylov) 1922 I 0.68 0.014 1.11
Eucalanidae
Subeucalanus subcrassus (Giesbrecht) 1888 I 13.14 0.272 1.11
Paracalanidae
Acrocalanus gracilis Giesbrecht 1888 I 15.37 ± 3.15 0.636 2.22
Paracalanus aculeatus Giesbrecht 1888 I 3.93 ± 1.7 0.244 3.33
Parvocalanus crassirostris (Dahl) 1893 I 473.19 ± 708.37 29.379 3.33
Pseudodiaptomidae
Pseudodiaptomus annandalei Sewell 1919 I 15.93 ± 17.04 0.989 3.33
Pseudodiaptomus serricaudatus (Scott T) 1894 I 365.22 ± 622.25 22.676 3.33
Temoridae
Temora turbinata (Dana) 1849 I 93.1 ± 128.56 3.854 2.22
Cyclopoida
Cyclopidae
Acanthocyclops sp. H 0.75 ± 1.06 0.031 1.11
Apocyclops borneoensis Lindberg 1954 A, B, H, I 36.81 ± 47.42 4.570 6.67
Cyclops sp. A, H, I 30.79 ± 31.81 1.912 3.33
Eucyclops sp. F, I 6.4 ± 7.38 0.265 2.22
Mesocyclops pehpeiensis Hu 1943 G, I 6.06 ± 2.12 0.251 2.22
Mesocyclops sp. G, H, I 0.97 ± 0.82 0.060 3.33
Microcyclops sp. A, B, H, I 6.63 ± 7.94 0.549 4.44
Thermocyclops kawamurai Kikuchi 1940 H, I 380.91 ± 514.96 15.766 2.22
Cyclopinidae
Cyclopina sp. I 56.98 ± 68.13 2.358 2.22
Oithona rigida Giesbrecht 1896 I 116.14 ± 133.15 4.807 2.22
Oithona similis Claus 1866 I 13.14 0.272 1.11
Harpacticoida
Euterpinidae
Euterpina acutifrons (Dana) 1847 I 289.15 5.984 1.11
Poecilostomatoida
Corycaeidae
Corycaeus (Ditrichocorycaeus) erythraeus Cleve 1901 I 13.14 0.272 1.11
Corycaeus (D.) subtilis M. Dahl 1912 H, I 20.38 ± 5.07 0.844 2.22
Corycaeus (Farranula) concinna (Dana) 1847 H 0.78 0.016 1.11
Corycaeus (F.) gibbula Giesbrecht 1891 I 2.2 0.046 1.11

Total 53.69 ± 367.2 100.0

166 Zoological Studies 51(2): 160-174 (2012)

sp., Eucyclops sp., Mesocyclops sp., Mesocyclops 193.50 ind./m 3 in June 2004 at the estuarine
peheiensis, Microcyclops sp., and Thermocyclops station. Only the estuarine station showed
kawamurai. significant differences in copepod abundances
The highest abundance (3410.05 ind./m3) was during the sampling period. Copepod abundances
recorded in Aug. 2004, followed by 745.04 ind./m3 were < 5.0 ind./m3 at all freshwater stations (Fig.
in Feb. 2005, and the 3rd highest record was 3A). The number of species at the 9 sampling

(A)
3500
3000
2500 Gah-Siang Bridge
2000
Sh-Gu-Fuh River
Copepod abundance (ind. m-3)

1500
1000 Ji-Kwan Bridge
500
Ga-Yuan Bridge
Niour-Douh Bridge
150
Sung-Luo River

100 Tsu-Keng River

Lanyang Bridge
50 Estuary

(B) (C)
14 1.6
Number of species (no. station-1)

12 1.4

10 1.2

1.0
Richness index

8
0.8
6
0.6
4
0.4
2
0.2
0
0.0
(D) (E)
1.8
1.0
1.6

0.8 1.4
Evenness index

1.2
Diversity index

0.6 1.0

0.8
0.4
0.6

0.4
0.2
0.2
0.0 0.0
June Aug. Oct. Dec. Feb. Apr. June Aug. Oct. Dec. June Aug. Oct. Dec. Feb. Apr. June Aug. Oct. Dec.
2004 2005 2004 2005
Sampling month Sampling month

Fig. 3. Total copepod abundance (A), number of species (B), indices of richness (C), evenness (D), and Shannon-Wiener diversity (E)
of each sampling month at each sampling station.
 Dahms et al. – Lanyang River Copepods 167

stations ranged 0-12/station. The highest species Temporal and spatial variations in the copepod
number (of 12 species/station) was recorded at community structure
the estuarine station in Aug. 2004 (Fig. 3B). The
record of identified species was < 2 at the LY As for seasonal differences, the highest
Bridge station. In the remaining 8 stations of the record of average copepod abundances
freshwater zone, the number of identified species (379.06 ind./m3) was in Aug. 2004, and the 2nd
ranged 0-1/station during the sampling period (Fig. highest record was 84.65 ind./m3 in Feb. 2005.
3B). The indices of richness (Fig. 3C), evenness The remaining sampling months showed values
(Fig. 3D), and diversity (Fig. 3E) showed high of < 25 ind./m 3 (Fig. 4A). The highest species
variations at the estuarine station. Indices could number (13) was also recorded in Aug. 2004
not be calculated at the remaining 8 sampling (Fig. 4B), whereas the lowest record of 1 was in
stations due to species numbers being < 2. Feb. 2005. Most sampling months presented

(A) (B)
1600 14

Number of species (no. month-1)

Copepod abundance (ind. m-3)

1500
12

10
400
8
300
6
200
4
100 2

0 0
(C) (D)
1.2 2.0

1.0
1.6
0.8
Evenness index
Richness index

1.2
0.6
0.8
0.4

0.2 0.4

0.0 0.0
June Aug. Oct. Dec. Feb. Apr. June Aug. Oct. Dec.
(E)
0.8 2004 2005
Sampling month

0.6
Diversity index

0.4

0.2

0.0
June Aug. Oct. Dec. Feb. Apr. June Aug. Oct. Dec.

2004 2005
Sampling month

Fig. 4. Average copepod abundance (A), total copepod species number (B), indices of richness (C), evenness (D) and Shannon-
Wiener diversity (E) in each sampling month.
168 Zoological Studies 51(2): 160-174 (2012)

communities with < 4 species. Indices of richness
(Fig. 4C), evenness (Fig. 4D), and diversity (Fig.
4E) showed some temporal variability without a
clear trend. There were no significant differences
in abundance, species number, or indices of
richness, evenness, and diversity of copepods
among sampling months (Fig. 4, p > 0.05, one-way
ANOVA).
When stations were compared, the highest
mean abundance (462.40 ind./m3, Fig. 5A) was
found at the estuarine station. Accumulated
records of copepod species provided the highest
species numbers (26 species/station, Fig. 5B) at
the estuarine station. The 7 upstream stations
(GS Bridge, SGF River, JK Bridge, GY Bridge,
ND Bridge, SL River, and TK River) showed
lower values of abundance, species number, and
richness, evenness, and diversity indices than the
downstream stations (LY Bridge and estuarine
stations) throughout the year (Fig. 5A-E).

(A) (B)
1600 30

Number of species (no. station-1)

Copepod abundance (ind. m-3)

1500
25
500
20
400
15
300
10
200

100 5

0 0
(C) (D)
1.6 1.0
1.4
0.8
1.2
Evenness index
Richness index

1.0 0.6
0.8
0.6 0.4

0.4
0.2
0.2
0.0 0.0
Br r

Br r
Es ge

y
F e

Y ge

Br e
SL dge

r
JK ive

TK ive
LY ive

(E)
ar
SG ridg

N ridg

id
G id

tu

1.4
R

R
R
i
B

B
S

D
G

1.2
Sampling month
1.0
Diversity index

0.8

0.6

0.4

0.2

0.0
F e

B e
D ge

SL dge

Es ge

y
Br r

TK iver

Br r
e

LY ive

ar
SG ridg

G idg
JK iv

N rid

id
tu
R

R
i
Br
B

Y
S
G

Sampling month

Fig. 5. Average copepod abundance (A), total copepod species number (B), indices of richness (C), evenness (D) and Shannon-
Wiener diversity (E) at each sampling station.
 Dahms et al. – Lanyang River Copepods 169

Copepod community structure cluster results indicated that copepod communities

A cluster analysis using Bray-Curtis simi-
larities of taxonomic abundances provided
3 groups of stations: IB (SL River), IIB (TK
River), IIIA (estuarine station and LY Bridge),
and IIIB (SGF River and GS Bridge) (Fig. 6).
Accumulations of major 75% copepods in each
group according to Bray-Curtis similarity cluster
results are given in table 3. The grouping results
allocated downstream and upstream stations
to different groups. The TK River and SL River
stations were separated in single groups due to
the appearance of rare species and low copepod
abundances. In the upstream area of the Lanyang
River, Apocyclops borneoensis and Microcyclops
sp. were major species at the SGF River and GS
Bridge stations. The dominant species at the SL
River and TK River stations were Eucyclops sp.
and Mesocyclops peheiensis, respectively. The
III A
II A
III B
IA
II B
IB
0 20 40 60 80
Bray-Curtis similarity
Fig. 6. Cluster analysis of Bray-Curtis similarities. Stations
fell into 3 groups: IB (SL River), IIB (TK River), IIIA (estuarine
station and LY Bridge), and IIIB (SGF River and GS Bridge).
were affected by intruding seawater at the
estuarine and LY Bridge stations (Table 3). Thus,
copepod communities clearly differed in the upper,
middle, and downstream areas.
Estuarine station: relationships of copepod
abundance and species number with environ-
mental factors
We found 2 peculiarities of copepod assem-
blages in the Lanyang River. First, there was a
low abundance and low diversity in the freshwater
zone. Second, seawater intrusions transported
oceanic copepods to the estuarine area which
raised the abundance and diversity of the copepod
communities. Only the estuary station was
significantly affected by seawater. Based on
these results, we focused on data of the estuarine
station to reveal the effects of intruding seawater
(Fig. 7). The highest abundance (3410.05 ind./m3)
and species number (12 species/station) (Fig. 7A)
Estuary
corresponded to the highest measured salinity
LY Bridge (37.0 psu) in the estuary (Fig. 7C). Pearson’s
SGF River product moment correlation analysis confirmed
GS Bridge the positive correlation of copepod abundances
TK River with salinity (r = 0.880, p = 0.001) (Table 4).
The species number was significantly positive
SL River
correlated with Chl a (r = 0.790, p = 0.007), salinity
100
(r = 0.780, p = 0.008), and copepod abundance
(r = 0.785, p = 0.007). These results indicated
that copepod assemblages at the estuarine station
were strongly affected by intruding seawater which
changed the hydrography and biota.

Table 3. Accumulated of major 75% copepods in each group according to the Bray-Curtis similarity cluster
results

Copepod species Group

IB IIB IIIA IIIB

Acartia erythraea 6.70

Apocyclops borneoensis 47.30
Eucyclops sp. 100.00
Euterpina acutifrons 10.53
Mesocyclops peheiensis 98.99
Microcyclops sp. 41.21
Pavocalanus crassirostris 17.22
Pseudodiaptomus serricaudatus 13.29
Thermocyclops kawamurai 27.73

Cumulative contribution (%) 100.00 98.99 75.47 88.50

170 Zoological Studies 51(2): 160-174 (2012)

DISCUSSION show diverse rotifer assemblages, whereas

microcrustaceans are almost always absent
Progress in understanding environmental (Richardson 1992, Sluss et al. 2008). This is
conditions that control the distribution and in contrast to lakes where copepods and large
abundance of riverine zooplankton and their cladocerans most frequently dominate the system,
ecological significance has lagged far behind with relative abundances often influenced by
that of lentic environments (Casper and Thorp biotic (e.g., chaoborid dipteran and fish predation)
2007). Hydrologically dynamic rivers commonly and abiotic factors (e.g., inorganic turbidity)

abundance species no.

(A) (B)
4000 30
12
Copepod abundance (ind. m-3)

27
Number of species

Temperature (°C)
3000
9
24
2000
6
21

1000 3 18

0 0 15
(C) (D)
40 7

6
30
5
Chlorophyll a (μg/L)
Salinity (PSU)

4
20
3

10 2

0 0
June Aug. Oct. Dec. Feb. Apr. June Aug. Oct. Dec. June Aug. Oct. Dec. Feb. Apr. June Aug. Oct. Dec.

2004 2005 2004 2005

Sampling month Sampling month

Fig. 7. Variations in copepod abundances and number of species (A), temperature (B), salinity (C), and chlorophyll a (D) of the estuary
station in each sampling month.

Table 4. Correlation between water temperature, chlorophyll a, salinity, copepod abundance and species
number at the estuary station during June 2004 to Dec. 2005

Pearson correlation

Temperature Chlorophyll a Salinity Abundance

Chlorophyll a 0.429
Salinity 0.346 0.531
Abundance 0.390 0.596 0.880 **
Species number 0.575 0.790 ** 0.780 ** 0.785 **

** Correlation is significant at the 0.01 level (2-tailed).

 Dahms et al. – Lanyang River Copepods
(Wetzel 2001). Slack waters are suggested to
be critical for sustaining a high biomass and
diversity of zooplankton (Thorp and Casper
2003). Where those are lacking, and the river
channel is steep and presents a high current
velocity, the zooplankton is affected by turbulence
and commonly shows low biomass (Sluss et al.
2008). This may also hold for the Lanyang River,
particularly the 3 stations (JK Bridge, GY Bridge,
and ND Bridge) in the midstream section where we
found no copepods in any sampling period.
A study by Hsu et al. (2004) further explored
the positive relationship between observed
sediment fluxes and runoff in the Lanyang River
which may also have ultimately affected the
zooplankton communities and caused the lack of
copepods at these stations. According to those
authors, the annual sediment discharge and
sediment yield of the Lanyang River are 8.0 Mt
and 8154 Mt/km2, respectively. The annual runoff
is 2773 × 10 6 m 3, and transient runoff always
rises sharply from the baseline level of several
tens of m 3 /s (with a mean rate of 62 m 3 /s) to
an abnormal level of thousands of m 3/s after a
heavy rainfall (with an average annual rainfall of
3256 mm). Since 1949, the maximal records of
daily runoff and suspended sediment concentration
are 4580 m3/s and 118 g/L, respectively (Water
Resources Bureau 1997a b). Sediment discharges
depend on river runoff, and a function relating the
2 parameters was established (Kao 1996).
According to Sluss et al. (2008), it is uncertain
which abiotic factors control both the relative
abundance of major groups and the relative size
of the zooplankton community of rivers. Biological
factors were addressed relatively rarely, but field
surveys and in situ experiments suggest that
competition and predation play roles in regulating
river plankton at least in slack waters (Casper and
Thorp 2007).
Our study demonstrated that marine zoo-
plankton substantially contributed to the
estuarine section of the Lanyang ecosystem.
Here, the highest abundance (3410.05 ind./m3)
and species number (12 species/station) cor-
responded with the highest salinity (37.0 psu),
demonstrating the marine role in shaping and
maintaining the estuarine planktonic community.
As mentioned, the zooplankton abundance
of the Lanyang River estuarine station was
significantly affected by seawater intrusions, and
the number of zooplankton groups was affected
by water temperature (as affected by the seasonal
monsoon; see Hsieh et al. 2011). Hence, the
171
marine compartment may determine the dynamics
of the zooplankton communities in the estuary of
the Lanyang River (Tan et al. 2004). The estuary
of the Lanyang River is next to nearshore waters
off the northeastern coast of Taiwan. Hwang et
al. (1998) found that copepods represented the
dominant zooplankton group along the northern
coast of Taiwan.
River flow and tidal motions respectively
drive the riverine and marine communities towards
estuaries (Hsieh and Chiu 1997, Waniek et al.
2005, Zhang et al. 2010) and hence shape the
diversity and density of estuarine communities
(Waniek 2003, Froneman 2004). However, there
is the possibility that resident dormant stages
contribute to estuarine populations as well, once
they emerge from their sedimentary depositions
(Dahms and Qian 2004, Dahms et al. 2006). The
biology and mesozooplankton including copepod
assemblages of the East China Sea and Kuroshio
Current are little known (Hsiao et al. 2004 2011),
even though there was an interdisciplinary
study of the Kuroshio Current (Marr 1970) and
several oceanographic research programs, such
as KEEP (Kuroshio-East China Sea Exchange
Process) in the last decade by several research
institutions and universities in Taiwan (Liu 1997,
Hwang et al. 2006). In conclusion, copepod up-
stream assemblages were characterized by low
abundances and low species diversities, whereas
the estuarine station showed a high abundance
and a high number of species which were cor-
related with intruding seawater.
Acknowledgments: Assistance by laboratory
members of J.S. Hwang at various stages of
sample collection and manuscript preparation is
gratefully acknowledged. We are thankful to the
captain and crew of local ships in the Lanyang
River estuary. We acknowledge the initiative of
Drs. K.T. Shao and H.J. Lin to help in getting the
present research underway.
REFERENCES
Bienfang PK, DA Ziemann. 1992. The role of coastal high
latitude ecosystems in global export production. In PG
Falkowski, AD Woodhead, eds. Primary productivity and
biogeochemical cycles in the sea. New York: Plenum, pp.
285-297.
Bonnet D, C Frid. 2004. Seven copepod species considered
as indicators of water mass influence and changes:
results from a Northumberland coastal station. S. Afr. J.
Mar. Sci. 61: 485-491.
172 Zoological Studies 51(2): 160-174 (2012)
Box GEP, DR Cox. 1964. An analysis of transformations. J.
Roy. Stat. Soc. Ser. B 26: 211-246.
Carlsson P, E Graneli, P Tester, L Boni. 1995. Influences
of riverine humic substances on bacteria, protozoa,
phytoplankton, and copepods in a coastal plankton
community. Mar. Ecol. Prog. Ser. 127: 213-221.
Carry AE, CA Nezat, WB Lyons, SJ Kao, DM Hicks, JS Owen.
2002. Trace metal fluxes to the ocean: the importance of
high-standing oceanic islands. Geophys. Res. Lett. 29:
2099.
Casper AF, JH Thorp. 2007. Diel and lateral patterns of
zooplankton distribution in the St. Lawrence River. River
Res. Appl. 23: 73-85.
Cearreta A, MJ Irabien, E Leorri, I Yusta, IW Croudace,
AB Cundy. 2000. Recent anthropogenic impacts on
the Bilbao Estuary, northern Spain: geochemical and
microfaunal evidence. Estuar. Coast. Shelf Sci. 50: 571-
592.
Chang WB, LC Tseng, HU Dahms. 2010. Abundance,
distribution and community structure of planktonic
copepods in surface waters of a semi-enclosed
embayment of Taiwan during monsoon transition. Zool.
Stud. 49: 735-748.
Chen QC. 1992. Zooplankton of China seas. Beijing: Science
Press, pp. 1-87.
Chen QC, SZ Zhang. 1965. The planktonic copepods of the
Yellow Sea and the East China Sea. I. Calanoida. Stud.
Mar. Sin. 7: 20-131, 53 plates. (in Chinese, with English
summary)
Chen QC, SZ Zhang, CS Zhu. 1974. On planktonic copepods
of the Yellow Sea and the East China Sea. II. Cyclopoida
and Harpacticoida. Stud. Mar. Sin. 9: 27-76, 24 plates. (in
Chinese, with English summary)
Chihara M, M Murano. 1997. An illustrated guide to marine
plankton in Japan. Tokyo: Tokyo Univ. Press, 1574 pp.
Chullasorn S, HU Dahms, KW Lee, JS Ki, N Schizas, P Kangtia
et al. 2011. Upscaled description for Tisbe alaskensis
sp. nov. (Crustacea, Copepoda) combining structural and
molecular traits. Zool. Stud. 50: 103-117.
Chullasorn S, WX Yang, HU Dahms, P Kangtia, M Holinska,
W Anansatitporn et al. 2009. Naupliar development of
Eucyclops cf. serrulatus tropicalis, Euc. cf. spatulatus, and
Ectocyclops medius Kiefer, 1930 (Copepoda: Cyclopidae).
Zool. Stud. 48: 12-32.
Clarke KR, RM Warwick. 1994. Change in Marine Com-
munities: An Approach to Statistical Analysis and
Interpretation. Primer User Manual. (Plymouth Routines
in Multivariate Ecological Research). Plymouth Marine
Laboratory, Bournemouth, UK. Bourne Press Limited.
Dahms HU, X Li, G Zhang, PY Qian. 2006. Resting stages of
Tortanus forcipatus (Crustacea, Calanoida) in sediments
of Victoria Harbor, Hong Kong. Estuar. Coast. Shelf Sci.
67: 562-568.
Dahms HU, PY Qian. 2004. Drift pump and drift net. J. Exp.
Mar. Biol. Ecol. 301: 29-37.
Dussart BH, D Defaye. 2006. World directory of Crustacea
Copepoda of inland waters. II. Cyclopiformes. Leiden,
the Netherlands: Backhuys Publishers, 354 pp.
Einsle E. 1996. Copepoda: Cyclopoida. Genera Cyclops,
Megacyclops, Acanthocyclops. In HJF Dumont, ed.
Guides to the identification of the microinvertebrates of
the continental waters of the world. Vol. 10. Leiden, the
Netherlands: Backhuys Publishers, 82 pp.
Froneman PW. 2004. Zooplankton community structure and
biomass in a South African temporarily open/closed
estuary. Estuar. Coast. Shelf Sci. 60: 125-132.
Ho CS. 1975. An introduction to the geology of Taiwan. Taipei,
Taiwan: Ministry of Economic Affairs, 153 pp.
Hsiao SH, TH Fang, C Shih, JS Hwang. 2011. Effects of the
Kuroshio Current on copepod assemblages in Taiwan.
Zool. Stud. 50: 475-490.
Hsiao SH, CY Lee, CT Shih, JS Hwang. 2004. Calanoid
copepods of the Kuroshio Current east of Taiwan, with
notes on the presence of Calanus jashnovi Hulseman,
1994. Zool. Stud. 43: 323-331.
Hsieh CH, TS Chiu. 1997. Copepod abundance and species
composition of Tanshui River estuary and adjacent waters.
Acta Zool. Taiwan. 8: 75-83.
Hsieh HY, WT Lo, LJ Wu, DC Liu, WC Su. 2011. Comparison
of distribution patterns of larval fish assemblages in the
Taiwan Strait between the northeasterly and southwesterly
monsoons. Zool. Stud. 50: 491-505.
Hsu SC, FJ Lin, WL Jeng, YC Chung, LM Shaw, KW Hung.
2004. Observed sediment fluxes in the southwesternmost
Okinawa Trough enhanced by episodic events: flood
runoff from Taiwan rivers and large earthquakes. Deep-
Sea Res. I 51: 979-997.
Hsu SC, FJ Lin, WL Jeng, TY Tang. 1998. The effect of a
cyclonic eddy on the distribution of lithogenic particles in
the southern East China Sea. J. Mar. Res. 56: 813-832.
Hwang JS, QC Chen, CK Wong. 1998. Taxonomic composition
and grazing rate of calanoid copepods in coastal waters
of northern Taiwan. Crustaceana 71: 378-389.
Hwang JS, QC Chen, CK Wong. 2003. Taxonomic com-
position, density and biomass of free-living copepods in
the coastal waters of southwestern Taiwan. Crustaceana
76: 193-206.
Hwang JS, HU Dahms, LC Tseng, QC Chen. 2007. Intrusions
of the Kuroshio Current in the northern South China Sea
affect copepod assemblages of the Luzon Strait. J. Exp.
Mar. Biol. Ecol. 352: 12-27.
Hwang JS, R Kumar, HU Dahms, LC Tseng, QC Chen. 2010.
Interannual, seasonal, and diurnal variation in vertical
and horizontal distribution patterns of 6 Oithona spp.
(Copepoda: Cyclopoida) in the South China Sea. Zool.
Stud. 49: 220-229.
Hwang JS, R Kumar, CW Hsieh, AY Kuo, S Souissi, MH Hsu et
al. 2010. Patterns of zooplankton distribution along the
marine, estuarine and riverine portions of the Danshuei
ecosystem in northern Taiwan. Zool. Stud. 49: 335-352.
Hwang JS, R Kumar, CS Kuo. 2009a. Impact of predation
by the copepod, Mesocyclops pehpeiensis on life table
demography and population dynamics of four cladoceran
species: a comparative laboratory study. Zool. Stud. 48:
738-752.
Hwang JS, S Souissi, HU Dahms, LC Tseng, FG Schmitt,
QC Chen. 2009b. Rank-abundance allocations as a
tool to analyze planktonic copepod assemblages off the
Danshuei River estuary (northern Taiwan). Zool. Stud.
48: 49-62.
Hwang JS, S Souissi, LC Tseng, L Seuront, FG Schmitt, LS
Fang et al. 2006. A 5-year study of the influence of
the northeast and southwest monsoons on copepod
assemblages in the boundary coastal waters between the
East China Sea and the Taiwan Strait. J. Plankt. Res. 28:
943-958.
Hwang JS, YY Tu, LC Tseng, LS Fang, S Souissi, TH Fang
et al. 2004. Taxonomic composition and seasonal
 Dahms et al. – Lanyang River Copepods
distribution of copepod assemblages from waters adjacent
to nuclear power plant I and II in northern Taiwan. J. Mar.
Sci. Technol. 12: 380-391.
Hwang JS, CH Wang, TY Chan, eds. 2000. Proceedings of
the International Symposium on Marine Biology in Taiwan-
Crustacean and Zooplankton Taxonomy, Ecology and
Living Resources. Natl. Taiwan Mus. Spec. Publ. Ser. 10:
1-200.
Hwang JS, CK Wong. 2005. The China Coastal Current as a
driving force for transporting Calanus sinicus (Copepoda:
Calanoida) from its population centers to waters off
Taiwan and Hong Kong during the winter NE monsoon
period. J. Plankt. Res. 27: 205-210.
Jan S, J Wang, CS Chern, SY Chao. 2002. Seasonal variation
of the circulation in the Taiwan Strait. J. Mar. Syst. 35:
249-268.
Jerling HL, TH Wooldridge. 1995. Plankton distribution and
abundance in the Sundays River estuary, South Africa,
with comments on potential feeding interactions. S. Afr. J.
Mar. Sci. 15: 169-184.
Kâ S, JS Hwang. 2011. Mesozooplankton distribution and
composition on the northeastern coast of Taiwan during
autumn: effects of the Kuroshio Current and hydrothermal
vents. Zool. Stud. 50: 155-163.
Kao SJ. 1996. Total organic carbon freight of a subtropical
mountainous river (Lanyang-Hsi) in Taiwan. Acad. Bull.
Chin. Naval Acad. 6: 19-70.
Kao SJ, KK Liu. 1996. Particulate organic carbon export from
a subtropical mountainous river (Lanyang-Hsi) in Taiwan.
Limnol. Oceanogr. 41: 1749-1757.
Kao SJ, KK Liu. 1997. Fluxes of DOC and fossil POC
from an Oceania small river (Lanyang-Hsi) in Taiwan.
Biogeochemistry 39: 255-269.
Kao SJ, KK Liu. 2002. Exacerbation of erosion induced by
human perturbation in a typical Oceania watershed:
insight from 45 years of hydrological records from
the Lanyang-Hsi River, northeastern Taiwan. Global
Biogeochem. Cycles 1: 1-7.
Lee HJ, SY Chao. 2003. A climatological description of
circulation in and around the East China Sea. Deep-Sea
Res. II 50: 1065-1084.
Liang WD, TY Tang, YJ Yang, MT Ko, WS Chuang. 2003.
Upper-ocean currents around Taiwan. Deep-Sea Res. II
50: 1085-1106.
Liu KK, ed. 1997. Kuroshio edge exchange processes
collected papers. Vol. 4. Taipei, Taiwan: Joint Global
Ocean Flux Study (China-Taipei) Committee.
Liu KK, TH Peng, PT Shaw, FK Shiah. 2003. Circulation and
biogeochemical processes in the East China Sea and
the vicinity of Taiwan: an overview and a brief synthesis.
Deep-Sea Res. II 50: 1055-1064.
Marr J, ed. 1970. Symposium on the Cooperative Study of
the Kuroshio and Adjacent Regions (1968: East-West
Center). Honolulu, HI: East-West Center Press.
Nittrouer CA, GJ Brunskill, AG Figueiredo. 1995. Importance
of tropical coastal environments. Geo-Mar. Lett. 15: 121-
126.
Parsons TR, Y Maita, CM Lalli. 1984. A manual of chemical
and biological methods for seawater analysis. Oxford,
UK: Pergamon Press, 1-173.
Pielou EC. 1966. The measurement of diversity in different
types of biological collections. J. Theor. Biol. 13: 131-144.
Richardson WB. 1992. Microcrustacea in flowing water:
experimental analysis of washout times and a field test.
173
Freshw. Biol. 28: 217-230.
Schminke HK. 2007. Entomology for the copepodologist. J.
Plankt. Res. 29: 149-162.
Shiah FK, SJ Kao, KK Liu. 1996. Spatial variability of
phytoplankton production and the implications of its
controlling mechanisms in the coastal zone near the river
mouth of Lanyang-Hsi. Bot. Bull. Acad. Sin. 37: 9-15.
Shih CT, SS Young. 1995. A checklist of free-living copepods,
including those associated with invertebrates, reported
from the adjacent seas of Taiwan. Acta Zool. Taiwan 6:
64-81.
Sluss TD, GA Cobbs, JH Thorp. 2008. Impact of turbulence on
riverine zooplankton: a mesocosm experiment. Freshw.
Biol. 53: 1999-2010.
Tan Y, L Huang, Q Chen, X Huang. 2004. Seasonal variation
in zooplankton composition and grazing impact on
phytoplankton standing stocks in the Pearl River Estuary,
China. Cont. Shelf Res. 24: 1949-1968.
Thor P, TG Nielson, P Tiselius, T Juul-Pederson, C Michel,
EF Møller et al. 2005. Post-spring bloom community
structure of pelagic copepods in the Disko Bay, western
Greenland. J. Plankt. Res. 27: 341-356.
Thorp JH, AR Black, KH Haag, JD Wehr. 1994. Zooplankton
assemblages in the Ohio River: seasonal, tributary, and
navigation dam effects. Can. J. Fish. Aquat. Sci. 51:
1634-1643.
Thorp JH, AF Casper. 2003. Importance of biotic interactions
in large rivers: an experiment with planktivorous fish,
dreissenid mussels and zooplankton in the St. Lawrence
River. River Res. Appl. 19: 265-279.
Thorp JH, S Mantovani. 2005. Zooplankton in turbid and
hydrologically dynamic, prairie rivers. Freshw. Biol. 50:
1474-1491.
Tseng LC, HU Dahms, QC Chen, JS Hwang. 2009. Copepod
feeding study in the upper layer of the tropical South
China Sea. Helg. Mar. Res. 63: 327-337.
Tseng LC, R Kumar, HU Dahms, QC Chen, JS Hwang. 2008a.
Copepod gut contents, ingestion rates and feeding impact
in relation to their size structure in the southeastern
Taiwan Strait. Zool. Stud. 47: 402-416.
Tseng LC, R Kumar, HU Dahms, CH Wu, QC Chen, JS Hwang.
2008b. Monsoon-driven seasonal succession of copepod
assemblages in the coastal waters of the northeastern
Taiwan Strait. Zool. Stud. 47: 46-60.
Vandromme P, FG Schmitt, S Souissi, EJ Buskey, JR Strickler,
CH Wu et al. 2010. Symbolic analysis of plankton
swimming trajectories: case study of Strobilidium sp.
(Protista) helical walking under various food conditions.
Zool. Stud. 49: 289-303.
Waniek JJ. 2003. The role of physical forcing in the initiation of
spring blooms in the Northeast Atlantic. J. Mar. Syst. 39:
57-82.
Waniek JJ, NP Holliday, R Davidson, L Brown, SA Henson.
2005. Freshwater control of onset and species
composition of Greenland shelf spring bloom. Mar. Ecol.
Progr. Ser. 288: 45-57.
Water Resources Bureau. 1997a. Hydrological year book of
Taiwan, Republic of China, 1995. Taipei, Taiwan: Ministry
of Economic Affairs, 399 pp.
Water Resources Bureau. 1997b. Hydrological year book of
Taiwan, Republic of China, 1995. Taipei, Taiwan: Ministry
of Economic Affairs, 400 pp.
Weaver W, CE Shannon. 1949. The mathematical theory of
communication. Urbana, IL: Univ. of Illinois.
174 Zoological Studies 51(2): 160-174 (2012)

Wetzel RG. 2001. Limnology: lake and river ecosystems, 3rd
ed. New York: Academic Press/Elsevier, 1006 pp.
Wu CH, HU Dahms, EJ Buskey, JR Strickler, JS Hwang. 2010.
Behavioral interactions of Temora turbinata with potential
ciliate prey. Zool. Stud. 49: 157-168.
Zhang G, S Sun, ZL Xu, QL Zhang. 2010. Unexpected
dominance of the subtropical copepod Temora turbinata
in the temperate Changjiang river estuary and its possible
causes. Zool. Stud. 49: 492-503.
Zoological Studies 51(2): 175-184 (2012)

Changes in Oak Gall Wasps Species Diversity (Hymenoptera:

Cynipidae) in Relation to the Presence of Oak Powdery Mildew (Erysiphe
alphitoides)
Mohammed Reza Zargaran1,*, Nadir Erbilgin2, and Youbert Ghosta1
1
Plant Protection Department- Sero Road- Agricultural Faculty, Urmia Univ., PO Box 165, Urmia, Iran
2
4-42 Earth Sciences Building, Department of Renewable Resources, Univ. of Alberta, Edmonton T6G 2E3, AB Canada

(Accepted September 29, 2011)

Mohammed Reza Zargaran, Nadir Erbilgin, and Youbert Ghosta (2012) Changes in oak gall wasps species
diversity (Hymenoptera: Cynipidae) in relation to the presence of oak powdery mildew (Erysiphe alphitoides).
Zoological Studies 51(2): 175-184. Plant-mediated interactions usually lead to multipartite interactions in a
community of organisms. To evaluate the impact of oak powdery mildew Erysiphe alphitoides infestation on
the distributions of cynipid oak gall wasps (Hymenoptera: Cynipidae), a field survey was conducted in West
Azerbaijan Province, Iran, in 2 consecutive years of 2009-2010. Multiple samples were taken from both infected
and uninfected trees (Quercus spp.) at 4 different sites where maximum activity of E. alphitoides occurred and
cynipid galls exhibited complete development. The species diversity and richness of gall-forming wasps were
estimated and also parameters such as Simpson’s index, Shannon’s H’, and the Sorensen similarity quotient
were calculated. Data were also analyzed by independent-samples t-test to compare the mean numbers of galls
occurring on infected and uninfected trees. Results clearly indicated that the highest richness and diversity of
oak gall-forming wasps were consistently found on uninfected trees at all study sites in the 2 consecutive years.
Further, the number and diversity of gall-forming wasps were negatively correlated with the extent (percentage)
of pathogen infection, and trees with the heaviest E. alphitoides infection had the lowest numbers of gall-forming
wasps. In addition, E. alphitoides decreased the rate of Sorensen’s coefficient between regions where oak trees
infected with E. alphitoides were sampled. This study demonstrates plant-mediated interactions between a
native pathogen and a community of gall-forming insects on oak trees.
http://zoolstud.sinica.edu.tw/Journals/51.2/175.pdf

Key words: Cynipid gall wasps, Tree-mediated interactions, Species diversity, Abundance, Oak forest.

P
lant-mediated interactions were commonly
reported in many studied systems (Karban and
Baldwin 1997, Nakamura et al. 2003, Foss and
Rieske 2004, Eyles et al. 2010, Staley et al. 2010).
Such interactions usually lead to multipartite
interactions in a community through indirect
interactions as 1 organism may change a host’s
suitability for others, and hosts become more or
less suitable for subsequent attackers (Karban
and Baldwin 1997). Most current knowledge of
plant-mediated interactions was obtained through
studies on herbaceous annuals or short-lived
perennials, but much less is known about trees,
either angiosperms or gymnosperms (Eyles et al.
2010).
In recent years, plant-mediated interactions
in forest ecosystems were documented (see
review by Eyles et al. 2010, Colgan and Erbilgin
2011), although the implications of those studies
are limited, in part, due to associations of a large
diversity of insects and pathogens with diff-
erent growth stages of trees (Eyles et al. 2010).

*To whom correspondence and reprint requests should be addressed. E-mail:Zargaran391@yahoo.com

175
176 Zargaran et al. – Powdery Mildew Reduced Diversity of Cynipid Gall Wasps
Further, it is difficult to compare different systems
because their relationships usually depend on the
interacting organisms, the intensity of damage,
and the time since induction (Herms and Mattson
1992, Eyles et al. 2007, Colgan and Erbilgin 2011).
Nevertheless, those studies clearly demonstrated
indirect interactions between species based on
the initial damage they caused to trees. In some
cases, 1 organism lowered the host suitability
to a subsequent organism (Kosaka et al. 2001,
Eyles et al. 2007), whereas other researchers
found increased host susceptibility to subsequent
attackers (Raffa et al. 1998, Wallin and Raffa
2001). For example, defoliation by the pine looper
(Bupalus piniaria L.) resulted in a strong decline
in the resistance of Scots pine to the blue-stain
fungus Leptographium wingfieldii (Långström et
al. 2001). Trees in the lowest-defoliation classes
were less susceptible to L. wingfieldii than those in
higher-defoliation classes.
An overwhelming majority of such studies
focused on interactions between a few organisms
at the same or different trophic levels, and roles of
plant-mediated interactions among a community
of organisms have seldom been documented,
although in nature, trees serve as foci for com-
munities of insects and diseases. This study
provides an example of plant-mediated interactions
in naturally occurring groups of organisms in
natural oak (Quercus spp.) forests.
We focused on interspecific interactions
between a native tree disease, oak powdery
mildew (Erysiphe alphitoides), and a community
of native oak gall-forming wasps (Hymenoptera:
Cynipidae). We were particularly interested in
whether prior infection of oaks by E. alphitoides
influenced the spatial abundance and richness of
oak gall-forming wasps on these oaks.
Large populations of many western Palearctic
species, including oaks, are commonly found in
Eastern Europe, Turkey, the Caucasus, and Iran
(Hewitt 1999). In the northern, southern and
western Iran, Q. pubescens Willd, Q. cerris L., Q.
infectoria Olivier, and Q. macranthera Fisch are
predominant; while junipers and oak forests such
as Q. infectoria, Q. brantii, and Q. pubescens are
predominant in the eastern region (Zargaran et al.
2008).
Oaks are reported to be primary hosts for
a larger number of plant pathogens and insect
herbivores (Stone et al. 2002). Among pathogens,
powdery mildew fungi infestations, including
species of Erysiphales are very common (Braun
1995). Detailed on world-wide distributions of
powdery mildew species were reported by Farr
and Rossman (2010). Erysiphe alphitoides is a
common fungal disease that appears on many oak
species (Griffon and Maublanc 1912, Mougou-
Hamdane et al. 2010).
Oaks are also commonly attacked by gall-
forming insects (Ronquist and Liljeblad 2001).
Cynipid wasps (Hymenoptera) are the 2nd most
diverse family after cecidomyiid midges, and the
majority of cynipid wasps are obligate parasites on
oaks (Stone et al. 2002). There are about 1300
species of cynipid oak gall-forming wasps globally,
with the majority occurring in the Nearctic (Cornell
1983, Ronquist and Liljeblad 2001, Stone et al.
2002).
Several studies documented the abundance
and richness of gall wasps with respect to the
richness and abundance of host plants (Chodjai
1980, Starzomski et al. 2008, Zargaran et al.
2008), plant quality (Genimar-Reboucas et al.
2003, Egan and Ott 2007), other herbivorous
insects and natural enemies (Veldtman and
Mcgeoch 2003, Cuevas-Reyes et al. 2004, Prior
and Hellmann 2010), and abiotic factors, such
as water stress (Stone et al. 2002). Few studies
focused on community-level interactions in gall-
forming insects. For example, Nakamura et al.
(2003) demonstrated that gall-formers had a
positive plant-mediated effect on other insect
herbivores and reported that the stem gall
midge Rabdophaga rigidae, and adults of 2 leaf
beetles, Plagiodera versicolora and Smaragdina
semiaurantiaca, on Salix eriocarpa were more
abundant on lateral shoots and leaves of galled
shoots than on ungalled shoots, respectively.
However, roles of other organisms, particularly
diseases, on cynipids are largely unstudied (Foss
and Rieske 2004). Further, how gall-forming
species are locally distributed and what biological
factors affect their local distributions (Veldtman and
McGeoch 2003) are largely unknown, given that
the suitability of oviposition sites has the potential
to generate indirect interspecific competition
between gall-forming insects and other species
(Stone et al. 2002).
In this study, we attempted to determine
whether prior E. alphitoides infestation of oak trees
affected the community of oak gall-forming wasps.
Specifically, we addressed whether gall-forming
wasp abundance and diversity were affected by E.
alphitoides infestations.
 Zoological Studies 51(2): 175-184 (2012) 177

MATERIALS AND METHODS optimal number of trees was determined to be 20

per site. Twenty trees infected with E. alphitoides
Study sites and 20 trees without infection were surveyed at
each site and in each of 2 consecutive years. All
Sampling was performed in West Azerbaijan cynipid galls were counted on 4 randomly selected
Province, Iran in 2009-2010 (Table 1). branches per tree. Galls found on plant surfaces
(branches and leaves) were identified based on
Studied species their morphology.

Chodjai (1980) reported 36 oak gall wasp Statistical analysis

species associated with the oak Q. infectoria
from Iran. Recent surveys were conducted on The Shannon-Weiner diversity index uses the
the cynipid fauna of Iran (Tavakoli et al. 2008, following formula:
Zargaran et al. 2008) and according to the latest
results, so far 82 species of oak gall wasps No

were recorded in oak forests of Iran, of which 25 Shannon’s H’ = - Σ [p *log p ]

i=1
i i

species were reported for the 1st time (Sadeghi

et al. 2010). Those surveys confirmed that the
cynipid fauna of Iran includes widespread western
Palaearctic species such as Andricus kollari
(Hartig) and Cynips quercusfolii (Hartig) (Chodjai
1980, Zargaran et al. 2008). Oak powdery mildew
Erysiphe alphitoiides L. was reported on Q.
infectoria Oliv. for the 1st time in Iran (Tavanaei
2006).
Sampling methods
At each site, oak cynipid galls were
collected from Q. infectoria in late Sept. when the
maximum activity of E. alphitoides occurred and
the development of cynipid galls was completed.
The optimal number of trees (sample unit) per
site was determined according to Southwood
and Henderson’s (2000) formula of N = [(t × s) /
(D × m)]2, where t is Student’s t-test from standard
statistical tables, D is the predetermined
confidence limit for estimation of the mean as
a decimal, m is the sampling mean and s is the
standard deviation. Based on this analysis, the
where pi is the proportion of the total number of
individuals belonging to a morphotype, and N o
is the total number of morphotypes seen in that
sample. Simpson’s diversity index is calculated
using the following formula:
N
ni (ni - 1)
Simpson’s D = 1 - Σ N(N - 1)
i=1
where ni is the number of individuals of a particular
morphotype and N is the total number seen in the
sample (Magurran 2004).
Diversity indices like the Shannon’s entropy
(“Shannon-Wiener index”) and the Gini-Simpson
index are not in themselves diversities. The
number of equally-common species required to
impact a particular value to an index is called the
“effective number of species”. This is the true
diversity of the community. Converting indices
to true diversities gives them a set of common
behaviors and properties. After conversion,
diversity is always measured in units of the number

Table 1. Characteristics at 4 study sites selected to investigate the effect of powdery mildew infestation on
oak gall wasp species diversity and richness on oaks in West Azerbaijan Province, Iran in 2009-2010

Site

Characteristic Ghabre-hossein Mirabad Rabat Dare-ghabr

Quercus species Q. infectori Q. infectoria Q. infectoria Q. infectori

Q. brantii Q. brantii Q. brantii Q. brantii
Q. libani Q. libani
Latitude 36°28'N 36°15'N 36°14'N 36°11'N
Longitude 45°18'W 45°22'W 45°33'W 45°24'W
Weather Very humid and cold Very humid and cold Humid, mildly cold Humid, mildly cold
178 Zargaran et al. – Powdery Mildew Reduced Diversity of Cynipid Gall Wasps
of species (Jost 2006). Conversion of common
indices to true diversities can be achieved as
described in table 2.
Evenness, the other information-statistical
index, is affected by both the number of species
and their equitability or evenness compared to a
community’s actual diversity, and the value of E
is constrained to 0-1.0. Shannon’s evenness is
calculated by the formula: H’/ Hmax.
Beta diversity is generally thought of as the
change in diversity among various alpha diversities
(variation in species composition among sites in a
geographic region) (Koleff et al. 2003, Magurran
2004). The classical Sorensen index is based on
both the number of species present in the total
sample and numbers only seen in each individual
sample (Koleff et al. 2003). Sorenson’s measure
is regarded as one of the most effective presence/
absence similarity measures. The Sorensen
similarity index is calculated by Cs = 2a/(2a + b + c),
where a is the number of species common to both
sites, b is the number of species at site B but not at
A, and c is the number of species at site A but not
in B (Magurran 2004). It is used when research
is conducted on more than 1 site and begins with
a table or matrix giving the similarity between
each pair of sites (using any similarity coefficient).
The 2 most similar sites are combined to form
a single cluster. The analysis then proceeds by
successively combining similar sites until all are
combined into a single cluster (a dendrogram).
Cluster analysis measured using the hierarchical
cluster and cluster method are based on Ward’s
method. Sorensen’s similarity index value was
used in a cluster analysis to illustrate similarity
patterns at the 4 sites. Also, data were analyzed
with an independent-samples t-test to compare
mean numbers of galls occurring on infected and
uninfected trees. The surface of the infected
leaves was measured by a leaf area meter and
Pearson’s correlation coefficient was used to test
the relationship between percent leaf infection and
number of leaf galls.
Table 2. Conversion of common indices to true diversities
Index x
N N
Shannon entropy x ≡ - Σ p ln p
i = 1
i i
S S
Gini-Simpson index x ≡ 1- Σp
i = 1
2
i 1/(1-x)
RESULTS
At 4 sites, 25 species of oak gall wasps
(asexual generation) were collected and identified
as the following species groups: Andricus (20
species), Cynips (3 species), and Neuroterus (2
species) (Table 3). Overall, stem gall wasps were
more abundant (20) than leaf gall wasps (5). All
stem gall wasps belonged to a single genus,
Andricus. Leaf-causing gall wasps were members
of Cynips and Neuroterus. The Andricus species
group had the highest abundance among species
groups collected from oaks.
Distributions of oak gall wasps among sites
differed (Table 3). Ghabre-hossein had the highest
number of species among sites, with 21 species
in 2009 and 17 species in 2010. Mirabad had the
lowest species abundance, with 6 in 2009 and
5 in 2010. Naturally some species overlapped
between sites. There was a slight decline in
species abundances from 2009 to 2010.
The highest and lowest number of Andricus
species were observed at Ghabre-hossein (16)
and Mirabad (4), respectively, in 2009 (Table 3),
whereas in 2010, the highest number of Andricus
species was found at Dare-ghabr (13) and the
lowest number was collected at Mirabad (3) (Table
3). All species belonging to the genera Cynips and
Neuroterus were only found at Ghabre-hossein.
Table 3 also shows the distributions of
species between infected and uninfected oak
trees at 4 sites. Two Cynips species (C. quercus
and C. quercusfolii) and 2 Neuroteus species
(N. numismalis and N. quercus-baccarum) were
commonly found on uninfected trees, occasionally
found on slightly infected trees, and virtually
absent from highly infected trees. Other species,
exclusively stem galls, were found on both infe-cted
and uninfected trees; however, they were more
commonly found on infected trees. It was inter-
esting to note that between C. quercus and C.
quercusfolii, the latter was overall more abundant.
Likewise between N. numismalis and N. quercus-
Diversity in terms of x Diversity in terms of pi
exp (x) exp (- Σ p ln p )
i = 1
i i
1/ Σp
i = 1
2
i
 Zoological Studies 51(2): 175-184 (2012) 179

baccarum, the later was more abundant. Several
stem gall wasps and 1 leaf gall wasp, C. cornifex,
were found equally on infected and uninfected
trees. None of the leaf gall wasps were found
together on the same leaf in the current study.
Overall, oak trees with oak powdery mildew
infection had reduced richness and diversity of
oak gall wasps (Table 4). The highest species
richness was found on uninfected trees in all study
sites in the 2 consecutive years. Gini-Simpson
indices were 1.75 in 2009 and 1.71 in 2010 for
infected and 2.33 in 2009 and 2.13 in 2010 for
uninfected trees. Gini-Simpson indices were lower
in 2010 compared to values in 2009. Among
sites, uninfected trees at Ghabre-hossein had
the highest Gini-Simpson indices in both years,
followed by Dare-ghabr and Rabat. Mirabad
had the lowest Gini-Simpson indices. Likewise,
Shannon’s H index indicated that uninfected trees
had the highest oak gall wasp diversity compared
to infected trees at all sites (Table 4). Differences
among sites were similar to the Gini-Simpson
index, with Ghabre-hossein having the highest
Shannon’s indices, followed by Dare-ghabr and

Table 3. Gall wasps species associated with infected and uninfected oak trees in 2009 and 2010. We
collected 25 gall wasps species in this research from 4 sites. The presence and absence of any gall wasps
are shown by (+) and (-), respectively. The 1st sign indicates that the specimen was either present on (+) or
absent from (-) uninfected trees, while the 2nd sign indicates that the specimen was either present on (+) or
absent from (-) infected trees. For example, Andricus aries was present on uninfected trees (+), but absent
from infected trees at Ghabre-hossein in 2009

Gall wasp species Site

Ghabre-hossein Mirabad Rabat Dare-ghabr

2009 2010 2009 2010 2009 2010 2009 2010

Stem gall
Andricus aries2 + (-) + (-) - (-) - (-) - (-) - (-) + (+) + (+)
A. askewi 2 + (+) + (+) - (-) - (-) - (-) - (-) + (-) + (-)
A. caputmedusae2 + (+) + (+) - (-) - (-) + (+) + (+) - (-) - (-)
A. conglomerates2 + (-) + (-) - (-) - (-) - (-) - (-) + (-) + (+)
A. coriarius2 - (-) - (-) - (-) - (-) - (-) - (-) + (+) + (+)
A. galeatus2 + (+) + (-) - (-) - (-) - (-) - (-) + (-) + (-)
A. hystrix2 + (+) + (+) - (-) - (-) - (-) - (-) - (-) - (-)
A. kollari 2 + (+) + (+) - (-) - (-) - (-) - (-) + (+) + (-)
A. lucidus2 - (-) - (-) + (+) + (+) - (-) - (-) + (+) + (+)
A. mediterraneae2 - (-) - (-) - (-) - (-) - (-) - (-) + (+) + (+)
A. megalucidus2 + (+) - (-) + (+) + (+) - (-) - (-) + (+) + (+)
A. panteli 2 + (+) + (+) - (-) - (-) - (-) - (-) + (+) + (-)
A. polycerus2 + (+) + (+) - (-) - (-) - (-) - (-) + (+) - (-)
A. quercuscalicis2 + (+) + (+) - (-) - (-) - (-) - (-) + (+) + (+)
A. quercustozae2 + (+) - (-) + (+) - (-) + (+) + (+) + (-) - (-)
A. seckendorffi 2 + (+) + (+) - (-) - (-) - (-) - (-) - (-) - (-)
A. sternlichtii 2 + (-) + (-) - (-) - (-) - (-) - (-) + (+) + (+)
A. theophrastea2 + (+) - (+) - (-) - (-) + (+) - (+) - (-) - (-)
A. tomentosus2 - (-) - (-) - (-) - (-) + (+) + (+) + (+) + (+)
A. megatruncicolus2 + (+) - (-) + (+) + (+) + (+) + (+) - (-) - (-)
Leaf gall
Cynipis cornifex2 + (+) + (+) - (-) - (-) - (-) - (-) - (-) - (-)
C. quercus1 + (-) + (-) + (-) + (+) + (-) + (-) - (-) - (-)
C. quercusfolii 1 + (-) + (-) + (-) + (-) + (-) + (-) + (-) + (-)
Neuroterus numismalis1 + (-) + (-) - (-) - (-) + (+) + (-) + (-) + (-)
N. quercus-baccarum1 + (-) + (-) - (-) - (-) + (-) + (-) + (-) + (-)
1
Species commonly found on uninfected trees, rarely found on lightly infected trees, and virtually absent from highly infected trees.
2
Species commonly found on both infected and uninfected oak trees.
180 Zargaran et al. – Powdery Mildew Reduced Diversity of Cynipid Gall Wasps

Rabat. Mirabad had the lowest Shannon’s indices.
An increase in either the Gini-Simpson index or
Shannon’s H index reduced the evenness of gall
wasps (Table 4). The highest and lowest species
evenness values were found at Mirabad and
Ghabre-hossein, respectively.
Cluster analysis dendrogram are shown
in figures 1 and 2. Dendrograms cluster sites
according to how strongly correlated the sites are,
and if sites are highly correlated, they will have a
correlation value of 1 or close to 1. In the current
study, the highest value of the Sorensen similarity
between Dare-ghabr and Ghabre-hossein was 0.72
for uninfected trees in 2009, while the similarity
between these 2 sites measured from infected
trees was 0.41. The lowest index of similarity
was recorded between Rabat and Dare-ghabr
on infected trees in 2009. In 2010, Sorensen
similarity indices of infected and uninfected trees
at Ghabre-hossein and Rabat were 0.27 and 0.40
respectively. The mean number of oak galls was
generally higher on uninfected trees than infected
trees at all sites in 2 the consecutive years (Table
5). Trees infected with powdery mildews showed
the lowest mean number of cynipid galls. Among
uninfected trees, the mean number of galls on
uninfected trees ranged from14.8 at Rabat in
2010 to 61.2 at Ghabre-hossein in 2009, while the
same means for infected trees ranged from 6.2
at Rabat to 25.4 at Ghabre-hossein in 2009. The
maximum and minimum uninfected: infected ratios
were 2.8 and 2.1 in Dare-ghabr in 2009 and 2010,
respectively.
Pearson’s correlation coefficients between
the number of galls and percent infection showed
significant negative correlations in 2009 (r = -0.714,
n = 20, p < 0.01) and 2010 (r = -0.581, n = 20,
p < 0.01) (Fig. 3). On infected oak trees, leaf gall
abundances declined with increasing levels of
powdery mildew.

Table 4. Species diversity indices and the true diversity of oak gall wasps in 2009 and 2010

Diversity indices

Species
Sites Simpson D Gini-Simpson True diversity Shannon’s H’ True diversity Evenness
richness

Infected oak trees (2009)

Ghabre-hossein 0.9012 11.95 2.48 13.21 0.91 14 0.08
Dare-ghabr 0.9104 7.38 1.99 9.95 0.89 11 0.10
Rabat 0.9188 3.91 1.36 5.33 0.81 6 0.19
Mirabad 0.9342 3.21 1.16 3.49 0.71 4 0.29
All sites pooled 6.61 1.75 8.00 0.84 8.7 0.16
Uninfected oak trees (2009)
Ghabre-hossein 0.8958 18.36 2.91 19.65 0.95 21 0.05
Dare-ghabr 0.9197 16.96 2.83 17.21 0.94 18 0.06
Rabat 0.9176 7.32 1.99 7.79 0.87 9 0.13
Mirabad 0.9419 4.86 1.58 5.27 0.81 6 0.19
All sites pooled 11.875 2.33 12.48 0.89 13.5 0.11
Infected oak trees (2010)
Ghabre-hossein 0.8311 8.86 2.18 8.12 0.89 10 0.11
Dare-ghabr 0.8422 8.29 2.12 8.42 0.88 9 0.12
Rabat 0.9140 3.98 1.38 4.39 0.77 5 0.23
Mirabad 0.9205 3.25 1.18 3.29 0.69 4 0.30
All sites pooled 6.095 1.71 6.06 0.81 7 0.19
Uninfected oak trees (2010)
Ghabre-hossein 0.8957 12.72 2.54 16.27 0.94 17 0.06
Dare-ghabr 0.9114 11.46 2.44 15.48 0.94 16 0.06
Rabat 0.9272 8.79 2.17 7.04 0.85 8 0.14
Mirabad 0.9380 3.86 1.35 4.31 0.77 5 0.24
All sites pooled 9.21 2.13 10.78 0.87 11.5 0.13
 Zoological Studies 51(2): 175-184 (2012) 181

100
Rescaled Distance Chister Combine (A)
80 60 40 20 0
DISCUSSION
Mirabad & Rabat
Ghabre-hossein & Dare-ghabr Our results clearly demonstrated an indirect
Rabat & Dare-ghabr
Mirabad & Dare-ghabr plant-mediated interaction between Erysiphe
Ghabre-hossein & Mirabad
Ghabre-hossein & Rabat
alphitoides and a community of cynipid oak gall-
Rescaled Distance Chister Combine (B) forming wasps, and we found that pathogen
100 80 60 40 20 0 infection significantly reduced the abundance and
Ghabre-hossein & Mirabad
Rabat & Dare-ghabr
species richness of the native oak gall wasps.
Mirabad & Dare-ghabr Although there were differences among sites, the
Ghabre-hossein & Rabat
Mirabad & Rabat highest and lowest abundance and richness values
Ghabre-hossein & Dare-ghabr
were always respectively associated with healthy
Fig. 1. Sorensen cluster analysis dendrogram of similarity and diseased oak trees, at any given site.
coefficients for oak gall wasps occurring on infected (A, above) This is the 1st study to demonstrate plant-
and uninfected (B, below) trees in 2009. mediated interactions between a leaf pathogen
and a community of gall-forming wasps. It was
commonly reported that pathogen or insect attacks
Rescaled Distance Chister Combine (A)
100 80 60 40 20 0 can affect the composition of insect and pathogen
Ghabre-hossein & Dare-ghabr communities associated with plants and mediate
Mirabad & Rabat
Ghabre-hossein & Rabat
the incidences and abundances of subsequent
Ghabre-hossein & Mirabad
Mirabad & Dare-ghabr
attackers (Stout et al. 2006, Eyles et al. 2010).
Rabat & Dare-ghabr In the current study, the mechanism of the plant-
Rescaled Distance Chister Combine (B) mediated interaction between E. alphitoides and
100 80 60 40 20 0
gall-forming wasps is not known although, based
Mirabad & Dare-ghabr
Rabat & Dare-ghabr
on earlier publications on cynipid gall wasps, we
Ghabre-hossein & Mirabad
Ghabre-hossein & Rabat
Mirabad & Rabat
Ghabre-hossein & Dare-ghabr Table 5. t-test comparison of the mean number
Fig. 2. Sorensen cluster analysis dendrogram of similarity
(± S.E.) of oak galls per trees between infected
coefficients for oak gall wasps occurring on infected (A, above) and uninfected oak trees in 2009 and 2010. A sig-
and uninfected (B, below) trees in 2010. nificant difference was accepted at < 0.05

Site Year Mean number (± S.E.) of

20.00 year oak galls per tree
2009
Uninfected Infected
2010

Ghabre-hossein 2009 61.2 (±5.3) 25.4 (±6.1)

15.00 2010 32.4 (±7.1) 17.3 (±3.2)
Number of leaf gall wasps

Dare-ghabr 2009 54.2 (±8.2) 19.4 (±11.1)

2010 38.4 (±1.8) 18.3 (±4.3)
Mirabad 2009 22.5 (±9.4) 10.4 (±5.8)
10.00 2010 16.9 (±2.7) 7.1 (±4.6)
Rabat 2009 20.3 (±1.4) 8.7 (±3.3)
2010 14.8 (±6.3) 6.2 (±2.4)

5.00 Site Year Statistics

t-value p-value

0.00 Ghabre-hossein 2009 20.16 0.002

2010 12.23 0.036
0.00 10.00 20.00 30.00 40.00 50.00 60.00 70.00 Dare-ghabr 2009 19.05 0.001
Percent of leaf infection 2010 15.64 0.026
Mirabad 2009 14.52 0.031
Fig. 3. Correlation between the number of leaf gall wasps and
2010 9.37 0.048
percent of leaf infection by oak powdery mildew on oaks in
Rabat 2009 7.78 0.043
2009 and 2010. An increased percent of disease infection led
2010 8.44 0.029
to a decrease in leaf oak gall wasp numbers.
182 Zargaran et al. – Powdery Mildew Reduced Diversity of Cynipid Gall Wasps
suspect that E. alphitoides can influence gall-
formers in 2 possible ways. First, E. alphitoides-
infection of oak trees most likely systemically
changes the host plant suitability, particularly host
nutrients and host secondary compounds, as
chemical interactions between gall-formers and
their host plants are important for both the success
and avoidance of gall formation. For example,
nutrients of plant tissues play critical roles in the
selection of oviposition sites and subsequent gall
development (Hartley 1998, Stone and Schönrogge
2003). Female cynipid gall wasps prefer host
tissues with high nutritional quality (Stone et al.
2002), and it is likely that an E. alphitoides infection
may diminish nutritional substances in oak tissues
(Stone and Schönrogge 2003). Decreasing
nitrogen or increasing carbon, due to increased
metabolism of carbon-based metabolites such as
tannins and lignin (Scriber and Slansky 1981, Wold
and Marquis 1997) or increases in photosynthesis
(Bagatto et al. 1996) may alter carbon: nitrogen
(C:N) ratio of plant tissues. Scriber and Slansky
(1981) suggested that tissues with high C: N ratios
provide low-quality food for developing immature
wasps inside galls. Additional investigations in this
system should focus on changes in plant nutritional
quality due to E. alphitoides infestations to fully
understand interactions between the disease and
cynipid oak gall wasps.
Further, our study cannot rule out the role
of secondary metabolites, particularly tannin
levels, in reducing the abundance and richness of
cynipid gall wasps in diseased oak trees. Tannin
is a phenolic compound used for defense against
a variety of organisms and is also induced by
pathogen infestation (Stone et al. 2002). Tannins
in cynipid galls are known to be concentrated in
the outer layers, where they may protect the gall
from endophytic fungi (Taper et al. 1986, Taper
and Case 1987, Wilson and Carroll 1997). For
example, the endophytic fungus Discula quercina
(Coelomycetes) was shown to cause almost 100%
cynipid gall wasp mortality in artificial infection
experiments (Wilson and Carroll 1997). Tannins
may also protect gall-formers against parasitoids
(Cornell 1983, Taper and Case 1987). This close
association of cynipid gall wasp and tannin levels
could explain the observed positive relationships
of oak tannin levels with cynipid diversity and
abundance (Taper and Case 1987, Wold and
Marquis 1997, Stone et al. 2002).
Although we do not know how different
severities of Erysiphe alphitoides infestation
affect tannin contents, we suspect that pathogen
infection either increases tannin contents in all
tissues such that high tannin contents in the
inner layers of galls might not be suitable for the
developing larvae, or significantly reduces tannin
contents such that developing larvae are not
protected from endophytic fungi, or a combination
of both, as tannin content are very likely to vary
with the severity of pathogen infection (Bonello et
al. 2006).
A 2nd possible alternative to explain the plant-
mediated interaction between E. alphitoides and
cynipid gall wasps is that the presence of a fungus
may prohibit oviposition by female wasps, as we
observed that hyphae of E. alphitoides covered
the surface of host leaves such that females could
not lay eggs. An E. alphitoides infestation on
leaves initially appears as light-green to yellow
spots. As the disease severity progresses, spidery
or threadlike white patches typically develop
with scattered small, black fruiting bodies. The
presence of an infestation of plant tissues by
E. alphitoides could also indirectly increase
competition for suitable oviposition sites among
leaf gall wasps (Gilbert et al. 1994). However,
this avoidance mechanism might only explain the
reduction in leaf gall wasp diversity and species
richness, not stem galls, as E. alphitoides is only
present on oak leaves.
Plant-mediated interactions between patho-
gens and herbivorous insects were commonly
reported in other systems (Krause and Raffa
1992, Felton and Korth 2000, Eyles et al. 2007).
For example, Krause and Raffa (1992) found that
infection of larch Larix decidua with the fungal
pathogen Mycosphaerella laricina induced a
systemic reduction in host quality for the larch
sawfly Pristiphora erichsonii. Likewise, Eyles et
al. (2007) reported that an infection by Diplodia
pinea elicited resistance against the defoliating
European pine sawfly Neodiprion sertifer in
Austrian pine Pinus nigra. In our system, plant
defensive responses apparently seem to be
operating against only the gall wasp community
and not E. alphitoides because we observed
continuous colonization of the same oak trees
by E. alphitoides after an initial infection. This
suggests a possible “cross-talk” between defensive
pathways against E. alphitoides (associated with
salicylic acid) and cynipid gall wasps (associated
with jasmonic acid) in oaks (Bostock 2005, Heil
and Ton 2008). Even though we do not have a
complete understanding of the defensive pathways
induced in oaks by either oak gall wasps or E.
alphitoides, further studies of oak systems should
 Zoological Studies 51(2): 175-184 (2012)
identify these pathways and determine whether
induction of 1 pathway prevents synthesis of the
other, thereby leading to an impaired capacity of
a plant to respond to either pathogen infection or
insect damage (Bostock 2005).
We currently do not know why the abundance
and species diversity of gall wasps were higher
at the Ghabre-hossein and Dare-ghabr sites
compared to the others. Despite differences in
climate, both sites have similar vegetative cover
and similar species abundance and richness levels
of cynipid gall wasps. Despite their similarities
in climate, Ghabre-hossein and Mirabad had
different vegetative cover, and the former had a
larger species complex. This suggests that the
distribution of host plant species may be highly
critical for determining patterns of herbivore
abundances (Starzomski et al. 2008) along with
factors like climate and phenological synchrony of
herbivores with host plants. This is not surprising
considering the fact that the abundance and
richness of gall wasps are related to the richness
and abundance of host plant species (Starzomski
et al. 2008). Likewise, the species richness of oak
gall wasps in Mexico was highly correlated to their
host plants (Cuevas-Reyes et al. 2004). Further,
Stone et al. (2002) suggested that geographical
differences in the oak gall wasp fauna were related
to oak distribution patterns in different regions. The
current study also added further complexity to host
plant-gall wasp interactions; i.e., the role of plant
pathogens in the spatial distribution of herbivorous
insects. Specifically, as indicated by the cluster
analysis dendrogram, although the Sorensen
similarity value for uninfected trees was fairly high
between Dare-ghabr and Ghabre-hossein (0.72),
E. alphitoides infestations dramatically reduced the
similarity index between these 2 sites to 0.41.
Acknowledgments: The authors are grateful to
the head of the Agricultural and Natural Resources
Research Center of West Azerbaijan (Dr. R.
Sokouti Oskouii) for his scientific and financial
support.
REFERENCES
Bagatto G, LC Paquette, D Shorthouse. 1996. Influence of
galls of Phanacis taraxaci on carbon partitioning within
common dandelion, Taraxacum sp. Entomol. Exp. Appl.
79: 111-117.
Bonello P, TR Gordon, DA Herms, DL Wood, N Erbilgin. 2006.
Nature and ecological implications of pathogen-induced
systemic resistance in conifers: a novel hypothesis.
183
Physiol. Mol. Plant. Pathol. 68: 95-104.
Bostock RM. 2005. Signal crosstalk and induced resistance:
straddling the line between cost and benefit. Annu. Rev.
Phytopathol. 43: 545-580.
Braun U. 1995. The powdery mildews (Erysiphales) of Europe.
New York: Gustav Fischer Verlag Press.
Chodjai M. 1980. Study of the gall forming cynipid
Hymenoptera of Iranian oak forests. J. Entomol. Soc.
Iran. 3: 1-44.
Colgan LJ, N Erbilgin. 2011. Tree-mediated interactions
between the jack pine budworm and a mountain pine
beetle fungal associate. Ecol. Entomol. (in press)
Cornell H. 1983. The secondary chemistry and com-
plex morphology of galls formed by the Cynipinae
(Hymenoptera): why and how? Am. Midl. Nat. 110: 225-
236.
Cuevas-Reyes P, M Quesada, P Hanson, R Dirzo, K Oyama.
2004. Diversity of gall-inducing insects in a Mexican
tropical dry forest: the importance of plant species
richness, life-forms, host plant age and plant density.
Ecology 92: 707-716.
Egan SP, JR Ott. 2007. Host plant quality and local adaptation
determine the distribution of a gall-forming herbivore.
Ecology 88: 2868-2879.
Eyles A, P Bonello, R Ganley, C Mohammed. 2010. Induced
resistance to pests and pathogens in trees. New Phytol.
185: 893-908.
Eyles A, R Chorbadjian, C Wallis, R Hansen, D Cipollini, D
Herms et al. 2007. Cross-induction of systemic induced
resistance between an insect and a fungal pathogen in
Austrian pine over a fertility gradient. Oecologia 153:
365-374.
Farr DF, AY Rossman. 2010. Fungal databases. Available at
http://nt.ars-grin.gov/ fungal databases Accessed 10 Apr.
2011.
Felton GW, KL Korth. 2000. Trade-offs between pathogen and
herbivore resistance. Curr. Opin. Plant. Biol. 3: 309-314.
Foss LK, LK Rieske. 2004. Stem galls affect oak foliage with
potential consequences for herbivory. Ecol. Entomol. 29:
273-280.
Genimar-Reboucas J, V Eduardo-Martins, F Wilson. 2003.
Richness and abundance of gall-forming insects in the
Mamiraua Varzea, a flooded Amazonian forest. Uakari 1:
39-42.
Gilbert F, C Astbury, J Bedingfield, B Ennis, S Lawson, T Sitch.
1994. The ecology of the pea galls of Cynips divisa. In
MAJ Williams, ed. Plants galls: organisms, interactions,
populations. Oxford, UK: Clarendon, pp. 331-349.
Griffon E, A Maublanc. 1912. Les Microsphaera des chênes.
Bull. Soc. Mycol. 28: 88-104.
Hartley SE. 1998. The chemical composition of plant galls: Are
levels of nutrients and secondary compounds controlled
by the gall-former? Oecologia 113: 492-501.
Heil M, J Ton. 2008. Long-distance signalling in plant defence.
Plant. Sci. 13: 264-272.
Herms DA, MJ Mattson. 1992. The dilemma of plants: grow or
defend. Rev. Biol. 67: 283-335.
Hewitt GM. 1999. Post-glacial re-colonization of European
biota. J. Linn. Soc. 68: 87-112.
Jost L. 2006. Entropy and diversity. Oikos 113: 363-375.
Karban R, IT Baldwin. 1997. Induced responses to herbivory.
Chicago, IL: Univ. of Chicago Press.
Koleff P, K Gaston, J Lennon. 2003. Measuring beta diversity
for presence/absence data. J. Anim. Ecol. 72: 367-382.
184 Zargaran et al. – Powdery Mildew Reduced Diversity of Cynipid Gall Wasps
Kosaka H, T Aikawa, N Ogura, K Tabata, T Kiyohara. 2001.
Pine wilt disease caused by the pine wood nematode: the
induced resistance of pine trees by a virulent isolates of
nematode. Eur. J. Plant. Pathol. 107: 667-675.
Krause SC, KF Raffa. 1992. Comparison of insect, fungal, and
mechanically induced defoliation of larch-effects on plant
productivity, host susceptibility. Oecologia 90: 411-416.
Långström B, H Solheim, C Hellqvist, P Krokene. 2001. Host
resistance in defoliated pine: effects of single and mass
inoculations using bark beetle-associated blue-stain fungi.
Agric. For. Entomol. 3: 211-216.
Magurran AE. 2004. Measuring biological diversity. Oxford,
UK: Blackwell Publishing Press.
Mougou-Hamdane A, X Giresse, C Dutech, ML Desprez-
Loustau. 2010. Spatial distribution of lineages of oak
powdery mildew fungi in France, using quick molecular
detection methods. Ann. For. Sci. 67: 449-457.
Nakamura N, Y Miyamoto, T Ohgushi. 2003. Gall initiation
enhances the availability of food resources for herbivorous
insects. Funct. Ecol. 17: 851-857.
Prior KM, JJ Hellmann. 2010. Impact of an invasive oak
gall wasp on a native butterfly: a test of plant-mediated
competition. Ecology 91: 3284-3293.
Raffa KF, SC Krause, PB Reich. 1998. Long-term effects of
defoliation on red pine suitability to insects feeding on
diverse plant tissues. Ecology 79: 2353-2364.
Ronquist F, J Liljeblad. 2001. Evolution of the gall wasp-host
plant association. Evolution 55: 2503-2522.
Sadeghi SE, G Melika, GN Stone, M Assareh, H Askary, M
Tavakoli et al. 2010. Oak gall wasps fauna of Iran
(Hymenoptera: Cynipidae: Cynipini). 7th International
Congress of Hymenoptroists 112, Hungary: The inter-
national society of Hymenopterists.
Scriber JM, F Slansky. 1981. The nutritional ecology of
immature insects. Annu. Rev. Entomol. 26: 183-211.
Southwood TR, PA Henderson. 2000. Ecological methods.
Oxford, UK: Blackwell.
Staley JT, A Stewart-Jones, TW Pope, DJ Wright, SR Leather,
P Hadley et al. 2010. Varying responses of insect
herbivores to altered plant chemistry under organic and
conventional treatments. Proc. R. Soc. Lond. Biol. 277:
779-786.
Starzomski BM, RL Parker, DS Srivastava. 2008. On the
relationship between regional and local species richness:
a test of saturation theory. J. Ecol. 89: 1921-1930.
Stone GN, K Schönrogge. 2003. The adaptive significance of
insect gall morphology. Trends in Ecol. Evol. 18: 512-522.
Stone GN, K Schönrogge, R Atkinson, D Bellido, J Pujade-
Villar. 2002. The population biology of oak gall wasps
(Hymenoptera: Cynipidae). Annu. Rev. Entomol. 47: 633-
668.
Stout MJ, JS Thaler, BPHJ Thomma. 2006. Plant-mediated
interactions between pathogenic microorganisms and
herbivorous arthropods. Annu. Rev. Entomol. 51: 663-
689.
Taper ML, TJ Case. 1987. Interactions between oak tannins
and parasite community structure: Unexpected benefits of
tannins to cynipid gall-wasps. Oecologia 71: 254-261.
Taper ML, EM Zimerman, TJ Case. 1986. Sources of
mortality for a cynipid gall-wasp Dryocosmus dubiosus
(Hymenoptera: Cynipidae): the importance of tannin/
fungus interaction. Oecologia 68: 437-495.
Tavakoli M, G Melika, SE Sedeghi, ZS Penzes, M Assareh, R
Atkinson et al. 2008. New species of Cynipid gall wasps
from Iran and Turkey (Hymenoptera: Cynipini). Zootaxa
1699: 1-64.
Tavanaei GH. 2006. Report of two fungi causing powdery
mildew on oak trees in forests of west Azarbaijan and
Kordistan Provinces. Iran. J. For. Range Prot. Res. 4: 46-
47.
Veldtman R, MA McGeoch. 2003. Gall-forming insect
species richness along a non-scleromorphic vegetation
rainfall gradient in South Africa: the importance of plant
community composition. Aust. Ecol. 28: 1-13.
Wallin KF, KF Raffa. 2001. Effects of folivory on subcortical
plant defences: Can defence theories predict interguild
processes? Ecology 82: 1387-1400.
Wilson D, GC Carroll. 1997. Avoidance of high-endophyte
space by gall-forming insects. Ecology 78: 2153-2163.
Wold EN, RJ Marquis. 1997. Induced defence in white oak:
effects on herbivores and consequences for the plant.
Ecology 78: 1356-1369.
Zargaran MR, SE Sadeghi, S Hanifeh. 2008. Identification of
oak gall wasps and their parasitoids in west Azerbaijan
Province of Iran, Urmia, Iran. Institute of Forests and
Rangelands, 108 pp. (in Persian with English abstract)
Zoological Studies 51(2): 185-194 (2012)

Age and Growth of Oxygymnocypris stewartii (Cyprinidae:

Schizothoracinae) in the Yarlung Tsangpo River, Tibet, China
Bin Huo, Cong-Xin Xie*, Bao-Shan Ma, Xue-Feng Yang, and Hai-Ping Huang
College of Fisheries, Huazhong Agricultural Univ., Wuhan, Hubei 430070, China

(Accepted September 14, 2011)

Bin Huo, Cong-Xin Xie, Bao-Shan Ma, Xue-Feng Yang, and Hai-Ping Huang (2012) Age and growth of
Oxygymnocypris stewartii (Cyprinidae: Schizothoracinae) in the Yarlung Tsangpo River, Tibet, China. Zoological
Studies 51(2): 185-194. To better understand the biology of Oxygymnocypris stewartii and its relationship with
management considerations, the age and growth of O. stewartii were examined using sectioned otoliths of 712
specimens collected from Aug. 2008 to Aug. 2009. The standard length (SL) ranged 45-587 mm. The location
of the 1st annulus was validated by a daily growth increment (DGI) analysis of otoliths. Monthly changes in
the marginal increment ratio of the otoliths with 1-8 annuli indicated that an annulus forms once a year, from
Mar. to June. The index of the average percentage error (IAPE) of the sectioned otoliths was 0.5%, and the
coefficient of variation (CV) for the age estimation was 0.7%. Estimated ages ranged 3-17 yr for males, 2-25 yr
for females, and 1-6 yr for those of undetermined sex. The SL-BW relationship was described as BW = 6.108
× 10-6 SL3.126 for females, BW = 9.872 × 10-6 SL3.052 for males, and BW = 3.203 × 10-5 SL2.821 for undetermined.
The von Bertalanffy function was used to model the observed length-at-age data as Lt = 526.8 {1 - exp[-0.141
(t - 0.491)]} for males, and Lt = 618.2 {1 - exp[-0.106(t - 0.315)]} for females. Females grew at a slower rate but
attained a larger size than males. Knowledge of this species’ characteristics of slow growth and a long life will
be useful for establishing reasonable management practices for its conservation.
http://zoolstud.sinica.edu.tw/Journals/51.2/185.pdf

Key words: Age, Growth, Otolith, Oxygymnocypris stewartii, Tibet.

T he subfamily Schizothoracinae is the stewartii is one of the endemic species that

predominant group of endemic fishes living in
high-elevation rivers and lakes on the Qinghai-
Tibetan Plateau (Cao et al. 1981). They are
found in very localized populations. Another
characteristic of these species is that they
are affected by anthropogenic pressures of
indiscriminate fishing, habitat modification resulting
from dam construction, and biological invasions.
In view of such alterations to the environment,
it has become a priority to make efforts towards
better understanding the biology of the subfamily
Scizothoracinae and its relationship with mana-
gement considerations.
Among the Schizothoracinae fishes inhabiting
in the Yarlung Tsangpo River, Oxygymnocypris
lives only in the clear, cool waters at elevations
of 3600-4200 m (Wu and Wu 1992, Chen and
Cao 2000). Moreover, as a result of the extreme
plateau environment, such as low temperatures
and poor food availability, O. stewartii is slow-
growing and long-lived. These characteristics of a
limited distribution, slow growth, and long life make
O. stewartii populations particularly vulnerable
to excessive exploitation. However, the rapidly
increasing demands for fish attributed to enhanced
immigration and gradual changes in many tra-
ditional customs have led to the overexploitation
of fish resources. The immoderate exploitation
has ultimately resulted in O. stewartii populations
rapidly declining, and this species is listed in the

*To whom correspondence and reprint requests should be addressed. Tel: 86-27-87286862. Fax: 86-27-87282114.
E-mail:xiecongxin@mail.hzau.edu.cn

185
186 Huo et al. – Age and Growth of Oxygymnocypris stewartii
IUCN’s Red List of Threatened Species as a near-
threatened fish (Ng 2010). However, attempts
to develop effective population management
strategies have been obstructed by a lack of basic
biological information. The available information
mainly focuses on its taxonomy (Lloyd 1908,
Cao and Deng 1962), its origin and evolution
(Cao et al. 1981), and aspects of its phylogenetic
development and biogeography (Chen 1998, Chen
2000, He and Chen 2007). There have been few
studies on the otolith microstructure, age and
growth of O. stewartii (Jia and Chen 2009 2011).
Accurate age determination and estimates of
growth and mortality parameters are fundamental
requirements for understanding population
dynamics and provide essential data needed to
maintain sustainable yields by fisheries (Campana
and Thorrold 2001). Results from the only 2
studies of age and growth of O. stewartii (Jia and
Chen 2009 2011) validated the periodicity of otolith
increment formation. However, no studies have
provided evidence to validate the 1st annual ring
or the precision of aging methods. This is often an
overlooked but necessary component of any aging
study (Campana 2001). The objectives of this
study were first to describe annulus characteristics
of otoliths, second to validate annuli and verify
annual periodicity in otoliths, and finally to estimate
the age and growth of O. stewartii.
88°W
Xiang Qu River
Namling
Xaitongmoin
Yarlung Tsangpo River Xigaze Rinbung
Nyang Qu River
Bainang
Gyangze
Fig. 1. Sampling locations of Oxygymnocypris stewartii in the Yarlung Tsangpo River.
MATERIALS AND METHODS
In total, 712 O. stewartii individuals were
obtained from the Yarlung Tsangpo River and a
tributary (Xiang Qu) monthly from Aug. 2008 to
Aug. 2009 by means of floating gillnets, bottom
gillnets, and trap nets (Fig. 1). More than 30 fish
were collected each month. The standard length
(SL) of each fresh specimen was measured to the
nearest 1 mm using a tapeline, and body weight
(BW) was measured to the nearest 0.1 g with an
electronic balance. Lapillus otoliths were extracted
from each fish, washed with 95% ethanol, air-
dried, and then stored in labeled tubes.
Both right and left lapillus otoliths were
removed from each individual, but usually just the
left otolith was used for the analysis. Otoliths were
mounted with the ventral face on a glass slide
using thermoplastic glue, with the dorsoventral
axis perpendicular to the slide plane. The otoliths
were then ground from the dorsal face using wet
sandpaper (600-1500 grit) and polished with
alumina paste (3 μm) until the core was visible
under a compound microscope. The otoliths were
removed by dissolving the glue with xylene, and
then the otoliths were re-affixed to the glass slides
using nail polish, with the polished face down. The
ventral face was then ground and polished until the
core was exposed again.
90°W
Lhasa River
Nyemo Quxu
Yarlung Tsangpo River
29°N
0 25 50 100 km
90°W
 Zoological Studies 51(2): 185-194 (2012)
The presumed daily growth increments
(DGIs) of otoliths of 6 young fish of 45-63 mm
SL (5 individuals caught on 4 Jan. 2009 and 1
individual caught on 23 Feb. 2009) were counted,
and the radius of the otoliths was measured along
the posterior axis (Sequeira et al. 2009). The DGI
periodicity was validated to be daily (Jia and Chen
2009).
A marginal increment ratio (MIR) analysis was
used to verify the period of opaque zone formation
in the otoliths. Monthly variations of the MIR
(1-8 annuli) were established using the following
equation:
MIR = (R - Rn) / (Rn - Rn-1);
where R is the otolith radius, Rn is the distance
from the focus to the outer edge of the last annulus
formed, and Rn-1 is the distance from the focus to
the outer edge of the penultimate complete annulus
(Haas and Recksiek 1995). Measurements were
conducted along the posterior growth axis using an
image analysis system (Leica Application Suite EZ,
Heerbrugg, Switzerland) with a direct data feed
between the dissecting microscope (Leica EZ 4D)
and a computer.
Each fish was assigned to an age class
assuming 1 Mar. as the birth date, which appro-
ximately corresponds to the peak spawning
season. A new ring mark found on the otolith of
a fish captured before 1 Mar. was not considered
to be an annulus in the age assignment, whereas
when a fish sampled after the assumed birth
date had no new ring mark, an annulus that was
supposed to have formed was considered in the
age estimation (Granada et al. 2004).
Otolith readings were made along the
posterior growth axis. The reader had no prior
knowledge of the length, sex, or time of capture
before the age estimation. All ages were deter-
mined twice by the same interpreter after a
considerable time (3 wk). Readings were only
accepted if both counts by the same examiner
were in agreement. If the 2 readings differed,
then the otolith was recounted, and the final count
was then accepted as the agreed age. If the
3rd reading had no consensus with either of the
previous 2 readings, the sample was discarded.
SL-BW relationgships were calculated by the
power relationship: BW = a × SLb; where a and
b are parameters. The standard von Bertalanffy
growth function (von Bertalanffy 1938) was used
to describe the observed body length-at-age using
the following formula:
187
Lt = L∞ {1 - exp[-k (t - t0)]};
where L ∞ is the asymptotic body length-at-age,
which represents the average body length-at-age
individuals would attain if they grew indefinitely, k
is the growth coefficient, and t0 is the age at length
0.
The growth performance index, Ø (Ø =
log 10 k + 2log 10 L ∞), was calculated to compare
the growth parameters obtained in the present
paper with values reported by other authors for
schizothoracine fishes (Munro and Pauly 1983).
The index of the average percentage error
(IAPE) and coefficient of variation (CV) were
calculated to measure the ageing precision
between the 2 readings. The equations (Campana
2001) are expressed as follows:
Σ Σ |X X- X | ) × 100%, and
1
N
1
R
ij j
IAPE = N
(R
j = 1 i = 1 j
R
Σ (XR--X1)
N 2
Σ(
ij j
CV =
1
) × 100%;
N i = 1
j = 1
Xj
where N is the number of fish aged, R is the
number of times each fish is aged, Xij is the ith age
determination of the jth fish, and Xj is the mean
age calculated for the jth fish.
The BW-SL relationship and von Bertalanffy
function were calculated by a non-linear regression
analysis (the Levenberg-Marquardt method;
Levenberg 1944). The difference in the BW-SL
relationship between the sexes was compared by
an analysis of covariance (ANCOVA). Deviation
of the allometric coefficient, b, from the theoretical
value of isometric growth (b = 3) was tested by
a t-test (Pauly 1984). A residual sum of squares
analysis (ARSS) was used to determine whether
any significant difference existed in the von
Bertalanffy equations for males and females (Chen
et al. 1992).
The analysis was carried out using SPSS
16.0 (Chicago, IL, USA), Originpro 8.0 (Originlab,
Northampton, USA), Microsoft Excel 2003
(Redmon, WA, USA), and Photoshop CS4
Extended (Adobe, San Jose, USA). Statistical
significance was accepted when p < 0.05.
188 Huo et al. – Age and Growth of Oxygymnocypris stewartii

RESULTS 15
Female
Male
Length-frequency distributions Undetermined
12

Of the 712 O. stewartii sampled, 373 were

Frequency (%)
females with SLs of 116-587 mm, 206 were 9
males with SLs of 167-455 mm, and 133 were an
undetermined sex with SLs of 45-260 mm. Length- 6
frequency distributions signnificantly differed
between sexes (Kolmogorov-Smirnov; D = 4.371,
3
p < 0.001) (Fig. 2). SLs of the captured fish were
mainly 100-450 mm (87.5%), and females were
significantly larger than males. 0
0 100 200 300 400 500 600
Age validation and annual periodicity Standard length (mm)

Fig. 2. Distributions of the standard length frequency of O.

The lapilli of 6 young O. stewartii with
stewartii.
SLs ranging 44.5-63.2 mm showed the typical
pattern of translucent and opaque zones, which
were respectively equivalent to the accretion
and discontinuous zones, composing a daily (A)
growth increment (Fig. 3). A continuous series
of concentric rings of declining size ranging from
4.6 to 0.5 μm was observed from the core to the
otolith margin. The 6 young specimens showed N
no transition zones (annuli). The estimated ages
were 178-202 (195 ± 9) d, which validated the 6
specimens to be young-of-the-year (YOY) fish. P
The mean radius of the lapilli was 669.97 (standard D-zone
deviation (S.D.) = 71.29) μm, while that of the 1st L-zone DGI
annulus was 676.04 (S.D. = 64.61) μm for older
fish.
For otolith sections with 1-8 annuli, monthly
changes of the MIR gradually increased from
June to Jan., and appeared to peak at 0.815 in
Jan. followed by a gradual decrease to 0.482 in
May (Fig. 4). Significant differences were found
in MIR values among months (one-way ANOVA, (B)
F = 13.326, p < 0.001). Tukey’s post-hoc pair-
wise comparisons revealed that the MIR in Jan.
significantly differed from those from Mar. to June
(p < 0.001). These results indicated that the
opaque band of the otoliths was laid down once a
year from Mar. to June.

Age structure

Sectioned otoliths of O. stewartii showed the

typical pattern of teleosts under transition light,
with an alternating sequence of broad opaque and
narrow hyaline bands that became progressively
narrower and of similar widths as the number
of bands increased (Fig. 5). Of the 712 otoliths
examined, only 10 (approximately 1.4%) were
Fig. 3. Daily growth increments (DGIs) in the lapillus of O.
stewartii with 45 mm SL. (A) Daily growth increments of core
region, scale bar = 20 μm. (B) Daily growth increments of
peripheral area, scale bar = 20 μm. N, nucleus; P, primordial;
L-zone, translucent zone; D-zone, opaque zone.
 Zoological Studies 51(2): 185-194 (2012) 189

discarded due to fragmentation and unidentifiable
annulus deposition. The reliablility of the age
estimates had low IAPEs (0.5%) and CVs (0.7%),
reflecting concordance in the readings. The
estimated age ranged 1-25 yr. Registered ages
of fish of undetermined sex ranged 1-6 yr, males
ranged 3-17 yr, and females ranged 2-25 yr. The
maximum estimated ages were 17 yr (455 mm SL)
for males and 25 yr (502 mm SL) for females.
Standard lengths at age of the 712 specimens
are given in table 1. Significant variation was
observed in the SL of individuals at the same age,
and the variation increased with elapsed years.
SL-BW relationships
SL-BW relationships were separately
calculated for males, females, and undetermined
(Fig. 6). Significant differences were found in
SL-BW relationships between sexes (ANCOVA,
F = 14.764, p < 0.0001). The regression equations
are shown as follows:
Female BW = 6.108 × 10-6 SL3.126 (R 2 = 0.955,
n = 373);

1.0 Male BW = 9.872 × 10-6 SL3.052 (R 2 = 0.957,

0.9
n = 206); and

0.8 Undetermined: BW = 3.203 × 10 -5 SL 2.821

0.7
(R = 0.981, n = 133).
2

0.6 The allometric index value (b) obtained

MIR (%)

from the function significantly differed from 3 for

0.5
females (t-test, t = 6.011, p < 0.01) and exhibited
0.4 no statistical difference from 3 for males (t-test,
t = 1.972, p > 0.05).
0.3

0.2
Jan. Feb. Mar. Apr. May June July Aug. Sept. Oct. Nov. Dec.
Month
Fig. 4. Mean monthly MIR for O. stewartii lapillus otoliths with
1-8 annuli, error bars represent the S.D.
Growth
The mean length-at-age did not significantly
differ between sexes for age classes 3-5 (unpaired
t-test, all p > 0.05). Therefore, the length-at-

(A) (B)

11 12
13 19

3
2 4 5 6
1
11

Fig. 5. Sectioned lapillus of O. stewartii with mm SL, estimated to be 19 yr old under transmitted light using the dissecting microscope.
Dots represent annuli. Scale bars: A = 0.5 mm; B = 0.3 mm.
 190 Huo et al. – Age and Growth of Oxygymnocypris stewartii

age data of undetermined specimens (except 0.893) for males and

for two 6-yr-old individuals) were included in the
von Bertalanffy models for both sexes. The von Lt = 618.2 {1 - exp[-0.106(t - 0.315)]} (R 2 =
Bertalanffy functions fitted to the observed length- 0.911) for females.
at-age are given as follows:
Lt = 526.8 {1 - exp[-0.141(t - 0.491)]} (R2 = The growth curve descirbed a trend of
relatively slow growth based on the obseved
length-at-age data between sexes (Fig. 7). Growth
3500 parameters suggested that the growth rate of
Female females was lower than that of males. The length-
3000 Male at-age rapidly increased during the 1st 4 yr (Table 1,
Undetermined
2500 Fig. 7). The growth performances indices (Ø) of O.
stewartii were 4.6076 for females and 4.5925 for
Body weight (g)

2000 males.
1500

1000 DISCUSSION
500
Based on the capture date (4 Jan. and
0 23 Feb.) and the hatching time (1 Mar.), the
0 100 200 300 400 500 600 700 daily growth increnements of YOY fish were
Standard length (mm)
supposededly about 300 or 360. However, DGIs
of only 178-202 were observed in this study.
Fig. 6. Length-weight relationships of O. stewartii. This phenomenon was also mentioned in other

Table 1. Number of specimens and Mean ± S.D. and range of standard length at age of O. stewartii

Age (yr) Female Male Undetermined

n Mean ± S.D. (mm) Range (mm) n Mean ± S.D. (mm) Range (mm) n Mean ± S.D. (mm) Range (mm)

1 14 44.5-87.1 56.4 ± 11.0

2 1 116 116 14 87.1-124.0 103.8 ± 8.7
3 14 151-199 171.6 ± 14.6 3 167-211 183.7 ± 23.9 89 87.9-205 142.9 ± 30.3
4 42 184-273 234.6 ± 23.1 21 208-273 244.4 ± 16.0 7 151-233 194.7 ± 31.6
5 70 210-327 273.4 ± 25.4 66 239-312 275.2 ± 16.7 6 237-258 250.8 ± 8.4
6 55 253-409 318.9 ± 29.8 58 249-363 296.9 ± 23.1 2 246-260 253.0 ± 9.9
7 33 280-440 344.1 ± 34.3 29 263-365 313.2 ± 25.2
8 14 245-483 372.7 ± 59.5 7 323-393 353.9 ± 24.8
9 21 351-520 425.8 ± 36.0 7 324-382 351.1 ± 21.1
10 11 388-556 449.6 ± 46.7 3 343-369 352.3 ± 14.5
11 6 405-488 429.0 ± 30.3 1 374 374
12 20 378-521 436.4 ± 41.7 1 350 350
13 26 349-528 432.4 ± 38.1 2 365-372 368.5 ± 4.9
14 23 396-524 444.3 ± 34.2 3 406-425 417.7 ± 10.2
15 9 410-555 467.2 ± 42.8
16 4 426-508 470.0 ± 34.8
17 3 493-562 517.0 ± 39.0 1 455 455
18 2 485-518 501.5 ± 23.3
19 2 444-507 475.5 ± 44.5
20 6 504-587 541.8 ± 30.4
21 2 539-559 549.0 ± 14.1
22 2 537-562 549.5 ± 17.7
23
24 2 496-536 516 ± 28.3
25 1 502 502
 Zoological Studies 51(2): 185-194 (2012)
Scizothoracinae fishes. Numbers of DGIs within
the 1st annulus were 137-154 in Ptychobarbus
dipogon (Li et al. 2009), 121-184 in O. stewartii
(Jia and Chen 2009), and 130-168 in Schizothorax
o’connori (Ma et al. 2011). The daily width
increment declined from 4.6 to 0.5 µm between
the core and the otolith margin for YOY fishes.
The variability in the daily width increment may
be related to several factors, especially the water
temperature (Marshall and Parker 1982, Neilson
and Geen 1982, Campana 1984). The variation in
the daily width increment within the annulus was
consistent with that of water temperature. The
daily width increment declined near the translucent
zone when water temperatures decreased
indicating that O. stewartii growth became slower at
that time (Jia and Chen 2009). Translucent zones
representing slow or no growth periods in the year
may be composed of many fine increments, but
these increments were too fine to be seen under a
light microscope. Therefore, the above-described
phenomenon could be attributed to the theoretical
resolution limit of the light microscope (Campana
and Neilson 1985). Perhaps, a scanning electron
microscope would be a more-accurate method.
Identification of the 1st or innermost growth
increment is an important component of any age
validation study. Validation of the 1st increment
is a mandatory adjunct to an age determination;
without a correctly defined starting point, age
determinations will be consistently wrong by
a constant amount. In species with a clearly
interpretable otolith microstructure, daily increment
700
600
Standard length (mm)
500
400
300
Female
200
Male
100
0 and Buxton 1997, Pajuelo et al. 2003, Bustos
0 5 10 15 20
Age (yr)
Fig. 7. The von Bertalanffy growth curve of O. stewartii with
the observed standard length at age estimated from otoliths.
The length-at-age data of the undetermined specimens were
included in models fitting both sexes (except for two 6 yr old
individuals)
191
counts can often be used to confirm the identity
of the 1st annulus (Waldron 1994, Lehodey and
Grandperrin 1996). DGIs can be used to estimate
the expected radius of the 1st annulus in otoliths
(Campana 2001). In this study, the mean radius
of older fish approached that of the 1st annulus in
YOY fish, indicating that the 1st annulus of lapilli
was validated in O. stewartii.
Mechanisms of growth increment formation
are poorly understood. There are several possible
explanations for band formation; temperature,
feeding regimes, and the reproductive cycle may
be factors affecting growth increment formation
(Beckman and Wilson 1995, Morales-Nin 2000,
Tserpes and Tsimenides 2001, Grandcourt et al.
2006, Liu et al. 2009). In this study, deposition
of the opaque zone in otoliths of O. stewartii
occurred in the spring and early summer, whereas
the hyaline zone was formed in winter months. A
similar phenomenon for Scizothoracinae fishes
was also demonstrated for O. stewartii (Jia and
Chen 2009), P. dipogon (Li et al. 2009), S. waltoni
(Qiu and Chen 2009), and S. o’connori (Ma et
al. 2011). Deposition of the opaque zone occurs
from Mar. to June (Fig. 4), corresponding to the
reproductive activity and water temperature
variations, and deposition of the hyaline zone
occurs during winter months, when there is
reduced metabolic activity. Formation of the
opaque zone in otoliths of O. stewartii appeared to
be partially associated with reproductive activity.
Peak reproductive activity (which occurs in Mar.,
unpubl. data) may promote a redirection of energy
to reproduction, with a consequent reduction in
somatic growth, which could possibly affect the
physiology of otolith growth. Moreover, formation
of the opaque zone in otoliths of O. stewartii may
be partially related to water temperature. The
average water temperature around the Xigaze is
2°C in Feb., rising abruptly to 15°C in June (Li et
al. 2010). This abrupt rise in water temperature
could cause changes in metablolic activities of
fish and could result in opaque zone deposition.
Similar influences by reproducion and water
temperature on other fishes were reported in past
studies (Morales-Nin and Ralston 1990, Mann
25 30
et al. 2009). Also, in a review of otolith studies
including 94 species from 36 families, Beckman
and Wilson (1995) found that the formation of
opaque and hyaline zones might be related to
water temperatures and spawning activity. Thus,
formation of the annulus in otoliths of O. stewartii
could be due to an interaction between water
192 Huo et al. – Age and Growth of Oxygymnocypris stewartii

temperature and reproduction. (Hofer et al. 1985). O. stewartii is piscivorous, and

The maximum ages estimated for O. stewartii
in this study were 25 yr for females and 17 yr
for males, which were comparable to values
obtained by Jia and Chen (2011) for both sexes,
indicating that females live longer than males.
Li and Chen (2009) recorded 45 yr for females
and 24 yr for males of P. dipogon based on an
interpretation of sectioned otoliths. Chen et al.
(2009) found 18 yr for females and 16 yr for males
of S. younghusbandi younghusbandi by means
of otoliths. Yao et al. (2009) obtained 24 yr for
females and 18 yr for males of S. o’connori using
otoliths. Ma et al. (2010) also reported 50 yr
for females and 40 yr for males of S. o’connori
based on otolith observations. Those studies
revealed that great longevity is a common trait in
schizothoracine fishes.
Comparing results of the growth of O.
stewartii with those of Jia and Chen (2011), the k
value obtained in this study was smaller (Table 2).
Differences among all of the estimated parameters
could be attributed to several factors: (1) different
sampled locations, (2) different age groups
employed to fit the VBGF function (the previous
study used 1-6 age groups), and (3) different size
distributions.
The growth performance indices (Ø) of
O. stewartii were larger than those of other
schizothoracines (Table 2), indicating that the
growth of O. stewartii is relatively greater than
other fishes of the Schizothoracinae, which inhabit
the same elevatons and region. These growth
differences might be related to feeding (Jia and
Chen 2011). Carnivorous fishes can obtain more
energy than that gained by other feeding habits
its food could contain more energy than that of
other Schizothoracinae fishes mentioned above.
The von Bertalanffy growth coefficient (k)
is a useful index for addressing the potential
vulnerability of stocks to excessive mortality
(Musick 1999). Comparing the parameters of
some Schizothoracinae fishes, Li and Chen (2009)
suggested that they are slow-growing species
with k values of around 0.1. Slow-growing, long-
lived fishes tend to be particularly vulnerable to
excessive exploitation and exhibit rapid stock
collapse, after which population turnover may
be lower than expected, and their responses to
rehabilitation measures slower than predicted
(Musick 1999). In this study, the estimated
maximum age was 25 yr, and the k value was
around 0.1, indicating that O. stewartii is a slow-
growing, long-lived species. Therefore, it is
essential to establish reasonable management
practices for this species to allow for its sustain-
able use. First, more scientific work such as
biological studies, resource investigations, and
life history studies should be vigorously carried
out in the near future to accumulate fundamental
biological data in order to properly manage this
species; and 2nd, based on these data, new
fishery regulations should be established, and the
effectiveness of these regulations assessed by
the continuous monitoring of stocks. These new
fishery regulations should focus on a sustainable
fishing intensity, a minimum catch size, and proper
fishing methods to prevent overfishing. Prohibiting
fishing and marketing during the spawning season
may be 1 way to protect the older classes of O.
stewartii; finally, local governments should properly

Table 2. Comparison of growth characters of Schizothoracinae fishes in different studies

Species Region Age material Sex L∞ (mm) k (year -1) t0 Ø Sources

Schizothorax o’connori Yarlung Tsangpo River Otolith F 492.4 0.1133 -0.5432 4.4389 Yao et al. (2009)
M 449.0 0.1260 -0.4746 4.4049
Yarlung Tsangpo River Otolith F 576.9 0.081 -0.946 4.4307 Ma et al. (2010)
M 499.7 0.095 -0.896 4.3751
Schizothorax waltoni Yarlung Tsangpo River Otolith F 691.1 0.056 -2.466 4.4273 Qiu and Chen (2009)
M 689.8 0.051 -3.257 4.3850
Ptychobarbus dipogon Lhasa River Otolith F 598.66 0.0898 -0.7261 4.5076 Li and Chen (2009)
M 494.23 0.1197 -0.7296 4.4659
Schizothorax younghusbandi Yarlung Tsangpo River Otolith F 471.4 0.0789 0.2 4.2439 Chen et al. (2009)
younghusbandi Lhasa River M 442.7 0.0738 -1.4 4.1603
Oxygymnocypris stewartii Yarlung Tsangpo River Otolith F 877.4821 0.1069 0.5728 4.9153 Jia and Chen (2011)
M 599.3939 0.1686 0.6171 4.7823
Yarlung Tsangpo River Otolith F 618.2 0.106 0.315 4.6076 Present study
M 526.8 0.141 0.491 4.5926
 Zoological Studies 51(2): 185-194 (2012)
guide local customs like “Releasing Day” to protect
spawning populations and recruitment.
Acknowledgments: The authors thank the
Institute of Hydrobiology, Chinese Academy of
Sciences, Wuhan, China for providing the otolith
image analysis system.
REFERENCES
Beckman DW, CA Wilson. 1995. Seasonal timing of opaque
zone formation in fish otoliths. In DH Secor, JM Dean,
SE Campana, eds. Recent developments in fish otolith
research. Columbia, SC: Univ. of South Carolina Press,
pp. 27-43.
Bustos R, Á Luque, JG Pajuelo. 2009. Age estimation and
growth pattern of the island grouper, Mycteroperca fusca
(Serranidae) in an island population on the northwest
coast of Africa. Sci. Mar. 73: 319-328.
Campana SE. 1984. Lunar cycles of otolith growth in the
juvenile starry flounder Platichthys stellatus. Mar. Biol.
80: 239-246.
Campana SE. 2001. Accuracy, precision and quality control
in age determination, including a review of the use and
abuse of age validation methods. J. Fish Biol. 59: 197-
242.
Campana SE, J Neilson. 1985. Microstructure of fish otoliths.
Can. J. Fish Aquat. Sci. 42: 1014-1032.
Campana SE, SR Thorrold. 2001. Otoliths, increments, and
elements: keys to a comprehensive understanding of fish
populations? Can. J. Fish Aquat. Sci. 58: 30-38.
Cao WX, YY Chen, YF Wu, SQ Zhu. 1981. Origin and
evolution of schizothoracine fishes in relation to the
upheaval of the Xizang Plateau. In CAS, ed. The
comprehensive scientific expedition to the Qinghai-Xizang
Plateau, studies on the period, amplitude and type of the
uplift of the Qinghai-Xizang Plateau. Beijing: Science
Press, pp. 118-130. (in Chinese)
Cao WX, ZL Deng. 1962. The schizothoracine fishes in the
region and adjacent area of western Szechwan Province.
Acta Hydrobiol. Sin. 2: 27-53. (in Chinese)
Chen F, YF Chen, DK He. 2009. Age and growth of
Schizopygopsis younghusbandi younghusbandi in the
Yarlung Zangbo River in Tibet, China. Environ. Biol. Fish.
28: 155-162.
Chen Y, DA Jackson, HH Harvey. 1992. A comparison of von
Bertalanffy and polynomial functions in moldelling fish
growth data. Can. J. Fish Aquat. Sci. 49: 1228-1235.
Chen YF. 1998. Study on phylogenetic and distribution pattern
of the schizothoracine fishes. I Phylogenetic. Acta
Zootaxon. Sin. 23: 17-25. (in Chinese)
Chen YF. 2000. Systermatics and resource biology of the
schizothoracine fishes. PhD dissertation, Chinese
Academy of Science, Wuhan, China. (in Chinese)
Chen YF, WX Cao. 2000. Schizothoracinae. In PQ Yue,
ed. Fauna sinica osteicthtyes cypriniformes III. Beijing:
Scince Press, pp. 273-388. (in Chinese)
Granada VP, Y Masuda, T Matsuoka. 2004. Age and growth of
the yellowbelly threadfin bream Nemipterus bathybius in
Kagoshima Bay, southern Japan. Fish. Sci. 70: 497-506.
Grandcourt EM, TZA Abdessalaam, F Francis. 2006. Age,
193
growth, mortality and reproduction of the blackspot
snapper, Lutjanus fulviflamma (Forsskål, 1775), in the
southern Arabian Gulf. Fish. Res. 78: 203-210.
Haas RE, CW Recksiek. 1995. Age verification of winter
flounder in Narragansett Bay. Trans. Am. Fish. Soc. 124:
103-111.
He DK, YF Chen. 2007. The molecular phylogeny and bio-
geography of highly specialized Schizothoracinae. Chin.
Sci. Bull. 52: 303-311. (in Chinese)
Hofer R, G Krewedl, F Koch. 1985. An energy budget for an
omnivorous cyprinid: Rutilus rutilus (L.). Hydrobiologia.
122: 53-59.
J i a Y T, Y F C h e n . 2 0 0 9 . O t o l i t h m i c r o s t r u c t u r e o f
Oxygymnocypris stewartii (Cypriniformes, Cyprinidae,
Schizothoracinae) in the Lhasa River in Tibet, China.
Environ. Biol. Fish. 28: 45-52.
Jia YT, YF Chen. 2011. Age structure and growth chara-
cteristics of the endemic fish Oxygymnocypris stewartii
(Cypriniformes: Cyprinidae: Schizothoracinae) in the
Yarlung Tsangpo River, Tibet. Zool. Stud. 50: 69-75.
Lehodey P, R Grandperrin. 1996. Age and growth of the
alfonsino Beryx splendens over the seamounts off New
Caledonia. Mar. Biol. 125: 249-258.
Levenberg K. 1944. A method for the solution of certain non-
linear problems in least squares. Q. Appl. Math. 2: 164-
168.
Li HJ, N Zhang, XT Lin. 2010. Spatio-temporal characteristics
of Yarlung Zangbo River in Tibet. J. Henan Norm. Univ.
38: 126-130. (in Chinese)
Li XQ, YF Chen. 2009. Age structure, growth and mortality
estimates of an endemic Ptychobarbus dipogon (Regan,
1905) (Cyprinidae: Schizothoracinae) in the Lhasa River,
Tibet. Environ. Biol. Fish. 28: 97-105.
Li XQ, YF Chen, DK He, F Chen. 2009. Otolith characteristics
and age determination of an endemic Ptychobarbus
dipogon (Regan, 1905) (Cyprinidae: Schizothoracinae) in
the Yarlung Tsangpo River, Tibet. Environ. Biol. Fish. 86:
53-61.
Liu KM, ML Lee, SJ Joung, YC Chang. 2009. Age and growth
estimates of the sharptail mola, Masturus lanceolatus, in
waters of eastern Taiwan. Fish. Res. 95: 154-160.
Lloyd RE. 1908. Report on the fish collected in Tibet by Capt
FH and Stewart IMS. Rec. Indian Mus. 2: 341-344.
Ma BS, CX Xie, B Huo, XF Yang, HP Huang. 2010. Age and
Growth of a Long-Lived Fish Schizothorax o’connori in the
Yarlung Tsangpo River, Tibet. Zool. Stud. 49: 749-759.
Ma BS, CX Xie, B Huo, XF Yang, P Li. 2011. Age validation,
and comparison of otolith, vertebra and opercular bone for
estimating age of Schizothorax o’connori in the Yarlung
Tsangpo River, Tibet. Environ. Biol. Fish. 90: 159-169.
Mann B, C Buxton. 1997. Age and growth of Diplodus sargus
capensis and D. cervinus hottentotus (Sparidae) on the
Tsitsikamma coast, South Africa. Cybium 21: 135-147.
Marshall SL, SS Parker. 1982. Pattern identification in the
microstructure of sockeye salmon (Oncorhynchus nerka)
otoliths. Can. J. Fish Aquat. Sci. 39: 542-547.
Morales-Nin B, S Ralston. 1990. Age and growth of Lutjanus
kasmira (Forskål) in Hawaiian waters. J. Fish Biol. 36:
191-203.
Morales-Nin B. 2000. Review of the growth regulation
processes of otolith daily increment formation. Fish. Res.
46: 53-67.
Munro JL, D Pauly. 1983. A simple method for comparing the
growth of fishes and invertebrates. ICLARM Fishbyte 1:
194 Huo et al. – Age and Growth of Oxygymnocypris stewartii
5-6.
Musick JA. 1999. Ecology and Conservation of Long-lived
Marine Animals. Am. Fish. Soc. Symp. 23: 1-10.
Neilson JD, GH Geen. 1982. Otoliths of chinook salmon
(Oncorhynchus tshawytscha): daily growth increments
and factors influencing their production. Can. J. Fish
Aquat. Sci. 39: 1340-1347.
Ng HH. 2010. Oxygymnocypris stewartii. IUCN Publishing
Red List of Threatened Species. Available at http://www.
iucnredlist.org/apps/redlist/details/168408/0 Accessed
Apr. 2010.
Pajuelo JG, JM Lorenzo, R Domínguez-Seoane. 2003. Age
estimation and growth of the zebra seabream Diplodus
cervinus cervinus (Lowe, 1838) on the Canary Islands
shelf (central-East Atlantic). Fish. Res. 62: 97-103.
Pauly D. 1984. Fish population dynamics in tropical waters: a
manual for use with programmable calculators. Manila,
the Philippines: International Center for Living Aquatic
Resources Management.
Qiu H, YF Chen. 2009. Age and growth of Schizothorax
waltoni in the Yarlung Tsangpo River in Tibet, China.
Ichthyol. Res. 56: 260-265.
Sequeira V, A Neves, AR Vieira, I Figueiredo, LS Gordo. 2009.
Age and growth of bluemouth, Helicolenus dactylopterus,
from the Portuguese continental slope. ICES J. Mar. Sci.
66: 524-531.
Tserpes G, N Tsimenides. 2001. Age, growth and mortality of
Serranus cabrilla (Linnaeus, 1758) on the Cretan shelf.
Fish. Res. 51: 27-34.
von Bertalanffy L. 1938. A quantitative theory of organic
growth (inquiries on growth laws II). Hum. Biol. 10: 181-
213.
Waldron ME. 1994. Validation of annuli of the South African
anchovy, Engraulis capensis, using daily otolith growth
increments. ICES J. Mar. Sci. 51: 233-234.
Wu YF, CZ Wu, eds. 1992. The fishes of the Qinghai-Xizang
Plateau. Chengdu, China: Sichuan Publingshing House
of Science and Technology. (in Chinese)
Yao JL, YF Chen, F Chen, DK He. 2009. Age and growth of
an endemic Tibetan fish, Schizothorax o’connori, in the
Yarlung Tsangpo River. J. Freshw. Ecol. 24: 343-345.
Zoological Studies 51(2): 195-203 (2012)

Collection of Pollen Grains by Centris (Hemisiella) tarsata Smith (Apidae:

Centridini): Is C. tarsata an Oligolectic or Polylectic Species?
Lia Gonçalves1, Cláudia Inês da Silva2, and Maria Luisa Tunes Buschini1,*
1
Departamento de Biologia, Univ. Estadual do Centro-Oeste, Rua Presidente Zacarias 875, CEP: 85010-990, Guarapuava (PR), Brasil
2
Departamento de Biologia, Univ. de São Paulo, Faculdade de Filosofia Ciências e Letras, Av. Bandeirantes, 3900, 14040-901, Ribeirão
Preto-SP, Brasil. E-mail:claudiainess@gmail.com

(Accepted September 21, 2011)

Lia Gonçalves, Cláudia Inês da Silva, and Maria Luisa Tunes Buschini (2012) Collection of pollen grains
by Centris (Hemisiella) tarsata Smith (Apidae: Centridini): Is C. tarsata an oligolectic or polylectic species?
Zoological Studies 51(2): 195-203. Among pollinator species, bees play a prominent role in maintaining
biodiversity because they are responsible, on average, for 80% of angiosperm pollination in tropical regions.
The species richness of the bee genus Centris is high in South America. In Brazil, these bees occur in many
types of ecosystems. Centris tarsata is an endemic species occurring only in Brazil. No previous studies
considered interactions between plants and this bee species in southern Brazil, where it is the most abundant
trap-nesting bee. Accordingly, the goals of this study were to investigate plants used by this species for its larval
food supply and determine if this bee is polylectic or oligolectic in this region. This work was conducted in the
Parque Municipal das Araucárias, Guarapuava (PR), southern Brazil, from Mar. 2002 to Dec. 2003. Samples of
pollen were collected from nests of these bees and from flowering plants in grassland and swamp areas where
the nests were built. All of the samples were treated with acetolysis to obtain permanent slides. The family
Solanaceae was visited most often (71%). Solanum americanum Mill. (28.6%) and Sol. variabile Mart. (42.4%)
were the primary pollen sources for C. tarsata in the study area. We found that although C. tarsata visited
20 species of plants, it preferred Solanum species with poricidal anthers and pollen grains with high protein
levels. This selective behavior by females of C. tarsata indicates that these bees are oligolectic in their larval
provisioning in this region of southern Brazil. http://zoolstud.sinica.edu.tw/Journals/51.2/195.pdf

Key words: Centris (Hemisiella) tarsata, Solanum variabile, Solanum americanum, Provision of pollen grains.

B
ees of the family Apidae can fly long
distances in tropical forests in search of preferred
plant species, thus promoting cross-pollination
(Frankie et al. 1983, Roubik 1993). The plant-
pollinator relationship is symbiotic and establishes
a beneficial relationship between 2 species with
different levels of dependency (Boucher et al.
1982, Del-Claro 2004). According to Faegri and
Van der Pijl (1979) and Proctor et al. (1996), plant-
pollinator interactions are considered to result from
natural selection, which produces a wide variety of
adaptations in plants, allows the transfer of pollen
grains, and increases gene flow within a species.
Among pollinator species, bees play an
important role in maintaining biodiversity. On
average, they are responsible for 80% of angio-
sperm pollination in tropical regions (Kevan and
Baker 1983, Bawa 1990). The higher efficiency of
bees as pollinators results from their high numbers
compared to other pollinators and from their
superior adaptations to complex floral structures.
For example, their bodies and mouthparts are
adapted to collect and transport resources, such as
nectar and pollen, respectively (Kevan and Baker

*To whom correspondence and reprint requests should be addressed. E-mail:liagoncalves22@hotmail.com; isatunes@yahoo.com.br

195
196 Gonçalves et al. – Pollen Sources of Centris tarsata
1983, Michener 2000).
Some bee species belonging to the tribes
Tapnotaspidini and Centridini exhibit reproductive
cycles and nesting activities that are synchronized
with the flowering periods of certain species
of plants (Rocha-Filho et al. 2008, Aguiar and
Melo 2009, Bezerra et al. 2009, Gaglianone et
al. 2011). These bees visit flowers to obtain oil,
pollen, nectar, and resin (resources needed to
build parts of their nests) to feed the larvae and
maintain adults and their reproductive activities
(Vogel 1974, Buchmann 1987, Roubik 1989,
Vinson et al. 1996). Some studies showed the
importance of these bees as pollinators of various
species of Neotropical plants (Frankie et al.
1976, Gottsberger et al. 1988, Freitas 1997),
including those producing oil, such as species of
the Malpighiaceae (Rêgo and Albuquerque 1989,
Freitas et al. 1999) and Scrophulariaceae (Vogel
and Machado 1991).
The genus Centris is typically tropical, and
its species belong to 12 subgenera. The species
richness of Centris is high in South America. In
Brazil, these bees are found in various eco-
systems, such as dunes and sandbanks (Silva
and Martins 1999, Silva et al. 2001, Viana and
Alves-dos-Santos 2002), caatinga (Martins 1994,
Zanella 2000, Aguiar and Almeida 2002, Aguiar et
al. 2003), grasslands, and savannas (Silveira and
Campos 1995, Albuquerque and Mendonça 1996).
Centris tarsata has only been recorded from
Brazil. The distribution of C. tarsata in Brazil is
based on Aguiar and Garófalo (2004), information
from specimens deposited in entomological
collections (J.M.F. Camargo, pers. commun.),
samples of females and/or males collected on
flowers (Camargo and Mazucato 1984, Vogel
and Machado 1991, Martins 1994, Silveira and
Campos 1995, Albuquerque and Mendonça 1996,
Freitas 1997, Schlindwein 1998, Zanella 2000),
and the location of nests (Chandler et al. 1985,
Camilo et al. 1995, Silva et al. 2001, Viana et al.
2001, Aguiar and Martins 2002). This information
indicates that C. tarsata occurs in the states of PA,
MA PI, CE, PB, PE, BA, MG, SP, PR, RS, MS, MT,
and GO.
In the savanna area of Uberlândia (Minas
Gerais State, Brazil), C. tarsata was recorded
as one of the principal pollinators of West Indian
cherry Malpighia emarginata DC (Malpighiaceae)
(Vilhena and Augusto 2007). This bee is solitary
and tends to nest in preexisting cavities. Its
nests can be built in trap-nests (Silva et al. 2001,
Aguiar and Garófalo 2004, Buschini and Wolff
2006). In southern Brazil, C. tarsata is the most
abundant bee species (Buschini 2006). It prefers
open habitats and shows greater nesting activity
during the hot season, especially in Dec. and Jan.
(Buschini and Wolff 2006).
Several studies were conducted in Brazil
to identify sources of pollen used by different
species of bees and to understand the degree
of association between bees and the plants that
they visit. Through an analysis of pollen grains,
it is possible to identify the main floral resources
used by bees. This information allows the asse-
ssment of resource availability in the field and
the identification of times of resource scarcity
(Salgado-Labouriau 1961, Ortiz 1994, Bastos et al.
2003).
An analysis of the pollen spectrum of C.
tarsata based on samples from nests in the
northeastern micro-region of Bahia State, Brazil
indicated the presence of 17 pollen types from 7
plant families. These samples, representing an
assemblage of 5-11 pollen types, identified plants
used by the bees to feed their offspring (Dórea et
al. 2009). In Maranhão State, also in northeastern
Brazil, pollen analyses of C. tarsata showed
relatively high quantities of pollen grains from
Banisteriopsis sp. (Malpighiaceae) and Cassia sp.
(Caesalpiniaceae).
Centris tarsata is endemic to Brazil. No
previous studies of the interactions of plants with
this bee species have been conducted in southern
Brazil, where it is the most abundant trap-nesting
bee. The goals of this study were to investigate
the plants that constitute the larval food supply for
C. tarsata and determine whether this bee has a
polylectic or an oligolectic tendency in this region.
MATERIALS AND METHODS
This study was carried out in the Parque
Municipal das Araucárias, located in the
municipality of Guarapuava, Paraná State,
southern Brazil (25°21'06"S, 51°28'08"W). The
area of the park is approximately 104 ha. The
vegetation is composed of mixed ombrophilous
forest (42.75%), gallery forest (10.09%), fields
(6.8%), swamps (7.13%), and altered areas
(33.23%). The grasslands are physionomically
characterized by areas of low grasses and no
bushes. Species of the families Cyperaceae,
Leguminosae, Verbenaceae, Compositae, and
Umbelliferae are the principal plants in this habitat.
The grasslands are surrounded by Araucaria
 Zoological Studies 51(2): 195-203 (2012)
forests, dominated by Araucaria angustifolia
(Coniferae: Araucariaceae). The swamps are
located in the lowest-elevation regions of the
park and are primarily composed of grasses and
members of the Compositae (Buschini and Fajardo
2010). According to the Köppen classification, the
climate is humid mesothermic, with no dry season
and mild summers because of the elevation. The
winter is moderate, with the frequent occurrence
of frost. The annual mean temperature is appro-
ximately 16°C.
In this study, pollen grains were removed
from provisioned cells of 11 nests of C. tarsata
from a total of 128 trap-nests installed in the
swamp and grassland areas. Centris tarsata has
a seasonal pattern of nesting activity from Nov.
to May (Buschini and Wolff 2006), so the pollen
grains used in this study were collected from Mar.
2002 to Dec. 2003. Pollen collected from the
nests was preserved on permanent slides with the
acetolysis method (Erdtman 1960). Five pollen
grain slides were made for each nest to produce a
total of 55 slides. Pollen grains were also collected
from flowering plants from May 2006 to Apr. 2007.
Pollen was collected throughout the area in which
nests were built. The pollen was removed from
the flowers and/or buttons of each plant to obtain
2 slides per plant. All pollen grain slides from
both nests and plants were examined using light
microscopy to identify plants used by the bee. The
pollen was quantified by consecutively counting
300 pollen grains per slide. Total numbers of
pollen grains counted were 1500 grains per
nest and 16,500 grains in all. Subsequently, we
determined the percentages of occurrence of each
species and botanical family in C. tarsata nests
according to the classification of Barth (1970) and
Louveaux et al. (1970 1978). Thus, pollen types
were classified as dominant (> 45% of total grain
on the slides), accessory (15%-45%), important
isolates (3%-14%), and occasional isolates (< 3%).
RESULTS
We collected 99 flowering plant species in the
study area during the activity period of C. tarsata.
Overall, 20 pollen types from 17 plant families
were collected by this bee (Fig. 1, Table 1).
The family Solanaceae was visited most
often (71%). Solanum americanum Mill. (28.6%)
and Sol. variabile Mart. (42.4%) were the primary
pollen sources for C. tarsata in the study area.
The 2nd most frequently visited family was the
197
Phytolaccaceae. Phytollaca dioica L. supplied
15.4% of the pollen in the samples. The family
Malpighiaceae represented 4.5% of the pollen
in the samples, whereas the families Lauraceae
(3.2%), Myrthaceae (2.8%), Melastomataceae
(1.01%), Lythraceae (0.9%), Campanulaceae
(0.4%), Convolvulaceae (0.2%), Caesalpiniaceae
(0.16%), Asteraceae (0.1%), Amaranthaceae
(0.08%), and Polygalaceae (0.07%) occurred at
low percentages. Erythroxylum deciduum A. St.
Hil. (0.03%), of the family Erythroxilaceae, and
another species not yet identified (Undetermined-1)
(0.01%) appeared in more than 1 sample but at
low occurrence percentages. Although the pollen
types of Styrax leprosum Hook and Am. (0.09%)
and another unidentified species (Undetermined-2)
(0.04%) were recorded in only 1 sample, their
percentages were higher than those of Ery.
deciduum and Undetermined-1.
The frequencies of occurrence of pollen
types in the 11 samples analyzed showed that Sol.
americanum and Sol. variabile (100%) were the
most consistent, followed by Janusia guaranitica
and Cinnamomum amoenum (Ness and Mart.)
Kosterm (60%), Gomphrena elegans Mart., and
Ipomoea grandifolia Lam. (40%). The 14 other
pollen types occurred in 10%-30% of samples:
Phytollaca dioica, Vernonia sp. Schreb, Senna
multijuga (Rich.) H. S., Cuphea sp. P. Browne
(30%), Ery. deciduum, Janusia sp. A. Juss,
Tibouchina cerastifolia Cong, and Undetermined-1
(20%), and Baccharis sp. L., Lobelia sp. Pohl,
Ipomoea purpurea (l.) Roth, Campomanesia
adamantium O. Berg, Polygala sp. L., Sty.
leprosum Hook. and Arn, and Undetermined-2
(10%).
DISCUSSION
Although C. tarsata used 20 types of pollen
grains, pollen of Sol. americanum, Sol. variabile,
and Phy. dioica were most common in the larval
diet. The importance of the family Solanaceae as
a source of pollen for C. tarsata was also reported
by Aguiar et al. (2003) and Dórea et al. (2009) in
the Caatinga, xerophytic vegetation predominant
in semi-arid northeastern Brazil. According to
Buchmann (1983), the presence of poricidal
anthers in flowers of the Solanaceae establishes
a close relationship with females of Centris. In
this plant-pollinator relationship, pollination by
vibration (buzz-pollination) is an effective method
of extracting pollen from these plants (Buchmann
198 Gonçalves et al. – Pollen Sources of Centris tarsata

(A) (B) (C)

(D) (E) (F)

(G) (H) (I)

(J) (K) (L)

(M) (N) (O)

(P) (Q) (R)

(S) (T) (U)

2 μm

Fig. 1. Pollen grains found in nests of Centris tarsata. (A) Gomphrena elegans, (B) Baccharis sp., (C) Vernonia sp., (D) Lobelia sp., (E)
Ipomoea grandifolia, (F) I. purpurea, (G) Erythroxylum deciduum, (H) Senna multijuga, (I) Cinnamomum amoenum, (J) Cuphea sp., (K)
Janusia guaranítica, (L) Janusia sp., (M) Tibouchina cerastifolia, (N) Campomanesia adamantium, (O) Phytolacca dioica, (P) Polygala
sp., (Q) Solanum americanum, (R) Sol. variabile, (S) Styrax leprosum, (T) Undetermined-1, (U) Undetermined-2.
 Zoological Studies 51(2): 195-203 (2012) 199

1983). The collection of pollen by vibration
also occurs on flowers of the Melastomataceae
(Buchmann and Hurley 1978) and Caesalpiniaceae
(Moure and Castro 2001). This method of polli-
nation is associated with small pollen grains, as
in Solanum species. These grains have a high
amount of protein, which is important for larval
development (Roulston et al. 2000).
Studies in different Brazilian biomes also
highlighted the importance of the families
Solanaceae (Dórea et al. 2009), Malpighiaceae,
Caesalpiniaceae, and Myrtaceae (Mendes and
Rego 2007) as sources of pollen for C. tarsata.
Aguiar et al. (2003), in Itatim (BA), northeastern
Brazil, recorded the presence of pollen of the
Caesalpiniaceae in the diet of C. tarsata offspring.
Senna Mill. was also found to represent a frequent
source of pollen and nectar for this bee. Plants of
this genus are also associated with a mechanism
of pollination by vibration (Santos et al. 2004,
Anacleto and Marchini 2005, Andena et al. 2005).
Moreover, Aguiar et al. (2003) stated that solitary
bees, such as species of Centris, are more likely to
act as generalists during foraging for nectar than

Table 1. Occurrence of pollen grain types in nests of Centris tarsata: pollen accessory (PA), pollen important
isolate (PII), pollen occasional isolate (POI)

Pollen type Resources Life form Local occurrence Month of Percent Classification of
available collection occurrence on pollen types
slides

Amaranthaceae
Gomphrena elegans Mart. - Herb Swamp Mar. 0.08% POI
Asteraceae
Baccharis sp. L. Pollen, nectar Shrub Grassland Feb., Mar. 0.04% POI
Vernonia sp. Schreb. Pollen, nectar Tree Forest Oct. 0.06% POI
Campanulaceae
Lobelia sp. Pohl. Herb Swamp Feb. 0.4% POI
Convolvulaceae
Ipomoea grandifolia Lam. Pollen Liana Swamp Mar. 0.13% POI
Ipomoea purpúrea (l.) Roth. Pollen Liana Swamp Mar. 0.07% POI
Erythroxilaceae
Erythroxylum deciduum A. St. Hil. Pollen, nectar Shrub Grassland Sept. 0.03% POI
Caesalpiniaceae
Senna multijuga (Rich.) H. S. Pollen, nectar Tree Grassland Feb. 0.16% POI
Lauraceae
Cinnamomum amoenum (Nees) Kosterm. Pollen Tree Forest Oct. 3.2% PII
Lythraceae
Cuphea sp. P. Browne. Pollen, nectar Herb Swamp Apr. 0.9% POI
Malpighiaceae
Janusia guaranítica (A. St.-Hil.) A. Juss. Pollen, oil Herb Grassland Dec. 3.2% PII
Janusia sp. A. Juss. Pollen, oil - - - 1.3% POI
Melastomataceae
Tibouchina cerastifolia Cong. Pollen, oil Herb Grassland Jan., Feb. 1.01% POI
Myrtaceae
Campomanesia adamantium O. Berg. Pollen Tree Grassland Oct. 2.8% POI
Phytolaccaceae
Phytolacca dioica L. Pollen Tree Grassland, forest Oct. 15.4% PA
Polygalaceae
Polygala sp. L. Pollen, nectar - - - 0.07% POI
Solanaceae
Solanum americanum Mill. Pollen Herb Grassland, Mar. 28.6% PA
swamp
Solanum variabile Mart. Pollen Tree Grassland Nov. 42.4% PA
Styracaceae
Styrax leprosum Hook. and Arn. Pollen, nectar Tree Forest Oct. 0.09% POI
Undetermined-1 - - - - 0.04% POI
Undetermined-2 - - - - 0.01% POI
200 Gonçalves et al. – Pollen Sources of Centris tarsata
during foraging for pollen and oil. Those authors
also stated that the exploitation of resources from
the families Caesalpiniaceae and Malpighiaceae is
frequently found in different biomes.
In Salinas (MG), southeastern Brazil,
Guimarães (2006) found that the family Myrtaceae
was visited by several species of Centris. Similar
results were obtained in the São Francisco Valley
of Brazil by Siqueira et al. (2005), who reported a
high frequency of Centris and Xylocopa visitation
to flowers of this family. Pollen is the primary
resource provided by this family for bees, which
are probably its most efficient pollinators (Gressler
et al. 2006). In these plants, pollination also
occurs by vibration, although the anthers exhibit
longitudinal dehiscence and are not poricidal
(Proença 1992).
Although the percentage of pollen from plants
of the family Malpighiaceae in the diet of C. tarsata
in Guarapuava was low (4.5%), this finding does
not mean that these plants have little importance
as resource suppliers for these bees. According to
Anderson (1979), Vogel (1990), and Ramalho and
Silva (2002), a close relationship between bees
of the tribe Centridini and plants of this family can
be interpreted as a product of a long evolutionary
history of interactions between the 2 groups. This
history could even explain the high reproductive
success of these plants in the Americas. The
plants provide both oil and pollen to feed the larvae
of these bees. They bloom almost year-round, but
the flowers are more highly abundant during the
warm and rainy period (Silberbauer-Gottsberger
and Gottsberger 1988). In the Brazilian savanna
(i.e., the cerrado), the nesting and foraging
activities of the Centridini are generally more
frequent during the period of peak flowering of
the Malpighiaceae (Rocha-Filho et al. 2008). The
Centridini is considered to be key pollinators of
this plant family (Michener 2000, Machado 2002
2004, Machado et al. 2002, Alves dos Santos et al.
2007). The system of oil production in these plants
and collection of the oil by the bees require a
series of morphological adaptations in both groups
and behavioral adaptations by the bees (Simpson
and Neff 1977). The oil, the primary resource
that attracts the bees to the plants, is secreted by
glands called elaiophores (Vogel 1974, Simpson
and Neff 1981) and is included in the diet of larval
bees.
In the Malpighiaceae, pollen grain sizes
usually range from medium to large. Pollen of
Janusia occurred in small quantities in bee nests,
but these quantities were considerably higher than
those found for Sol. americanum, Sol. variabile,
and Phy. dioica. According to Severson and
Parry (1981), measurements of a pollen sample
should be representative of the mass of pollen by
including the average number of grains counted
and should also reflect the estimated volumetric
contribution of the grain type. Thus, the degree
of importance of 1 type of pollen grain should
not be based solely on its percentage but should
also include both its numeric and volumetric
representation in the sample.
The sporadic presence of pollen of the
Melastomataceae in nests of C. tarsata in
Guarapuava may reflect the tendency of the bees
to seek the oil of these plants to build their nests.
When collecting the oil, they place their ventral
abdomen and thorax on the stigma and anthers of
the flowers. This behavior facilitates the transfer
of pollen to the stigma (Gimenes and Lobão
2006) and also results in the transport of small
amounts of pollen from the plants to the bees’
nests. In studies in Camaçari (BA), northeastern
Brazil, Oliveira-Rebouças and Gimenes (2004)
observed that medium- and large-sized species
of Centris were highly efficient at collecting pollen
from flowers of the Melastomataceae. In the study
region, the use of pollen of the Convolvulaceae
(e.g., Ipomoea) by Centris may be related to the
morphology of the pollen grains. These grains
are large-sized, are porate and colpate with a
perforated exine, and are spiculated and hairy.
The spine characteristic of this genus assists in
the attachment of pollen grains to the hair of bees,
thereby optimizing the transport process (Machado
and Melhem 1987, Sengupta 1972, Leite et al.
2005).
The occurrence of pollen from the
Phytolaccaceae, Lauraceae, and Styracaceae
in the diet of C. tarsata may be related to the
bees’ search for resources in plants located in
transitional areas between the grassland and
Araucaria forest. These areas are close to sites
where the bees nest. Frankie et al. (1983) also
observed many species of Centris foraging in the
canopy of mass-flowering tree species.
Although C. tarsata was found to visit 20
species of plants, it preferred Solanum species
with poricidal anthers and pollen grains with high
amounts of protein. This selective behavior by
females of C. tarsata indicates that this bee is
oligolectic in its larval provisioning in this region
of southern Brazil. Because C. tarsata occurs in
areas of natural grasslands and collects pollen from
plants in transition zones between these areas and
 Zoological Studies 51(2): 195-203 (2012)
Araucaria forest, these bees undoubtedly promote
the pollination of various plant species in these
areas that are currently suffering from severe
exploitation and fragmentation in southern Brazil.
Further studies should be conducted to investigate
the ability of these bees to explore different
grassland fragments in this region and transport
pollen grains between them, thereby increasing
the genetic variability of the region’s plants.
Acknowledgments: Partial support was provided
by Fundação Araucária (The State of Paraná
Research Foundation) and UNICENTRO (Univ.
Estadual do Centro-Oeste). We thank Prof. Dr. J.
Cordeiro for plant identification.
REFERENCES
Aguiar AJC, CF Martins. 2002. Abelhas e vespas solitárias
em ninhos-armadilha na Reserva Biológica Guaribas
(Mamanguape, Paraíba, Brasil). Rev. Bras. Zool. 19:
101-116.
Aguiar AJC, GAR Melo. 2009. Notes on oil sources for
the bee genus Caenonomada (Hymenoptera, Apidae,
Tapinotaspidini). Rev. Bras. Entomol. 53: 154-156.
Aguiar CML, GF Almeida. 2002. Atividade diária de coleta de
óleo por espécies de Centris (Hymenoptera: Apidae) em
Mcvaughia bahiana W.R. Anderson (Malpighiaceae) na
caatinga. Acta Biol. Leop. 24: 131-140.
Aguiar CML, CA Garófalo. 2004. Nesting biology of Centris
(Hemisiella) tarsata Smith (Hymenoptera, Apidae,
Centridini). Rev. Bras. Zool. 21: 477-486.
Aguiar CML, FCV Zanella, CF Martins, CAL Carvalho. 2003.
Plantas visitadas por Centris spp. (Hymenoptera: Apidae)
na caatinga para obtenção de recursos florais. Neotrop.
Entomol. 32: 247-259.
Albuquerque PMC, JAC Mendonça. 1996. Anthophoridae
(Hymenoptera, Apoidea) e Flora associada em uma
formação de cerrado no município de Barreirinhas, Ma,
Brasil. Acta Amazôn. 26: 45-54.
Alves-dos-Santos I, IC Machado, MC Gaglione. 2007. História
natural das abelhas coletoras de óleo. Oecol. Bras. 11:
242-255.
Anacleto DA, LC Marchini. 2005. Análise faunística de abelhas
(Hymenoptera, Apoidea) coletadas cerrado do Estado de
São Paulo. Acta Sci. Biol. Sci. 27: 277-284.
Andena SR, LR Bego, MR Mechi. 2005. A Comunidade de
abelhas (Hymenoptera, Apoidea) de uma área de cerrado
(Corumbataí, SP) e suas visitas às flores. Rev. Bras.
Zoociênc. 7: 55-91.
Anderson WR. 1979. Floral conservatism in Neotropical
Malpighiaceae. Biotropica 11: 219-223.
Barth OM. 1970. Análise microscópica de algumas amostras
de mel. Pólen dominante. Anais Acad. Bras. Ciênc. 42:
351-366.
Bastos EMAF, VM Silveira, AEE Soares. 2003. Pollen
spectrum of honey produced in cerrado areas of Minas
Gerais State (Brazil). Braz. J. Biol. 63: 599-615.
Bawa KS. 1990. Plant-pollinator interactions in tropical rain
forests. Ann. Rev. Ecol. Syst. 21: 399-422.
201
Bezerra ES, AV Lopes, IC Machado. 2009. Biologia
reprodutiva de Byrsonima gardnerana A. Juss.
(Malpighiaceae) e interações com abelhas Centris
(Centridini) no Nordeste do Brasil. Rev. Bras. Bot. 32: 95-
108.
Boucher DH, S James, KH Keeler. 1982. The ecology of
mutualism. Ann. Rev. Ecol. Syst. 13: 315-347.
Buchmann SL. 1983. Buzz pollination in angiosperms. In
CE Jones, RJ Little, eds. Handbook of experimental
pollination biology. New York: Scientific and Academic
Editions, pp. 73-113.
Buchmann SL. 1987. The ecology of oil flowers and their
bees. Ann. Rev. Ecol. Syst. 18: 343-369.
Buchmann SL, JP Hurley. 1978. A biophysical model for buzz
pollination in angiosperms. J. Theor. Biol. 72: 639-657.
Buschini MLT. 2006. Species diversity and community
structure in trap-nesting bees in southern Brazil.
Apidologie 37: 58-66.
Buschini MLT, S Fajardo. 2010. Biology of the solitary wasp
Trypoxylon (Trypargilum) agamemnon Richards 1934
(Hymenoptera: Crabronidae) in trap-nests. Acta Zool-
Stockh. 91: 426-432.
Buschini MLT, LL Wolff. 2006. Biologia de nidificação de
Centris (Hemisiella) tarsata Smith no Sul do Brasil
(Hymenoptera, Apidae, Centridini). Braz. J. Biol. 66:
1091-1101.
Camargo JMF, M Mazucato. 1984. Inventário da apifauna e
flora apícola de Ribeirão Preto, SP - Brasil. Dusenia 14:
55-87.
Camillo E, CA Garófalo, JC Serrano, G Muccillo. 1995. Diver-
sidade e abundância sazonal de abelhas e vespas
solitárias em ninhos-armadilha (Hymenoptera, Apocrita,
Aculeata). Rev. Bras. Entomol. 39: 459-470.
Chandler L, JAF Barrigosi, EBS Diaz. 1985. The first defi-
nitive host record for Leucospis cayennensis Westwood
(Hymenoptera: Leucospidae). Rev. Ceres 32: 170-174.
Del-Claro K. 2004. Multitrophic relationships, conditional
mutualisms, and the study of interaction biodiversity in
tropical savannas. Neotrop. Entomol. 33: 665-672.
Dórea MC, FAR Santos, LCL Lima, LER Figueroa. 2009.
Análise Polínica do Resíduo Pós-Emergência de
Ninhos de Centris tarsata Smith (Hymenoptera: Apidae,
Centridini). Neotrop. Entomol. 38: 197-202.
Erdtman G. 1960. The acetolized method. A revised
description. Svensk Bot. Tidskrift 54: 561-564.
Faegri K, L Van der Pijl. 1979. The principles of pollination
ecology. 3rd ed. Oxford: Pergamon Press, 244 pp.
Frankie GW, PA Opler, KS Bawa. 1976. Foraging behaviour of
solitary bees: implications for outcrossing of a Neotropical
forest tree species. J. Ecol. 64: 1049-1057.
Frankie GW, SB Vinson, PA Opler. 1983. Centris aethyctera
and Centris fuscata in the lowland dry forest of Costa Rica
(Hymenoptera: Anthophoridae). In DH Janzen, ed. Costa
Rica Natural History. Chicago, Univ. Press, pp. 707-708.
Freitas BM. 1997. Number and distribution of cashew
(Anacardium occidentale) pollen grains on the bodies
of its pollinators, Apis mellifera and Centris tarsata. J.
Apicult. Res. 36: 15-22.
Freitas BM, JE Alves, GF Brandão, ZB Araujo. 1999.
Pollination requirements of West Indian cherry (Malpighia
emarginata) and its putative pollinators, Centris bees. J.
Agric. Sci. 133: 303-311.
Gaglianone MC, AJC Aguiar, F Vivallo, I Alves-dos-Santos.
2011. Checklist das Abelhas Coletoras de Óleos do
202 Gonçalves et al. – Pollen Sources of Centris tarsata
Estado de São Paulo, Brasil. Biot. Neotrop., 11: 1-10.
(Online Edição em Inglês)
Gimenes M, CS Lobão. 2006. A polinização de Krameria
bahiana B.B. Simpson (Krameriaceae) por abelhas
(Apidae) na restinga, BA. Neotrop. Entomol. 35: 440-445.
Gottsberger G, JMF Camargo, I Silberbauer-Gottsberger.
1988. A bee-pollinated tropical community: the beach
dune vegetation of Ilha de São Luís, Maranhão. Braz.
Bot. Jahrb. Syst. 109: 469-500.
Gressler E, MA Pizo, L Patrícia, C Morellato. 2006. Polinização
e dispersão de sementes em Myrtaceae do Brasil. Rev.
Bras. Bot. 29: 509-530.
Guimarães RA. 2006. Abelhas (Hymenoptera: Apoidea)
visitantes das flores da goiaba (Psidium guajava L.),
laranja (Citrus sinensis L.) e tangerina (Citrus reticulata
B.) em pomares comerciais em Salinas - MG. Vitória da
Conquista, BA: Univ. Estadual do Sudoeste da Bahia.
Kevan PG, HG Baker. 1983. Insects as flower visitors and
pollinators. Ann. Rev. Entomol. 28: 407-453.
Leite KRB, R Simão-Bianchini, FAR Santos. 2005. Morfologia
polínica do gênero Merremia Dennst. (Convolvulaceae)
ocorrentes no Estado da Bahia, Brasil. Acta Bot. Bras.
19: 313-321.
Louveaux J, A Maurizio, G Vorwohl. 1970. Methods of
melissopalynology. Bee World 51: 25-138.
Louveaux J, A Maurizio, G Vorwohl. 1978. Methods of
melissopalynology. Bee World 59: 139-157.
Machado IC. 2002. Flores de óleo e abelhas coletoras de óleo
floral: conceitos, distribuição e revisão. Ribeirão Preto,
SP, BR.: Anais do V Encontro Sobre Abelhas, pp. 129-
135.
Machado IC. 2004. Oil-collecting bees and related plants: a
review of the studies in the last twenty years and case
histories of plants occurring in NE Brazil. In BM Freitas,
JOP Pereira, eds. Solitary bees. Conservation, rearing
and management for pollination. Fortaleza, CE, BR:
Imprensa Univ., pp. 255-281.
Machado IC, TS Melhem. 1987. Morfologia polínica de
Ipomoea hederifolia L. e I. quamoclit L. (Convolvulaceae).
Hoehnea 14: 25-30.
Machado IC, S Vogel, AV Lopes. 2002. Pollination of
Angelonia cornigera Hook. (Scrophulariaceae) by long-
legged oil-collecting bees in NE Brazil. Plant. Biol. 4: 352-
359.
Martins CF. 1994. Comunidade de abelhas (Hym., Apoidea)
da caatinga e do cerrado com elementos de campo
rupestre do Estado da Bahia, Brasil. Rev. Nord. Biol. 9:
225-257.
Mendes FN, MMC Rêgo. 2007. Nidificação de Centris
(Hemisiella) tarsata Smith (Hymenoptera, Apidae,
Centridini) em ninhos-armadilha no Nordeste do
Maranhão, Brasil. Rev. Bras. Entomol. 51: 382-388.
Michener CD. 2000. The bees of the World. Baltimore, MD:
Johns Hopkins Univ. Press, 913 pp.
Moure JS, MS Castro. 2001. Uma nova espécie de Centris
Fabricius (Hymenoptera, Apoidea, Anthophoridae) do
Nordeste do Brasil. Rev. Bras. Zool. 18: 329-333.
Oliveira-Rebouças P, M Gimenes. 2004. Abelhas (Apoidea)
Visitantes de Flores de Comolia ovalifolia DC Triana
(Melastomataceae) em uma Área de Restinga na Bahia.
Neotrop. Entomol. 33: 315-320.
Ortiz PL. 1994. The Cistaceae as food resources for honey
bees in SW Spain. J. Apic. Res. 33: 136-144.
Proctor M, P Yeo, A Lack. 1996. The natural history of
pollination. London: Harper Collins Publishers, 496 pp.
Proença CEB. 1992. Buzz pollination older and more
widespread than we think? J. Trop. Ecol. 8: 115-120.
Ramalho M, M Silva. 2002. Flora oleífera e sua guilda de
abelhas em uma comunidade de restinga tropical. Rev.
Univ. Estadual Feira Santana Bras. 2: 34-43.
Rêgo MMC, PMC Albuquerque. 1989. Comportamento das
abelhas visitantes de murici, Byrsonima crassifolia (L.).
Kunth, Malpighiaceae. Bol. Mus. Para. Emílio Goeldi.
Sér. Zool. 5: 179-193.
Rocha-Filho LC, CI Silva, MC Gaglianone, SC Augusto. 2008.
Nesting behavior and natural enemies of Epicharis
(Epicharis) bicolor Smith, 1854 (Hymenoptera: Apidae).
Trop. Zool. 21: 227-242.
Roubik DW. 1989. Ecology and natural history of tropical
bees. Cambridge Tropical Biology Series. New York:
Cambridge Univ. Press, 514 pp.
Roubik DW. 1993. Tropical pollinators in the canopy and
understory - field data and theory for stratum preferences.
J. Insect Behav. 6: 659-673.
Roulston TH, JH Cane, SL Buchmann. 2000. What governs
protein content of pollen: pollinator preferences, pollen-
pistil interactions, or phylogeny? Ecol. Monogr. 70: 617-
643.
Salgado-Labouriau ML. 1961. Pollen grains of plants of the
“Cerrado” I. Ann. Acad. Bras. Cienc. 33: 119-130.
Santos FM, CAL Carvalho, RF Silva. 2004. Diversidade
de abelhas (Hymenoptera: Apoidea) em uma área de
transição Cerrado-Amazônia. Acta Amazon. 34: 319-328.
Schlindwein C. 1998. Frequent oligolecty characterizing a
diverse bee-plant community in a xerophytic bushland of
subtropical Brazil. Stud. Neotrop. Fauna E 33: 46-59.
Sengupta S. 1972. On the pollen morphology of Convol-
vulaceae with special reference to taxonomy. Rev.
Palaeobot. Palynol. 13: 157-212.
Severson DW, JE Parry. 1981. A chronology of pollen
collection by honeybees. J. Apic. Res. 20: 97-103.
Silberbauer-Gottsberger I, G Gottsberger. 1988. A polinização
de plantas do cerrado. Rev. Bras. Biol. 48: 651-663.
Silva FO, BF Viana, EL Neves. 2001. Biologia e Arquitetura
de ninhos de Centris (Hemisiella) tarsata Smith
(Hymenoptera: Apidae: Centridini). Neotrop. Entomol. 30:
541-545.
Silva MCM, C Martins. 1999. Flora apícola e relações tróficas
de abelhas (Hymenoptera: Apoidea) em uma área
de duna (praia de Interlagos, Cabedelo - PB, Brasil).
Principia 7: 40-51.
Silveira FA, MJO Campos. 1995. A mellissofauna de
Corumbataí (SP) e Paraopeba (MG) e uma análise
da biogeografia das abelhas do cerrado brasileiro
(Hymenoptera, Apoidea). Rev. Bras. Entomol. 39: 371-
401.
Simpson BB, JL Neff. 1977. Krameria, free-fatty acids and oil-
collecting bees. Nature 267: 150-151.
Simpson BB, JL Neff. 1981. Floral rewards: alternatives to
pollen and nectar. Ann. MO Bot. Gard. 68: 301-322.
Siqueira KMM, CF Martin, C Schlindwein, LHP Kill. 2005.
Polinizadores efetivos da goiaba (Psidium guajava) em
área irrigada do submédio do vale do São Francisco.
Ann. Enc. Zool. Nord. 15: 260.
Viana BF, I Alves-dos-Santos. 2002. Bee diversity of the
coastal sand dunes of Brazil. In PG Kevan, VL Imperatriz-
Fonseca, eds. Pollinating bees: the conservation link
between agriculture and nature. Brasília, BR: Ministério
 Zoological Studies 51(2): 195-203 (2012)
do Meio Ambiente, pp. 135-153.
Viana BF, FO Silva, AMP Kleinert. 2001. Diversidade e
sazonalidade de abelhas solitárias (Hymenoptera:
Apoidea) em dunas litorâneas no nordeste do Brasil.
Neotrop. Entomol. 30: 245-251.
Vilhena AMGF, SC Augusto. 2007. Polinizadores da aceroleira
Malpighia emarginata DC (Malpighiaceae) em área de
Cerrado no Triângulo Mineiro. Biosci. J. 23: 14-23.
Vinson SB, GW Frankie, HJ Williams. 1996. Chemical ecology
of bees of the genus Centris (Hymenoptera: Apidae). Fla.
Entomol. 79: 109-129.
203
Vogel S. 1974. Ölblumen und ölsammelnde Bienen. Trop.
Subtrop. Pflanz. 7: 285-369.
Vogel S. 1990. History of the Malpighiaceae in the light of
pollination ecology. Memoir. NY Bot. Gard. 55: 130-142.
Vogel S, ICS Machado. 1991. Pollination of four sympatric
species of Angelonia (Scrophulariaceae) by oil-collecting
bees in NE, Brasil. Plant. Syst. Evol. 178: 153-178.
Zanella FCV. 2000. The bees of the Caatinga (Hymenoptera,
Apoidea, Apiformes): a species list and comparative notes
regarding their distribution. Apidologie 31: 579-592.
Zoological Studies 51(2): 204-212 (2012)

Monogamous System in the Taiwan Vole Microtus kikuchii Inferred from

Microsatellite DNA and Home Ranges
Jung-Sheng Wu, Po-Jen Chiang, and Liang-Kong Lin*
Department of Life Science, Tunghai Univ., 181 Taichung Port Road, Sec. 3, Taichung 407, Taiwan

(Accepted September 14, 2011)

Jung-Sheng Wu, Po-Jen Chiang, and Liang-Kong Lin (2012) Monogamous system in the Taiwan vole
Microtus kikuchii inferred from microsatellite DNA and home ranges. Zoological Studies 51(2): 204-212. The
Taiwan vole Microtus kikuchii is considered socially monogamous based on indirect information of captive
behaviors and home-range ecology. Genetic components of its mating system were not previously examined.
We tested the hypotheses that M. kikuchii is both socially and genetically monogamous by combining field
information of home ranges with genetic analysis of relationships among individuals. Trapping was conducted
in the Hehuan Mt. area of Taroko National Park, central Taiwan, from June 2004 to Aug. 2005. We chose 16
microsatellite loci using primers designed for M. oeconomus and M. montebelli to amplify M. kikuchii DNA.
Eleven loci produced clear, polymorphic banding patterns and were used for the genetic analysis. The home-
range sizes of adults did not significantly differ between sexes or among seasons. For the 14 social units
indicated by overlapping home ranges, 11 (78.6%) were male-female pairs. The other 3 social units involved
more than 2 individuals. In two of these, ranges of a male-female pair overlapped ranges of their offspring and
other individuals. The genetic analysis revealed that some of the male-female pairs identified by overlapping
home ranges did not reproduce. Information based on the home-range data was not powerful enough to
identify genetic components of M. kikuchii ʼs mating system and may provide misleading results. A parentage
analysis based on microsatellite genotyping revealed litters (with a total 31 of offspring) sired by 18 males and
20 females. The only 2 males that fathered more than 1 litter did so in different years when their 1st mate was
no longer present. None of the 9 litters with multiple offspring had more than 1 father. Home-range overlap was
mostly between a single male and a single female and with their offspring. All pairs producing offspring were
genetically monogamous. Our results strongly support the hypotheses that M. kikuchii is socially and genetically
monogamous. http://zoolstud.sinica.edu.tw/Journals/51.2/204.pdf

Key words: Genetic monogamy, Home range, Parentage analysis, Social monogamy.

M ating systems of mammals can be defined
as monogamous, polygynous, polyandrous, or
promiscuous based on the number of partners
each individual has (Wittenberger 1979, Clutton-
Brock 1989). In the past, a species’ mating system
was indirectly determined by sexual dimorphism,
space use, and mating behaviors (Emlen and
Oring 1977, Getz and Hofmann 1986, Carter et
al. 1995). Monogamy occurs in < 3% of mammal
species (Kleiman 1977) and has attracted much
research interest. Traditional ways to determine
monogamy include 1) pair bonds between males
and females in reproductive and non-reproductive
seasons (Carter et al. 1995), 2) aggressive
behaviors toward strange individuals (Carter et
al. 1995, Back et al. 2002), 3) bi-parental care
(Solomon 1993a, Patris and Baudoin 2000), 4)
regulation of social factors (e.g., estrus induction)
(Carter et al. 1995), 5) the same home range sizes
for males and females (Gaulin and FitzGerald
1988), and 6) shared use of a territory (e.g.,
strong overlap in home ranges) (Reichard 2003).

*To whom correspondence and reprint requests should be addressed. Po-Jen Chiang and Liang-Kong Lin contributed equally to this
work. Tel: 886-4-23595845. Fax: 886-4-23595845. E-mail:lklin@thu.edu.tw

204
 Wu et al. – Monogamy of Taiwan Voles
These traditional methods provide clues for social
monogamy, but not for genetic monogamy. Social
monogamy indicates that 1 male and 1 female
show social living behavior, but it infers no sexual
or reproductive patterns (Reichard 2003). Genetic
monogamy is when 1 male and 1 female have an
exclusive reproductive relationship, and there are
no extra-pair copulations (Reichard 2003).
With the development of molecular genetic
techniques, genetic data are being used to exa-
mine mating systems (Queller et al. 1993, Avise
1994). Social mating systems may differ from
genetic mating systems (Clutton-Brock and Isvaran
2006). In small mammals, for example, Neotoma
cinerea in North America was thought to be socially
polygynous based on sexual dimorphism and
female clustering (Finley 1958, Escherich 1981). It
was identified as genetically monogamous using
DNA fingerprinting techniques (Topping and Millar
1998). In contrast, socially monogamous species
were found to engage in extra-pair copulations
suggesting they are not genetically monogamous.
These include the genetically promiscuous
Apodemus sylvaticus in the UK (Baker et al. 1999)
and the genetically polygynous A. argenteus
in Japan (Ohnishi et al. 2000). Peromyscus
polionotus (Foltz 1981) and P. californicus (Ribble
1991); however, are both socially and genetically
monogamous. To distinguish between social and
genetic components of a mating system (Hughes
1998), monogamous mating systems should be
examined with field observations and genetic
analyses that indicate parentage of offspring and
rule out extra-pair copulations (Kraaijeveld-Smit
2004).
In Microtus species occurring in the New
and Old Worlds (Hoffmann and Koeppl 1985),
promiscuousness and polygyny are common,
but monogamy is rare (Wolff 1985). Microtus
kikuchii is the only Microtus species endemic to
Taiwan. It is the southernmost Old World Microtus
species (Hoffmann and Koeppl 1985). It lives in
diverse habitats, such as grasslands, scrub, and
forests, including coniferous, broadleaf, and mixed
coniferous and broadleaf forests. It is mainly
found at elevations of > 2000 m in alpine habitats
of coniferous forests and Yushan cane (Yushania
niitakayamensis) grasslands. The reproductive
season is from Mar. to Aug. (Lu 1991). Chen et
al. (2006) studied the behavior of M. kikuchii in
captivity, and found that when given the freedom
to choose, it spent significantly more time with its
mated partner than with strange individuals. They
also observed paternal care of offspring. Wu
205
(1998) studied the home ranges of M. kikuchii
using radio-tracking and field trapping. He found
that home range sizes did not differ between males
and females, only opposite sexes had overlapping
ranges, and each range overlapped with only
1 individual of the opposite sex. Those results
suggest social monogamy. To date; however, there
has been no study of the genetic components of M.
kikuchii mating systems.
Parental care by both parents and pairing
exclusivity are supporting behavioral components
of monogamy (Solomon 1993b, Carter et al.
1995, Patris and Baudoin 2000). These home-
range and behavioral observational studies led
us to hypothesize that M. kikuchii is socially and
genetically monogamous. Since microsatellite
DNA can be sensitive enough to identify parental
relationships, relatedness, and dispersal rates
(Scribner and Pearce 2000), we used microsatellite
DNA and capture-recapture methods to identify
consistencies between the social and genetic
mating systems of M. kikuchii. We tested the
following predictions: 1) adult home-range sizes
do not significantly differ between sexes, 2) home-
range overlaps among adults during reproductive
seasons are extensive or exclusive to a single
individual of the opposite sex, and 3) there is a
lack of evidence of males mating with more than 1
female at the same time (both females alive) or of
litters sired by multiple males.
MATERIALS AND METHODS
Trapping
Trapping was carried out from June 2004 to
Sept. 2005 in the Hehuan Mt. area (121°17'17.4"E,
24°08'36.4"N) of Taroko National Park, central
Taiwan. The elevation is about 3000 m. The
dominant vegetation is Yushan cane grassland.
Sherman traps baited with sweet potato were
set up in a 4-ha (200 × 200-m) grid. To reduce
trapping mortality and help retain warmth in cold
months, balled-up wads of shredded paper were
put in front of the trigger of each Sherman trap.
Previous trapping results with 10-m trap spacing in
the same Hehuan Mt. area revealed mean home
range sizes of 447.9 m 2 in spring, 423.4 m 2 in
summer, 258.3 m2 in fall, and 210.6 m2 in winter
with no statistical differences among seasons or
between sexes (Wu 1998). Radio-tracking of 8
individuals for at least 24 h of continuous tracking
revealed a mean home range size of 843 m 2
206 Zoological Studies 51(2): 204-212 (2012)
(202-2260 m2) and a > 20-m movement distance
between radio locations in a single day (average
47.5 m, range 22-89 m) (Wu 1998). To maximize
the number of individuals trapped, we chose a
20-m trap spacing to form a 40,000-m2 grid with
121 Sherman traps. In an attempt to catch all
members of each family group, we trapped 5
successive nights each month using the capture-
recapture method. Because activity of M. kikuchii
peaks at 04:00-10:00 and 16:00-21:00 (Wu 1998),
we checked traps 3 times a day at 05:00-07:00,
09:00-11:00, and 19:00-22:00.
We used toe-clipping to mark each vole
the 1st time it was caught. Clipped toes and
additionally clipped pieces of the left ear and tail
were preserved in 99.9% alcohol for the genetic
analysis. For each capture, we recorded the trap
site, individual number, sex, age, body weight,
and reproductive status (e.g., whether the female
nipple size indicated it was pregnant or lactating
and whether a male had descended testes).
Adults were distinguished from immature voles by
the reproductive condition or weight (adult > 30 g).
Home-range size and overlap
Adult individuals captured more than 10 times
(Swilling and Wooten 2002) and residing over a
month within the trapping grid (i.e., trapped in at
least 2 different but not necessarily consecutive
monthly trap sessions) were considered residents
and used for the home-range analysis. Although
Seaman et al. (1999) suggested ≥ 30 relocations
for a home range to reduce bias, it was almost
impossible to achieve this number of trapping
locations because of our regime of 4 trap nights
per month and the short life of voles. Therefore,
we calculated home ranges for individuals
captured in at least 2 monthly trap sessions and
with ≥ 10 locations. Individuals caught in only
1 monthly trapping session were not included
in the home-range analysis. Home-range sizes
were estimated for the reproductive and non-
reproductive seasons. ArcView GIS 3.3 (ESRI
1996) with animal movement analysis (Hooge
and Eichenlaub 2000) was used to estimate the
minimum convex polygon (MCP) home-range
size (m2) from trapping locations. The percentage
of home-range overlap between males and
females was subsequently calculated. Because
of the small sample sizes and dependency of
some individuals on home ranges in both the
reproductive and non-reproductive seasons, the
nonparametric Mann-Whitney U-test was used to
compare home-range sizes between adult males
and females and between reproductive and non-
reproductive seasons. Home-range overlap was
examined for each 3-mo period in the reproductive
seasons to ensure that home ranges overlapped
at least part of the time. For example, examination
of home-range overlap from Mar. to May could
guarantee temporal overlap because individuals in
the home-range analysis were captured in at least
one of the following scenarios: Mar.-Apr., Apr.-May,
or Mar.-May. Thus, ranges of individuals trapped in
two of these 3 sessions would overlap temporally.
The percentage of home-range overlap was only
calculated between resident females and males.
Selection of microsatellite primers
There are no primers designed for Microtus
kikuchii. Microtus montebelli and M. oeconomus
are the most closely related species to M. kikuchii
(Conroy and Cook 2000). Therefore, we tested
primers for loci MSMM-1-8 designed for M.
montebelli (Ishibashi et al. 1999) and for loci Moe1-
8 designed for M. oeconomus (Van de Zande et al.
2000) to determine their suitability for analyzing M.
kikuchii microsatellite DNA sequences.
Genomic DNA was extracted from left-
ear tissue with a DNA purification kit (Epicentre,
Madison, WI, USA). The above primers amplified
specific sequences. A polymerase chain reaction
(PCR) used a total volume of 50 μl with 1 μl of a
fluorescence-labeled forward primer (25 mM), 1 μl
of an unlabeled reverse primer (25 mM), 5 μl PCR
buffer (10x), 0.6 μl DNA, 0.6 μl DNTP (10 mM),
0.6 μl Taq, and 41.2 μl water. Because the lengths
of these PCR products were too short for direct
sequencing, they were excised from the agarose
gel for TA cloning. Each specific sequence was
ligated with a vector (Invitrogen, Grand I., NY,
USA) and put into competent cells for TA cloning.
All clones were further re-amplified with M13
primers (forward and reverse) which were supplied
with TA cloning kit (Invitrogen) for length check.
PCR protocol of the TA cloning check started
from denaturation at 94°C for 10 min. Twenty-five
cycles were performed at the following conditions:
1 min at 94°C, 1 min at 55°C for annealing, and
1 min at 72°C for extension. Horizontal elec-
trophoresis with a 2% agarose gel was used to
check the sequence lengths of the clones.
Clones containing sequences of < 500 bp
(Schlotterer and Harr 2001) were sent to Mission
Biotech Company (Taichung, Taiwan) to be
sequenced on an ABI PRISM TM 3730xl DNA
 Wu et al. – Monogamy of Taiwan Voles
Analyzer (Applied Biosystems, Carlsbad, CA,
USA). Usable loci were determined using BioEdit
6.0.5 (Hall 1999) to check each sequence for
repeating units and whether both sides of the
sequence were conserved and stable. Primers of
usable microsatellite loci were used to synthesize
fluorescent primers for the microsatellite analysis.
Genetic data analysis
For the microsatellite analysis, protocols
for DNA extraction and the PCR were the same
as those described above. PCR products were
separated on an ABI 310 genetic analyzer (Applied
Biosystems). Individuals were genotyped using
Genotyper vers. 2.0 (Applied Biosystems).
Tests of pairwise linkage disequilibrium
between loci were conducted using GenePop
(Raymond and Rousset 1995). Allele diversity,
heterozygosity (observed Ho and expected He),
and Hardy-Weinberg equilibrium of each loci were
calculated and tested using GENALEX 6 (Peakall
and Smouse 2006).
C E RV U S 2 . 0 ( S l a t e e t a l . 2 0 0 0 ) a n d
GENALEX 6 (Peakall and Smouse 2006) were
used to estimate parentage. Since the real
parentage of any individual could not be assured
based on the capture data, we randomly compared
genotypes of all individuals to all males to identify
the most likely fathers. These offspring-father pairs
were randomly compared to all females to identify
the most likely mothers. An error rate of 1% was
incorporated into the simulation with 80% relaxed
and 95% strict confidence intervals. Parentage
was confirmed on the basis of mismatching
putative parentage at 0 loci or 1 locus, the LOD
score (log-likelihood of each candidate parent), and
the confidence of ΔLOD (the difference between
the 2 most likely parents). A ΔLOD score of > 3.0
confirmed parentage, while a ΔLOD score of < -3.0
rejected parentage (Slate et al. 2000). A ΔLOD
score was calculated by the difference in LOD
scores between the most likely and the 2nd most
likely candidate parents (either of which might be
the true parent). The most likely candidate parent
was the one with a ΔLOD score exceeding the
critical ΔLOD score with a 95% confidence inter-
val. Relationships of individuals with overlapping
home ranges were checked with results of the
parentage analysis to see whether they were
mates or family members.
207
RESULTS
In total, 169 voles (79 males and 90 females)
were caught in 1615 captures. One vole was
excluded from the parentage analysis due to
failure to amplify its DNA.
Polymorphism of microsatellite loci
In total, 11 microsatellite loci were chosen.
Seven loci (MSMM-2, -3, -4, -5, -6, -7, and -8)
used primers designed from Microtus montebelli
and 4 loci (Moe1, -2, -5, and -6) used primers
designed from M. oeconomus (Table 1). Except
for MSMS-7, numbers of alleles were > 10;
averaging 14.3 (range 8-19). All amplified loci
were polymorphic. The observed heterozygosity
(Ho) value of each locus was close to the expected
heterozygosity (He) value. Average values of Ho
and He were both 0.88. As a result, the mean
inbreeding coefficient, F, was essentially 0 at
-0.002. There were no departures from Hardy-
Weinberg equilibrium (Table 1), indicating that
the study population was under Hardy-Weinberg
equilibrium. Locus pairs MSMM-4/MSMM-7
and MSMM-4/Moe5 showed significant linkage
disequilibrium. Most loci showed no significant
linkage disequilibrium. Therefore, locus MSMM-4
was not used for the genetic analysis.
Parentage analysis
The critical ΔLOD with a 95% level of
certainty was 0.08 (with 95% of the parentage
resolved) if neither parent was known. Of the total
168 voles used for the parentage analysis, 20
mated pairs were found (18 males and 20 females)
to have 31 offspring. In total, 69 voles were
assigned parentage (Table 2). In other words,
41.1% (69/168) of the 168 voles, including adult
and immature voles, could be assigned parentage
with both parents identified. Except for 2 males
(49M and 50M), each male mated with only 1
female during the study period. The 2 males who
mated with more than 1 female did so in different
breeding seasons in different years and after the
1st female was no longer present.
Home-range size and overlap and their relation-
ships
The home-range sizes were 2763.6 ±
2228.5 m2 (n = 22) for adult males and 2170.0 ±
1341.3 m2 (n = 20) for adult females. No significant
208 Zoological Studies 51(2): 204-212 (2012)

difference was found between sexes (Mann-
Whitney U-test, p = 0.31). Average home-range
sizes in the reproductive season were 1826.1 ±
1463.9 m2 (n = 23) for adult males and 1684.2 ±
1256.7 m2 (n = 19) for adult females. The average
home-range sizes in the non-reproductive season
were 2072.7 ± 1637.7 m2 (n = 11) for adult males
and 1125.0 ± 874.6 m2 (n = 8) for adult females.
No significant difference in home-range sizes
was found between sexes (Mann-Whitney U-test,
p = 0.74 for the reproductive season and p = 0.16
for the non-reproductive season). The sample size
for the non-reproductive season was small.
We separated the reproductive season into
3 periods of 3 mo each and compared the home-
range overlap among resident individuals (Fig. 1).
Only those with ranges that overlapped ranges
of the opposite sex are shown. There were 21
pairwise combinations of home ranges overlapping
those of the opposite sex. Eleven (52.4%) showed
exclusive home-range overlap between 1 male
and 1 female (Fig. 1). Eight (38.1%) of these 21
pairwise combinations between male and female
ranges were detected as sexually paired partners.
Six (75%) of these 8 reproductive pairs had
exclusive, overlapping home ranges. The average
overlap of a male’s range with a female’s range
did not statistically differ from the average overlap
of a female’s range by a male’s range (Wilcoxon
rank-sum test, p = 0.126). Average percentages
of a female’s home range overlapped by a male’s
were significantly larger in mated pairs (77.1%,
n = 8) than in pairs not found to produce offspring
(41.8%, n = 11) (Mann-Whitney U-test, p = 0.0372).
Average percentages of a female’s home range
overlapped by a male’s were also significantly
larger than a male’s home range overlapped by his
sexual female partner’s (42.4%, Wilcoxon rank-
sum test, p = 0.0499, n = 8). In other words, a
female’s home range tended to overlap more with
that of her sexual partner, but males did not show
this trend.
In total, 14 social units were recognized
based on overlapping home ranges involving

Table 1. Microsatellites used in the study of Microtus kikuchii at Hehuan Mt., Taiwan, from June 2004 to
Sept. 2005. Microsatellite variations include the annealing temperature (Ta), number of alleles, observed
heterozygosity (Ho), expected heterozygosity (He), inbreeding coefficient (F), Hardy-Weinberg equilibrium
(H-W), and p value (p)

H-W
Locus Core Ta Sequence (5'-3') Allele size Number of Ho He F
sequence (°C) range (bp) alleles
p

MSMM-2a (CA)21 52 TAACCACAACCCCTCCAACTG 163-197 15 0.893 0.900 0.007 NS 0.627

TCATTTGGAGTTGCTGAGAAC
MSMM-3a (CA)15 52 TACGCCCTTCAAACTCATGTG 102-136 14 0.833 0.827 -0.008 NS 0.279
TCCTTTATCTTAGGTGATGGAG
MSMM-4a (CA)19 52 TGTTTCAAGGCAATAAGGTGG 143-187 18 0.913 0.872 -0.047 NS 0.512
TCGTTTCCCTGGAGATTGGG
MSMM-5a (CA)17 52 TCTAATACCCTCTTCCTTGGG 69-111 19 0.900 0.910 0.011 NS 0.247
TCCTATCAAGGGGCATTCATCT
MSMM-6a (CA)20 52 TCCTATCAAGGGGCATTCATCT 145-167 12 0.873 0.857 -0.019 NS 0.987
TACAAAGCCATTGTTCCCTGCT
MSMM-7a (CA)18 56 TAAGAAGGGCCACTAAGACCC 105-123 8 0.820 0.830 0.012 NS 0.915
TGGGATTAAAGGTGTGCACCA
MSMM-8a (CA)17 50 TGCTTAGTTCACTGCTGAACC 170-196 13 0.927 0.886 -0.046 NS 0.246
TCTTACTATCTGTCATTGAAGA
Moe1b (GT)18 60 TGGTTGTTCTGTGGTGAATACAG 93-133 15 0.847 0.890 0.049 NS 0.770
ACAGTAAGCAGTTTATCCACAAACC
Moe2b (GT)17 60 CATCTGATGAGTCCCTGAGG 145-185 15 0.887 0.889 0.002 NS 0.480
GCAACCTTCTTCTGACTTTTAC
Moe5b (TC)25 60 GGTCATGCTCCAAGAAGCTC 108-138 14 0.833 0.866 0.037 NS 0.663
AAAACCAAGGGTGCTGCTC
Moe6b (GT)25 60 GGTTTTCTGATTCAGGCAGG 210-246 14 0.913 0.998 -0.017 NS 0.420
CCTCTTCTGGCCTCTCCAG
a
Ishibashi et al. (1999). bVan de Zande et al. (2000). NS, non-significant.
 Wu et al. – Monogamy of Taiwan Voles 209

June-Aug. 2004 Mar.-May 2005 June-Aug. 2005

58F
41M 102M
41M 60M
60M 138F
80F 135F
60M
42F 25F 80F
44M
44M
44M
132F
8F 136F 41M
52M
34F
120M
91F

45M 109M 36F

36F 108M
101M
65F 50M
50M 72F 64F

Fig. 1. Overlapping home ranges of adult Microtus kikuchii at Hehuan Mt., Taiwan, during 3 consecutive reproductive periods.
Numbers indicate different individuals. Letters indicate the sex (M for males and F for females). Male home ranges are illustrated with
solid-line boundaries and lightly shaded interiors. Female home ranges have bold dotted boundaries. Only overlapping home ranges
between opposite sexes are shown.

Table 2. Parentage analysis of M. kikuchii at Hehuan Mt., Taiwan

Number of mismatched loci

Offspringa Date offspring captured Parentsa between offspring and parents LODb ΔLODc*
(female/male)

15M 2004 June 2F and 16M 1/1 8.06 4.84

4M 2004 June 17F and 13M 0/0 8.36 0.55
83F 2004 Sept. 0/0 9.64 7.79
5M 2004 June 10F and 50M 0/0 11.40 6.31
37M 2004 July 12F and 22M 0/0 9.66 7.26
42F 2004 July 24F and 35M 0/0 9.59 2.17
46M 2004 July 0/0 8.89 3.15
58F 2004 July 19F and 30M 0/0 8.03 3.03
43F 2004 July 48F and 49M 0/0 9.96 6.77
77M 2004 Sept. 65F and 54M 0/0 8.93 4.98
140F 2005 June 1/1 8.69 6.40
144M 2005 June 0/0 6.14 1.10
82F 2004 Sept. 74F and 90M 1/0 8.05 4.74
110F 2005 Mar. 0/0 7.39 1.64
70F 2004 Aug. 93F and 68M 1/0 9.07 0.46
81F 2004 Sept. 0/0 9.65 2.30
91F 2004 Oct. 0/0 10.20 2.91
102M 2005 Jan. 18F and 40M 0/0 6.84 4.62
111F 2005 Mar. 0/0 11.70 10.40
135F 2005 May 80F and 44M 0/0 12.80 9.21
136F 2005 May 0/0 8.86 4.20
146M 2005 June 91F and 109M 0/0 10.10 2.67
157M 2005 July 0/0 11.60 3.03
148F 2005 June 58F and 49M 1/1 8.45 6.70
150M 2005 June 72F and 101M 1/1 8.46 3.80
160F 2005 July 1/0 8.03 1.90
153M 2005 July 36F and 50M 0/0 9.96 5.52
158F 2005 July 25F and 60M 0/0 9.87 7.89
156F 2005 July 152F and 100M 0/0 10.50 5.17
164M 2005 Aug. 111F and 102M 0/0 6.47 1.08
167F 2005 Aug. 142F and 210M 0/0 10.50 8.23
a
Sex indicated by M (male) and F (female). bLOD score, log of product of likelihood ratios at each locus. cΔLOD, difference in LOD
score between the most likely candidate parent and the 2nd most likely candidate parent. *All ΔLOD values were significant (p < 0.05).
210 Zoological Studies 51(2): 204-212 (2012)
both sexes (Fig. 1). Eleven (78.6%) had overlaps
exclusively between 1 male and 1 female. Six of
these 11 social units (54.5%) were paired sexually
and monogamously produced offspring, and 5
social units were not detected to have produced
any offspring. For the 3 social units with > 2
individuals, only the social unit with 2 males and
2 females (June-Aug. 2004) was not detected to
have reproduced. The 2nd social unit (Mar.-May
2005) had a mated pair (44M-80F). The 3rd (June-
Aug. 2005) had a family group (parents 44M-80F
and 2 daughters: 135F and 136F). No sexual
pairing or family groups were found in June-Aug.
2004. In 2005, all social units found to reproduce (8
of 10, 80%) practiced monogamy and were either
mated pairs or family groups. In total, 8 (57.1%)
social units were found to have produced offspring.
Six (75%) of these had exclusive home-range
overlap between 1 male and 1 female.
DISCUSSION
Our results strongly support the hypotheses
that Microtus kikuchii is socially and genetically
monogamous. We identified litters sired by 18
males and 20 females. The 2 males that fathered
more than 1 litter did so in different years and with
a different mate because their 1st mate was no
longer present. None of the 9 litters with multiple
offspring had more than 1 father (Table 2). We
found no significant differences in home-range
sizes between sexes. Most (78.6%) social units
consisted of only 1 male and 1 female. Home-
range overlaps between male and female pairs not
found to produce offspring were relatively small
(44M-58F, 41M-80F, and 120M-136F), except for
60M and 80F. Therefore, overlap in home range
was mostly by a single male, a single female,
and their offspring, strongly suggesting social
monogamy.
Microtus kikuchii was suspected of being
socially monogamous based on home ranges
(Wu 1998) and observations of captive individuals
(Chen et al. 2006). Combining our home-range
data with a genetic analysis of parentage provided
more in-depth information on the relationships of
voles observed to have overlapping home ranges.
Microtus ochrogaster is considered a socially
monogamous species (Hofmann et al. 1984, Carter
et al. 1995) even though not all adults live as male-
female pairs. Adults live in groups of single males
and females, and groups of 3 or more adults were
also documented (Getz et al. 1993, Cochran and
Solomon 2000, Lucia et al. 2008). We maintain
that M. kikuchii is socially monogamous because
of similar home-range sizes between sexes, the
very high proportion of social units comprised of
male-female pairs, and the relatively low overlap
in home ranges of individuals without reproductive
relationships (e.g., not sexual partners or family
members).
Only six of the 11 male-female pairs iden-
tified by overlapping home ranges were found
to be paired partners that had successfully
produced young. Male-female pairs determined
by overlapping home ranges might not truly
be breeding pairs. In M. ochrogaster, socially
monogamous, multiple paternity in five of 9 litters
was also identified by a genetic study (Solomon
et al. 2004). Thus, data from home-range overlap
cannot reflect true pairing conditions or whether M.
kikuchii is genetically monogamous. To determine
the mating system of a species, field data and
genetic data are both necessary (Hughes 1998,
Kraaijeveld-Smit 2004). As we found no litters
sired by multiple fathers, genetic monogamy of
M. kikuchii should be assured. The parentage
analysis showed survival of 1 or 2 offspring in each
litter. This is consistent with Lu’s (1991) data from
field trapping (range 1-3, average litter size 2.1)
and our own observations from captive breeding
(litter size 1-3 with 2 most frequent) (pers. unpubl.
data). With an average litter size of 2.1, it may
be more difficult to detect multiple paternity in M.
kikuchii than for species with larger litter sizes,
such as M. ochrogaster. Low detectability of
multiple paternity due to small litter size is unlikely
for M. kikuchii because Lu (1991) found a very
low post-implantation mortality rate (2 resorbed
embryos of 64 embryos). Moreover, we are con-
fident that we trapped most of the population
because our extensive trapping effort spanned
2 breeding seasons, and we had high recapture
rates (71.5%-96.5%). Even so, we still found no
litters sired by multiple fathers.
Some studies are beginning to show that
in a number of species considered to be mono-
gamous, females mate with multiple males. In
mammals, extra-pair copulation was found in
some socially monogamous species (Richardson
1987, Agren et al. 1989, Solomon et al. 2004,
Mabry et al. 2011). Previously, only 2 known
rodent species simultaneously showed social and
genetic monogamy, i.e., Peromyscus polionotus
(Foltz 1981) and P. californicus (Ribble 1991).
Genetic promiscuity and polygyny are common in
Microtus (e.g., M. pennsylvanicus, M. richardsoni,
 Wu et al. – Monogamy of Taiwan Voles
M. xanthognathus, and M. californicus), but
monogamy is rare (Wolff 1985). Our study has
added M. kikuchii to the list of rodent species
simultaneously showing social and genetic
monogamy.
Acknowledgments: The authors thank the
anonymous reviewers for their critical comments
and many valuable remarks on the original
manuscript. We thank Q.W. Zhu, J.K. Lin, G.H.
Zhen, S.L. Yuan, and many other volunteers for
help with fieldwork. We thank M.Y. Zhen, S.F.
Zhen, Z.X. Zhang, L.Y. Liu, R.P. Huang, W.Y.
Zhen, and Z.Y. Guo for help with genetic work. We
thank the high-elevation experimental station of
the Taiwan Endemic Species Research Institute
for accommodations during fieldwork. Trapping
protocols complied with government regulations.
Permits were obtained from Taroko National Park.
This research was funded by the National Science
Council of Taiwan (NSC96-2621-B-029-001-MY3)
and Tunghai Univ.
REFERENCES
Agren G, Q Zhou, W Zhong. 1989. Ecology and social beha-
viour of Mongolian gerbils, Meriones unguiculatus, at
Xilinhot, Inner Mongolia, China. Anim. Behav. 37: 11-27.
Avise JC. 1994. Molecular markers, natural history and
evolution. New York: Chapman and Hall.
Back SR, LA Beeler, RL Schaefer, NG Solomon. 2002.
Testing functional hypotheses for the behavior of resident
pine voles, Microtus pinetorum, toward non-residents.
Ethology 108: 1023-1039.
Baker RJ, KD Makova, RK Chesser. 1999. Microsatellites
indicate a high frequency of multiple paternity in
Apodemus (Rodentia). Mol. Ecol. 8: 107-111.
Carter CS, AC De Vries, LL Getz. 1995. Physiological
substrates of mammalian monogamy: the prairie vole
model. Neurosc. Biobehav. Rev. 19: 303-314.
Chen Y-C, Z-Y Shieh, L-K Lin. 2006. Mating system of Taiwan
vole: evidence of behavioral studying. Chin. Biosci. 49:
51-73. (in Chinese)
Clutton-Brock TH. 1989. Mammalian mating systems. Proc. R.
Soc. Lond. Ser. B Biol. Sci. 236: 339-372.
Clutton-Brock TH, K Isvaran. 2006. Paternity loss in con-
trasting mammalian societies. Biol. Lett. 2: 513-516.
Cochran GR, NG Solomon. 2000. Effects of food supple-
mentation on the social organization of prairie voles
(Microtus ochrogaster). J. Mammal. 81: 746-757.
Conroy CJ, JA Cook. 2000. Molecular systematics of a
Holarctic rodent (Microtus: Muridae). J. Mammal. 81:
344-359.
Emlen ST, LW Oring. 1977. Ecology, sexual selection, and the
evolution of mating systems. Science 197: 215-223.
Escherich PC. 1981. Social biology of the bushy-tailed
woodrat, Neotoma cinerea. Univ. CA Publ. Zoology 110:
1-132.
211
ESRI. 1996. Arcview GIS Software. Redlands, CA: Enviro-
nmental Systems Research Institute (ESRI).
Finley RB. 1958. The woodrats of Colorado: distribution and
ecology. Univ. KS Publ. Mus. Nat. Hist. 10: 213-552.
Foltz DW. 1981. Genetic evidence for long-term monogamy
in a small rodent, Peromyscus polionotus. Am. Nat. 117:
665-675.
Gaulin SJC, RW FitzGerald. 1988. Home-range size as a
predictor of mating systems in Microtus. J. Mammal. 69:
311-319.
Getz LL, JE Hofmann. 1986. Social organization in free-
living prairie voles, Microtus ochrogaster. Behav. Ecol.
Sociobiol. 18: 275-282.
Getz LL, B Mcguire, T Pizzuto, JE Hofmann, B Frase.
1993. Social-organization of the prairie vole (Microtus
ochrogaster). J. Mammal. 74: 44-58.
Hall TA. 1999. BioEdit: a user-friendly biological sequence
alignment editor and analysis program for Window 95/998/
NT. Nucleic Acids Res. Symp. Ser. 41: 95-98.
Hoffmann RS, JW Koeppl. 1985. Zoogeography. In RH
Tamarin, ed. Biology of New World Microtus. Boston,
MA: American Society of Mammalogist, pp. 84-115.
Hofmann JE, LL Getz, L Gavish. 1984. Home range overlap
and nest cohabitation of male and female prairie voles.
Am. Midl. Nat. 112: 314-319.
Hooge PN, B Eichenlaub. 2000. Animal movement extension
to Arcview. ver. 2.0. Alaska Science Center - Biological
Science Office, U.S. Geological Survey, Anchorage, AK,
USA.
Hughes C. 1998. Integrating molecular techniques with field
methods in studies of social behavior: a revolution results.
Ecology 79: 383-399.
Ishibashi Y, Y Yoshinaga, T Saitoh, S Abe, H Iida, MC Yoshida.
1999. Polymorphic microsatellite DNA markers in the field
vole Microtus montebelli. Mol. Ecol. 8: 163-164.
Kleiman DG. 1977. Monogamy in mammals. Q. Rev. Biol. 52:
39-69.
Kraaijeveld-Smit F. 2004. Behavioural ecology. In TJC
Beebee, G Rowe, eds. An introduction to molecular
ecology. Oxford, UK: Oxford Univ. Press, pp. 81-109.
Lu MS. 1991. Ecology and reproduction of Microtus kikuchii
in the Hohuanshan area. Master’s thesis, Tunghai Univ.,
Taichung, Taiwan. (in Chinese with English abstract)
Lucia KE, B Keane, LD Hayes, YK Lin, RL Schaefer, NG
Solomon. 2008. Philopatry in prairie voles: an evaluation
of the habitat saturation hypothesis. Behav. Ecol. 19:
774-783.
Mabry KE, CA Streatfeild, B Keane, NG Solomon. 2011.
avpr1a length polymorphism is not associated with either
social or genetic monogamy in free-living prairie voles.
Anim. Behav. 81: 11-18.
Ohnishi N, T Saitoh, Y Ishibashi. 2000. Spatial genetic
relationships in a population of the Japanese wood mouse
Apodemus argenteus. Ecol. Res. 15: 285-292.
Patris B, C Baudoin. 2000. A comparative study of parental
care between two rodent species: implications for
the mating system of the mound-building mouse Mus
spicilegus. Behav. Proc. 51: 35-43.
Peakall R, PE Smouse. 2006. GENALEX 6: genetic analysis
in Excel. Population genetic software for teaching and
research. Mol. Ecol. Notes 6: 288-295.
Queller DC, JE Strassmann, CR Hughes. 1993. Microsatellite
and kinship. Trends Ecol. Evol. 8: 285-288.
Raymond M, F Rousset. 1995. An exact test for population
212 Zoological Studies 51(2): 204-212 (2012)
differentiation. Evolution 49: 1280-1283.
Reichard UH. 2003. Monogamy: past and present. In UH
Reichard, C Boesch, eds. Monogamy. Cambridge, UK:
Cambridge Univ. Press, pp. 3-25.
Ribble DO. 1991. The monogamous mating system of
Peromyscus californicus as revealed by DNA finger-
printing. Behav. Ecol. Sociobiol. 29: 161-166.
Richardson PRK. 1987. Aardwolf mating system: overt
cuckoldry in an apparently monogamous mammal. S. Afr.
J. Sci. 83: 405-410.
Schlotterer C, B Harr. 2001. Microsatellite instability. In
Encyclopedia of life sciences (eLS). NJ: J. Wiley & Sons,
Ltd., pp. 1-4. (http://www.els.net)
Scribner KT, JM Pearce. 2000. Microsatellite: evolutionary and
methodological background and empirical applications
at individual, population and phylogenetic levels. In
AJ Baker, ed. Molecular methods in ecology. Berlin:
Blackwell Science, pp. 235-273.
Seaman DE, JJ Millspaugh, BJ Kernohan, GC Brundige, KJ
Raedeke, RA Gitzen. 1999. Effects of sample size on
kernel home range estimates. J. Wildl. Manage. 63: 739-
747.
Slate J, T Marshall, J Pemberton. 2000. A retrospective
assessment of the accuracy of the paternity inference
program CERVUS. Mol. Ecol. 9: 801-808.
Solomon NG. 1993a. Body size and social preferences of
male and female prairie voles, Microtus ochrogaster.
Anim. Behav. 45: 1031-1033.
Solomon NG. 1993b. Comparison of parental behavior in male
and female prairie voles (Microtus ochrogaster). Can. J.
Zool. Rev. Can. Zool. 71: 434-437.
Solomon NG, B Keane, LR Knoch, PJ Hogan. 2004. Multiple
paternity in socially monogamous prairie voles (Microtus
ochrogaster). Can. J. Zool. 82: 1667-1671.
Swilling WR, MC Wooten. 2002. Subadult dispersal in a
monogamous species: the Alabama beach mouse
(Peromyscus polionotus ammobates). J. Mammal. 83:
252-259.
Topping MG, JS Millar. 1998. Mating patterns and reproductive
success in the bushy-tailed woodrat (Neotoma cinerea) as
revealed by DNA fingerprinting. Behav. Ecol. Sociobiol.
43: 115-124.
Van de Zande L, RC van Apeldoorn, AF Blijdenstein, D de
Jong, W van Delden, R Bijlsma. 2000. Microsatellite
analysis of population structure and genetic differentiation
within and between populations of the root vole, Microtus
oeconomus in the Netherlands. Mol. Ecol. 9: 1651-1656.
Wittenberger JF. 1979. The evolution of mating systems in
birds and mammals. In P Marler, JG Vandenbergh, eds.
Handbook of behavioral neurobiology. New York: Plenum,
pp. 271-349.
Wolff JO. 1985. Behavior. In RH Tamarin, ed. Biology of
New World Microtus. Boston, MA: American Society of
Mammalogists, pp. 340-372.
Wu PJ. 1998. Population ecology and activity pattern of
Kikuchii’s field vole (Microtus kikuchii) in Hohuanshan
area. Master’s thesis, Tunghai Univ., Taichung, Taiwan. (in
Chinese with English abstract)
Zoological Studies 51(2): 213-221 (2012)

A New Shallow-Water Species, Polycyathus chaishanensis sp. nov.

(Scleractinia: Caryophylliidae), from Chaishan, Kaohsiung, Taiwan
Mei-Fang Lin1, Marcelo V. Kitahara2, Hiroyuki Tachikawa3, Shashank Keshavmurthy1, and Chaolun
Allen Chen1,4,5,6,*
1
Biodiversity Research Center, Academia Sinica, Nangang, Taipei 115, Taiwan
2
Comparative Genomic Centre and ARC Centre of Excellence for Coral Reef Studies, James Cook Univ., Townsville 4810, Australia
3
Natural History Museum and Institute, Chiba 955-2, Japan
4
Institute of Oceanography, National Taiwan Univ., Taipei 106, Taiwan
5
Department of Life Science, National Taitung Univ., Taitung 950, Taiwan
6
Taiwan International Graduate Program (TIGP)- Biodiversity, Academia Sinica, Nangang, Taipei 115, Tawian

(Accepted October 12, 2011)

Mei-Fang Lin, Marcelo V. Kitahara, Hiroyuki Tachikawa, Shashank Keshavmurthy, and Chaolun Allen
Chen (2012) A new shallow-water species, Polycyathus chaishanensis sp. nov. (Scleractinia: Caryophylliidae),
from Chaishan, Kaohsiung, Taiwan. Zoological Studies 51(2): 213-221. A small population of a new species of
zooxanthellate scleractinian coral, Polycyathus chaishanensis sp. nov., is described from shallow water (< 3 m)
off Chiashan, Kaohsiung, an uplifted Pleistocene reef located on the southwest coast of Taiwan. Polycyathus
chaishanensis sp. nov. is a zooxanthellate coral associated with Symbiodinium C1 and forms small encrusting
colonies. Polycyathus chaishanensis sp. nov. differs from other Polycyathus by having (1) the smallest corallites
(2.0-3.7 mm in calicular diameter) reported in the genus Polycyathus; (2) septa hexamerally arranged in 4
incomplete cycles displaying dentate or laciniate axial edges; (3) crispate and well-developed pali before the
secondary septa; and (4) light brown pigmented pali/columellar elements. When expanded, vivid-red to brown
polyps rise considerably above the calice, and long and slender tentacles are covered with white nematocyst
batteries. Polycyathus chaishanensis is the only species of Polycyathus known from Taiwanese waters and
appears to be endemic to a small region at Chaishan. The small population of this new species raises concerns
as to its vulnerability to natural and anthropogenic threats.
http://zoolstud.sinica.edu.tw/Journals/51.2/213.pdf

Key words: Scleractinia, Polycyathus chaishanensis, Zooxanthellae, Chaishan, Shallow water.

D escribed from specimens collected near
St. Helena, in the South Atlantic Ocean, the
genus Polycyathus Duncan, 1876 (Anthozoa:
Scleractinia: Caryophylliidae) is characterized
by small reptoid to plocoid colonies that form
through corallites that grow close to the base of
their neighbors and become sparser with age.
The corallites are cylindrical to slightly conical
in shape, bud from a common coenosteum or
from stolons (Cairns 1995), and are epithecated.
There are 3-5 irregularly arranged septal cycles,
of which the last is usually incomplete, and the
1st and 2nd are the most distinct and exsert. Two
crowns of well-developed pali (P1 and P2) are
present before the 2nd and 3rd septal cycles, of
which P2 is usually more difficult to distinguish
from columellar elements than P3 (Wijsman-Best
1970). The fossa is deep and contains a papillose
columella. According to Duncan (1876), septa that
are not incised and the absence of endotheca are
diagnostic characters of this genus.
Ranging from shallow to waters deeper

*To whom correspondence and reprint requests should be addressed. Tel: 886-2-27899549. Fax: 886-2-28958059.
E-mail:cac@gate.sinica.edu.tw

213
214 Lin et al. – New Scleractinian Coral From Taiwan
than 400 m (Cairns 1999), the vast majority of
Polycyathus representatives are reported from the
Pacific Ocean (Fig. 1), of which 5 are known to
occur in southern Pacific waters (P. verrilli Duncan
1876, P. octuplus Cairns 1999, P. fulvus Wijsman-
Best 1970, P. norfolkensis Cairns 1995, and P.
andamanensis Alcock 1893). In the northwestern
Pacific, 3 Polycyathus species are described from
the Philippines, in waters deeper than 35 m (Verheij
and Best 1987). Among them, P. hodgsoni Verheij
& Best 1978 and P. marigondoni Verheij & Best
1978 have the lowest and highest number of septal
cycles (3 and 5, respectively) compared to their
congeners.
In the present study, a new species of
Polycyathus is described. This new species
inhabits a shallow-water area of Chaishan, an
uplifted reef developed about 0.6 Mya (Fig. 2).
Chaishan is about 6 km long and is home to
about 15.62% of coastal habitats of Kaohsiung
City, southwestern Taiwan (CPAMI 2008). The
beach at Chaishan is composed of scattered hard
substrates of carbonaceous rocks of various sizes
which originated from nearby coastal hills. The
water column contains a high concentration of
particles which increases the turbidity of the water
and might be one of the contributing factors to
the low number of scleractinian corals reported
in this area. Nonetheless, the new species of
Polycyathus described herein appears to be
endemic to this small Taiwanese region, as so far,
it has not been found anywhere else in Taiwan.
30°N
0° PACIFIC OCEAN
0 - 10 m
30°S 11 - 50 m
51 - 100 m ATLANTIC OCEAN
> 100 m
150°W 90°W
Fig. 1. Worldwide distribution and depths of Polycyathus spp.
Mitochondrial (mt) 16S ribosomal (r)RNA
gene sequences were amplified and aligned
with previously published sequences from 8
representatives of morphologically related
caryophylliid genera (including P. muellerae
Abel 1959) and 13 representatives of non-
caryophylliid families to investigate the validity of
this genus. Following Kitahara et al. (2010a b),
the phylogenetic analysis did not indicate that the
Caryophylliidae is a monophyletic family, and also
raises concerns about the validity of Polycyathus,
which is one of the less-understood scleractinian
genera.
MATERIALS AND METHODS
Specimens examined in the present study
were collected by snorkeling in 2000, 2005, and
2008 from a tidal pool (< 3 m in depth) at Chaishan,
Kaohsiung, Taiwan (22°38'18"N; 120°15'19"E) (Fig.
2). Colonies were photographed in situ using an
Olympus SP350 camera (Center Valley, PA, USA)
with an underwater housing. Collected specimens
were bleached to remove soft tissues, rinsed with
fresh water, thoroughly dried, and photographed
using a Nikon D200 (Tokyo, Japan) camera.
Morphological observations were carried out using
an Olympus SZ-ST stereomicroscope equipped
with an ocular graticule. Scanning electron
microscopy (SEM) was performed on a FEI Quanta
200/Quorum PP2000TR FEI, 2007 (Hillsboro, OR,
PACIFIC OCEAN
INDIAN OCEAN
°W 0°E 60°E 120°E
 Zoological Studies 51(2): 213-221 (2012) 215

N (A) (B)

24°N
Kaohsiung City TAIWAN

22°N

(C) (D)
n

120°E 122°E
i-sha
Cha

(E) (F)

Taiwan Strait

Fig. 2. Map of sampling localities of Polycyathus chaishanensis sp. nov. (A) Landscape of an uplifted coral reef; (B) patches of
limestone dominated by Ulva sp., chiton, and barnacles; (C) ancient coral; (D) Anthopleura sp.; (E) Psammocora sp.; (F) Porites
okinawanensis.

(A) (B)

(C) (D)

1 cm 2 mm

Fig. 3. Polycyathus chaishanensis sp. nov. (A) Caulerpa racemosa and calcified red algae; (B) a specimen with brown tissues
indicating the presence of zooxanthellae; (C) colony view of the holotype (NMNS-6309-001) consisting of 73 corallites in different
stages of development; (D) calicular view (SEM) of 1 corallite of the holotypic colony (NMNS-6309-001).
216 Lin et al. – New Scleractinian Coral From Taiwan
USA) instrument.
Skeleton vouchers were deposited at the
National Museum of Natural Science (NMNS),
Taichung, Taiwan and at the Museum of Tropical
Queensland (MTQ), Townsville, Australia. In
the morphological description, the following
abbreviations were used: CD, calicular diameter;
GCD, great CD; Sx, septa of the x order; Px, pali
of the x order; and H, height. Tissue samples
preserved in CHAOS solution (Fukami 2004) were
used for DNA extraction.
Symbiodinium identification
Following LaJeunesse (2002), denaturing
gradient gel electrophoresis (DGGE) of the internal
transcribed spacer (ITS)-2 region was performed
to identify the Symbiodinium clade present in P.
chaishanensis sp. nov. The ITS-2 region was
amplified using primers ITS2 clamp and ITSintfor
2 developed by LaJeunesse and Trench (2000). A
polymerase chain reaction (PCR) was performed
with a touch-down cycle according to LaJeunesse
(2002). PCR products were subjected to elec-
trophoresis for 15-16 h on denaturing gradient gels
(45%-80%) using a CBS Scientific System (Del
Mar, CA, USA). Gels were stained with SYBR
green (Molecular Probes, Eugene, OR, USA) for
20 min, and photographed for further analysis.
Bands were excised from the gel and sent for
direct sequencing. Resulting sequences were
deposited in the NCBI database (with accession
nos.: 180016-180021)
Sequence analysis and phylogeny
Forty mt16S rDNA and the cytochrome c
oxidase subunit I (COI) sequences, including
these 2 regions from the complete mt genome of
P. chaishanensis sp. nov (Lin et al. 2011), were
retrieved from GenBank. This dataset contained
11 robust and 4 complex scleractinian families.
Phylogenetic analyses were performed using
MEGA 4.0 (Tamura et al. 2007) for Neighbor-
joining (NJ) and MrBayes 3.1.2 (Huelsenbeck
and Ronquist 2001) for Bayesian inference (BI).
The most appropriate model of nucleotides was
determined to be HKY+I using MrModeltest vers.
2.3 (Nylander 2004). The NJ analyses were
performed with 500 replicates, and for the BI, 2
runs each of 10 6 generations were calculated
for each marker with topologies saved every
100 generations. The 1st quarter of the saved
topologies were discarded as burn-in, and the
remaining ones were used to calculate posterior
probabilities.
RESULTS
Systematic description
Subclass Hexacorallia.
Order Scleractinia Bourne, 1900.
Suborder Caryophylliina Vaughan & Wells, 1943.
Family Caryophylliidae Dana, 1846.
Genus Polycyathus Duncan, 1876.
Polycyathus chaishanensis sp. nov.
Illustrations of the holotype are given in figures 3C, D, 4A-C;
and illustrations of the paratype are given in figure 4D, E.
Materials examined: Holotype: NMNS-6309-
001 (Taichung, Taiwan). Paratypes: NMNS-6309-
002, NMNS-6309-003 (Taichung, Taiwan), and
MTQ G64703 (Queensland, Australia, 1 specimen).
Type locality: 22°38'18''N, 120°15'19''E (Taiwan),
3 m in depth.
Description: Small reptoid colonies formed by
closely spaced cylindrical corallites arising from a
common coenosteum or from stolons. Holotypic
colony consisting of approximately 70 corallites.
Extratentacular budding common; however, some
corallites displaying intratentacular division. Calice
circular to slightly elliptical. Largest corallite
examined 3.65 × 3.73 mm in CD and 4.0 mm in
H. Theca thick. Costae more prominent near
calicular edge. All costae equal in width (about
0.21 mm wide), slightly convex, and bearing low,
coarse granules. Intercostal striae deep and
flat near calicular edge, becoming less distinct
in direction of base. Coenosteum and theca
white, but columellar elements usually light-brown
pigmented. Vivid-red to dark brown sub-pellucid
polyps considerably expanded above calicular
edge; tentacles long, slender, with knobby end,
and covered by small white verruca.
Septa hexamerally arranged in 4 incomplete
cycles, according to formula: S1 ≥ S2 > S3 > S4.
Corallites < 2 mm in GCD with 12 or fewer septa,
but larger corallites (up to 3.7 mm in GCD) with
several pairs of S4 totaling up to 34 septa. S1
exsert (0.5-0.7 mm), with straight and almost-
vertical axial edges sometimes bearing small,
cylindrical (0.24 mm in diameter) palus. S2 only
slightly less exsert and equal or narrower than
S1. S3 less exsert, thinner, and about 2/3 width
of S2. Axial edges of S1-S2 dentate, those of S3
laciniated. S4 1/2-2/3 width of S3. Well-developed
P3 (sometimes bilobated) present before S3. If
 Zoological Studies 51(2): 213-221 (2012) 217

(A)

1 mm

(B) (C)

1 mm 2 mm

(D) (E)

500 μm 1 mm

Fig. 4. (A) Calicular view of 1 corallite of the holotypic colony (NMNS-6309-001) undergoing extratentacular budding; (B) calicular
view of 1 corallite of the holotypic colony (NMNS-6309-001) undergoing intratentacular budding; (C) calicular view of 1 corallite of the
holotypic colony (NMNS-6309-001); (D) lateral view of a corallite from the paratype colony (NMNS-6309-002); (E) detail of columellar
elements MTQ G64703.
 218 Lin et al. – New Scleractinian Coral From Taiwan
present, P2 difficult to distinguish from columellar
elements. Septal and palar faces bearing several
pointed granules aligned perpendicular to septal/
palar edges. Fossa moderately deep, containing
elongate papillose columella. Columella composed
of 5-7 slender, irregularly shaped rods.
Remarks: Polycyathus chaishanensis sp. nov.
differs from all other known species of this genus
by having a much smaller corallite. Twenty-one
corallites examined from the holotype colony had
a mean CD of 3.05 ± 0.26 mm (Fig. 5), whereas
corallites among the other 18 valid Polycyathus
species are significantly larger (mean CD of
4.38 ± 1.10 mm). In addition, P. chaishanensis
sp. nov. has one of the shallowest bathymetric
ranges known from representatives of this genus
(≤ 3 m) (Fig. 1), and all colonies were found to
inhabit tidal pools. Of the 18 extant Polycyathus
species, 3 were described from the Atlantic Ocean
(P. atlanticus Duncan, 1876 [depth unknown], P.
senegalensis Chevalier 1966 [12-143 m], and
P. mayae Cairns 2000 [110-309 m]; 5 from the
Indian Ocean (P. persicus Duncan 1876 [depth
unknown], P. fuscomarginatus Klunzinger 1879
[depth unknown], P. verrilli [depth unknown], P.
difficilis Duncan 1889 [depth unknown], and P.
andamanensis [depth unknown]); 1 species from
the Mediterranean Sea (P. muellerae Abel 1959
[10-32 m]); and according to Cairns (1999), 9
8.00
7.00
Calicular Diameter (mm)
6.00
5.00
4.00
3.00
2.00
1.00
P. chaishanensis Polycyathus spp.
Fig. 5. Measurement of the calicular diameter (CD) of P.
chaishanensis sp. nov. (21 corallites) and extant Polycyathus
species (18 species). The CD of each P. chaishanensis
corallite and its congeners are indicated by black circles in
the box plot. The non-parametric Wilcoxon-Mann-Whitney
rank sum test showed no significant difference (p = 0.1135)
in calicular diameters between P. chaishanensis sp. nov. and
extant Polycyathus species.
species are known from Pacific waters (P. palifera
Verrill 1869 [reef depth], P. hondaensis (Durham
& Barnard 1952) [55-64 m], P. fulvus [30-50 m], P.
isabela Wells, 1982 [14-23 m], P. hodgsoni [> 35 m];
P. marigondoni [35 m]; P. furanaensis Verheij &
Best 1987 [6-52 m], P. norfolkensis [10-20 m], and
P. octuplus [90-441 m]).
Among Pacific and Indian congeners that
have small corallites, P. chaishanensis sp. nov. is
most similar to P. difficilis (Mergui Archipelago).
Both species have an exserted S1, indistinct
P1, and S2 and S3 with dentate/laciniate axial
edges. However, P. chaishanensis sp. nov. differs
in having 4 incomplete cycles of septa, while P.
difficilis has 3 cycles of septa.
Interestingly, DGGE from the ITS-2 confirmed
the presence of Symbiodinium subclade C1
associated with P. chaishanensis sp. nov. Although
Wijsman-Best (1970) described the association
of zooxanthellae with P. fulvus, to date, all other
representatives of this genus are considered
azooxanthellate (Cairns et al. 1999). However, to
reinvestigate this important ecological aspect of
shallow-water Polycyathus, new samples enabling
the examination of their tissue must be collected.
Etymology: This species is named for the
uplifted reef in southern Taiwan (Chaishan) from
which it was collected and to which it is possibly
endemic.
Distribution: Known only from the sublittoral
zone (< 3 m deep) near Chaishan, Kaohsiung,
Taiwan (22°37'13"N, 120°15'56"E to 22°38'18"N,
120°15'19"E).
DISCUSSION
Phylogeny of Polycyathus
To test the hypothesis that Polycyathus
is a natural genus, a 16S rRNA sequence
was extracted from the P. chaishanensis sp.
nov. mt genome (accession no.: NC 015642;
Lin et al. 2011) and aligned with previously
published sequences from 8 representatives
of morphologically related caryophylliid genera
and 13 representatives of non-caryophylliid
families. Results of the phylogenetic analysis are
summarized in figure 6, and following Romano
and Cairns (2000), Le-Goff Vitry et al. (2004),
and Fukami et al. (2008), sequences from 4
scleractinian species in the “complex” coral clade
were used as an outgroup. Despite the fact that
only 2 Polycyathus species were represented in
 Zoological Studies 51(2): 213-221 (2012) 219

Colpophyllia natans
Faviidae
86/64 Favia fragum
Lobophyllia hemprichii
75/78 Mussiidae
Cynarina sp.
100/96
Hydnophora rigida Merulinidae
95/88
100/96 Pectinia alcicornis Pectiniidae
Caulastraea furcata Faviidae
Trochocyathus efateensis
76/67 Caryophylliidae
Tethocyathus virgatus
100/94
Dichocoenia stokesi Meandrinidae
Phyllangia mouchezii Rhizangiidae
Rhizosmilia maculata
100/95 Caryophylliidae
Paracyathus pulchellus
88/71 74/70
Cladocora caespitosa Faviidae
69/71
Oculina patagonica Oculinidae
77/87
Astrangia sp. Rhizangiidae
Polycyathus muellerae Caryophylliidae
Polycyathus chaishanensis Caryophylliidae
Coscinaraea sp. Siderastreidae
98/99
Zoopilus echinatus Fungiidae
56/58
Psammocora stellata Siderastreidae
Leptastrea bottae Faviidae
Fungia scutaria
Fungiidae
Fungia vaughani
Caryophyllia grayi
Caryophyllia atlantica
Caryophyllia diomedeae
100/100 69/69
Crispatotrochus rugosus
Caryophyllia planilamellata
Caryophyllia rugosa Caryophylliidae
Caryophyllia unicristata
Dasmosmilia lymani
71/73
Caryophyllia scobinosa
Caryophyllia grandis
100/95 Caryophyllia transversalis
Madracis mirabilis
100/100 Pocilloporidae
Pocillopora damicornis
Porites porites

100/100 Acropora tenuis

Complex corals
Siderastrea radians
Fungiacyathus stephanus

0.01

Fig. 6. Phylogenetic analyses based on Bayesian inference and Neighbor-joining analyses of the partial mitochondrial sequence of
the 16S rRNA gene and the cytochrome oxidase subunit I gene from 41 scleractinian species. Numbers at the nodes correspond
to Bayesian posterior probabilities and bootstrap support of the Neighbor-joining analysis, respectively. The scale unit is 0.01
substitutions per site.
220 Lin et al. – New Scleractinian Coral From Taiwan
the analysis, both the BI and NJ analyses indicated
that this genus is not monophyletic. Polycyathus
chaishanensis sp. nov. was not grouped with
any other congener (P. muellerae) or any other
caryophylliid representative. Instead, our results
show that P. chaishanensis sp. nov. has a genetic
immediacy to some representatives of the
Siderastreidae (Coscinaraea and Psammocora),
Fungiidae (Zoopilus and Fungia), and Faviidae
(Leptastrea) (Fig. 6). In addition, our results
support P. muellerae having a close relationship
with Paracyathus pulchellus (Kitahara et al. 2010b)
but not with Rhizosmilia maculata. These results
were also supported by the COI sequence data
(data not shown).
In previous molecular studies, many of the
morphologically defined families, especially those
composed of zooxanthellate species, showed
extensive polyphyly (Romano and Cairns 2000,
Le Goff-Vitry et al. 2004, Fukami et al. 2008,
Kitahara et al. 2010a). In an attempt to clarify
the validity of morphology-based taxonomy,
additional taxon sampling, more-comprehensive
morphological analyses, and additional molecular
data are required (Fukami et al. 2008). Therefore,
molecular data from other Polycyathus species are
needed to clarify the phylogenetic status of this
genus.
Ecology of Polycyathus chaishanensis sp. nov.
The rare distribution and the small-sized
population of this new species raise several
concerns as to its vulnerability to natural and
anthropogenic threats, in a period of intense urban
development at Chaishan.
Chaishan is an uplifted reef formed during the
late Pleistocene (2.59-0.01 Mya; Gong et al. 1998).
The Pleistocene reef limestone in southwestern
Taiwan occurs in the Gutingkeng Formation near
Kaohsiung (Gong et al. 1998). A debris avalanche
and sandy substrate form the main characteristics
of the Chaishan area and have contributed to
the benthic communities of this region. Among
hermatypic organisms reported from Chaishan’s
formation, the most important are scleractinian
corals (such as Acropora, Porites, Favia, and
Favites), mollusks, and encrusting calcareous
red algae (Gong et al. 1998). Hard surfaces
exposed to light in the Chaishan area were found
to be heavily dominated by algae, primarily the
green algae Ulva fasciata and U. lactuca, and
some turf algae such as Chaetomorpha antennina
(Huang 2003). Colonial zooxanthellate corals,
Psammocora sp. and Porites sp., were found on
limestone or among fleshy algae. However, most
of these scleractinian species were found in tidal
pools of < 5 m deep. Reefs in shallow water with
less light are usually dominated by zoanthus and
sea anemones, probably including Anthopleura sp.
(Fig. 2). In addition, overhangs and overhanging
surfaces with less light are primarily dominated by
encrusting sponges. Polycyathus chaishanensis
sp. nov. was only found on well-lit reefs dominated
by green and encrusting calcareous red algae, and
was generally rare (Fig. 2). However, this area
is dominated by green algae and turbid waters
caused by erosion, which may have inhibited the
occurrence of most other scleractinians. The small
population of this new species raises concerns as
to its vulnerability to natural and anthropogenic
threats.
Acknowledgments: We thank Dr. K. Soong
(National Sun-Yat Sen Univ., Kaohsiung, Taiwan)
for ecological information on the Chaishan area,
Mr. L.C. Wang (National Taiwan Univ., Taipei,
Taiwan) for assistance with the SEM technology,
and Dr. Y. Nozawa for help with photography.
Constructive comments from the members of the
Coral Reef Evolutionary and Ecological Genetics
(CREEG) Laboratory, Biodiversity Research
Center, Academia Sinica (BRCAS) and 3 anony-
mous reviewers are especially appreciated. SK
is the recipient of a postdoctoral fellowship from
Academia Sinica (2010-2012). This study was
supported by a BRCAS Thematic grant (2006-
2008) and one from the National Science Council,
Taiwan (NSC94-2621-B-001-005) to CAC. This is
the CREEG Laboratory contribution no. 73.
REFERENCES
Cairns SD. 1995. The marine fauna of New Zealand:
Scleractinia (Cnidaria: Anthozoa). NZ Oceanogr. Inst.
Mem. 103: 1-216.
Cairns SD. 1999. Cnidaria, Anthozoa: deep-water
azooxanthellate Scleractinia from Vanuatu, and Wallis and
Futuna Islands. Mém. Mus. natl. d'Hist. nat. 180: 31-167.
Cairns SD, BW Hoeksema, J van der Land. 1999. Species
richness of recent Scleractinia. Atoll Res. Bull. (National
Museum of Natural History Smithsonian Institution
Washington, DC, USA) 459: 1-46.
CPAMI. 2008. The program of sustainable coastal deve-
lopment. Taipei, Taiwan: Construction and Planning
Agency Ministry of the Interior (CPAMI) Press. (in
Chinese)
Duncan PM. 1876. Notices of some deep-sea and littoral
corals from the Atlantic Ocean, Caribbean, Indian, New
 Zoological Studies 51(2): 213-221 (2012)
Zealand, Persian Gulf, and Japanese etc. seas. Proc.
Zool. Soc. Lond. 44: 428-442.
Fukami H, AF Budd, DR Levitan, J Jara, R Kersanach, N
Knowlton. 2004. Geographic differences in species
boundaries among members of the Montastraea annularis
complex based on molecular and morphological markers.
Evolution 58: 324-337.
Fukami H, CA Chen, AF Budd, A Collins, C Wallace, YY Chuang
et al. 2008. Mitochondrial and nuclear genes suggest
that stony corals are monophyletic but most families of
stony corals are not (order Scleractinia, class Anthozoa,
phylum Cnidaria). PLoS One 3: e3222.
Gong SY, SW Wang, TY Lee. 1998. Pleistocene coral reefs
associated with claystones, southwestern Taiwan. Coral
Reefs 17: 215-222.
Huang SF. 2003. Seaweeds of northeastern Taiwan. Taipei,
Taiwan: National Taiwan Museum Press. (in Chinese)
Huelsenbeck JP, F Ronquist. 2001. MrBayes: Bayesian
inference of phylogeny. Bioinformatics 17: 754-755.
Kitahara MV, SD Cairns, DJ Miller. 2010a. Monophyletic
origin of Caryophyllia (Scleractinia, Caryophylliidae), with
descriptions of six new species. Syst. Biodivers. 8: 91-
118.
Kitahara MV, SD Cairns, J Stolarski, D Blair, DJ Miller. 2010b.
A comprehensive phylogenetic analysis of the Scleractinia
(Cnidaria, Anthozoa) based on mitochondrial CO1
sequence data. PLoS One 5: e11490.
LaJeunesse TC. 2002. Diversity and community structure of
symbiotic dinoflagellates from Caribbean coral reefs. Mar.
221
Biol. 141: 387-400.
LaJeunesse TC, RK Trench. 2000. The biogeography of two
species of Symbiodinium (Freudenthal) inhabiting the
intertidal anemone, Anthopleura elegantissima (Brandt).
Biol. Bull. 199: 126-134.
Le Goff-Vitry MC, AD Rogers, D Baglow. 2004. A deep-sea
slant on the molecular phylogeny of the Scleractinia. Mol.
Phylogen. Evol. 30: 167-177.
Lin MF, KS Luzon, WY Licuana, MC Ablan-Lagman, CA Chen.
2011. Seventy-four universal primers for characterizing
the complete mitochondrial genomes of scleractinian
corals (Cnidaria; Anthozoa). Zool. Stud. 50: 513-524.
Nylander JAA. 2004. MrModeltest v2. Program distributed by
the author. Evolutionary Biology Centre, Uppsala Univ.
Romano SL, SD Cairns. 2000. Molecular phylogenetic
hypotheses for the evolution of scleractinian corals. Bull.
Mar. Sci. 67: 1043-1068.
Tamura K, M Dudley, M Nei, S Kumar. 2007. MEGA4:
Molecular Evolutionary Genetics Analysis (MEGA) soft-
ware vers. 4.0. Mol. Biol. Evol. 24: 1596-1599.
Verheij E, MB Best. 1987. Notes on the genus Polycyathus
Duncan, 1876 and a description of three new scleractinian
corals from the Indo-Pacific. Zool. Mededelingen 61: 147-
154.
Wijsman-Best M. 1970. A new species of Polycyathus
Duncan, 1876, from New Caledonia and a new record of
Polycyathus senegalensis Chevalier, 1966 (Madreporaria).
Beaufortia 227: 79-84.
Zoological Studies 51(2): 222-231 (2012)

Systematic Study of the Simocephalus Sensu Stricto Species Group

(Cladocera: Daphniidae) from Taiwan by Morphometric and Molecular
Analyses
Shuh-Sen Young1,*, Mei-Hui Ni2, and Min-Yun Liu3
1
Department of Applied Science, National Hsinchu University of Education, Hsinchu 300, Taiwan
2
Hsinchu Municipal Hsinchu Elementary School, Hsinchu 300, Taiwan
3
National Applied Research Laboratories, Taiwan Ocean Research Institute, Taipei 106, Taiwan

(Accepted September 9, 2011)

Shuh-Sen Young, Mei-Hui Ni, and Ming-Yun Liu (2012) Systematic study of the Simocephalus sensu stricto
species group (Cladocera: Daphniidae) from Taiwan by morphometric and molecular analyses. Zoological
Studies 51(2): 222-231. There is some controversy regarding the traditional taxonomy of the Simocephalus
sensu stricto species group. We conducted molecular and morphometric analyses to differentiate the 3 species
from this group found in Taiwan: S. vetulus (O.F. Müller, 1776), S. vetuloides Sars, 1898, and S. mixtus Sars,
1903. The landmark method was employed, followed by a transfer into 24 characteristic values for a principal
component analysis (PCA), the results of which indicated morphometric overlap among these species. The
dorsal angle, brood size, and body length were smallest in S. vetulus, medium in S. vetuloides, and largest
in S. mixtus. In the Simocephalus sensu stricto group from Taiwan, the dorsal angle and body length were
significantly correlated with brood size in a quadratic manner. In the molecular analysis, 98 specimens of
Simocephalus were used, and the 641-bp mitochondrial DNA cytochrome oxidase subunit 1 sequence was
employed as a marker to analyze the genetics of S. vetulus, S. vetuloides, S. mixtus, S. serrulatus (Koch, 1841),
and S. heilongjiangensis Shi and Shi, 1994. Simocephalus vetulus, S. vetuloides, and S. mixtus shared several
haplotypes, and the interspecific genetic distance was merely 0.00671-0.00785, which is within the range of
intraspecific differences. We concluded that S. vetulus, S. vetuloides, and S. mixtus in Taiwan belong to the
same species and should be treated as S. cf. vetulus. The number of species of Simocephalus in Taiwan is thus
reduced to 3: S. cf. vetulus, S. serrulatus, and S. heilongjiangensis.
http://zoolstud.sinica.edu.tw/Journals/51.2/222.pdf

Key words: Systematics, Biodiversity, Simocephalus, Freshwater zooplankton.

T he general morphologies of Simocephalus 1998, Orlova-Bienkowskaja 2001, Tuo 2002).

vetulus (O.F. Müller, 1776), S. vetuloides Sars
1898, and S. mixtus Sars 1903 are very similar.
Sars (1916) first discriminated S. vetulus and
S. vetuloides based on the dorsoposterior valve
angle. After that, many authors defined S.
vetuloides by a more-protruding dorsal valve
margin and more-numerous and larger denticles
on the posterior dorsal valve margin compared
to S. vetulus (Uéno 1966, Chiang and Du 1979,
Yoon and Kim 1987 2000, Shi and Shi 1996, Kim
Other authors treated S. vetuloides as a local form
(Johnson 1953) or as a synonym of S. vetulus
(Fryer 1957, Harding 1961, Sharma 1978, Negrea
1983, Michael and Sharma 1988). Sars (1903)
described S. mixtus as having a more-protruding (to
the rear) dorsal valve margin and larger denticles
on the posterior dorsal valve margin compared to
S. vetulus and S. vetuloides. Flössner (1972) and
Negrea (1983) treated S. mixtus as a synonym of
S. vetulus. After that, Orlova-Bienkowskaja (1998)

*To whom correspondence and reprint requests should be addressed. E-mail:shuh@mail.nhcue.edu.tw

222
 Young et al. – Study of Simocephalus from Taiwan
made a more-detailed revision and treated S.
mixtus as a valid species.
Orlova-Bienkowskaja (2001) proposed a
different method of discriminating S. vetulus, S.
vetuloides, and S. mixtus. She drew an inner circle
along the shell posterior, the diameter of which
and the prominence of the dorsal valve being key
features for identification. The shell posterior of
S. vetulus ends without an extending shell spine,
the inner circle is larger than in S. vetuloides and
S. mixtus, while S. mixtus has more-protruding
dorsal valves than S. vetuloides. The diameter
of the inner circle of S. mixtus is larger than the
prominence portion, and S. vetuloides differs from
S. mixtus in that the diameter of the inner circle of S.
vetuloides is smaller than the prominence portion.
In the past, many authors proposed S. vetulus
to be a cosmopolitan species first des-cribed from
the Old World, as it was found in many areas,
with the exception of New Zealand and Australia
(Werestschagin 1923, Uéno 1927, Rylov 1930,
Hemsen 1952, Harding 1961, Manuilova 1964,
Uéno 1966, Chiang and Du 1979, Rajapaksa and
Fernando 1982, Boonsom 1984, Yoon and Kim
1987, Kim 1998, Mizuno and Takahashi 1991, Du
1993, Hann 1995, Shi and Shi 1996, Michael and
Sharm 1998, Tuo 2002). Orlova-Bienkowskaja
(2001) indicated that the distribution of S. vetulus
was limited to northern Africa and Europe,
while S. vetuloides had a limited distribution in
eastern Siberia. Outside of Africa, Europe, and
eastern Siberia, Simocephalus sensu stricto
comprises S. punctatus Orlova-Bienkowskaja,
1998, S. elizabethae (King, 1853), and S.
mixtus. Simocephalus mixtus is a cosmopolitan
species distributed in Asia, Eastern Europe,
North Africa, and North America. Simocephalus
(Coronocephalus) serrulatus (Koch, 1841) is
also regarded as a cosmopolitan species, as
it is distributed in Asia, Europe, Africa, North
America, South America, and Australia (Orlova-
Bienkowskaja 2001).
Based on the description by Orlova-
Bienkowskaja (2001) and other morphological
comparisons, Tuo (2002) described 3 species
of Simocephalus from Taiwan, S. serrulatus,
S. vetulus, and S. vetuloides. Since then, this
extensive collection has increased to include S.
heilongjiangensis Shi and Shi, 1994 and S. mixtus
Sars from southern Taiwan (Ni 2005). At some
collection sites, S. vetulus and S. vetuloides were
found simultaneously as were S. vetuloides and
S. mixtus (Ni 2005). Morphological similarities
among S. vetulus, S. vetuloides, and S. mixtus are
223
large, with the exception of the shape of the dorsal
valve. However, the shape of the dorsal valve
of cladocerans may be affected by the brooding
status, with growing embryos pushing the valve
more prominently outwards, than in individuals
without eggs.
The species level is recognized as the
basic unit of biodiversity (Mayer and Ashlock
1991). Nowadays, alpha taxonomy is still based
mainly on morphology. Morphometry is one of
several possible methods to determine species
and analyze morphological differences between
closely related species and populations (Chen et
al. 2010). With the advent of molecular technology
for DNA sequencing, morphologically cryptic
species have been increasingly revealed, and the
use of DNA markers as a new tool to overcome
morphological impediments was suggested (Tautz
et al. 2003). The ideal DNA-based identification
system (DNA barcodes) would employ a single
gene, and be suitable for any organism in the
taxonomic hierarchy. Folmer et al. (1994) de-
signed a universal primer for the mitochondrial
cytochrome oxidase subunit I (COI) gene, which
subsequently became a popular marker to study
invertebrates. Hebert et al. (2003), Tautz et al.
(2003), Blaxter (2004), Lefébure et al. (2006),
and Costa et al. (2007) suggested that the COI
gene appears to be an appropriate molecular
marker (as a DNA barcode) on several taxonomic
scales, but particularly at the species level. We
attempted to clarify the taxonomic status of S.
vetulus, S. vetuloides, and S. mixtus in Taiwan
by morphometric comparisons and used the
mitochondrial (mt)DNA COI gene marker as a new
character.
This paper is our 1st step dealing with
vetulus-like populations of Simocephalus in Taiwan,
which are currently regarded as conspecific to
the Palaearctic cosmopolitan species. We thus
attempted to improve the taxonomy of the genus
Simocephalus by solving a small piece of the
puzzle from the overall picture.
MATERIALS AND METHODS
Samples were taken from many temporary
freshwater bodies throughout Taiwan using a
plankton net. Each sample was fixed in 70%
ethanol (EtOH), later preserved in 95% EtOH and
stored at a low temperature (< -20°C). Within 72 h,
each raw sample was sorted and identified under
a stereomicroscope. In total, 187 individuals (170
224 Zoological Studies 51(2): 222-231 (2012)

with eggs) were collected in 2003 and 2004 and
used for the morphometric analysis: 45 individuals
of S. vetulus from 8 sites, 72 individuals of S.
vetuloides from 11 sites, and 70 individuals of
S. mixtus from 10 sites. From this set of 187
individuals, 72 individuals, including 22 individuals
of S. vetulus from 8 sites, 28 individuals of S.
vetuloides from 11 sites, and 22 individuals of S.
mixtus from 10 sites, were selected for the DNA
analysis. Additionally, 7 individuals of S. serrulatus
(Fig. 1) from 3 sites and 19 individuals of S.
heilongjiangensis (Fig. 1) from 5 sites were also
included in the DNA analysis. Daphnia similoides
Hudec, 1991 (Daphniidae) and Diaphanosoma
dubium Manuilova, 1964 (Sididae) from Taiwan
were analyzed in order to obtain outgroup
sequences.
Morphometric analysis
Lateral-view images of S. vetulus, S.
vetuloides, and S. mixtus were taken using a
digital camera under a stereomicroscope for the
morphometric study. Morphometric characters
were extracted from the photographic images, and
8 morphometric data points were used to construct

(A) (B) (C)

(D) (E)

Fig. 1. General morphology of female Simocephalus with summer eggs found in Taiwan (all drawings are original). (A) S. vetulus; (B)
S. vetuloides; (C) S. mixtus; (D) S. serrulatus; (E) S. heilongjiangensis. The valve shape is the major difference among S. vetulus, S.
vetuloides, and S. mixtus; S. serrulatus has teeth on the top of its head, and S. heilongjiangensis has a different posterior end of the
valve. Scale bars = 0.1 mm.
 Young et al. – Study of Simocephalus from Taiwan
24 length measurements, each of which was
divided by body length to obtain size-free ratios
(Fig. 2). The body length and dorsal valve angle
(Fig. 2) were also measured on the photographic
images, and the clutch size of each individual
was assessed under a microscope. SPSS vers.
10.0.1 (Chicago, IL, USA) was used to analyze
the numerical data. The data matrix was tested
using the Kaiser-Meyer-Olkin (KMO) measure
of sampling adequacy and by the Bartlett X2 test
prior to the principle component analysis (PCA).
For individuals with eggs, Pearson’s correlation
analyses and non-linear regressions among the
dorsal angle, body length, and clutch size were
carried out.
DNA extraction, amplification, and sequencing
Total genomic DNA was extracted using
Chelex (InstaGene Matrix BIO-RAD 7326030,
Bio-Rad Laboratories, Hercules USA) from single
A
B 45 s, and extension at 72°C for 45 s; followed by
G H
60° 60°
F
E D
Fig. 2. Morphometry of each specimen extracted from 8
data points (A-H), from which we constructed 24 length
measurements; each length measurement was then divided by
body length (AE) to obtain size-free ratios. The angle between
lines AE and ED was taken as the dorsal valve angle.
225
animals. Each animal was taken from 95% EtOH
and placed into pure water for 1 h for cleaning.
After that, each animal was placed at the bottom
of a 0.5-ml centrifuge tube for 30 min to dry in a
speed vacuum-drying system. Dried samples
were then ground up by needles, and 50 μl of a 5%
Chelex solution was used to extract the DNA by
incubation at 56°C for 2-3 h, followed by incubation
at 90°C for 8 min. For each polymerase chain
reaction (PCR), 5 μl of upper cleaning was used as
the DNA template after centrifugation at 104 rpm
(9168g) for 3 min.
We employed the universal primers, LCO
1490 (5'-GGTCAACAAATCATAAAGATATTGG-3')
and HCO2918 (5'-TAAACTTCAGGGTGACCAA
AAAATCA-3') (Folmer et al. 1994), to amplify the
mitochondrial COI gene by a PCR. Each PCR
sample had a total volume of 50 μl and consisted
of pH 9.2 buffer solution (50 mM Tris-HCl, 16 mM
ammonium sulfate, 2.5 mM MgCl 2 , and 0.1%
Tween 20), 5 pM of each primer, 50 μM of dNTPs,
2 units of Taq DNA polymerase (super Therm
DNA polymerase, Bio-Taq, BioKit Biotechnology,
Miaoli Taiwan), and 10-50 ng of genomic DNA.
The PCRs were performed in an Eppendorf
Mastercycler gradient 384 machine (Eppendorf,
Hamburg, Germany). Thermocycling began with
5 min of preheating and continued with 35 cycles
at 94°C for 30 s, primer annealing at 51°C for
incubation at 72°C for 10 min for full extension
of the DNA and ended with 4°C holding. PCR
products were electrophoresed in 2% agarose
gels, after which the gels were stained with
ethidium bromide (EtBr) and photographed under
C an ultraviolet light box. DNA fragments were
excised from the gel and extracted using a 1-4-
3 DNA extraction kit (Gene-Spin, Protech, Taipei,
Taiwan) to obtain purified DNA. Sequences of
DNA fragments were resolved on an ABI3730
automated sequencer (Applied Biosystems,
Carlsbad, California USA) using 20-50 ng of tem-
plate with 5 pM of the LCO1490 primer.
Alignment, genetic diversity, and phylogeny
After a search of GenBank, all COI sequences
of Simocephalus were downloaded and aligned
with our sequences. The download sequences
included S. vetulus from the UK (accession no.,
DQ889172: Costa et al. 2007), S. cf. punctatus
from Mexico and Guatemala (EU702310 and
EU702282, Elias-Gutierrez et al. 2008), S.
cf. exspinosus from Mexico and Guatemala
226 Zoological Studies 51(2): 222-231 (2012)
(EU702296 and EU702279, Elias-Gutierrez et al.
2008), S. cf. mixtus from Mexico and Guatemala
(EU702305 and EU702281, Elias-Gutierrez et
al. 2008), and S. serrulatus from Mexico and
Guatemala (EU702312, Elias-Gutierrez et al.
2008). COI gene sequences were aligned by eye
using the BioEdit program vers. 7.0.2 (Hall 1999).
We calculated the haplotype diversity (Hd, Nei
1987), nucleotide diversity (π, Nei 1987), genetic
distance (D xy, Nei 1987), and average genetic
distances between each pair of species using
MEGA 3 vers. 3.0 (Kumar et al. 2004). Daphnia
similoides and Diaphanosoma dubium were used
as outgroups, and the phylogenetic tree was
derived using all sequences by the Neighbor-
joining (NJ) and maximum-parsimony (MP)
methods (Saitou and Nei 1987) based on Kimura
2-parameter (K2P) distances with 1000 bootstraps
using MEGA 3.
RESULTS
Morphometric comparisons of Simocephalus
vetulus, S. vetuloides, and S. mixtus
The KMO value for the morphometric data
matrix was 0.81, and Bartlett’s X 2 was 2583.96
(d.f. = 276; p = 0.000), demonstrating the
suitability of the PCA. After the PCA, 91% of
the variance was explained by the 1st, 2nd, and
3rd components combined. On the 1st and 2nd
component plots, S. vetulus and S. mixtus were
separated from each other, but S. vetuloides was
mixed with both groups; thus, they did not separate
very well into 3 different species (Fig. 3).
Simocephalus vetulus individuals with eggs
(n = 170) (clutch sizes ranged 1-4, dorsal valve
angle ranged 39.5°-74.8°) had fewer eggs than the
2 other species; S. vetuloides (clutch sizes ranged
1-12, dorsal valve angle ranged 41.5°-69.5°) was
intermediate; and S. mixtus (clutch sizes ranged
1-30; dorsal valve angle ranged 63.4°-97.5°) had
the most eggs. In a pooled analysis of these
3 species, Pearson’s correlation between the
dorsal valve angle and clutch size was r = 0.725
(p = 0.000), and between body length and clutch
size was r = 0.70 (p = 0.000). The relationship
between clutch size (Y) and dorsal valve angle (X)
fit a quadratic function Y = 0.0088X2 - 0.9091X +
25.3361 (r 2 = 0.53), and the one between clutch
size (Y) and body length (X) also fit a quadratic
function Y = 9.81X2 - 20.48X + 12.00 (r 2 = 0.49).
Hence, irrespective of the species, clutch size was
positively correlated with the dorsal valve angle
and body length. The valve shape was not a
species-specific character, but rather it depended
on the clutch size.
Molecular analysis of COI sequences
We used 110 COI sequences from S. vetulus
(n = 22), S. vetuloides (n = 28), S. mixtus (n = 10),
S. serrulatus (n = 7), S. heilongjiangensis (n = 19),
Daphnia similoides (n = 5), and Diaphanosoma
dubium (n = 7) for the phylogenetic analysis. Each
sequence was 641 bp long. Twelve haplotypes
were detected for the 5 species of Simocephalus
with 151 segregation sites; the genetic diversity,
Hd, was 0.891, and the nucleotide diversity, π, was
0.07049. Simocephalus vetulus had 4 haplotypes
from 8 sites (Hd = 0.576), S. vetuloides had 6
haplotypes from 11 sites (Hd = 0.802), S. mixtus
had 4 haplotypes from 9 sites (Hd = 0.636),
S. serrulatus had 2 haplotypes from 3 sites
(Hd = 0.571), and S. heilongjiangensis had 3
haplotypes from 6 sites (Hd = 0.374) (Table 1).
Genetic distances (Dxy) between each pair of
species based on the COI gene ranged 0.00671-
0.1604 (Table 2). Genetic distances among S.
vetulus, S. vetuloides, and S. mixtus were all
< 0.01, while those between S. serrulatus and the
other species were > 0.15, and those between
3 S. vetulus S. mixtus S. vetuloides
2
1
PCA 1
0
-1
-2
-4 -3 -2 -1 0 1 2 3 4
PCA 2
Fig. 3. Results of the principal component analysis of the
morphometric dataset: 1st and 2nd principle component plot.
Simocephalus vetulus and S. mixtus were well separated with
a distribution gap, while S. vetuloides filled the gap and mixed
with those 2 species.
 Young et al. – Study of Simocephalus from Taiwan 227

S. heilongjiangensis and the other species were
> 0.14.
In the phylogenetic NJ tree (Fig. 4), S.
vetulus, S. vetuloides, and S. mixtus (hap a-g)
were mixed together as a well-supported group
with a bootstrap value of 99%. Simocephalus
serrulatus (hap k-l) and S. heilongjiangensis (hap
h-j ) were well separated, with each group being
supported by a 99% bootstrap value. The dorsal
valve shape variation was not associated with
genetic differences based on the COI gene. The
most protruding valve shape (S. mixtus) was
common in haplotypes a and b. Valve shapes of
S. vetulus and S. vetuloides were also common

Table 1. Haplotypes (Hap) of each species of Simocephalus and their collection sites

Haplotype n Collection sites

S. vetulus 22 8 collection sites; HD = 0.576; π = 0.00806

Hap a 14 scA (3), scB (3), scC (2), zb (6)
Hap e 2 dgA (2)
Hap f 3 dy (3)
Hap g 3 scE (2), hsB (1)

S. vetuloides 28 11 collection sites; HD = 0.802; π = 0.00777

Hap a 10 hsA (3), scD (3), sf (1), xse (3)
Hap b 5 dd (1), khC (1), mn (3)
Hap c 3 lj (3)
Hap d 6 bs (6)
Hap e 2 khB (2)
Hap g 2 gA (2)

S. mixtus 22 10 collection sites; HD = 0.636; π = 0.00535

Hap a 12 gA (2), dh (3), dy (3), hsA (1), scF (3)
Hap b 6 dd (2), dy (1), tt (3)
Hap e 3 dgB (3)
Hap g 1 gs (1)

S. serrulatus 7 3 collection sites; HD = 0.571; π = 0.00357

Hap k 4 mf (4)
Hap l 3 gs (1), sf (2)

S. heilongjiangensis 19 6 collection sites; HD = 0.374; π = 0.00140

Hap h 15 pjA (3), pjB (4), pjC (4), pjD (4)
Hap i 2 khA (2)
Hap j 2 khA (2)

bs: Baoshan (Hsinchu County); dd: Dadu (Taichung County); dgA-B: Dongang A-B (Pingtung County); dh: Dahu
(Miaoli County); dy: Dayuan (Taoyuan County); gA: Green Grass Lake (Hsinchu City); gs: Guanxi (Hsinchu
County); hsA-B: Hengshan A-B (Hsinchu County); khA-C: Kaohsiung City A-C; lj: Longjing (Taichung County);
mf: Minfu (Hsinchu city); mn: Meinong (Kaohsiung County); pjA-D: Pingzhen A-D (Taoyuan County); scA-E:
Hsinchu City A-E; sf: Shinfeng (Hsinchu County); tt: Taitung city; xse: Xiangshan (Hsinchu City); zb: Zhubei
(Hsinchu County).

Table 2. Genetic distances (Dxy) among Simocephalus species from Taiwan based on
mitochondrial DNA cytochrome oxidase subunit I sequences

S. vetuloides S. mixtus S. vetulus S. serrulatus

S. vetuloides - - - -
S. mixtus 0.00671
S. vetulus 0.00785 0.00698
S. serrulatus 0.15550 0.15572 0.15473
S. heilongjiangensis 0.16017 0.16046 0.15945 0.14391
228 Zoological Studies 51(2): 222-231 (2012)
in haplotype b. Haplotypes e and g were shared
by all 3 morphospecies (Fig. 5, Table 1). We
reconstructed the phylogenetic trees by including
both our sequences and downloaded sequences,
and obtained NJ and MP phylogenetic trees with
similar tree structures (Fig. 6). Haplotypes a-g
from Taiwan were all placed in the same group.
DISCUSSION
DNA barcoding can be helpful in species
identification within cryptic species groups (Hebert
et al. 2004, Belyaeva and Taylor 2009). In general,
sequence divergences are much lower among
individuals of a species than between closely
related species. For example, congeneric species
of moths exhibit an average sequence divergence
of 6.5% in the mitochondrial COI gene, whereas
divergences among conspecific individuals
average only 0.25% (Moore 1995, Hebert et al.
2004). Similar values were obtained in birds,
with intraspecific divergences of COI averaging
0.27%, whereas congener divergences averaged
7.93% (Hebert and Stoeckle et al. 2004). Among
1781 congeneric species pairs of crustaceans,
only 1.3% had COI gene divergences of < 2%,
13.4% had COI gene divergences ranging 4%-
8%, and 81.8% had COI gene divergences
ranging 8%-32% (Hebert et al. 2003). In a study
86
Daphnia similoides
99
0.02
Fig. 4. Phylogenetic tree for Simocephalus species in Taiwan, derived using the Neighbor-joining (NJ) method based on mitochondrial
(mt)DNA cytochrome oxidase subunit I (COI) sequences. The numbers indicate support values for 1000 bootstrap calculations.
of the scale of intercontinental divergence for the
cladoceran genus Daphnia, Adamowicz et al.
(2009) observed a pairwise sequence divergence
within the D. obtusa complex of up to a maximum
of 16.9%, with divergences of up to 19% within
the D. longispina complex. In our study, S.
serrulatus and S. heilongjiangensis showed 14%-
16% COI divergence from each other and from
Simocephalus sensu stricto. These interspecific
differences were similar to most crustaceans
(Hebert et al. 2003).
Based on the morphological differences
described by Orlova-Bienkowskaja (2001), 3
species - S. vetulus, S. vetuloides, and S. mixtus
- were previously recorded in Taiwan. Indeed, our
morphometric analysis of the valve shape revealed
a significant difference between S. vetulus
and S. mixtus from Taiwan, which appeared
to support their taxonomic status as different
species. However, when all 3 putative species
were included in the analysis, the PCA did not
separate S. vetulus, S. vetuloides, and S. mixtus
from one another, as they formed a morphological
continuum. This is consistent with a single
morphologically variable species. Furthermore,
differences in valve shape among S. vetulus, S.
vetuloides, and S. mixtus collected in Taiwan were
not associated with genetic variations. The genetic
distances in COI among them were very small
(0.6%-0.8%), a divergence level that corresponds
Hap d: S. vetuloides, S. mixtus, S. vetulus
57 Hap e: S. vetuloides, S. mixtus, S. vetulus
69 Hap g: S. vetuloides
99 Hap f : S. vetulus
Hap a: S. vetuloides, S. mixtus, S. vetulus
99 Hap b: S. vetuloides, S. mixtus
71 Hap c: S. vetuloides
Hap k: S. serrulatus
99 Hap l : S. serrulatus
Hap h: S. heilongjiangensis
99 Hap i : S. heilongjiangensis
79 Hap j : S. heilongjiangensis
Daphnia similoides
Diaphanosoma dubium
 Young et al. – Study of Simocephalus from Taiwan 229

bs-1
scC-1
scA-2

Hap d

sf-1 ha-5
scD-1
Hap a
scf-1 dgA-1
dh-1
khB-4
hs3-1 dgB-1
Hap e

dd-3
dy-2
Hap f
khC-5
mn-2 dd-1
tt-2 Hap b gA-1
scE-3
hs3-3

1j-2 gs-1
Hap g

Hap c

Fig. 5. Dorsal valve shapes of different haplotypes belonging to Simocephalus vetulus, S. vetuloides, and S. mixtus. Haplotypes a, b, e,
and g have different valve shapes with large-scale variations.

Hap d (*) Hap d (*)

62
66 Hap g ( ) Hap g (*)
* 73
87 Hap e ( ) Hap e (*)
*
99 Hap f (*)
100 Hap f (*)
Hap a (*) Hap a (*)
99 95
Hap b (*)
63 77 Hap b (*) 75
Hap c (*) Hap c (*)
Simocephalus vetulus (+)
Simocephalus vetulus (+) 99
100
Simocephalus cf. punctatus (#) 85 Simocephalus cf. punctatus (#)
97
Simocephalus punctatus (#) Simocephalus punctatus (#)
Simocephalus cf. exspinosus (#) 63 Simocephalus cf. exspinosus (#)
78
Simocephalus cf. exspinosus (#) Simocephalus cf. exspinosus (#)
Simocephalus cf. mixtus (#) Simocephalus cf. mixtus (#)
69
Simocephalus cf. mixtus (#) Simocephalus cf. mixtus (#)
Hap k (*) 99 Hap k (*)
100
62 Hap l (*)
Hap l (*)

100
Hap h (*) 99 Hap h (*)
76 Hap i (*)
92 Hap i (*)
Hap j (*) Hap j (*)
Simocephalus serrulatus (#) Simocephalus serrulatus (#)
D. similoides (*) D. dubium (*)
D. similoides (*)
D. dubium (*)

*: Taiwan +: UK #: Mexico and Guatemala *: Taiwan +: UK #: Mexico and Guatemala

0.02 20

Fig. 6. Reconstructed phylogenetic trees of Simocephalus. Sequences from GenBank were included in this analysis: S. vetulus
(accession no., DQ889172) from the UK, S. cf. punctatus (EU702310 and EU702282) from Mexico and Guatemala, S. cf. exspinosus
(EU702296 and EU702279) from Mexico and Guatemala, S. cf. mixtus (EU702305 and EU702281) from Mexico and Guatemala, and
S. serrulatus (EU702312) from Mexico and Guatemala. Both the Neighbor-joining (NJ) and maximum-parsimony (MP) trees shared
similar branching structures. Haplotypes a-f from our study were all grouped together.
230 Zoological Studies 51(2): 222-231 (2012)
to intraspecific variations. Therefore, we prefer
to treat all morphotypes of Simocephalus sensu
stricto from Taiwan as a single species, S. cf.
vetulus, as the publication time of S. vetulus was
earlier than those of the other 2 species.
COI sequence comparison of S. cf. vetulus
from Taiwan with the European S. vetulus showed
that these were not conspecific (Fig. 6). As no
sequences of S. mixtus or S. vetuloides from the
areas of their primary distribution were available
for comparison, it remains unclear whether the
species found in Taiwan are conspecific with those
species. It is possible that Simocephalus found
in Taiwan is either S. mixtus or S. vetuloides or a
new undescribed species. Future studies should
compare sequences of S. vetulus, S. mixtus, and
S. vetuloides collected from the type locations with
sequences of S. cf. vetulus from Taiwan to verify
its taxonomic status.
According to allozymic studies by Hann
(1995), intraspecific differentiation within S. cf.
vetulus in North America was very slight. North
American and European populations were gene-
tically distinct according to the allozyme data, but
no morphological distinctiveness was identified.
In the past, conspecific populations from different
continents were believed to be widespread
within the Cladocera based on morphological
identifications. An intercontinental distribution of
a species is generally presumed to be a result
of passive transport by migratory birds or other
dispersal mechanisms (Dumont and Negrea 2002,
Adamowicz et al. 2009). The alternative hypothesis
of geographical isolation assumes that gene flow
among populations of cosmopolitan species on
different continents is interrupted, and therefore
the question is how large their genetic divergence
is relative to the geographical dis-continuum
scale. For example, Xu et al. (2009) explored the
global phylogeography of the non-cosmopolitan
freshwater cladoceran Polyphemus pediculus
(Linnaeus, 1761) (Crustacea, Onychopoda) using
2 mitochondrial genes, COI and 16s ribosomal (r)
RNA, and 1 nuclear marker, 18s rRNA. The P.
pediculus complex represents an assemblage of at
least 9 largely allopatric, cryptic species. The Far
East harbors exceptionally high levels of genetic
diversity at both the regional and local scales.
In contrast, little genetic subdivision is apparent
across the formerly glaciated regions of Europe
and North America.
Similar to Xu et al. (2009) and many other
previous studies on cosmopolitan cladoceran
species (Ishida et al. 2006, Rowe et al. 2007,
Belyaeva and Taylor 2009, Abreu et al. 2010), our
results indicate that S. cf. vetulus from Taiwan
is probably not the same species as S. vetulus
from the UK, and S. serrulatus from Taiwan is not
conspecific with S. cf. serrulatus from Mexico.
Simocephalus cf. vetulus from Taiwan appears to
be geographically isolated from populations on
other continents. Future studies should collect
barcodes of all morphospecies of Simocephalus
from different locations around the world in order
to reconstruct their systematic relationships.
Acknowledgments: We thank the National Sci-
ence Council of Taiwan for their grant (NSC87-
2311-B-134-001) to support part of this work. We
are very grateful to the anonymous reviewers for
their critical and constructive comments on our
manuscript.
REFERENCES
Abreu MJ, MJ Santos-Wisniewski, O Rocha, TC Orlando.
2010. The use of PCR-RFLP to genetically distinguish
the morphologically close species: Ceriodaphnia dubia
Richard, 1894 and Ceriodaphnia silvestrii Daday, 1902
(Crustacean Cladocera). Braz. J. Biol. 70: 121-124.
Adamowicz SJ, A Petrusek, JK Colbourne, PDN Hebert, JDS
Witt. 2009. The scale of divergence: a phylogenetic
appraisal of intercontinental allopatric speciation in a
passively dispersed freshwater zooplankton genus. Mol.
Phylogen. Evol. 50: 423-436.
Belyaeva M, DJ Taylor. 2009. Cryptic species within the
Chydorus sphaericus species complex (Crustacea:
Cladocera) revealed by molecular markers and sexual
stage morphology. Mol. Phylogen. Evol. 50: 534-546.
Blaxter ML. 2004. The promise of a DNA taxonomy. Phil.
Trans. R. Soc. B Biol. Sci. 359: 669-679.
Boonsom J. 1984. The freshwater zooplankton of Thailand
(Rotifera and Crustacea). Hydrobiologia 113: 223-229.
Chen CS, CH Tzeng, TS Chiu. 2010. Morphological and
molecular analyses reveal separations among
spatiotemporal populations of anchovy (Engraulis
japonicus) in the southern East China Sea. Zool. Stud.
49: 270-282.
Chiang SC, NS Du. 1979. Freshwater Cladocera. Fauna
Sinica. Crustacea. Beijing, China: Science Press,
Academia Sinica, 297 pp.
Costa FO, JR deWarrd, J Boutillier, S Ratnasingham, RT
Dooh, M Hajibabaei, PDN Hebert. 2007. Biological
identifications through DNA barcodes: the case of the
Crustacea. Can. J. Fish. Aquat. Sci. 64: 272-295.
Du NS. 1993. Crustacean biology. Beijing, China: Science
Press, 1003 pp. (in Chinese)
Elias-Gutierrez M, FM Jeronimo, NV Ivanova, M Valdez-
Moreno, PDN Hebert. 2008. DNA barcodes for
Cladocera and Copepoda from Mexico and Guatemala:
highlights and new discoveries. Zootaxa 1893: 1-42.
Dumont HJ, SV Negrea. 2002. Introduction to the class
Branchiopoda. (No. 19) Guides to the identification of the
 Young et al. – Study of Simocephalus from Taiwan
microinveryebrates of the continental waters of the world.
Leiden, the Netherlands: Backhuys Publishers, 398 pp.
Flössner D. 1972. Krebstiere, Crustacea. Kiemen- und
Blattfüsser, Branchiopoda. Fischläuse, Branchiura.
Tierw. Deutschl. 60. Gustav Fischer Verlag, Jena. 501 p.
(in German)
Folmer O, M Black, W Hoeh, R Lutz, R Vrijenhoek. 1994. DNA
primers for amplification of mitochondrial cytochrome c
oxidase subunit I from diverse metazoan invertebrates.
Mol. Mar. Biol. Biotechnol. 3: 294-299.
Fryer G. 1957. Freeliving freshwater Crustacea from Lake
Nyassa and adjoining waters. Part II. Cladocera and
Conchostraca. Arch. Hydrobiol. 53: 223-239.
Hall TA. 1999. BioEdit: a user-friendly biological sequence
alignment editor and analysis program for Windows 95/98/
NT. Nucleic Acids. Symp. Ser. 41: 95-98.
Hann BJ. 1995. Genetic variation in Simocephalus
(Anomopoda: Daphniidae) in North America: patterns and
consequences. Hydrobiologia 307: 9-14.
Harding JP. 1961. Some South African Cladocera collected by
Dr. A. D. Harrison. Ann. S. Afr. Mus. 46: 35-46.
Hebert PDN, CA Ball, SL deWaard Jr. 2003. Biological identi-
fications through DNA barcodes. Proc. R. Soc. Lond. B
Biol. Sci. 270: 313-321.
Hebert PDN, EH Penton, JM Burns, DH Janzen, W Hallwachs.
2004. Ten species in one: DNA barcoding reveals cryptic
species in the Neotropical skipper butterfly Astraptes
fulgerator. PNAS 101: 14812-14817.
Hebert PDN, S Ratnasingham, SL deWarrd JR. 2003. Bar-
coding animal life: cytochrome c oxidase subunit 1
divergences among closely related species. Proc. R. Soc.
Lond. B 270(Supplement): 96-99.
Hebert PDN, MY Stoeckle, TS Zemlak, CM Francis. 2004.
Identification of birds through DNA barcodes. Plos Biol. 2:
1657-1663.
Hemsen J. 1952. Ergebnisse Österreichischen Iran Expedition
1949/50. Cladoceren und freilebende Copepoden, usw.
Österr. Akad. Wiss. Mathem.-nat. Kl. Abt. I 161: 585-644.
(in German)
Ishida S, AA Kotov, DJ Taylor. 2006. A new divergent lineage of
Daphnia (Cladocera: Anomopoda) and its morphological
and genetical differentiation from Daphnia curcirostris
Eylmann, 1887. Zool. J. Linn. Soc. 146: 385-405.
Johnson DS. 1953. On some Cladocera from South African
muds. Ann. Mag. Nat. Hist. 12: 923-928.
Kim IH. 1998. Key to the Korean freshwater Cladocera. Kor. J.
Syst. Zool. Spec. Issue 2: 43-65. (in Korean)
Kumar S, K Tamura, M Nei. 2004. MEGA 3: integrated
software for molecular evolutionary genetic analysis
sequence alignment. Bioinformatics 5: 150-163.
Lefébure T, CJ Douady, M Gouy, J Gibert. 2006. Relationships
between morphological taxonomy and molecular
divergence within Crustacea: proposal of a molecular
threshold to help species delimitation. Mol. Phylogen.
Evol. 40: 435-447.
Manuilova EF. 1964. Vetvistousye rakoobraznye fauny SSSR.
Opredeliteli po faune SSSR 88. Moscow and Leningrad:
Nauka, 327 pp. (in Russian)
Mayer E, PD Ashlock. 1991. Principles of systematic zoology,
2nd ed. New York: McGraw-Hill, pp. 42-51.
Michael RG, BK Sharma. 1998. Indian Cladocera: fauna of
Indian and adjacent countries. Shillong, Meghalaya,
India: Department of Zoology, North-Eastern Hill Univ.,
262 pp.
Mizuno T, E Takahashi. 1991. An illustrated guide to freshwater
231
zooplankton in Japan. Hiroshibacho, Suita-shi, Osaka,
Japan: Hoikusha Publishing, 532 pp.
Moore WS. 1995. Inferring phylogenies from mtDNA variation:
mitochondrial-gene trees versus nuclear-gene trees.
Evolution 49: 718-726.
Negrea S. 1983. Cladocera. Fauna Republicii Socialiste
Romania. Volume IV. Fascicula 12. Cladocera. Editura
Academiei Republicii Romania. Bucuresti, 399p.
Nei M. 1987. Molecular evolutionary genetics. New York:
Columbia Univ. Press.
Ni MH. 2005. Systematic studies of Simocephalus (Cladocera:
Daphnnidae) from Taiwan. Master’s thesis, Department
of Applied Science, National Hsinchu Univ. of Education,
Hsinchu, Taiwan, 69 pp.
Orlova-Bienkowskaja MY. 1998. A revision of the clarocen
Genus Simocephalus (Crustacea: Daphniidae). Bull. Nat.
Hist. Mus. Lond. (Zool.) 64: 1-62.
Orlova-Bienkowskaja MY. 2001. Cladocera: Anomopoda
Daphniidae: genus Simocephalus. Leiden, the Nether-
lands: Backhuys, 130 pp.
Rajapaksa R, CH Fernando. 1982. The Cladocera of Sri Lanka
(Ceylon), with remarks on some species. Hydrobiologia
94: 49-69.
Rowe CL, SJ Adamowicz, PDN Hebert. 2007. Three new
cryptic species of the freshwater zooplankton genus
Holopedium (Crustacea: Branchiopoda: Ctenopoda),
revealed by genetic methods. Zootaxa 1656: 1-49.
Rylov WM. 1930. Cladocera et Copepoda in Abhandlungen
der Pamir-Expedition 1928. II. Zoologie: 105-133.
Saitou N, M Nei. 1987. Neighbor-joining method. Mol. Biol.
Evol. 4: 406-425.
Sharam BK. 1978. A note on freshwater cladocerans from
West Bengal. Bangladesh J. Zool. 6: 139-151.
Sars GO. 1903. On the crustacean fauna of central Asia. Part
II. Appendix. Cladocera. Ann. Mus. Zool. Acad. Sci. St.
Petersburg 8: 157-264.
Sars GO. 1916. The freshwater Entomostraca of Cape
Province (Union of South Africa) Part I. Cladocera. Ann.
S. Afr. Mus. 15: 303-351.
Shi XL, XB Shi. 1996. On the species and distribution of
Simocephalus in Heilongjiang Province, China. Acta
Zootax. Sin. 21: 263-276. (in Chinese)
Tautz D, P Arctander, A Minelli, RH Thomas, AP Vogler. 2003.
A plea for DNA taxonomy. Trends Ecol. Evol. 18: 70-74.
Tuo YS. 2002. Freshwater Cladocera of Taiwan. Master’s
thesis, Institute of Mathematics and Science Education,
Hsinchu Education College, Hsinchu, Taiwan, 60 pp.
Uéno M. 1927. On some freshwater branchiopods from China.
Annot. Zool. Jpn. 11: 157-163.
Uéno M. 1966. Cladocera and Copepoda from Nepal. Jpn. J.
Zool. 26: 95-100.
Werestschagin G. 1923. Notiz über die Süsswasserfauna
des Pamirs. Bull. I`Inst. Hydrobiol. Russie 6: 1-40. (in
German)
Xu S, PDN Hebert, AA Kotov, ME Cristescu. 2009. The non-
cosmopolitanism paradigm of freshwater zooplankton:
insights from the global phylogeography of the predatory
cladoceran Polyphemus pediculus (Crustacea,
Onychopoda). Mol. Ecol. 18: 5161-5179.
Yoon SM, HS Kim. 1987. A systematic study on the freshwater
Cladocera from Korea. Kor. J. Syst. Zool. 3: 175-207. (in
Korean)
Yoon SM, W Kim. 2000. Taxonomic review of the cladoceran
genus Simocephalus (Branchiopoda, Anomopoda,
Daphnidae). Kor. J. Limnol. 33: 152-161.
Zoological Studies 51(2): 232-247 (2012)

Two New Species of Amphipods of the Superfamily Aoroidea (Crustacea:

Corophiidea) from the Strait of Malacca, Malaysia, with a Description of
a New Genus
Bin Abdul Rahim Azman* and Bin Haji Ross Othman
Marine Ecosystem Research Centre, Faculty of Science and Technology, Universiti Kebangsaan Malaysia, 43600 UKM Bangi, Selangor,
Malaysia

(Accepted September 22, 2011)

Bin Abdul Rahim Azman and Bin Haji Ross Othman (2012) Two new species of amphipods of the
superfamily Aoroidea (Crustacea: Corophiidea) from the Strait of Malacca, Malaysia, with a description of a
new genus. Zoological Studies 51(2): 232-247. A taxonomic study on the amphipods collected from muddy
bottom habitats of the west coast of Peninsular Malaysia (Strait of Malacca) revealed 2 new species from the
superfamily Aoroidea. Klebang barnardi gen. nov., sp. nov., and Grandidierella melakaensis, sp. nov., are
described below. Klebang barnardi sp. nov. differs from the rest of its congeners in the combination of (1) a
unique carpal configuration of gnathopod 2, (2) a largely expanded posterior margin of the carpus of gnathopod 1,
and (3) a densely setose mandibular palp. Grandidierella melakaensis sp. nov., on the other hand, can be easily
distinguished from other Grandidierella species in having (1) a distinctly projecting rostrum, (2) pereopod 5 with
a merus and ischium of equal length, and (3) epimerons 1 and 2 with long plumose setae posteroventrally.
http://zoolstud.sinica.edu.tw/Journals/51.2/232.pdf

Key words: Amphipoda, Klebang barnardi, Grandidierella melakaensis, New genus, Strait of Malacca.

I n their revision of the suborder Corophiidea, MATERIALS AND METHODS

Myers and Lowry (2003) divided the superfamily
Aoroidea into the 2 families of the Aoridae
and Uniciolidae. We found 2 new species of
amphipods each belonging to these families. The
new species were discovered in benthic fauna
samples from muddy bottom habitats of the Strait
of Malacca at a depth range of 15-20 m in the
vicinity where Listriella longipalma was described
by Othman and Morino (2006). Complete
drawings of the appendages of the male and some
important characters of the female are presented.
In addition, comparisons of the new species with
related species are made.
This study is based on benthic materials
collected from the muddy-sand substrata in the
vicinity of the Port of Sungai Udang, Melaka
(Fig. 1). Samples were collected using a Smith-
McIntyre grab (0.05 m2) at depths ranging 15-20 m.
Once hauled in, the contents of the grab were
emptied into a container and wet sieved through
a 0.05-mm-mesh sieve. The materials retained
on the sieve were then carefully transferred into
plastic containers and fixed with a 4% buffered
formaldehyde-seawater solution. In the laboratory,
animals were examined under a compound
microscope and later selected for dissection. The
appendages of the dissected specimens were
examined and figures were produced under a Leica
DMLB light microscope using a camera lucida.

*To whom correspondence and reprint requests should be addressed. Tel: 60-3-89213038. Fax: 60-3-89253357.
E-mail:abarahim@gmail.com

232
 Azman and Othman – A New Genus and Species of Aoroid Amphipod
The following abbreviations are used: A,
antenna; ABD, abdomen; G, gnathopod; HD, head;
l, left; LL, lower lip; MD, mandible; MX, maxilla;
MP, maxilliped; P, pereopod; PL, pleopod; r, right;
T, telson; U, uropod; UR, urosome; UL, upper
lip; ♂ , male; ♀, female. The type materials of
the new species are deposited at the Universiti
Kebangsaan Malaysia Muzium Zoologi (UKMMZ),
Bangi, Malaysia.
RESULTS
Corophiida Leach, 1814
Aoroidea Stebbing, 1899
Diagnosis (description based on Myers
and Lowry 2003): Head rectangular, anterodistal
margin recessed, lateral cephalic lobe weakly
extended, eye, if present, situated proximal to
lobe; anteroventral margin weakly recessed,
moderately excavate. Mandible palp 3-articulated
or absent, article 3, when present, asymmetrical,
distally rounded, with setae extending along most
of posterodistal margin, or approximately parallel-
sided with distal setae only; posterior margin with
setae of variable length, or with comb of short
N 1 km
THAILAND KLEBANG
Port of Sungai Udang
6°
PENINSULAR MALAYSIA
Strait of Malacca
KEPULAUAN ANAMBAS
SINGAPORE
SUMATERA
0°
Fig. 1. Map showing the sampling area.
233
setae and a few long, slender setae. Gnathopod
1 enlarged in both sexes, or only in males; coxa 1
enlarged, larger than coxa 2. Merus of gnathopod
2 not enlarged. Pereopods 5-7 without accessory
spines on anterior margin. Pereopod 7 longer or
much longer than pereopod 6. Urosomites not
coalesced. Uropods 1 and 2 without a dense array
of robust setae. Peduncle of uropod 3 relatively
short, length usually ≤ 2 times breadth; with 2, 1,
or no rami. Telson without hooks or denticles.
Aoridae Stebbing, 1899
Diagnosis: Anteroventral margin of head
moderately excavate. Pereopod 7 very elongate,
entire propodus extending beyond pereopod 6.
Grandidierella Coutière, 1904
Diagnosis: Eyes small to medium. Accessory
flagellum of antenna 1 minute, 1-segmented.
Inner plate of maxilla 1 vestigial. Coxae very
small, relatively short, of various sizes and shapes.
Gnathopod 1 (male) complexly subchelate and
much larger than gnathopod 2. Gnathopod 2
subchelate. Dactylus of pereopods 6 and 7
elongate, falcate. Uropods 1 and 2 biramous;
rami slightly subequal; peduncle with ventrodistal
process. Uropod 3 uniramous. Telson entire.
Species composition: Grandidierella contains
40 species of G. africana Schellenberg, 1936; G.
bispinosa Schellenberg, 1938; G. bonnieroides
Stephensen, 1948; G. cabindae (Schellenberg,
1925); G. chelata K.H. Barnard, 1951; G.
chaohuensis Hou and Li, 2002; G. dentimera
Myers, 1970; G. elongata (Chevreux, 1926); G.
exilis Myers, 1981; G. fasciata Ariyama, 1996; G.
gilesi Chilton, 1921; G. gravipes K.H. Barnard,
1935; G. grossimana Ledoyer, 1967; G. indentata
Ledoyer, 1979; G. insulae Myers, 1981; G.
ischienoplia Bochert and Zettler, 2010; G. japonica
Stephensen, 1938; G. kanakensis Myers, 1998;
G. koa J.L. Barnard, 1977; G. lignorum K.H.
Barnard, 1935; G. longidactyla Ledoyer, 1982;
G. lutosa K.H. Barnard, 1952; G. macronyx K.H.
Barnard, 1935; G. mahafalensis Coutière, 1904
(type species); G. makena J.L. Barnard, 1970; G.
melakaensis sp. nov.; G. nottoni Shoemaker, 1935;
G. nyala Griffiths, 1974; G. osakaensis Ariyama,
1996; G. palama J.L. Barnard, 1977; G. perlata
Schellenberg, 1938; G. propodentata Moore, 1986;
G. rhizophorae Myers, 2009; G. robusta Ledoyer,
1982; G. spinicoxa Myers, 1972; G. taihuensis
Morino and Dai, 1990; G. teres Myers, 1981; G.
234 Zoological Studies 51(2): 232-247 (2012)
trispinosa Bano and Kazmi, 2010; G. unidentata
Ren, 2006; and G. vietnamica Dang, 1968.
Grandidierella melakaensis sp. nov.
(Figs. 2-5)
Material examined: Holotype. ♀, Malaysia,
Strait of Malacca, Melaka, Port of Sungai
Udang, (2°14'3"N, 102°7'43"E), 17 m, muddy
bottom, 26 May 1995, C. Zaidi, M. Soed, S.
Zuhaimi (Smith-McIntyre grab). UKM I.D. 3611
(UKMMZ-1273). Allotype. ♂ , data same as for
holotype. (UKMMZ-1274). Paratypes. Data same
as for holotype, UKMMZ-1275 (2 ♂♂ , 2 ♀♀);
UKMMZ-1276 (3 ♂♂ , 6 ♀♀); UKMMZ-1277
(5 ♂♂ , 3 ♀♀).
Description: Female (holotype). Total body
length 2.6 mm (from tip of rostrum to apex of
telson). Head (HD, Fig. 3) with short, pointed
rostrum, about as long as pereonites 1 and 2
combined, with triangular-shaped anterior head
lobe, inferior antennal sinus deep, beyond middle
of head. Eye small, oval, placed just behind
anterior head lobe. Antenna 1 (A1, Fig. 3) much
longer than antenna 2, with peduncle longer than
flagellum, length ratio of 9: 11: 4; flagellum shorter
than peduncle, composed of 15 articles, distal one
of which vestigial, each article distally provided
with tuft of long and short setae. Antenna 2 (A2,
Fig. 3) short and stout, 4-segmented in ratio of 5:
7: 16: 14; 1st and 2nd peduncular articles very
short, their combined length subequal to article
3, broader than those of antenna 1; flagellum
very short, slightly longer than 1/2 length of
peduncular article 4, 3-articulate, all articles
setiferus, distalmost article apically armed with 2
stout spines surrounded by a tuft of setae. Apical
margin of upper lip (UL, Fig. 3) broad, slightly
concave medially, bearing minute bristle. Inner
plate of lower lip (LL, Fig. 3) developed, broad
and angular, minutely pubescent, outer plate with
rounded shoulder, densely pubescent, and with
strongly developed, rounded mandibular process.
Incisor of mandible (MD, Fig. 3) well-developed,
with 4 teeth on left mandible and 5 teeth on right
one; lancinia mobilis armed with 4 teeth on both
left and right mandibles; accessory blades 8
on left mandible and 7 on right one; right molar
process developed, with circular apex, fringed with
apically branched processes; palp triarticulate.
Inner plate of maxilla 1 (MX1, Fig. 3) small and
short, with setae; outer plate distally truncate;
palp biarticulate, extending slightly beyond outer
plate, with rounded apex. Inner plate of maxilla
2 (MX2, Fig. 3) broad medially, pointed distally,
outer margin naked; outer plate extending just
beyond inner one, both outer and inner margins
naked. Inner plate of maxilliped (MP, Fig. 3)
elongate, extending well beyond proximal article
of palp, medially narrow, apically truncate; outer
plate almost reaching end of palp article 2, inner
margin straight and outer margin evenly convex,
dense bristles on outer margin; palp consisting of
4 articles, article 4 small, subtriangular, tapering to
truncate tip and ending in stout spine. Pereonites
1-5 subequal to each other in length, 6 and 4 of
equal length, and 5-7 deeper than preceding ones,
pereonite 1 anteroventrally roundly produced.
Coxal plates small, shallow, separated.
Gnathopod 1 (G1 ♀, Fig. 2) subequal in
size with gnathopod 2, length ratio of articles from
basis to dactylus approximately 16: 3: 4: 15: 9: 7;
basis stout, anterior margin straight; ischium short,
subrectangular, anterior margin distally weakly
produced and naked; merus slightly longer than
ischium, distally tapering to become subtriangular,
posterior margin and submargin throughout with
numerous setae which are peculiarly very long and
bristly; carpus about as long as basis, elongate,
posterior margin weakly convex but crenulate and
both its margin and submargin throughout densely
covered with very long bristly setae; propodus
narrower and slightly longer than 1/2 of carpus,
slightly curved but with uniform width, densely
covered with very long setae both anteriorly
and posteriorly; dactylus shorter than propodus,
stout, falcate, tapering to pointed tip, grasping
margin minutely serrated medially. Length ratio
of articles of gnathopod 2 (G2 ♀, Fig. 2) from
basis to dactylus approximately 14: 3: 4: 9: 12: 3;
brood plate narrow and elongate, about as broad
as basis and about 1/2 as long as gnathopod 2;
basis elongate and parallel-sided; ischium short,
with distally slightly produced anterior margin and
naked posterior margin; merus slightly longer than
ischium, subcircular, as long as broad; carpus
shorter than propodus, naked along its length;
propodus elongate, as broad as and subequal
to basis in length, palm transverse, defined by 3
stout spines, palm margin possessing some robust
setae; dactylus stout, short, as long as palm, claw-
like, grasping margin with a hump near proximal
end. Pereopod 3 (P3, Fig. 4) longer than pereopod
4; brood plate elongate and lanceolate; length ratio
of articles from basis to dactylus approximately
13: 3: 6: 4: 5: 8; basis linear; ischium short,
subrectangular, anterior margin medially concave;
merus longer than carpus, distally slightly broader;
 Azman and Othman – A New Genus and Species of Aoroid Amphipod 235

G2 ♀

G1 ♂

G1 ♀

G2 ♂

Fig. 2. Grandidierella melakaensis sp. nov., holotype, female (UKMMZ-1273), 2.6 mm, allotype, male (UKMMZ-1274), 2.9 mm. Port of
Sungai Udang, Melaka. Scale bars: G1 ♀ and G2 ♀ = 0.25 mm; G1 ♂ and G2 ♂ = 0.2 mm.
236 Zoological Studies 51(2): 232-247 (2012)

A2

HD

A1
MP

MX1

MX2

MD

LL
UL

Fig. 3. Grandidierella melakaensis sp. nov., holotype, female (UKMMZ-1273), 2.6 mm. Port of Sungai Udang, Melaka. Scale bars:
A2 = 0.25 mm; A1 and HD = 0.5 mm; MP and MD = 0.2 mm; UL, LL, MX1, and MX2 = 0.1 mm.
 Azman and Othman – A New Genus and Species of Aoroid Amphipod 237

carpus shorter than propodus, anterior margin tip, both anterior and posterior margins naked.
naked; propodus shorter than dactylus, rather Pereopod 4 (P4, Fig. 4) larger than pereopod 5;
narrower than preceding articles; dactylus very brood plate lanceolate, rather large, with row of
long and thin, slightly curved, slightly tapering to very long setae; length ratio of articles from basis

P3

P4

P7

P5

P6

Fig. 4. Grandidierella melakaensis sp. nov., holotype, female (UKMMZ-1273), 2.6 mm. Port of Sungai Udang, Melaka. Scale bars: P3
and P5 = 0.2 mm; P4, P6, and P7 = 0.5 mm.
238 Zoological Studies 51(2): 232-247 (2012)
to dactylus approximately 11: 3: 7: 5: 6: 8; basis a
little more than 1/3 length of pereopod 4; ischium
short, anterior margin slightly convex; merus
larger than carpus, distally slightly broader; carpus
subequal to propodus, slightly wider than propodus,
anterior margin gently concave; propodus longer
but narrower than carpus; dactylus rather long and
thin, longer than propodus, falcate, tapering to
pointed tip, anteroproximally with a seta. Pereopod
5 (P5, Fig. 4) shortest and smallest among all
pereopods; length ratio of articles from basis to
dactylus approximately 13: 6: 5: 5: 7: 3; basis 1/3
as long as pereopod 5, proximally wider; ischium
longer than merus, rectangular, anterodistally
with pair of setae; merus as long as carpus but
broader; carpus narrower than merus; propodus
rather long and narrow; dactylus short, abruptly
curved at apex, anterodistally with spine tooth and
posterodistally submargin (grasping submargin)
with seta. Pereopod 6 (P6, Fig. 4) reaching end of
telson, much longer than pereopod 5 but shorter
than pereopod 7; length ratio of articles from
basis to dactylus approximately 11: 2: 8: 6: 9: 3;
basis slightly expanded anteriorly; ischium very
short, rectangular, slightly narrower than basis;
merus elongate, rectangular, longer than carpus;
carpus narrower than merus but as broad as
propodus; propodus elongate, longer than both
carpus and merus; dactylus short and falcate,
pointed anteriorly, proximally wider and tapering to
pointed distal end, anteriorly grasping margin and
posteriorly convex margin armed with a spine each
at about subapex. Pereopod 7 (P7, Fig. 4) very
long, extending well beyond telson, length ratio
of articles from basis to dactylus approximately
11: 2: 9: 9: 12: 3; basis 1/4 as long as pereopod
7, anteriorly slightly expanded; ischium very short
and rectangular; merus rather long, rectangular;
carpus almost as long as merus, but narrower;
propodus longest among articles, narrow and
rectangular; dactylus short, stout, falcate, pointed
forward, anteriorly grasping margin and posteriorly
convex margin with thin spine each at subapex
and plumose seta at proximal end of posterior
margin. Pleopods (PL1, PL2, PL3, Fig. 5) well-
developed. Pleonites 1 and 2 equally elongate, but
each obviously shorter than pleonite 3. Epimerons
1 and 2 (ABD, Fig. 5) rectangular, but epimeron 3
obtusely produced to rear at posteroventral angle
and dorsomedially posterior end with an acute
process, posteroventral margins of epimerons 1
and 2 respectively bearing 4 and 7 plumose setae.
Uropod 1 (U1, Fig. 5) extending slightly
beyond uropod 2; peduncle longer than rami; outer
ramus a little longer than inner one, with row of
5 robust setae on outer margin, row of 4 robust
setae on inner margin, and 3 robust setae on
apex; inner ramus with row of 5 robust setae on
outer margin, and 3 stout spines on apex, middle
one of which distinctly shorter. Peduncle of uropod
2 (U2, Fig. 5) a little longer than rami, outer margin
bearing 2 robust setae, one at middle and one at
distal end, distal 1/2 of inner margin with row of
3 long stout robust setae; outer ramus distinctly
shorter and narrower than inner one, with row of 3
robust setae on outer margin, apex with 3 robust
setae; inner ramus with row of 4 robust setae on
outer margin, 2 robust setae on distal 1/2 of inner
margin, and 3 long robust setae on apex, middle
one longer. Uropod 3 (U3, Fig. 5) extending a little
beyond uropod 2, uniramous, peduncle short and
about 1/2 as long as ramus, with slightly convex
lateral margins; ramus biarticulate but distal article
vestigial, proximal article medially gently broader
than its proximal and distal parts, both outer and
inner margins with row of 4 long stiff setae each,
and apically rounded margin with cross row of 3
submarginal robust setae; distally small article
armed with 1 very long stiff seta. Combined length
of urosomites 1-3 almost as long as pleonite 3,
and successively smaller in size. Telson (T, Fig.
5) proximally wider, apical margin truncate, with a
spine near dorsolateral angle.
Male (sexually dimorphic characters):
(allotype – UKMMZ-1274) Total body length
2.9 mm (from tip of rostrum to apex of telson).
Gnathopod 1 (G1 ♂ , Fig. 2) carpochelate,
stouter and larger than gnathopod 2, coxal plate
subquadrangular, length ratio of articles from
basis to dactylus approximately 8: 2: 3: 11: 4: 3;
basis stout, anterior margin straight and naked,
posteriorly gently developed except at proximal
end where basis narrowed, posterior margin with
a seta in middle and another at distal end; ischium
short, anterodistally slightly produced; merus
longer than ischium, proximally broadest and
tapering to tip, anterior margin naked, posterior
margin rather convex; carpus very strong and
massive, much longer than basis, nearly 2 times
as long as broad, proximally narrow and distally
uniformly broad, both anterior and posterior
margins convex and carpus subovate, anterior
margin naked except for minute seta near distal
end, posterior margin covered throughout with
several plumose setae on margin and submargins,
posterodistal corner produced into very strong
and large process which is outwardly deflected
and ends in blunt tip, at base of which, on distal
 Azman and Othman – A New Genus and Species of Aoroid Amphipod 239

U2
U1 U3

ABD

PL2

PL3

PL1

Fig. 5. Grandidierella melakaensis sp. nov., holotype, female (UKMMZ-1273), 2.6 mm. Port of Sungai Udang, Melaka. Scale bars:
T = 0.1 mm; U1 = 0.25 mm; U2 and U3 = 0.2 mm; ABD, and PL1-PL3 = 0.5 mm.
240 Zoological Studies 51(2): 232-247 (2012)
margin, with a group of several plumose setae;
propodus much shorter and narrower than carpus,
proximally and distally broader than medial part,
anterior margin uneven, posterior concave margin
medially produced forming strong and apically
blunt process, throughout its length covered with
several plumose setae; dactylus rather stout,
somewhat straight, proximally wider and tapering
to blunt tip, grasping margin proximally bearing
single small tooth and subapically with 2 pairs of
small teeth, anterior margin with pair of setae near
proximal end. Gnathopod 2 (G2 ♂ , Fig. 2) in both
female and male rather similar except for length
of basis which in male is distinctly longer than
propodus (1.4 times as long as propodus).
Remarks: The genus Grandidierella Coutière,
1904, a member of amphipods of the family
Aoridae, is characterized by a subcylindrical
body, small- to medium-sized eyes, small
coxae, gnathopod 1 larger than gnathopod 2,
male gnathopod 1 carpochelate, and with a
uniramous uropod 3. According to Myers (1970),
Grandidierella presumably originated from the old
Tethys Sea and is considered to have a tropical
affinity. Records of this genus appear scattered
throughout the Caribbean Sea to Madagascar,
Tanzania, and India (Myers 1970). To date, the
genus Grandidierella is known to contain 40
species, with recent additions by Ren (2006),
Myers (2009), Bochert and Zettler (2010), and
Bano and Kazmi (2010). The present work
reports on the 1st record of this genus occurring in
Malaysian waters, along the coast of the state of
Melaka, Peninsular Malaysia.
Grandidierella melakaensis sp. nov. can be
easily distinguished from all other species in the
genus by a set of characters known only in this
species: (1) an obviously projecting rostrum, (2)
pereopod 5 having a merus and ischium of equal
lengths, and (3) epimerons 1 and 2 with long
plumose setae posteroventrally. Nonetheless,
the specimens examined resemble G. elongata
in having a triangular ocular lobe; and G. exilis,
G. gilesi, G. mahafalensis, G. palama, and G.
indentata in bearing several very long plumose
setae on the propodus, carpus, and merus of
gnathopod 2 of both the male and female and
possessing a single posterodistal spine on male
gnathopod 1, but clearly differ in many other
respects, especially in the form of gnathopod 1.
Furthermore, the inflated uropod 3 peduncle and
very short mandibular palp article 1 in G. elongata,
the much inflated uropod 3 peduncle in G. gilesi,
and the ventral pereon process on pereonite 1 in G.
exilis also differ.
Etymology: The new species of Grandidierella
is named after its type locality, Melaka as
melakaensis.
Unciolidae Myers and Lowry, 2003
Diagnosis (description from Myers and
Lowry 2003): Anteroventral margin of head
moderately excavate, or strongly excavate for
receiving enlarged antenna 2. Pereopod 7 not
very elongate, entire propodus not extending
beyond pereopod 6. Included subfamilies/genera.
Acuminodeutopinae: Acuminodeutopus J.L.
Barnard, 1959; Klebang gen. nov.; Rudilemboides
J.L. Barnard, 1959; and Wombalana Thomas
and Barnard, 1991. Unciolinae: Dryopoides
Stebbing, 1888; Janice Griffiths, 1973; Liocuna
Myers, 1981a; Neohela Smith, 1881; Orstomia
Myers 1998; Pedicorophium Karaman, 1981;
Pseudunciola Bousfield, 1973; Pterunciola Just,
1977; Ritaumius Ledoyer, 1978; Rildardanus J.L.
Barnard, 1969; Uncinotarsus L’Hardy and Truchot,
1964; Unciola Say, 1818; Unciolella Chevreux,
1911; and Zoedeutopus J.L. Barnard, 1979.
Remarks: Myers and Lowry (2003) esta-
blished the family Unciolidae and included it
together with the existing Aoridae Stebbing
in the superfamily Aoroidea. It can easily be
distinguished by a combination of characters that
includes a moderate or strong excavation along
the anteroventral margin of the head for receiving
the enlarged antenna 2; antenna 1 article 3 short,
≤ 1/2 the length of article 2; an enlarged gnathopod
1; pereopods 5, 6, and 7 in a regular length
progression; and all urosomites free. Currently,
the Unciolidae is composed of the 2 subfamilies
of the Acuminodeutopinae with 3 genera and
the Unciolinae with 14 genera and is distributed
worldwide in both cold and warm waters.
Klebang gen. nov.
Type species: Klebang barnardi sp. nov., present designation.
Included species: K. barnardi sp. nov.
Diagnosis: Rostrum short, ocular lobes
moderate, produced to front, pointed. Eyes
moderate. Antenna 1 slightly longer than antenna
2, both slender; peduncular article 3 slightly shorter
than article 1, article 2 longest, accessory flagellum
present. Peduncular article 3 of antenna 2 short,
flagellum with only 3 or 4 articles. Mandibular palp
 Azman and Othman – A New Genus and Species of Aoroid Amphipod
setose; article 2 longest. Male gnathopods 1 and
2 subequal, subchelate, and carpochelate. Outer
ramus of uropod 1 with brush setae. Uropod 3
uniramus; peduncle short; ramus elongate with
robust setae on both margins. Telson semicircular
and lobed.
Remarks: The diagnosis of the new genus
is based on the type-species described be-
low. Klebang gen. nov. is closely related to
Grandidierella Coutière, from which it shares
several generic characters in having a sub-
cylindrical body, an enlarged carpochelate
gnathopod 1, free urosomites, and a uniramus
uropod 3. A careful examination of the newly
acquired material on the other hand, although
closely similar morphologically to Grandidierella,
suggests that it represents a new genus in the
Aoroidea. Myers and Lowry (2003) provided
a valuable updated key to the families and
subfamilies of the Corophiidea. Some key chara-
cters show that our material naturally fits into the
Acuminodeutopinae, like the short article 3 of
antenna 1 at ≤ 1/2 the length of article 2, uropod
3 lacking recurved robust setae, gnathopods 1
and 2 not together forming a sieving basket, free
urosomites, an enlarged gnathopod 1, pereopods
5, 6, and 7 in a regular length progression, and
most importantly the acute lateral cephalic lobes
of the head. As shown by the excellent series of
head drawings of selected genera in Myers and
Lowry (2003), the acute head cephalic lobes are
of special importance in the classification of this
group (Fig. 4 in Myers and Lowry 2003). Currently,
the acuminodeutopine clade includes only the 3
genera of Acuminodeutopus, Rudilemboides, and
Wombalano, and all 3 share the characteristic
of having the acute, triangular, lateral cephalic
lobes. Clearly within this clade only Wombalano
possesses the same distinctive generic characters
shown in the Klebang gen. nov. material in
having a uniramus uropod 3. However, the
unique formation of the male gnathopod 2 (with
an expanded basis and carpus) in Wombalano
is an advanced character that separates it from
the Klebang gen. nov. material. At the same
time, Klebang gen. nov. is highly distinctive in
having this combination of characters: (1) the
unique carpal configuration of gnathopod 2, (2) a
largely expanded posterior margin of the carpus
of gnathopod 1, and (3) the densely setose
mandibular palp that has not yet been formulated.
Therefore, we consider the current species to be
representative of a new genus.
Etymology: The name Klebang refers to
241
Pantai Klebang, Melaka, Malaysia the general area
in Melaka where this genus was discovered.
Klebang barnardi sp. nov.
(Figs. 6-8)
Material examined: Holotype. ♂ , Malaysia,
Strait of Malacca, Melaka, Port of Sungai Udang,
St. CS, Petronas (2°14'43"N, 102°6'53"E), 20 m,
muddy bottom, 22 Oct. 2003, C. Zaidi, M. Soed,
S. Zuhaimi (Smith-McIntyre grab). UKM I.D.
7187 (ref: UKMMZ-1350). Paratypes. From the
same sample as holotype, UKMMZ-1352 (7 ♂♂ );
UKMMZ-1353 (4 ♂♂ ); UKMMZ-1354 (8 ♂♂ ).
Description: Male (holotype). Total body
length 6.7 mm (from tip of rostrum to apex of
telson). Body rather slender. Head (HD, Fig. 6)
broader and deeper than pereonite 1; rostrum
not developed, anterior lateral head lobe (ocular
lobe) extending forward and anteriorly pointed in
triangular shape; inferior antennal sinus deep and
straight vertically; eye distinct and located behind
anterior head lobe. Antenna 1 (A1, Fig. 6) slightly
longer than antenna 2, ratio of peduncular articles
1-3 as 1.1: 1.5: 1; article 1 with 4 postero-marginal
setae; flagellum with 5 articles, 2 times as long
as peduncle; accessory flagellum uni-articulate,
short. Peduncular article 3 of antenna 2 (A2, Fig.
6) with 3 long and 1 short setae posterodistally;
article 4 slightly shorter than article 5 with row of
long setae along posterior margin; flagellum short,
composed of 4 articles. Labrum of upper lip (UL,
Fig. 7) broad, its apical margin weakly concave
mid-ventrally and pubescent on each lobe. Inner
plates of lower lip (LL, Fig. 7) highly developed
and subtriangular, mandibular process narrow but
well-developed; outer plates with bristly shoulders.
Both mandibles (MD, Fig. 7) similar to each other
except for number of accessory blades with 4 on
right and 5 on left; incisor produced to interior,
broad, with 5 teeth; lacinia mobilis on both sides
4-toothed, followed by 4 or 5 accessory blades;
molar process medium, ridged distally and serrate
marginally, with a single seta; palp triarticulate.
Inner plate of maxilla 1 (MX1, Fig. 7) reduced;
outer plate with truncate apical margin; palp
extending beyond outer plate, biarticulate. Inner
plate of maxilla 2 (MX2, Fig. 7) slightly shorter than
outer one; outer plate larger than inner one, distally
broadest and with rounded apical margin. Inner
plate of maxilliped (MP, Fig. 7) short, not extending
beyond tip of palmer proximal article; outer plate
extending beyond 1/2 of palmer article 2, outer
margin naked, evenly convex; palp 4-articulated,
242 Zoological Studies 51(2): 232-247 (2012)
terminal article distally tapering and ending in a
long nail-like spine-tooth.
Gnathopod 1 (G1, Fig. 6) subchelate and
carpochelate, subequal to gnathopod 2 in size;
coxa plate shallow, rhomboidal, smaller than
others, anteroventral angle markedly produced;
length ratio of articles from basis to dactylus
approximately 14: 3: 4: 12: 10: 7; basis linear,
almost parallel-sided; ischium short, posterior
margin 3 times as long as anterior margin; merus
subrectangular, longer than wide, anterior margin
naked; carpus robust, mainly subrectangular
except proximally small and short, with triangular
ending, greatly wider and longer than propodus,
posterodistally 3/4 margin throughout equally
produced into widely expanded plate which is
proximoventrally oblique and distoventrally pointing
forwards and ending in a small tooth; propodus
shorter and narrower than carpus, anterior evenly
convex and posterior margin barely concave;
dactylus stout, fairly curved, about 1/2 as long
as carpus, tapering to pointed tip. Gnathopod 2
(G2, Fig. 6) longer than gnathopod 1, subchelate
and carpochelate; coxa plate shallow with
ventral margins medially produced into triangular
expansion; length ratio of articles from basis to
dactylus 18: 3: 5: 10: 14: 6; basis linear; ischium
shortest of all articles, subrectangular; merus
slightly longer than ischium, posterodistal angle
with triangular spine-tooth; carpus at mid-length
twice as long as merus but shorter than propodus,
proximally narrowing into triangular end and distally
widening with truncate apical margin, distal 1/2 of
posterior margin produced into large elongated
expansion which is proximally wider, distally
tapering to rounded tip, outwardly deflected,
and reaching near distal margin of propodus;
propodus rather long and stout, proximally narrow
and distally obviously broader, palm strongly
transverse, with minutely serrated marginal spines;
dactylus fitting on palm, stout, tapering to pointed
tip, anteroproximally with long setae. Pereopod 3
(P3, Fig. 8) thin and elongate; coxa plate shallow
with ventral margins medially produced into
triangular expansion; length ratio of articles from
basis to dactylus approximately 20: 3: 8: 9: 10: 8;
basis linear, almost uniform in width, 1/3 as long
as pereopod 3; ischium short, subrectangular;
merus shorter than carpus, anterodistally weakly
produced; carpus rectangular; propodus narrower
than carpus; dactylus rather long, 4/5 as long as
propodus, gently curved, tapering to pointed tip,
anteroproximally armed with a seta. Pereopod 4
(P4, Fig. 8) rather similar to pereopod 3 but slightly
shorter, coxa plate with ventral margins medially
produced into triangular expansion; length ratio of
articles from basis to dactylus approximately 15:
3: 7: 7: 8: 5; and with less setation than pereopod
3. Pereopod 5 (P5, Fig. 8) slightly longer than
pereopod 4; coxa posteroventrally expanded into
long and narrowly obtuse angle; length ratio of
articles from basis to dactylus approximately 20:
3: 11: 9: 9: 3; basis slightly expanded in proximal
part, about 1/3 as long as pereopod 5; ischium
short, anterior margin longer than posterior one;
merus longer than carpus, uniform in width, apical
margin anteriorly produced into triangular process;
carpus subequal to merus in width, parallel-sided
except near proximal end; propodus about as long
as carpus; dactylus short, 1/3 as long as propodus,
weakly curved, tapering to pointed tip, bearing 1
plumose seta at anterior proximal end and 1 thin
spine in middle of grasping margin. Pereopod 6
(P6, Fig. 8) rather long, extending well beyond
end of telson and uropods; coxa with posteriorly
produced fairly narrow and rounded lobe, anteriorly
and anteroventrally rounded; length ratio of
articles from basis to dactylus approximately 12:
2: 10: 5: 7: 4; basis almost linear and uniform in
width, about 1/3 as long as pereopod 6; ischium
short, posterodistally slightly produced; merus 2
times longer than carpus, distinctly narrower than
basis, twisted near distal end; carpus shorter than
propodus, anterior and posterior margins curved
forward forming a groove along its length; dactylus
about 1/2 as long as propodus, tapering to sharply
pointed tip, posterior margin with long slender
spine at 2/3 from proximal end. Pereopod 7 (P7,
Fig. 8) extremely long, extending well beyond end
of pereopod 6; coxa comparatively shallower;
length ratio of articles from basis to dactylus 7: 1:
8: 4: 6: 3; basis weakly expanded, 1/4 as long as
pereopod 7; ischium short, posterior margin distally
slightly produced; merus longer but narrower
than basis; anterior and posterior margins of
carpus curved to rear forming a groove; propodus
elongate and rather slender; dactylus 1/2 as long
as propodus, weakly curved, tapering to pointed
tip, with 1 long thin spine at 2/3 from proximal end.
Epimeron 1 (ABD, Fig. 6) subrectangular,
2 posteriorly evenly rounded, and 3 with roundly
produced antero- and posteroventral angles.
Urosomites 1-3 in combined length as long as
epimeron 3.
Pleopods 1-3 (PL1, PL2, PL3, Fig. 8)
similar to each other; peduncles cylindrical and
anterodistally with plumose setae, each one
distinctly shorter than inner ramus but equal to
 Azman and Othman – A New Genus and Species of Aoroid Amphipod 243

A1

G1

HD

A2

ABD

G2

Fig. 6. Klebang barnardi sp. nov., holotype, male (UKMMZ-1350), 6.2 mm. Port of Sungai Udang, Melaka. Scale bars: G1, G2, ABD,
and HD = 0.5 mm; A1 and A2 = 0.2 mm.
244 Zoological Studies 51(2): 232-247 (2012)

MX1

MX2

MD L MD R

LL

UL

U1

MP

U2

U3

Fig. 7. Klebang barnardi sp. nov., holotype, male (UKMMZ-1350), 6.2 mm. Port of Sungai Udang, Melaka. Scale bars: MP, MD L-R,
U2, U3, and T = 0.25 mm; MX1 and MX2 = 0.1 mm; UL and LL = 0.2 mm; U1 = 0.5 mm.
 Azman and Othman – A New Genus and Species of Aoroid Amphipod 245

P3
P4

P6

P7

PL2

P5

PL3

PL1

Fig. 8. Klebang barnardi sp. nov., holotype, male (UKMMZ-1350), 6.2 mm. Port of Sungai Udang, Melaka. Scale bar: P3-P7 =
0.5 mm; PL1-PL3 = 0.5 mm.
246 Zoological Studies 51(2): 232-247 (2012)
or slightly longer than outer one; rami densely
covered with rather long swimming setae.
Uropod 1 (U1, Fig. 7) extending well beyond
ends of other uropods; peduncle longer than both
rami; outer ramus slightly longer and broader than
inner ramus, outer margin lined with row of spines,
inner margin with row of robust setae, distal margin
rounded and bearing set of 1 long and 2 short
robust setae; inner ramus of almost uniform width,
outer margin with row of spines and apically with
group of 3 large and 1 very small robust setae.
Uropod 2 (U2, Fig. 7) slightly extending beyond
uropod 3; peduncle shorter than both rami, both
outer and inner margins with row of robust setae;
peduncular apex bearing triangular inter-ramal
process, outer ramus subequal to inner one in
length and apical margin with several robust setae;
apical margin of inner ramus with group of robust
setae. Uropod 3 (U3, Fig. 7) uniramous, peduncle
extremely short, about 1/10 as long as ramus;
ramus elongate, medially slightly wider, both outer
and inner margins with row of robust setae; apex
with 4 long stiff setae. Telson (T, Fig. 7) short, not
reaching tip of uropod 3 peduncle, semicircular,
ending in smaller medium circular lobe, with 2
telsonic angles bearding robust seta plus 2 or 3
plumose setae on each one.
Remarks: Although Klebang resembles
Acuminodeutopus, Rudilemboides, and
Wombalano as mentioned above, it can be readily
separated from the remaining genera by having
(1) the unique carpal configuration of gnathopod
2, (2) the largely expanded posterior margin of
the carpus of gnathopod 1, and (3) the setose
mandibular palp.
Etymology: The species is named in honor
of the late J. Laurens Barnard for his exceptional
work on world gammaridean amphipods.
Acknowledgments: This work was supported
by a grant (UKM-ST-08-FRGS0020-2009) from
the Ministry of Higher Education of Malaysia
and a UKM research grant (UKM-GGPM-
PLW-034-2010).
REFERENCES
Bano H, QB Kazmi. 2010. Grandidierella trispinosa, a new
species of amphipod from the Karachi coast, Pakistan
(Crustacea: Amphipoda: Aoridae). Turk. J. Zool. 34: 151-
157.
Barnard JL. 1970. Sublittoral Gammaridea (Amphipoda) of the
Hawaiian Islands. Smithson. Contrib. Zool. 34: 1-286.
Barnard JL. 1977. The cavernicolous fauna of Hawaiian lava
tubes 9. Amphipoda (Crustacea) from brackish lava
ponds on Hawaii and Maui. Pac. Insects 17: 267-299.
Barnard KH. 1935. Report on some Amphipoda, Isopoda, and
Tanaidacea in the collections of the Indian Museum. Rec.
Indian Mus. 37: 279-319.
Barnard KH. 1951. New records and descriptions of new
species of isopods and amphipods from South Africa.
Ann. Mag. Nat. Hist. 12: 698-709.
Barnard KH. 1952. Description of a new species of amphipod.
Trans. R. Soc. South Afr. 33: 279-282.
Bochert R, ML Zettler. 2010. Grandidierella (Amphipoda,
Aoridae) from Angola with description of a new species.
Crustaceana 83: 1209-1219.
Chevreux E. 1926. Amphipodes, 1: Gammariens (Cont.) In
Voyage Goelette Fr. 20: 365-398.
Chilton C. 1921. Fauna of the Chilka Lake. Amphipoda. Mem.
Indian Mus. 5: 519-558.
Griffiths CL. 1974. The Amphipoda of southern Africa. Part 3.
The Gammaridea and Caprellidea of Natal. Ann. South
Afr. Mus. 62: 209-264.
Hou ZE, SQ Li. 2002. A new species of the genus
Grandidierella from Lake Chaohu, China (Crustacea:
Amphipoda: Aoridae). Acta Zootaxon. Sin. 27: 225-234.
Ledoyer M. 1967. Amphipodes gammariens des herbiers
de phanerogammes marines de la region de Tulear
(Republique Malgache). Etude systematique et ecolo-
gique. Annales de l’Universite de Madagascar 5: 121-
170.
Ledoyer M. 1979. Expedition Rumphius II (1975). Crustaces
parasites, commensaux etc. VI. Crustaces Amphipodes
Gammariens. Bull. Mus. Natl. d'Hist. nat. Paris Sere 4 1:
137-181.
Ledoyer M. 1982. Crustaces Amphipodes Gammariens.
Famille des Acanthonozomatidae a Gammaridae. Faune
Madagascar 59: 1-598.
Moore PG. 1986. A new species in the genus Grandidierella
Coutière (Crustacea: Amphipoda) from an Australian solar
salt-works. J. Nat. Hist. 20: 1393-1399.
M y e r s A A . 1 9 7 0 . Ta x o n o m i c s t u d i e s o n t h e g e n u s
Grandidierella, with a description of G. dentimera sp. nov.
Bull. Mar. Sci. 20: 135-147.
M y e r s A A . 1 9 7 2 . Ta x o n o m i c s t u d i e s o n t h e g e n u s
Grandidierella Coutiére (Crustacea: Amphipoda) II. The
Malagasy species. Bull. Mus. Natl. d'Hist. nat. Paris Sere
3 Zool. 64: 789-796.
M y e r s A A . 1 9 8 1 . Ta x o n o m i c s t u d i e s o n t h e g e n u s
Grandidierella Coutière (Crustacea, Amphipoda). III.
Fijian, Australian and Saudi Arabian species. Bull. Mus.
Natl. d’Hist. nat. Paris Sere 4 3: 213-226.
Myers AA. 1998. The Amphipoda (Crustacea) of New
Caledonia: Aoridae. Rec. Aust. Mus. 50: 187-210.
Myers AA. 2009. Aoridae. In JK Lowry, AA Myers, eds.
Benthic Amphipoda (Crustacea: Peracarida) of the Great
Barrier Reef, Australia. Zootaxa 2260: 220-278.
Myers AA, JK Lowry. 2003. A phylogeny and a new classi-
fication of the Corophiidea (Amphipoda). J. Crust. Biol.
23: 443-485.
Othman BHR, H Morino. 2006. Listriella longipalma sp. nov.,
a new amphipod species (Crustacea: Liljeborgiidae) from
the Straits of Melaka, Malaysia. Zootaxa 1305: 21-32.
Ren X. 2006. Crustacea Amphipoda Gammaridea (I). Fauna
Sin. Invertebr. 41: 1-588.
Schellenberg A. 1925. Amphipoda, Beiträge zur Kenntnis der
Meeres fauna. Westafrikas 3: 113-204.
 Azman and Othman – A New Genus and Species of Aoroid Amphipod 247

Schellenberg A. 1936. Zwei neue Amphipoden des Stillen
Ozeans und zwei Berichtungen. Zool. Anzeiger 116: 153-
156.
Schellenberg A. 1938. Littoral Amphipoden des Topischen
Pazifiks. K. svenska Vetensk Akad. Handl. 16: 1-105.
Shoemaker CR. 1935. A new species of amphipod of the
genus Grandidierella and a new record for Melita nitida
from Sinaloa, Mexico. J. Wash. Acad. Sci. 25: 65-71.
Stebbing TRR. 1908. South African Crustacea (Part IV). Ann.
South Afr. Mus. 6: 1-96.
Stephensen K. 1938. Grandidierella japonica n. sp. A new
amphipod with stridulating organ from brackish water in
Japan. Annot. Zool. Jpn. 17: 179-184
Stephensen K. 1948. Amphipods from Curaçao, Bonaire,
Aruba and Margarita. Stud. Fauna Curaçao, Aruba,
Bonaire Venezuelan Islands 3: 1-20.
Zoological Studies 51(2): 248-258 (2012)

Leucosiid Crabs of the Genus Hiplyra Galil, 2009 (Crustacea: Brachyura:

Leucosiidae) from the Persian Gulf and Gulf of Oman, with Description
of a New Species
Reza Naderloo1,2,* and Michael Apel3
1
Research Institute and Natural Museum of Senckenberg, Senckenberganlage 25, 60325 Frankfurt am Main, Germany
2
School of Biology, College of Science, Univ. of Tehran, Tehran, Iran
3
Museum Mensch und Natur, Maria-Ward-Straße 1b, 80638 München, Germany. E-mail:apel@musmn.de

(Accepted September 26, 2011)

Reza Naderloo and Michael Apel (2012) Leucosiid crabs of the genus Hiplyra Galil, 2009 (Crustacea:
Brachyura: Leucosiidae) from the Persian Gulf and Gulf of Oman, with description of a new species. Zoological
Studies 51(2): 248-258. Four species of the leucosiid genus Hiplyra Galil, 2009, are reported here from the
Persian Gulf and Gulf of Oman. A new species, H. ramli sp. nov. was collected along the coast of Fujairah (United
Arab Emirates) in the western part of the Gulf of Oman. This new species differs from congeners in the shape
of the male 1st gonopod, the morphology of the female gonopore, and the armature of the 6th segment of the
male abdomen. Hiplyra elegans (Gravier 1920) is also recorded from the Iranian coast of the Gulf of Oman for
the 1st time. Two species from the area, H. variegata (Rüppell, 1830) and H. sagitta Galil 2009, are included in
this study, and a key is provided for the genus in the area. http://zoolstud.sinica.edu.tw/Journals/51.2/248.pdf

Key words: Brachyura, Leucosiidae, Hiplyra, Persian Gulf, Gulf of Oman.

C rabs of the family Leucosiidae are common
faunal elements of littoral and sublittoral soft-
sediment habitats in the Persian Gulf and Gulf of
Oman. They are the most diverse of all brachyuran
families (Stephensen 1946, Titgen 1982, Apel
2001). Apel (2001) listed 30 leucosiid species from
the Persian Gulf plus 2 additional species only
known from the Gulf of Oman and commented that
records of 4 more species reported from the region
are doubtful. Thus almost 1/6 of all brachyuran
crab species of the Persian Gulf belong to the
Leucosiidae (Apel 2001). Some additional new
species were also described and recorded from the
Persian Gulf in recent years, raising the number
of leucosiid crabs in the region to 35 species
(Naderloo and Sari 2005). Recently, Galil (2009)
added 2 more species (Hiplyra sagitta Galil, 2009,
and Lyphira perplexa Galil, 2009) when she revised
Philyra Leach, 1817. Therefore, the actual number
of leucosiid species recorded from the Persian Gulf
is currently 37. However, the leucosiid fauna of
the Gulf of Oman remains poorly known, and only
11 species of this group have been recorded there
(Nobili 1906, Stephensen 1946, Tan and Ng 1995,
Apel 2001). By adding 2 recorded species herein,
the number of known leucosiid species from the
Gulf of Oman rises to 13.
Galil (2009), in her recent treatment of
Philyra, divided the genus into 8 genera. Among
those, Hiplyra Galil, 2009, comprises 6 species
distributed in the Indo-West Pacific: H. elegans
Gravier, 1920; H. longimana A. Milne Edwards,
1874; H. michellinae Galil, 2009; H. platycheir
De Haan, 1841; H. sagitta Galil, 2009; and H.
variegata (Rüppell, 1830). The genus is chara-
cterized by elongate adult chelipeds with the

*To whom correspondence and reprint requests should be addressed. E-mail:rnaderloo@senckenberg.de

248
 Naderloo and Apel – Hiplyra in Northern Indian Ocean
cutting edge of the movable finger entire and
blade-shaped, the inner margin of the immovable
finger fringed with dense setae, and segments 2-6
of the male abdomen triangular being fused with
the lobate proximal margins (Galil 2009). Two
species of this genus, H. variegata (recorded by
Stephensen 1946) and H. sagitta (described by
Galil (2009) from the Persian Gulf) were previously
recorded from the Persian Gulf and adjacent
waters. A reexamination of the material identified
by Stephensen (1946) as H. variegata; however,
revealed that specimens collected from the Gulf of
Oman in Jask, differed from the descriptions and
illustrations of H. variegata provided by Rüppell
(1830) and Galil (2009). Those specimens are
assigned here to H. elegans, which was previously
known from Madagascar and Sri Lanka (Galil
2009). A new species is described from the east
coast of the United Arab Emirates (UAE) in the
Gulf of Oman. The number of species currently
placed in the genus Hiplyra is now raised to 7,
of which 4 occur in the Persian Gulf and Gulf of
Oman.
Drawings were made using a camera lucida
attached to a Leica MZ8 stereomicroscope (Leica,
Germany). The following abbreviations were used:
CL, carapace length; CB, carapace breadth; ML,
length of merus of male cheliped; G1, 1st male
gonopod; juv., juvenile; ovig., ovigerous; SMF,
Senckenberg Museum, Frankfurt am Main; ZMK,
Zoological Museum of Copenhagen.
SYSTEMATIC ACCOUNT
Hiplyra elegans (Gravier, 1920)
(Figs. 1, 2)
Philyra platychira Laurie 1906: 363.
Philyra variegata var. elegans Gravier, 1920: 379, figs. 1-7.
Philyra variegata Stephensen 1946: 89-93 (not Hiplyra
variegata (Rüppell, 1830) (part of the material from st. 73,
Jask, Iran)).
Philyra elegans Galil 2009: 292-293, 315 (in key), fig. 7.
Type locality: Madagascar.
Material examined: Gulf of Oman, 5 ♂♂
(ZMK CRU929, CL = 9.90 mm, CB = 9.33 mm),
tidal zone, St. 73, Jask, Gulf of Oman, 20 Apr.
1937, G. Thorson.
Additional material: 7 ♂♂ , 1 ♀ (SMF 11119),
Madagascar, Stumpf and Ebenau; 1 ♂ , 1 ♀ (SMF
11118), Madagascar.
Diagnosis: Carapace about as long as wide;
anterior margin of efferent channel straight,
249
separated from lateral granulated margin by deep
U-shaped incision; somite 6 of male abdomen
smooth, with no process, male telson elevated
on lateral portion; G1 widened distally, with very
small apical process subdistally, directed laterally;
1st somite of female abdomen not lobate; female
gonopore with membranous oval process directed
anteroposteriorly.
Redescription: Carapace (Fig. 2A) about as
long as wide, very slightly longer (CL/CB = 1.05),
distinctly convex; dorsal surface finely punctate
medially, laterally, and posteriorly; carapace re-
gions weakly defined, grooves delimiting cardiac
and intestinal regions distinct; branchiocardiac
grooves shallow; frontal region nearly smooth,
slightly depressed immediately behind frontal
ridge. Front nearly as wide as posterior margin of
carapace, produced, slightly extended medially;
shallow furrow extending posteriorly in frontal
region. Upper orbital margin finely granular, deep
fissure occurring laterally, short setae along inner
margin of upper orbital margin. Epibranchial
margin moderately swollen, with small granules;
anterolateral margin with large granules, becoming
smaller posteriorly; posterolateral margin regularly
granular, granules continuing to posterior margin,
small granules below posterior margin. Anterior
margin of efferent channel straight, separated
from lateral granulated margin by deep U-shaped
incision. Subhepatic and pterygostomial regions
minutely granular.
Ischium of 3rd maxilliped distinctly longer than
merus, about 1.5-times merus length, outer surface
faintly granular; merus elongated-triangular, outer
surface weakly granular, large granules on distal
margin; exopod large, wider distally, outer surface
smooth, margins minutely serrate. Thoracic
sternal plates granular, granules larger anteriorly;
anterior margin of abdominal sulcus regularly with
large granules.
Male chelipeds (Figs. 1A, 2A) long; merus
long, very slightly shorter than carapace breadth
(mean ML/CB = 0.95); upper surface proximally
granular, anterior margin with large granules,
becoming larger medially; anterior surface with
small granules; posterior margin with small
granules, becoming larger proximally. Anterior
lower and upper margins of carpus minutely
granular. Manus long, with smooth upper surface,
row of small granules on lower portion of inner
surface extending from proximal part almost to
base of fingers; lower margin granular, upper
margin faintly serrate. Movable finger distinctly
shorter than manus, about 2/3 of manus length,
250 Zoological Studies 51(2): 248-258 (2012)

arched, cutting edge blade-shaped; immovable
finger shorter than movable finger, curved gently
downward; cutting edge with small teeth along
edge, 2 distal ones large-triangular; short dense
setae along cutting edge of immovable finger,
shorter distally; short setae along round process of
distal margin of manus on articulation to movable
finger.
Male abdomen (Fig. 1B) elongated-triangular;
segments 2-5 completely fused, proximal margin
of fused somites 2-5 with large granules, medial
depression proximally, lateral margins with small
granules; somite 6 firmly merged with fused
segments 2-5, not freely movable; lateral margin
sharply diverging proximally, gently converging
distally along most of its length, outer surface
smooth, with no process; telson elongated-
triangular, distinctly shorter than somite 6, with
2 elevations basally on lateral portion, margins
smooth.
G1 (Fig. 1C, D) curved laterally in proximal
1/3 (Fig. 1D); apical portion wide; small apical
process subdistal, directed laterally; long setae
around apical process; sperm channel curving on

(A)

(B)
1 mm

(C) (D)

(E)
1 mm

1 mm

1 mm

Fig. 1. Hiplyra elegans (Gravier, 1920). Male holotype (ZMK CRU929) (A-D); female paratype (SMF 11119) (E). (A) cheliped of male
(left), upper surface; (B) male abdomen; (C) G1 (right) dorsal surface; (D) G1 (right), ventral surface; (E) female gonopore (right).
 Naderloo and Apel – Hiplyra in Northern Indian Ocean
dorsal surface.
Female gonopore (Fig. 1E) on inner anterior
edge of sternite 5, nearly round; large membranous
oval process directed anteroposteriorly. First
somite of female abdomen not distinctly trilobate,
with granular distal margin.
Remarks: Stephensen (1946) listed subs-
tantial material from the Persian Gulf and Gulf of
Oman under the name Philyra variegata (Rüppell,
1830). We had the opportunity to reexamine most
of Stephensen’s (1946) material, compared it with
Rüppell’s type material of H. variegata from the
Red Sea, and found that some specimens were
not H. variegata but H. elegans instead. Hiplyra
elegans is distinguished from H. variegata by the
carapace shape, morphology of the G1, the form
of the male abdomen, and the gonopore structure
of females. The carapace of H. elegans is slightly
longer than wide (mean CL/CB = 1.05), while the
carapace of H. variegata is as long as wide, and
even in large specimens is only slightly wider than
long. The apical process of G1 in H. elegans is
very small, subdistal, and directed laterally (Fig.
1D), while in H. variegata, the small apical process
is completely distal and directed ventrally (Fig. 7A).
(A)
(B)
Fig. 2. Hiplyra elegans (Gravier, 1920). Male holotype, CL =
9.90 mm, CB = 9.33 mm (ZMK CRU929). (A) dorsal surface; (B)
ventral surface.
251
For the male abdomen, the telson in H. elegans
has 2 distinct elevations proximally on the lateral
portions (Fig. 1D), while that of H. variegata is
smooth, with no elevation (Fig. 7C). In addition,
these 2 species have distinct morphologies of
the female gonopore allowing females of these
congeners to readily be distinguished. While H.
elegans has a distinct oval membranous process
which is directed anteroposteriorly (Fig. 1E), the
female gonopore of H. variegata has a large
opening (Fig. 7D) on the inner side of a prominent
elevation. As Galil (2009) discussed, the 1st
somite of the female abdomen of H. variegata
is distinctly trilobate, while this somite is simple
in females of H. elegans. It must be noted that
drawings provided by Stephensen (1946: 88, fig.
15F-K) clearly depict H. variegata, in particular,
the male abdomen which shows a smooth telson
lacking any elevation.
Distribution: Madagascar, Gulf of Oman, Sri
Lanka.
Hiplyra ramli sp. nov.
(Figs. 3, 4)
Type locality: Al Aqah, Fujairah, east coast of
UAE, Gulf of Oman.
Material examined: Holotype 1 ♂ (SMF
38466, CL = 7.4. mm, CB = 6.8 mm), Al Aqah, near
Sandy Beach Hotel, Fujairah, UAE, Gulf of Oman,
25°30'N, 56°22'E, sandy substrate, under stones
and corals, 3-4 m depth, 4 July 1995, M. Apel.
Paratypes: 6 ♂♂ , 9 ♀♀ (4 ovig.) (SMF
38467), same data as for holotype.
Diagnosis: Carapace slightly longer than wide;
anterior margin of efferent channel nearly straight,
separated from lateral granulated margin by deep
U-shaped incision; somite 6 of male abdomen with
triangular arrow-shaped process on distal portion;
telson wide-triangular, swollen on basal portion; G1
distally widened, with small apical process directed
dorsally; female gonopore obliquely directed
anterodorsally, small membranous oval process on
outer margin of opening.
Description: Carapace (Fig. 4A) slightly
longer than wide (CL/CB = 1.1), distinctly convex;
dorsal surface finely punctate medially, laterally,
and posteriorly; carapace regions weakly defined,
distinct grooves delimiting cardiac and intestinal
regions; branchiocardiac grooves shallow;
frontal region nearly smooth, slightly depressed
immediately behind frontal ridge. Front slightly
shorter than posterior margin of carapace,
produced, slightly extended medially; shallow
252 Zoological Studies 51(2): 248-258 (2012)

furrow extending to rear in frontal region. Upper regions minutely granular.

orbital margin finely granular, deep fissure present
laterally.
Epibranchial margin with small granules;
anterolateral margin with large granules anteriorly,
becoming smaller posteriorly; posterolateral margin
finely granular, granules continuing to posterior
margin, small granules below posterior margin.
Anterior margin of efferent channel nearly straight,
separated from lateral granulated margin by deep
U-shaped incision. Subhepatic and pterygostomial
Ischium of 3rd maxilliped slightly longer than
merus, outer surface nearly smooth, distal margin
minutely granular; merus long-triangular, outer
surface finely granular, granules larger distally
on outer surface; exopod large, wider distally,
outer surface smooth, margins minutely serrate.
Abdominal sternums granular, granules larger
anteriorly; anterior margin of abdominal sulcus
granular.
Male chelipeds (Figs. 3A, 4A) with moderately

(A)

1 mm

(B)

(C) (D)

(E)
1 mm

1 mm

Fig. 3. Hiplyra ramli sp. nov. Male holotype (SMF 38466) (A-D); female paratype (SMF 38466) (E). (A) cheliped of male (left), upper
surface; (B) male abdomen; (C) G1 (right) dorsal surface; (D) G1 (right), ventral surface; (E) female gonopore (right).
 Naderloo and Apel – Hiplyra in Northern Indian Ocean
long merus, distinctly shorter than carapace
breadth (mean ML/CB = 0.72); upper surface
granular proximally; anterior margin with large
granules, granules becoming larger medially;
posterior margin with small granules, proximally
moderately larger. Anterior lower and upper
margins of carpus minutely granular. Upper
surface of manus smooth, faint row of very small
granules on lower portion, extending parallel
to lower margin in proximal 1/2; lower margin
granular; upper margin faintly serrate. Movable
finger slightly shorter than manus, arched
medially, cutting edge blade-shaped; immovable
finger shorter than movable finger, curved gently
downward; cutting edge with small triangular teeth
distally, 2 or 3 distal ones larger; short dense setae
along cutting edge of immovable finger, shorter
distally; short setae along round process of distal
margin of manus on articulation to movable finger.
Male abdomen (Figs. 3B, 4B) long-triangular,
scarcely granular; somites 2-5 completely fused,
fused somites proximally with large granules,
medially with depression, lateral margins
(A)
(B)
Fig. 4. Hiplyra ramli sp. nov. male holotype, CL = 7.4. CB = 6.8
(SMF 38466). (A) dorsal surface; (B) ventral surface.
253
proximally with small granules; somite 6 firmly
merged to fused somites 2-5, not freely movable;
lateral margin sharply diverging proximally, gently
converging distally along most of its length,
prominent elevated arrow-shaped process distally
on outer surface; telson elongate-triangular, slightly
shorter than somite 6, with 2 processes at basis of
lateral portion, margins smooth.
G1 (Fig. 3C, D) slightly curved laterally,
narrowing medially; apical portion expanded, with
long setae on lateral margin, relatively short setae
on mesial margin; small apical process directed
dorsally; sperm channel curved on dorsal surface.
Female gonopore (Fig. 3E) on inner ante-
rior edge of sternite 5, obliquely directed antero-
dorsally; small membranous oval process on outer
margin of opening.
Remarks: Hiplyra ramli sp. nov. is a relatively
small-sized species, which is morphologically
closest to H. sagitta and H. elegans. With regard
to the lengths of the carapace and male chelipeds,
the new species; however, is clearly distinct from
H. elegans and more closely allied to H. sagitta.
Hiplyra elegans has a slightly wider carapace
(CL/CB = 1.05), while this ratio in the 2 other
species is 1.1. The relatively long adult chelipeds
of H. elegans with a long merus (ML/CB = 0.95)
distinguishes H. elegans from H. ramli sp. nov.,
which has only moderately long chelipeds
(ML/CB = 0.72). There are 3 further distinct
differences between the new species and all other
congeners. The male abdomen of H. ramli sp. nov.
has a wide triangular process on the distal portion
of segment 6 (Fig. 3B), while this process in H.
sagitta is distinctly elongate and arrow-shaped,
with a distinct groove (Fig. 5C), and the male
abdominal somite 6 of H. elegans is completely
smooth, with no process (Fig. 1B). Hiplyra ramli
sp. nov. has a telson with 2 proximal elevations
which are very similar to those of H. elegans
and clearly distinct from the narrow and smooth
telson of H. sagitta. G1 of the new species is
characterized by having a small distally broadened
apical process which is directed dorsally, while in
the 2 other species, the apical process is directed
laterally (Fig. 3C, D).
Furthermore, the distinctive form of the
female gonopore easily distinguishes this species
from its congeners (see “Remarks” for H. elegans).
It should be noted that the available
specimens of the new species revealed that the
chelipeds of males are slightly longer than those of
females, with the ratio of ML/CB in males ranging
0.64-0.77 (n = 10), while this ratio was 0.56-0.64
254 Zoological Studies 51(2): 248-258 (2012)
(n = 13) in females.
Distribution: Presently only known from the
Gulf of Oman coast of the UAE (Fujairah).
Etymology: The species is named after the
Arabic word “raml” for sand, since it was collected
in sandy substrate. The name is used as a noun
in apposition.
Hiplyra sagitta Galil, 2009
(Figs. 5, 6)
Philyra platychira Alcock 1896: 242.
Philyra platycheir Tirmizi and Kazmi 1986: 100, fig. 29.
Philyra variegata Stephensen 1946: 89-93 (not Hiplyra
variegata (Rüppell 1830) (part of the material from st. 27,
near Bushehr, Persian Gulf)).
Hiplyra sagitta Galil 2009: 296-297, 315 (in key), figs. 11, 12A.
(A) (B)
(D)
1 mm
Fig. 5. Hiplyra sagitta Galil, 2009. Male (ZMK CRU880) (A-C); female (SMF 38393) (D). (A) G1 (right) dorsal surface; (B) G1 (right),
ventral surface; (C) male abdomen; (D) female gonopore (right).
Type locality: Near Bushehr, Persian Gulf,
Iran.
Material examined: Persian Gulf: 2 ♂♂
(CL = 9.68, 16.04 mm, CB = 9.45, 14.56 mm)
(ZMK CRU880), St. 32, 7.5 m, N of Kharg I., G.
Thorson, 23 Mar. 1937; 1 ♂ (CL = 16.67 mm,
CB = 15.12 mm), 3 ♀♀ (CL = 15.42-19.07 mm,
CB = 14.09-17.66 mm) (SMF 38392), 22 m,
Kuwait, 28°53'N, 48°24'E, trawl, 24 Apr. 1995, F.
Krupp; 2 ♀♀ (CL = 15.55, 17.00 mm, CB = 14.23,
16.13 mm) (SMF 38393), 13-17 m, Kuwait,
29°10'N, 48°28'E, trawl, 23 Apr. 1995, F. Krupp.
Diagnosis: Carapace (Fig. 6A) slightly
longer than wide (CL/CB = 1.1); dorsal surface
finely punctate medially, laterally, and posteriorly.
Anterior margin of efferent channel straight,
(C)
1 mm
 Naderloo and Apel – Hiplyra in Northern Indian Ocean
separated from lateral granulated margin by
somewhat wide U-shaped incision. Ischium of
3rd maxilliped slightly longer than merus, about
1.2-times merus length. Male abdomen (Figs.
5C, 6B) elongate-triangular; segment 6 with long
arrow-shaped process, telson narrow and smooth.
Male chelipeds (Fig. 6) moderately long; movable
finger about as long as manus, or slightly shorter.
Immovable finger with small denticles along cutting
edge, 2 large triangular teeth subdistally, cutting
edge with dense short setae. G1 (Fig. 5A, B)
long, narrow, with moderately large apical process,
directed laterally. Female gonopore (Fig. 5D)
small, on inner side of large elevation.
Remarks: Hiplyra sagitta was recently
described from Bushehr in the Persian Gulf by
Galil (2009). She mentioned that H. sagitta differs
from its congeners by having a triangular incision
which separates the anterior margin of the efferent
channel from the lateral granulated margin (Galil
2009: 297), while this incision is narrow and
U-shaped in the 3 other species she examined.
Another distinct character mentioned by Galil
(2009) is the particular arrow-shaped process on
the distal part of male abdominal somite 6 which
(A)
(B)
Fig. 6. Hiplyra sagitta Galil, 2009, male, CL = 16.79, CB
= 15.13 mm (SMF 38392). (A) dorsal surface; (B) ventral
surface.
255
is absent in all other known congeners except
for the newly described H. ramli sp. nov. In the
latter, the process; however, is distinct from that
of H. sagitta in its short and triangular form (Fig.
3B). Apart from the 2 discriminative characters
presented by Galil (2009), we add the feature of
the distinctive apical part of G1 and the structure
of the female gonopore, which separate H. sagitta
from other congeners treated here (discussed
under “Remarks” of the new species, H. ramli sp.
nov.).
The holotype and paratypes described by
Galil (2009) are from Stephensen’s (1946) material
examined under H. variegata from Bushehr in the
Persian Gulf.
Hiplyra sagitta is one of the largest species in
the genus with the largest male found at Kharg I. in
the Persian Gulf (CL = 16.04 mm, CB = 14.56 mm)
and the largest female recorded from Kuwait in the
Persian Gulf (CL = 19.07 mm, CB = 17.66 mm).
Distribution: Persian Gulf, India, Andaman
Sea.
Hiplyra variegata (Rüppell, 1830)
(Figs. 7, 8)
Myra variegata Rüppell 1830: 17, pl. 4-4.
Philyra platycheira Paulson 1875: 83, pl. 10, fig. 3. Alcock
1896: 242 (specimens from the Persian Gulf).
Philyra variegata Nobili 1906: 169. Laurie 1915: 410. Balss
1915: 14. Stephensen 1946: 89, figs. 15f-k, 16. Serène
1968: 46. Guinot 1967: 249 (in list). Titgen 1982: 248 (in
list).
Philyra platychira Balss 1915: 14.
Hiplyra variegata Galil 2009: 287-299, 315 (in key), fig. 13.
Tape locality: Red Sea, Egypt.
Material examined: Lectotype 1 ♂ (CL =
7.66 mm, CB = 7.73 mm) (SMF 11121), among
corals, Sinai Peninsula, Egypt, id. E. Rüppell,
1827. Paralectotype: 9 ♂♂ , 4 ♀♀ (SMF 11121),
data same as for lectotype. Persian Gulf: 2 ♂♂
(CL = 6.05, 10.10 mm, CB = 5.68, 9.24 mm) (ZMK
CRU886), 56 m, sandy clay, 13 nautical miles W
of outermost light-buoy at Bushehr, Bushehr, G.
Thorson, 13 Mar. 1937; 27 ♂♂ , 11 ♀♀ (SMF
38462), sandy, 6 m depth, S of Rams, UAE,
25°50'N, 55°00'E, 11 July 1995, M. Apel; 1 ♂ ,
2 ♀♀ (ovig.) (SMF 38463), sandy, 0-6 m depth, N
coast of As Sham, Ras al Khaymah, UAE, 26°02'N,
55°05'E, 10 July 1995, M. Apel. Red Sea: 1 ♀
(ovig.) (SMF 38464), mangroves, Umm al Gamar I.,
Egypt, 27°22'N, 33°55'E, 13 Sept. 1994.
Diagnosis: Carapace (Fig. 8A) as long as
wide, slightly wider than long; dorsal surface
256 Zoological Studies 51(2): 248-258 (2012)

finely granular laterally and posteriorly. Anterior
margin of efferent channel straight, separated from
lateral granulated margin by narrow U-shaped
incision. Ischium of 3rd maxilliped distinctly longer
than merus, about 1.5-times merus length. Male
abdomen (Figs. 7C, 8B) elongate-triangular;
segment 6 and telson completely smooth, with
no process or elevation. Male cheliped (Fig. 8A)
moderately long; movable finger about 1.5-times
manus length. Immovable finger with small
denticles along cutting edge, 2 large triangular
teeth subdistally, cutting edge densely covered with
short setae. G1 (Fig. 7A, B) long, narrow; small
apical process distally, directed ventrally. Female
gonopore (Fig. 7D) on oval transverse elevation,
with large opening directed anteroposteriorly.
Remarks: This small-size species was briefly
described and illustrated by Rüppell (1830) from
the Red Sea. Galil (2009) redescribed the species
by examining the type material and mentioned
that the original description and illustration of the
species were not correct, as Rüppell (1830) did not
mention the minutely dentate row beyond the setae
along the cutting edge of the immovable finger.
This character; however, was mentioned by Alcock
(1896: 243), who examined material from the

(A) (B) (C)

1 mm
(D)

1 mm

Fig. 7. Hiplyra variegata (Rüppell, 1830). Male paratype (SMF 11121) (A-C); paratype (SMF 11121) (D). (A) G1 (right), ventral
surface; (B) G1 (right), dorsal surface; (C) male abdomen; (D) female gonopore (right).
 Naderloo and Apel – Hiplyra in Northern Indian Ocean
Persian Gulf and by Nobili (1906) who studied the
type material of H. variegata (Rüppell 1830). Such
an indentation was seen in the material examined
in the present paper. Galil (2009) recorded 2 other
characters as discriminative for distinguishing the
species from its congeners, including a marbled
color pattern of the carapace and 2 triangular
teeth distally on the cutting edge of the immovable
finger.
We found 3 additional morphological chara-
cters which we believe are even more significant
and readily distinguish it from all other congeners
treated here. Hiplyra variegata has a unique
G1 structure and female gonopore, and its male
abdomen is completely smooth with no elevation
or process on the 6th male abdominal somite
or telson (Fig. 7A-D). Detailed discussions on
the differences between H. variegata and its
congeners are presented under “Remarks” of the
former species.
Distribution: Kenya, Red Sea, Gulf of Aden,
Persian Gulf, Gulf of Oman.
(A)
(B)
Fig. 8. Hiplyra variegata (Rüppell, 1830), male lectotype,
CL = 7.66, CB = 7.73 mm (SMF 11121). (A) dorsal surface; (B)
ventral surface.
257
DISCUSSION
The genus Hiplyra Galil, 2009, was recently
separated from Philyra Leach, 1817, using the
following characters: apical process of G1 minute,
the presence of a thick fringe of setae on the inner
margin of the movable finger, and the cheliped
merus being longer than the carapace in males
(Galil 2009: 314). The 1st 2 characters clearly
distinguish Hiplyra from Philyra. The 3rd character
is rather confusing as in all known species of
Hiplyra, even the type species, H. platycheir (De
Haan, 1841), the merus is actually clearly shorter
than the carapace.
Hiplyra currently includes 7 species, which are
primarily distinguished from each other using the
morphology of the carapace and male abdomen
(Galil 2009). Apart from these 2 characters, 2
more-important discriminative characters including
G1 and the female gonopore were found to be
useful here to separate closely related species.
The morphology of the female gonopore allows
females of the different Persian Gulf species to be
distinguished and will probably work for other taxa
as well. The 4 species discussed in the present
study are morphologically close and all are found
in sandy substrates of the shallow subtidal zone.
Hiplyra variegata and H. elegans have patchy
distributions in the western Indian Ocean which
could be largely due to the lack of extensive
sampling, particularly in the subtidal zone. We
believe that the recently described species, H.
sagitta Galil, 2009, and H. ramli sp. nov. will be
found further westwards when further surveys are
done in those regions.
Key to the genus Hiplyra known from the
Persian Gulf and Gulf of Oman
1. Somite 6 of male abdomen with elevated arrow-shaped
process ............................................................................. 2
Somite 6 of male abdomen smooth, without a process .... 3
2. Somite 6 of male abdomen with long arrow-shaped
process, creating distinct groove; telson narrow, smooth,
with no elevation; G1 distally narrow, with relatively large
apical process directed laterally .................. Hiplyra sagitta
3. Somite 6 of male abdomen with triangular arrow-shaped
process on distal portion; telson wide-triangular, swollen
on basal portion; G1 distally widened, with small apical
process directed dorsally ................... Hiplyra ramli sp. nov.
4. Male telson elevated on lateral portion; G1 widened distally,
with very small apical process subdistally, directed laterally;
1st somite of female abdomen not lobate ............................
.................................................................... Hiplyra elegans
- Male telson completely smooth; G1 narrowing distally,
apical process located distally, directed ventrally; 1st somite
of female abdomen trilobate .................... Hiplyra variegata
258 Zoological Studies 51(2): 248-258 (2012)
Acknowledgments: We are grateful to J. Olesen
(ZMK) for kindly providing us with the valuable
brachyuran material collected from the “Danish
Scientific Expedition in Iran” conducted in 1937/38.
We are indebted to Prof. M. Türkay (SMF) for
his support as supervisor of both authors, and
to Deutscher Akademischer Austausch Dienst
(DAAD) for financial support in the form of a PhD
scholarship to R. Naderloo. Furthermore, we are
grateful to J.A. Khan and the team of the Arabian
Seas Expedition who gave great support to one of
the authors (M. Apel) during a survey of the UAE
coastline in 1995.
REFERENCES
Alcock A. 1896. Materials for carcinological fauna of India. N.
2: The Brachyura Oxystomata. J. Assoc. Soc. Beng. 65:
134-296.
Apel M. 2001. Taxonomie und Zoogeographie der Brachyura,
Paguridea und Porcellanidae (Crustacea: Decapoda) des
Persisch-Arabischen Golfes: 1-268. PhD dissertation,
Johann Wolfgang Goethe-Univ., Frankfurt am Main,
Germany.
Balss H. 1915. Anomuren, Dromiaceen und Oxystomen.
XXXI. Die Decapoden des Roten Meeres. Expeditionen
S.M. Schiff “Pola” in das Rote Meer nordliche und sudliche
halfte 1895/96-1897/98. Berichte der Kommission für
ozeanographische Forschungen, 18 pp.
Galil B. 2009. An examination of the genus Philyra Leach,
1817 (Crustacea, Decapoda, Leucosiidae) with description
of seven new genera and six new species. Zoosystema
31: 279-320.
Gravier C. 1920. Sur une collection de crustacés recueillis à
Madagascar par M. le Lieutenant Decary. Bull. Mus. Hist.
Nat. Paris 26: 376-383.
Guinot D. 1967. La faune carcinologique (Crustacea,
Brachyura) de l’Ocean Indien occidental et de la Mer
Rouge. Catalogue remarques biogéographiques et
bibliographie. Mém. Inst. Fond. Afr. 77: 235-352.
Laurie RD. 1915. 1906. Report on the Brachyura collected
by Professor Herdman at Ceylon, in 1902. In WA
Henderman ed. Report to the Government of Ceylon on
the Pearl Oyster Fisheries of the Gulf of Manaar. Part v.
Supplementary Report 40: 349-432, pls. 1, 2.
Laurie RD. 1915. Reports on the marine biology of the
Sudanese Red Sea. XXI. On the Brachyura. J. Linn.
Soc. Lond. 31: 407-475, figs. 1-5, pls. 42-45.
Naderloo R, A Sari. 2005. Iranian subtidal leucosiid crabs
(Crustacea: Decapoda: Brachyura) of the Persian Gulf:
taxonomy and zoogeography. Iran. J. Anim. Biosys. 1:
31-46.
Nobili G. 1906. Faune carcinologique de la Mer Rouge
décapodes et stomatopodes. Ann. Sci. Nat. (Zool.) 4:
1-347, figs. 1-12, pls. 1-11.
Paulson OM. 1875. Izsledovaniya rakoobraznykh Krasnago
Morya s zametkami otnositel’no rakoobraznykh drugikh
morei. Chast‘ I. Podophthalmata i Edriophthalmata
(Cumacea). Kiew, Kul’zhenko 1875: I-XIV + 1-144, pls.
1-21.
Rüppell E. 1830. Beschreibung und Abbildung von 24
Arten kurzschwanzigen Krabben, als Beitrag zur
Naturgeschichte des rothen Meers. Frankfurt, Germany:
H.L. Brönner, 28 pp.
Serène R. 1968. Prodromus for a check list of the non-
planctonic marine fauna of South East Asia. Sing. Nat.
Acad. Sci. Spec. Pub. 1: 1-122.
Stephensen K. 1946. The Brachyura of the Iranian
Gulf. Danish scientific investigations in Iran, part IV.
Copenhagen: E. Munksgaard, pp. 57-237.
Tan CGS, PKL Ng. 1995. A revision of the Indo-Pacific genus
Oreophorus Rüppel, 1830 (Crustacea: Decapoda:
Brachyura: Leucosiidae). In B Richer De Forges, ed.
Les fonds meubles des lagons de Nouvelle-Calédonie
(Sédimentologie, benthos). Etudes & Thèses, Vol. 2.
Paris: Orstom, pp. 101-189.
Tirmizi NM, QB Kazmi. 1986. Marine fauna of Pakistan. 4.
Crustacea: Brachyura (Dromiacea, Archaebrachyura,
Oxystomata, Oxyrhyncha). Publication I, BCCI Foun-
dation Chair, Institute of Marine Sciences, Univ. of
Karachi, Pakistan, pp. 1-244.
Titgen RH. 1982. The systematics and ecology of the
decapods of Dubai, and their zoogeographic relationships
to the Persian Gulf and the western Indian Ocean. PhD
dissertation, Texas A&M Univ., Texas, USA.
Zoological Studies 51(2): 259-271 (2012)

A Predictive Model to Differentiate the Fruit Bats Cynopterus brachyotis

and C. cf. brachyotis Forest (Chiroptera: Pteropodidae) from Malaysia
Using Multivariate Analysis
Vijaya K. Jayaraj1,*, Charlie J. Laman2, and Mohd T. Abdullah2
1
Faculty of Agro Industry and Natural Resources, Universiti Malaysia Kelantan, Locked bag 36, Pengkalan Chepa, Kelantan 16100,
Malaysia
2
Department of Zoology, Faculty of Resource Science and Technology, Universiti Malaysia Sarawak, Kota Samarahan 94300, Sarawak,
Malaysia

(Accepted September 7, 2011)

Vijaya K. Jayaraj, Charlie J. Laman, and Mohd T. Abdullah (2012) A predictive model to differentiate the
fruit bats Cynopterus brachyotis and C. cf. brachyotis Forest (Chiroptera: Pteropodidae) from Malaysia using
multivariate analysis. Zoological Studies 51(2): 259-271. Field discrimination of Cynopterus brachyotis and C.
cf. brachyotis Forest (as designated by Francis 2008) in southern Thailand, Peninsular Malaysia, and Borneo is
problematic. These 2 forms are sympatric in this region but are confined to different habitat types: C. brachyotis
inhabits open habitats, orchards, and agricultural areas, while C. cf. brachyotis Forest is confined to primary and
old secondary forests. In this study, we attempted to develop prediction models to identify both C. brachyotis
and C. cf. brachyotis Forest in this region based on multivariate statistics. Two predictive models were
generated using a canonical discriminant function, and it was found that 5 characters can be used to accurately
identify museum vouchers of C. brachyotis and C. cf. brachyotis Forest. Four characters are needed for field
identification of these 2 forms of Cynopterus in southern Thailand, Peninsular Malaysia, and Borneo. A review
of the current taxonomy and classification indicated that there is a need to describe the 6 existing forms of the C.
brachyotis complex in the Indo-Malayan region. This will aid conservationists, field ecologists, and taxonomists
in taxonomic- and conservation-related decisions about this species complex.
http://zoolstud.sinica.edu.tw/Journals/51.2/259.pdf

Key words: Cynopterus brachyotis, Discriminant function analysis, Habitat type.

T he genus Cynopterus F. Cuvier 1824, Discriminating between species in this genus is

commonly known as dog-faced fruit bats or short-
nosed fruit bats are widely distributed in the Indo-
Malayan region (Corbet and Hill 1992). The
taxonomic status of this genus has undergone
many revisions, and the most recent classification
by Simmons (2005) lists 7 species in this genus:
C. brachyotis (Müller, 1838); C. horsfieldii Gray,
1843; C. luzoniensis Peters, (1861); C. minutus
Miller, 1906; C. nusatenggara Kitchener and
Maharadatunkamsi, 1991; C. sphinx (Vahl,
1797); and C. tithaecheilus (Temminck, 1825).
often problematic given the many variations and
overlap between species representatives across
a geographical gradient. Work such as that by
Bumrungsri and Racey (2005) is often done to
discriminate similar sympatric species in this
genus.
The nominate C. brachyotis type specimen
was described by Müller (1838), but currently the
status of C. brachyotis is uncertain, as recent
studies indicated that it may actually be a complex
of species (Campbell et al. 2004). Corbet and Hill

*To whom correspondence and reprint requests should be addressed. Tel: 60-9-7717087. Fax: 60-9-7717232.
E-mail:jayaraj_vijayakumaran@yahoo.com

259
260 Jayaraj et al. – A Model to Differentiate Cynopterus brachyotis Forms
(1992) listed 19 synonyms of C. brachyotis, but
Simmons (2005) recognized only seven of them,
with most of them lacking data on their status and
current distribution. Abdullah (2003) compared
morphological measurements of Cynopterus
from various sources (Andersen 1912, Hill and
Thonglongya 1972, Lekagul and McNeely 1977,
Medway 1978, Hill 1983, Payne et al. 1985,
Kitchener and Maharadatunkamsi 1991 1996,
Ingle and Heaney 1992, Nor 1996) and found
that a lot of morphological measurements overlap
within and between species across its distribution.
This species is widely distributed throughout
Southeast Asia (Fig. 1) and can be found at areas
up to 1600 m in elevation in Borneo (Lekagul and
McNeely 1977, Medway 1978, Bergmans and
Rozendall 1988, Corbet and Hill 1992, Peterson
and Heaney 1993, Abdullah 2003). It can be found
in many habitats (but most frequently in disturbed
forest) including lower montane forest, dipterocarp
forest, gardens, mangroves, and strand vegetation.
Francis (1990) found that there were forearm
length differences in C. brachyotis caught in
primary forests and that from secondary habitats
in Sepilok, Sabah. This observation was later
investigated by Abdullah et al. (2000) and Abdullah
(2003) using molecular and external morphometric
e loading and aspect ratio was not an informative
b character that can be used to differentiate the 2
c forms of C. brachyotis. Another study on flight
a parameters also showed similar results (Menon
h 2007).
f
g
d
Fig. 1. Distribution of 8 subspecies of C. brachyotis in the
Indo-Malayan region (Mickelburgh et al. 1992, Simmons
2005). (a) C. b. altitudinis found in highlands of the Main
Range, Peninsular Malaysia; (b) C. b. brachysoma found in
the Andaman Is.; (c) C. b. ceylonensis found in Sri Lanka; (d)
C. b. concolor found on Enganno I.; (e) C. b. hoffeti found in
Vietnam; (f) C. b. insularum found in the Kangean Is. and Laut
Kecil Is.; (g) C. b. javanicus found in Bali, Java, Madura, and
Penidah; and (h) C. b. brachyotis found in Bangka, Belitung,
Borneo, Lombok, the Nicobar Is., Peninsular Malaysia, the
Philippines, Singapore, Sulawesi, Sumatra and Thailand.
data on samples from Borneo and Peninsular
Malaysia to the southern tip of Thailand. Results of
those studies showed that 2 forms of C. brachyotis
inhabited 2 contrasting habitats (in Peninsular
Malaysia and Borneo). The larger form was found
to inhabit open areas, whereas the smaller form
was confined to primary forests. Abdullah et al.
(2000) postulated that these differences found in
C. brachyotis are based on ecological differences
in the habitats they occupy. Later Campbell
et al. (2004) reexamined the species complex
using different genetic markers and discovered
4 additional distinct lineages in the C. brachyotis
complex scattered in the Indo-Malayan region.
These 4 lineages are respectively found in India,
Myanmar, Sulawesi, and the Philippines. Abdullah
and Jayaraj (2006) later performed a cluster
analysis on the type specimen of C. brachyotis
using morphological measurements described
by Müller (1938), and the results showed that the
nominate C. brachyotis was clustered with the
larger form of C. brachyotis.
A recent study using microsatellites and 2
mitochondrial (mt)DNA genes by Fong (2011)
showed congruent findings with Abdullah et al.
(2000), Abdullah (2003), Campbell et al. (2004
2006), and Julaihi (2005) of the existence of
2 C. brachyotis lineages in southern Thailand,
Peninsular Malaysia, and Borneo. The morpho-
metrics of this species also showed same
findings but there were misclassifications of some
samples (Jayaraj et al. 2004 2005). Campbell et
al. (2007) also reviewed the morphological and
ecomorphological aspects of this species using
multivariate statistics and found that the wing
Results from general descriptive statistics,
mtDNA, microsatellites, and morphometric studies
showed congruency of the existence of 2 divergent
forms of C. brachyotis. As Abdullah and Jayaraj’s
(2006) study showed that the larger form was
indeed the assigned C. brachyotis, it is apparent
that the small form may be a new species of
Cynopterus yet to be described. However, a recent
taxonomy of the Cynopterus by Simmons (2005)
did not include this new form, and Francis (2008)
assigned C. cf. brachyotis Sunda to the large form
of C. brachyotis commonly found in open areas
and C. cf. brachyotis Forest to the small form of C.
brachyotis commonly found in primary forests. For
 Zoological Studies 51(2): 259-271 (2012)
the easy interpretation of this paper, we assign C.
brachyotis as the known large form as verified by
Abdullah and Jayaraj (2006), and C. cf. brachyotis
Forest as the new undescribed form found in
primary forests.
In terms of forearm length differences,
Francis (1990) showed that C. brachyotis has a
mean forearm length of 62.1 mm (n = 22) and C.
cf. brachyotis Forest has a mean forearm length
of 58.4 mm (n = 21). Abdullah (2003) reported
that the forearm length of C. cf. brachyotis Forest
was 59.43 (± standard deviation (SD) 2.70) mm,
and C. brachyotis had a mean forearm length of
63.87 (± 5.02) mm. Campbell et al. (2006) used
a forearm length of 63.8 (± 1.6) mm to identify C.
brachyotis and a forearm length of 59.5 (± 1.7) mm
to identify C. cf. brachyotis Forest.
The high reliance on forearm length to
distinguish these 2 forms is problematic as
various authors have reported different forearm
length measurements used for differentiation.
The development of additional characters to
differentiate these 2 forms would be useful in the
field, as this will aid field ecologists in accurately
identifying both C. brachyotis and C. cf. brachyotis
Forest in southern Thailand, Peninsular Malaysia,
and Borneo. Thus, in this study, we attempted to
further describe detailed morphometric variations
that exist in the genus Cynopterus from Peninsular
Malaysia and Borneo using multivariate statistics.
The approach was to develop a classification
function that can be used to differentiate C.
brachyotis and C. cf. brachyotis Forest in the field
and verify museum specimens.
MATERIALS AND METHODS
In total, 74 specimens (10 individuals of
C. horsfieldii, 34 individuals of C. brachyotis,
29 individuals of C. cf. brachyotis Forest, and 1
individual of Eonycteris major) were used in this
study. These specimens were either collections
from field sampling done in various localities within
Borneo and Peninsular Malaysia or museum
samples from the zoological museum at Universiti
Malaysia Sarawak (Sarawak, Malaysia) and
the Department of Wildlife and National Parks
(PERHILITAN) Museum (Pahang, Malaysia). Due
to a limited number of samples of Cynopterus, we
opted to focus on the problem of differentiating
C. brachyotis and C. cf. brachyotis Forest using
multivariate statistics. Only specimens of C.
brachyotis and C. cf. brachyotis Forest previously
261
confirmed by Abdullah (2003) and Fong (2011)
using DNA sequences of the partial Cytochrome
b (700 bps) and Cytochrome Oxidase 1 (486 bps)
were used in this study.
Twenty-eight morphological measurements
(skull, dental, and external morphological
measurements; Fig. 2) were recorded following
Kitchner et al. (1995) and Jayaraj et al. (2004
2005). Abbreviations for the characters measured
are as follow: BL, bulla length; C1BW, canine tooth
basal width; C1C1B, breadth across both canine
outside surfaces; C1M3L, canine molar length or
maxillary tooth row length; CW, cranial width; DBC,
distance between cochleae; DL, dentary length;
D3MCL, 3rd digit metacarpal length; D4MCL,
4th digit metacarpal length; D5MCL, 5th digit
metacarpal length; D3P1L, 3rd digit 1st phalanx
length; D3P2L, 3rd digit 2nd phalanx length; EL,
ear length; GBPL, greatest basial pit length; GSL,
great skull length; IOW, interorbital width; M3L,
3rd molar tooth crown length; M3W, 3rd molar
tooth crown width; M3M3B, breadth across outside
surfaces of both 3rd molar teeth; MW, mastoid
width; PES, pes length; PL, palatal length; POW,
postorbital width; PPL, postpalatal length; RL,
radius length; TL, tibia length; TVL, tail to ventral
length; and ZW, zygomatic width. Bat skulls were
extracted after morphological data were collected
following Nargorsen and Peterson (1980).
A cluster analysis using Euclidean distances
with the unweighted pair-groups method
average (UPGMA) was performed to construct
a morphometrics-based phylogeny and confirm
the initial grouping of samples (Everitt 1993).
The E. major measurements were used as the
outgroup for this analysis. Data of confirmed
groupings were then subjected to a t-test to check
for sexual dimorphism. Levene’s test for equality
of variances was used as a selection criterion for
the assumption of equal or unequal variances
prior to the t-test (Zar 1984). The normality of
the data was checked using a normal probability
plot and the Shapiro-Wilk test. The assumption
of homoscedasticity was tested using Box’s M
test, and the assumption of multicolinearity was
checked by observing the tolerance value for all
independent variables (Joseph et al. 1992). Next,
the data were subjected to a stepwise discriminant
function analysis following Joseph et al. (1992)
and Manly (1994). Two separate analyses
were performed: 1) using a combination of all
available characters and 2) using only external
morphological characters. Data were analyzed
using Minitab 2002 v13.2 (2006 Minitab, Pine Hall
262 Jayaraj et al. – A Model to Differentiate Cynopterus brachyotis Forms

Rd State College, PA, USA) and SPSS vers. 13
(SPSS, Chicago, IL, USA).
RESULTS
The UPGMA cluster analysis (Fig. 3) shows
th groupings of Cynopterus spp. based on
morphological measurements. Based on the
phylogram, there are 3 major clades consisting
of C. horsfieldii, C. cf. brachyotis Forest, and C.
brachyotis. Of the 28 characters examined, 3
characters (IOW for C. horsfieldii and D3MCL
and D5MCL for C. brachyotis) were found to be
sexually dimorphic (Table 1). The means and
standard deviations (SDs) of all characters are
shown in table 2. PES was log 10-transformed
to achieve normality, whereas PPL, PL, and TL
were excluded from the analysis as these data did
not follow a normal distribution either prior to or
after transformation to achieve normality. Box’s
M statistics had a value of 23.406 (probability of
p = 0.483, p > 0.001) indicating homoscedasticity.
Thus, the data were analyzed using a pooled
covariance matrix for classification. Multicolinearity
among the independent variables was not present,
as tolerance values for all variables were > 0.10.
For analysis of all remaining characters, the
stepwise method identified 1 discriminant function
(Function 1) that was statistically significant based

C1C1B

C1BW
IOW
M3W
PL C1M3L
POW
M3L
M3M3B
CW GSL
GBPL

PPL BL
MW

ZW
DL
DBC

D3P1L
D3MCL
D3P2L

RL EL
D4MCL

D5MCL

TL

TVL

PES

Fig. 2. Skull, dental, and external measurements taken during this study. The abbreviations of body measurements please refer to
“MATERIALS AND METHODS” section.
 Zoological Studies 51(2): 259-271 (2012) 263

Distance
46.44 30.96 15.48 0.00

1 Lalang Dam, Bario

4 Salt Lake, Bario
7 Samunsam W.S.
2 Lalang Dam, Bario
3 Lalang Dam, Bario
5 Samunsam W.S.
10 Lalang Dam, Bario
6 P. Balak, Bangi
8 Sungai Dusun, Selangor
9 Sungai Dusun, Selangor
12 Batang Ai N.P.
11 Mount Penrisen
14 Batang Ai N.P.
20 Mount Penrisen C. cf. brachyotis
13 Mount Penrisen
15 Batang Ai N.P. Forest
28 Kubah N.P.
18 Mount Penrisen
17 Mount Penrisen
19 Mount Penrisen
21 Mount Penrisen
23 Mount Penrisen
22 Kubah N.P.
25 Mount Penrisen
24 Mount Penrisen
27 Kubah N.P.
16 Batang Ai N.P.
26 Kubah N.P.
29 Kubah N.P.
33 Kubah N.P.
35 Samunsam W.S.
34 Kubah N.P.
39 G. Pueh, Sematan
32 G. Silam, Lahat Datu
52 Samunsam W.S.
63 G. Silam, Lahat Datu
56 Samunsam W.S.
30 Mount Penrisen
55 Salt Lake, Bario
41 P. Talang Kecil
58 Salt Lake, Bario
62 G. Silam, Lahat Datu
47 P. Balak, Bangi
57 Salt Lake, Bario
45 P. Talang Kecil
46 P. Balak, Bangi
48 P. Balak, Bangi C. brachyotis
54 Samunsam W.S.
31 Lalang Dam, Bario
36 Samunsam W.S.
38 Samunsam W.S.
37 Samunsam W.S.
51 Samunsam W.S.
49 Samunsam W.S.
42 Gading N.P.
43 P. Talang Kecil
61 G. Silam, Lahat Datu
53 Samunsam W.S.
60 G. Silam, Lahat Datu
44 Lalang Dam, Bario
50 Gading N.P.
40 G. Pueh, Sematan
59 Salt Lake, Bario
65 P. Balak, Bangi
66 P. Balak, Bangi
67 Kubah N.P.
68 G. Pueh, Sematan
64 Kubah N.P. C. horsfieldii
69 G. Pueh, Sematan
70 G. Pueh, Sematan
72 Sungai Dusun, Selangor
73 Mount Penrisen
71 Sungai Dusun, Selangor
74 Mount Penrisen E. major

Fig. 3. UPGMA cluster analysis of Cynopterus spp.

264 Jayaraj et al. – A Model to Differentiate Cynopterus brachyotis Forms

on Wilks’ lambda (Table 3), and 6 characters
(GBPL, M3L, M3W, TVL, D3P2L, and RL; Table
4) were generated from the stepwise procedure.
The characters with the highest weight on function
1 were RL (0.957) and M3L (0.506), whereas
M3L (3.912) and M3W (-3.454) had the highest
discriminant loadings. All 6 characters determined
by the stepwise procedure produced a discriminant
function with an accuracy rate of 100% (see
accuracy rates, Table 5).
For analysis of only external morphological
characters, the stepwise method identified a
discriminant function (Function 1) that was stati-
stically significant (Table 6) with 4 characters (TVL,
D3P1L, D3MCL, and RL; Table 7) generated from
the stepwise procedure. The character with the
highest weight and loading was RL (weight = 1.240;
loading = 0.706), while D3MCL (-0.731) had the
2nd-highest weight. All 4 characters determined
by the stepwise procedure produced a discriminant
function with an accuracy rate of 96.8% (see
accuracy rates, Table 8).
A histogram of the discriminant scores
of the discriminant function for all characters

Table 1. Sexual dimorphism test using a t-test for equality of means (equal/unequal variances; only sexually
dimorphic characters are shown)

C. horsfieldii C. brachyotis

Character IOW D3MCL D5MCL

t 3.434 1.346 1.113
d.f. 8 32 32
Significance (2-tailed *) 0.009 0.021 0.090
Conclusion sexually dimorphic sexually dimorphic sexually dimorphic

Characters are defined in “MATERIALS AND METHODS”.

Table 2. Means and standard deviations (SDs) of all characters used in this analysis

Cynopterus cf. brachyotis Forest C. brachyotis Overall

Character Mean S.D. Mean S.D. Mean S.D.

GSL 27.39 0.76 28.45 0.88 27.92 0.97

IOW 5.60 0.32 5.92 0.36 5.76 0.37
POW 6.32 0.60 6.48 0.65 6.40 0.63
CW 12.06 0.39 12.39 0.36 12.23 0.41
MW 12.23 0.40 12.66 0.43 12.45 0.47
ZW 17.93 0.79 18.40 0.80 18.16 0.82
DBC 5.62 1.07 4.72 0.79 5.17 1.04
BL 2.60 0.58 2.18 0.42 2.39 0.55
GBPL 6.95 0.93 5.72 0.90 6.34 1.10
C1BW 1.61 0.24 1.44 0.20 1.52 0.24
C1C1B 5.92 0.29 6.05 0.30 5.99 0.30
M3M3B 8.37 0.36 8.44 0.36 8.40 0.36
C1M3L 8.94 0.39 9.10 0.28 9.02 0.34
M3L 1.83 0.11 1.95 0.14 1.89 0.14
M3W 1.25 0.15 1.18 0.13 1.21 0.14
TVL 11.40 1.95 11.47 2.63 11.43 2.29
EL 14.48 1.34 14.67 1.29 14.58 1.31
D3P1L 26.63 1.37 28.38 1.25 27.51 1.57
D3P2L 33.75 2.32 36.19 2.42 34.97 2.65
D3MCL 41.47 1.76 43.06 1.71 42.27 1.90
D4MCL 38.87 1.40 40.94 1.63 39.91 1.83
D5MCL 39.64 1.43 42.31 1.44 40.98 1.96
RL 58.08 1.40 63.55 2.04 60.82 3.26
LogPES 1.02 0.04 1.04 0.07 1.03 0.06

Characters are defined in “MATERIALS AND METHODS”.

 Zoological Studies 51(2): 259-271 (2012) 265

(Fig. 4) showed that C. brachyotis and C. morphological characters (Fig. 5) showed some
cf. brachyotis Forest formed distinct groups, misclassifications (2 individuals). The discriminant
whereas the histogram of the discriminant scores functions based on the unstandardized canonical
of the discriminant function for only external coefficient functions (Tables 3, 7) can be used

Table 3. Wilks’ lambda test of discriminant Table 6. Wilks’ lambda test of discriminant
function 1 (with all available characters) function 1 (with external morphological characters)

Percent of Percent of
Wilks’ lambda Chi-squared Eigenvalue Wilks’ lambda Chi-squared Eigenvalue
variance variance

0.157 94.412 5.368 100% 0.195 96.337 4.188 100%

Cumulative Canonical Cumulative Canonical

d.f. Significance ** d.f. Significance **
percent correlation percent correlation

100% 9.18 6 0.00 100% 0.897 4 0.00

Table 4. Standardized and unstandardized Table 7. Standardized and unstandardized

canonical discriminant function coefficients (with all canonical discriminant function coefficients (with
characters) external morphological characters)

Character Function 1 Character Function 1

Standardized Unstandardized Standardized Unstandardized

GBPL -0.371 -0.405 TVL -0.479 -0.216

M3L 0.506 3.912
D3P1L 0.572 0.442
M3W -0.481 -3.454
TVL -0.346 -0.149 D3MCL -0.731 -0.433
D3P2L 0.350 0.148
RL 1.240 0.706
RL 0.957 0.546
Constant - -37.326 Constant - -34.507

Characters are defined in “MATERIALS AND METHODS”. Characters are defined in “MATERIALS AND METHODS”.

Table 5. Classification results (pooled covariance Table 8. Classification results (pooled covariance
matrix) of the stepwise discriminant function matrix) of the stepwise discriminant function
analysis (with all available characters) analysis (with external morphological characters)

Predicted group Predicted group

membership membership
Group Total Group Total
1 2 1 2

Original Count 1 29 0 29 Original Count 1 32 2 34

2 0 34 34 2 0 29 29
Percent 1 100% 0% 100% Percent 1 94.1% 5.9% 100%
2 0% 100% 100% 2 0% 100% 100%
Cross-validateda Count 1 29 0 29 Cross-validateda Count 1 32 2 34
2 0 34 34 2 0 29 29
Percent 1 100% 0% 100% Percent 1 94.1% 5.9% 100%
2 0% 100% 100% 2 0% 100% 100%

Both 100% of the original and cross validated. agrouped cases Both 96.8% of the original and cross-validated. agrouped cases
were correctly classified. were correctly classified.
266 Jayaraj et al. – A Model to Differentiate Cynopterus brachyotis Forms

as a tool to determine whether a specimen is
C. brachyotis or C. cf. brachyotis Forest. The
predictive models are as follows:
for all remaining characters
ŷ = -0.405a + 3.912b - 3.454c - 0.149d +
0.148e + 0.546f - 37.326 (constant) (a)
and for only external morphological characters
ŷ = -0.216d + 0.442 - 0.433h - 0.706f - 34.507
(constant); (b)
where ŷ is the discriminant score (a negative score
indicates C. cf. brachyotis Forest and a positive
score indicates C. brachyotis), a is the GBPL, b
is the M3L, c is the M3W, d is the TVL, e is the
D3P2L, f is the RL, g is the D3P1L, and h is the
D3MCL.
DISCUSSION
General discussion of statistical results
Based on the cluster analysis, a clear division
(approximately 15.48% distance, based on
estimates from the graph) was observed between
C. brachyotis and C. cf. brachyotis Forest. Visual
observations of samples during field sampling
indicated that adult C. brachyotis can be vaguely
identified due to the brown fur with a pronounced
yellowish or reddish tinge, and these bats usually
have a forearm of > 60 mm. Adults of C. cf.
brachyotis Forest have a smaller body size with
duller coloration and usually have a forearm length
of < 60 mm.
Comparison with previous bat surveys
(Timoh 2006, Fukuda et al. 2008) and personal
observations indicate that C. brachyotis was
sampled across a wide variety of vegetation
types with different capture rates, whereas C.
cf. brachyotis Forest was confined to primary
forests. Capture rates of C. brachyotis were 44%
in secondary forests, 41% in orchards, and 72%
in oil palm plantations (Fukuda et al. 2008). The
high capture rate in oil palm plantations is probably
associated with the abundance of oil palm fruit,
i.e., a food source (Fukuda et al. 2008). We also
speculated that this abundant food source would
also likely increase the life expectancy of C.
brachyotis in oil palm plantations as many older
individuals were captured (with distinct reddish-
brown fur on their shoulders and worn out or
missing teeth in most individuals) in Timoh’s (2006)
study.
In this study, the analyses revealed that the
RL, M3L, and M3W had the highest discriminant
loading and weight, and this was reflected by
the importance of these characters during the
identification process. The RL or forearm length
is one of the characters useful in identifying bats,
especially fruit bats of the family Pteropodidae.
This character was also previously used to diffe-
rentiate C. brachyotis, C. cf. brachyotis Forest,
and other Cynopterus in Malaysia (Abdullah et

C. cf. brachyotis Forest C. cf. brachyotis Forest

Canonical Discriminant Function C. brachyotis Canonical Discriminant Function C. brachyotis
Mean = 2.16
8 Mean = -2.22 Mean = 2.34
Std. Dev. = 0.801
Std. Dev. = 1.004 Std. Dev. = 1.048
N = 29
N = 29 N = 34 10 Mean = 1.84
6 Std. Dev. = 1.142
8 N = 34
Frequency

Frequency

4 6

4
2
2

0 0
-5.0 -2.5 0.0 2.5 5.0 -4 -2 0 2 4
Discriminant Scores Discriminant Scores

Fig. 4. Histogram of discriminant scores of both C. brachyotis Fig. 5. Histogram of discriminant scores of both C. brachyotis
and C. cf. brachyotis Forest for all available characters. and C. cf. brachyotis Forest for external morphological
characters.
 Zoological Studies 51(2): 259-271 (2012)
al. 2000, Abdullah 2003, Campbell et al. 2004
2006 2007, Jayaraj et al. 2004 2005, Fong 2011).
Although M3L and M3W are not generally used
for species identification, molar differences in C.
horsfieldii, C sphinx, and C. brachyotis are a key
character which can be used to differentiate these
3 species of Cynopterus in Malaysia. The D3MCL
and D3P1L both contribute to the length and size
of the wings, and this may reflect the habitats that
both species occupy.
Th e cl uste r a n d d iscri mi n a nt fu ncti o n
analyses showed that C. brachyotis and C. cf.
brachyotis Forest populations are morphologically
distinct, congruent with previous results using
molecular methods (Abdullah 2003, Campbell et
al. 2004 2006, Julaihi 2005, Fong 2011). Although
the topology of the dendogram generated by
the cluster analysis was not similar to previous
molecular studies, it was able to differentiate
C. brachyotis from C. cf. brachyotis Forest.
The different topologies might be a reflection
of the morphological appearances of these
bats. Morphologically both C. brachyotis and
C. cf. brachyotis Forest look similar, whereas C.
horsfieldii is very much larger with distinct cusps
on the lower premolar and 1st lower molar; these
characteristics are not present in C. brachyotis or C.
cf. brachyotis Forest.
The prediction models developed will
be particularly useful in accurately identifying
C. brachyotis and C. cf. brachyotis Forest in
Malaysia. Specifically function (a) can be used
to verify museum specimens, and function (b)
will be more appropriate for field identification.
Function (a) requires cranial, dental, and external
morphological measurements; thus, unverified
museum specimens can be identified once the
skull is extracted, reducing the cost of validating
the species using molecular tools. Although having
a lower accuracy rate, function (b) can be used in
the field as only external morphological characters
are needed to identify the species. If needed,
however, a tissue sample via skin scraping or a
wing punch can be taken for species verification
in the lab. An accurate identification method will
definitely aid ecologists, conservationists, and law
enforcement officials in studying and conserving
this species complex.
Body sizes and relation to habitat types of
Cynopterus brachyotis and C. cf. brachyotis
Forest
Body size can be related to the flight perfor-
267
mance of bats as the total body mass is negatively
correlated with wing loading, a measure of the
ability to navigate around obstacles (Aldridge
1986, Aldridge and Rautenbach 1987, Jones et
al. 1993, Rhodes 2002) and maneuverability in
cluttered areas (Aldridge and Rautenbach 1987,
Jones et al. 1993, Kalcounis and Brigham 1995,
Brigham et al. 1997, Rhodes 2002). This can be
directly linked to the habitats of both species, with
C. brachyotis occupying less-cluttered habitats and
C. cf. brachyotis Forest occupying dense areas
(Abdullah et al. 2000, Abdullah 2003, Jayaraj
et al. 2004 2005). Body size seems to be the
discriminating factor in the cluster analysis for
effectively discriminating these 2 species, which
explains why both species can be separated, but
body size per se does not depict the entire picture
of the divergence of these bats. In terms of the
flight apparatus and dimensions, both species
apparently did not undergo the change in wing
shape indicated in a recent study by Campbell et
al. (2007), but rather a change in body size which
might have been due to selective pressures for C.
brachyotis and C. cf. brachyotis Forest to fit into
their respective habitats. Similarly, Menon (2007)
revealed that the aspect-ratio and wing-loading
indices cannot be used to differentiate these 2
species in Borneo.
Previous studies (Freeman 1981, Schluter
1993, Wain-Wright 1996) noted that there was a
relationship between the structure of the feeding
apparatus and diet in bats. As M3W and M3L
are associated with feeding and foraging, it was
speculated that the shape and dimension of the
dentition are associated with the diet. Current
knowledge of the diet and foraging behavior of C.
brachyotis in Malaysia was previously documented
by Lim (1970), Phua and Corlett (1989), Fujita
and Tuttle (1991), Francis (1994), Funakoshi and
Zubaid (1997), Tan et al. (1998), Mohd Azlan et al.
(2000), and Hodgkison et al. (2003), but none of
those authors focused on differences in the diets
and foraging behaviors of C. brachyotis and C.
cf. brachyotis Forest. Thus a more-detailed study
of the diets and foraging behaviors would shed
more light on ecological differences between C.
brachyotis and C. cf. brachyotis Forest.
Implications of recent studies for the taxono-
mic status of the Cynopterus brachyotis
complex
It was proven by various studies that C.
brachyotis is a species complex with 6 distinct
268 Jayaraj et al. – A Model to Differentiate Cynopterus brachyotis Forms
lineages. Genetically C. brachyotis has 6 forms; 4
geographically distinct lineages respectively from
India, Myanmar, Sulawesi, and the Philippines,
and 2 sympatric forms (recognized as C. cf.
brachyotis Forest and C. brachyotis in this study)
in southern Thailand, Peninsular Malaysia, and
Borneo. These 2 sympatric forms are found in
distinct habitats: C. brachyotis is found in open
areas, and C. cf. brachyotis Forest is found in
the primary and old secondary forests (Abdullah
2003, Campbell et al. 2004 2006, Jayaraj et al.
2004 2005, Julaihi 2005, Fukuda et al. 2008, Fong
2011). Cynopterus brachyotis is the ancestral
lineage of all Cynopterus in Peninsular Malaysia
and Borneo with nucleotide divergence ranging
8%-9%, whereas C. cf. brachyotis Forest is closely
related to C. horsfieldii, differing by only a genetic
divergence of 3.5% (Abdullah 2003).
This scenario is not new to the taxonomy
of Cynopterus as C. nusatenggara described by
Kitchner and Maharadatunkamsi (1991 1996)
was also found within Cynopterus populations
during field sampling on islands of Nusa Tenggara,
Indonesia. Cynopterus bats are currently repre-
sented by 7 species (Simmons 2005), but there
are a lot of variations in terms of body size and
coloration between and within species. These
variations were observed in island populations and
highland populations, and are due to differences
in vegetation and other ecological factors (see
Hill and Thonglongya 1972, Lekagul and McNeely
1977, Medway 1978, Payne et al. 1985, Kitchner
and Maharadatunkamsi 1991 1996, Ingle and
Heaney 1992, Schmitt et al. 1995, Nor 1996,
Abdullah et al. 2000, Storz et al. 2001, Abdullah
2003, Campbell et al. 2004 2006 2007, Jayaraj et
al. 2004, Menon 2007, Fukuda et al. 2008, Fong
2011).
Menon (2007) collected an unidentified
Cynopterus specimen from Satang I., Borneo,
Malaysia, and this specimen was later identified
using DNA techniques. The forearm length of
this Cynopterus specimen was 69 mm indicating
it was C. sphinx, but DNA identification indicated
that it was C. brachyotis (unpubl. data). A molar
examination of the specimen did not show a clear
distinction between C. brachyotis and C. sphinx.
Such an observation is the norm when individuals
from this genus are collected in a wide range of
vegetative types, which indicates that there are
high intra- and interspecific variations among
Cynopterus representatives. The lack of such
knowledge indicates the necessity for a current
large-scale study on inter- and intraspecific forms
of Cynopterus across their distribution. Although
C. brachyotis is widely distributed, information on
the current status of the 6 lineages of C. brachyotis
especially is not clear. Confounded by the non-
recognition of these new C. brachyotis lineages
(Forest, India, Myanmar, Sulawesi, and the
Philippines) as distinct species (see Abdullah and
Jayaraj 2006), the survival of these rare species
may be threatened if no clear and proper planning
for conservation is put in place.
In terms of biogeography, the existing reco-
gnized Cynopterus species of C. brachyotis,
C. horsfieldii, C. luzoniensis, C. minutus, C.
nusatenggara, C. sphinx, and C. tithaecheilus are
distributed in the Indo-Malayan region and their
distributions overlap. Simmons (2005) listed their
distributions as follow: C. brachyotis is distributed
in Sri Lanka, India, Nepal, Burma, Thailand,
Cambodia, Vietnam, South China, Malaysia, the
Nicobar and Andaman Is., Borneo, Sumatra,
Sulawesi, Magnole, Sanana, Sangihe I., and
Talaud I. with possible occurrence in the Palawan
region of the Philippines; C. horsfieldii is limited to
Thailand, Cambodia, Peninsular Malaysia, Borneo,
Java, Sumatra, the Lesser Sunda Is., and adjacent
small islands; C. luzoniensis is found in Sulawesi,
the Philippines, and adjacent small islands; C.
minutus is found in Sumatra, Java, Borneo, and
Sulawesi; C. nusatenggara is found in Lombok,
Moyo, Sumbawa, Sangeang, Komodo, Flores,
Sumba, Adonara, Lembata, Pantar, Alor, and the
Wetar Is.; C. sphinx is found in Sri Lanka, Pakistan,
Bangladesh, India, South China, Southeast Asia
including Burma, Vietnam, Cambodia, Peninsular
Malaysia, Sumatra, and possibly in Borneo; and
C. tithaecheilus is found in Sumatra, Java, Bali,
Lombok, Timor, and adjacent small islands.
In Malaysia; however, only 5 species of
Cynopterus coexist together, i.e., C. horsfieldii,
C. sphinx, C. brachyotis, C. minutus, and C. cf.
brachyotis Forest. Cynopterus horsfieldii, C.
brachyotis, and C. cf. brachyotis Forest have a
high geographic distributional overlap (Abdullah
2003, Campbell et al. 2004 2006 2007), but
C. cf. brachyotis Forest’s distribution extends
farther north into Thailand, Vietnam, and probably
Cambodia and Laos (Campbell et al. 2004
2006). Ecologically, C. sphinx and C. brachyotis
are common in open habitats, orchards, and
agricultural areas, whereas C. horsfieldii and C.
cf. brachyotis Forest are found in primary and old
secondary forests in Peninsular Malaysia and
southern Thailand. Cynopterus cf. brachyotis
Forest is also rare in Peninsular Malaysia, as its
 Zoological Studies 51(2): 259-271 (2012)
occurrence is dictated by the existence of primary
and old secondary forests. Cynopterus sphinx
is found in both habitat types, but declines in
number near forest edges (Campbell et al. 2006).
Similar observations were found in Borneo, but
with the exclusion of C. sphinx as records of this
species occurring in Borneo are only from Central
Kalimantan (Payne et al. 1985, Abdullah et al.
1997), and to date, there are no recent records of
this species in Malaysian Borneo. The occurrence
of C. minutus in Borneo is still in question, as
there is little information on it, but a recent survey
by Benda (2010) did record C. minutus in Sabah.
The forearm length of C. minutus captured in his
study was 54.3-58.1 (mean, 55.69, SD, 1.644) mm
(n = 5), which is slightly smaller but overlaps with
forearm length measurements of C. cf. brachyotis
Forest in Abdullah (2003), Campbell et al. (2004
2006 2007), Jayaraj et al. (2004 2005), Jayaraj
(2009), and Fong (2011).
As Abdullah and Jayaraj’s (2006) preliminary
investigation of the nominate specimen of C.
brachyotis revealed that the type specimen of C.
brachyotis described by Müller (1838) is the larger
form, it is apparent that the remaining C. brachyotis
lineages (Forest, India, Myanmar, Sulawesi, and
the Philippines) require further study to clarify their
phylogenetic positioning and taxonomic status. To
date, there are more than 10 studies (see Abdullah
et al. 2000, Abdullah 2003, Campbell et al. 2004
2006 2007, Jayaraj et al. 2004 2005, Julaihi 2005,
Abdullah and Jayaraj 2006, Jayaraj 2009, Fong
2011) that have validated the existence of C. cf.
brachyotis Forest, but there are no published
studies on the remaining C. brachyotis lineages
in the Indo-Malayan region. Thus, a complete
phylogenetic tree of all 7 recognized species and
recorded divergent forms of Cynopterus (including
the 6 divergent forms of C. brachyotis) should
be generated to clarify the taxonomic status of
all Cynopterus spp. in the Indo-Malayan region.
Clarification of C. luzoniensis from Sulawesi and
Palawan is also needed, as there is the possibility
that the Sulawesi and Philippine forms of C.
brachyotis previously described by Campbell et al.
(2004) could possibly be C. luzoniensis, or these 2
C. brachyotis forms may differ from C. luzoniensis
altogether. Finally, because C. minutus is
recognized as a distinct species (Simmons 2005),
there is a need to check the status of this species
in Borneo as little information is available.
Two models to differentiate C. brachyotis
and C. cf. brachyotis Forest were developed
using multivariate statistics with a high accuracy
269
rate of of identifying both C. brachyotis and C. cf.
brachyotis Forest. Based on the 1st prediction
model (function a), 6 chara-cters are needed
to accurately differentiate C. brachyotis from
C. cf. brachyotis Forest in southern Thailand,
Peninsular Malaysia, and Borneo. This model
would be more appropriate for use on museum
specimens as skull and dental characters are
needed for the calculation. The 2nd prediction
model (function b) can be used during field
sampling, as only external morphological
measurements are needed for identification.
These prediction models can subsequently be
used by bat biologists to correctly identify adult C.
brachyotis forms in southern Thailand, Peninsular
Malaysia, and Borneo, thus aiding in research and
conservation efforts of both C. brachyotis and C.
cf. brachyotis Forest in this region. Further sug-
gestions on taxonomic research of this species
complex should include verification of multiple
genetic markers, examination of detailed morpho-
metrics, and a review of the taxonomic status of
the 6 existing C. brachyotis forms in the Indo-
Malayan region. Conservation of this species
complex needs to be carefully planned in order to
ensure that all 6 divergent forms do not go extinct,
as these are suspected of being undescribed
species in the Indo-Malayan region.
Acknowledgments: We would like to thank Dr. L.
S. Hall for the inspiration to go further and realize
our potential, to F.A.A. Khan, A.K.H. Guan, B.
Ketol, F.P. Har and all staffs in Molecular Ecology
Laboratory Universiti Malaysia Sarawak (UNIMAS)
for all their support and comments to improve our
work and being great companions during field trips
and laboratory sessions. We would also like to
thank the Sarawak Forest Department for issuing
permit no. 04608 and the Sarawak Forestry
Corporation for their hospitality during our visit
to protected areas in Sarawak. We extend our
gratitude to the Faculty of Resource Science and
Technology, UNIMAS, Department of Wildlife and
National Parks (Kuala Lumpur), Sabah Parks,
Sabah Wildlife Department, and many other
individuals for various administrative and logistical
support throughout the course of the study. We
thank 2 critical anonymous reviewers who tre-
mendously helped us improve the taxonomic
perspectives of an initial draft of this paper. The
main author, JVK, would also like to thank M.
Muhamad for her comments on the statistical
analysis and the overall manuscript. This study
was funded by a Malaysian government IRPA
270 Jayaraj et al. – A Model to Differentiate Cynopterus brachyotis Forms
grant (09-02-09-1022-EA001) awarded to MTA and
colleagues, a UMK Short Term Grant (R/SGJP/
A03.00/00481A/001/2010/000037) awarded to JVK
and YCW and ARA, and a UNIMAS ZAMALAH
scholarship awarded to JVK.
REFERENCES
Abdullah MT. 2003. Biogeography and variation of Cynopterus
brachyotis in Southeast Asia. PhD dissertation, Univ. of
Queensland, St. Lucia, Australia.
Abdullah MT, VK Jayaraj. 2006. Preliminary investigation on
the relationship of the nominate C. brachyotis with the
small-sized and large-sized C. brachyotis using clustering
analysis. Sarawak Mus. J. 62: 223-236.
Abdullah MT, C Moritz, GC Grigg, LS Hall. 2000. Evidence
of cryptic species within Cynopterus brachyotis by using
mtDNA sequence. In Z Yaacob, S Moo-Tan, S Yorath,
eds. Proceedings of the International Conference on In-
Situ And Ex-Situ Biodiversity Conservation in the New
Millennium. Kota Kinabalu, Malaysia: Yayasan Sabah.
Abdullah MT, MA Rahman, LS Hall. 1997. New records for
bats in Sarawak, Malaysia. Malay. Nat. J. 50: 365-367.
Aldridge HDJN. 1986. Manoeuvrability and ecological
segregation in the little brown (Myotis lucifugus) and Yuma
(M. yumamensis) bats (Chiroptera: Vespertilionidae).
Can. J. Zool. 64: 370-382.
Aldridge HDJN, IL Rautenbach. 1987. Morphology, echo-
location and resource partitioning in insectivorous bats. J.
Anim. Ecol. 56: 763-778.
Andersen K. 1912. Catalogue of the Chiroptera in the
collections of the British Museum. Vol. I. Megachiroptera.
London: British Museum of Natural History.
Benda P. 2010. On a small collection of bats (Chiroptera) from
western Sabah (North Borneo, East Malaysia). Vespertilio
13-14: 45-76.
Bergmans W, FG Rozendaal. 1988. Notes on collections
of fruit bats from Sulawesi and some off-lying islands
(Mammalia, Megachiroptera). Zoologische Verhan-
delingen 248: 1-74.
Brigham RM, SD Grindal, MC Firman, JL Morissette. 1997.
The influence of structural clutter on activity patterns of
insectivorous bats. Can. J. Zool. 75: 131-136.
Bumrungsri S, PA Racey. 2005. Field discrimination between
lesser short-nosed fruit bat (Cynopterus brachyotis Müller,
1838) and greater short-nosed fruit bat (C. sphinx Vahl,
1797) (Chiroptera: Pterpodidae) in southern Thailand.
Nat. Hist. Bull. Siam Soc. 53: 111-121.
Campbell P, CJ Schneider, AM Adnan, A Zubaid, TH Kunz.
2004. Phylogeny and phylogeography of Old World
fruit bats in the Cynopterus brachyotis complex. Mol.
Phylogen. Evol. 33: 764-781.
Campbell P, CJ Schneider, AM Adnan, A Zubaid, TH Kunz.
2006. Comparative population structure of Cynopterus
fruit bats in Peninsular Malaysia and southern Thailand.
Mol. Ecol. 15: 29-47.
Campbell P, CJ Schneider, AM Adnan, A Zubaid, TH
Kunz. 2007. Morphological and ecological correlates
of coexistence in Malaysian fruit bats (Chiroptera:
Pteropodidae). J. Mammal. 88: 105-118.
Corbet GB, JE Hill. 1992. The mammals of the Indomalayan
region: a systematic review. Oxford, UK: Oxford Univ.
Press.
Cuvier F. 1824. Des dents des mammiferes. F. G. Strasbourg,
France: Levrault, 256 pp.
Everitt BS. 1993. Cluster analysis, 3rd ed. New York: Halsted
Press.
Fong PH. 2011. Genetic variation and population structure
of the genus Cynopterus in Malaysia. Master thesis,
Universiti Malaysia Sarawak, Kota Samarahan, Malaysia.
Francis CM. 1990. Trophic structure of bat communities in the
understorey of lowland dipterocarp rain forest in Malaysia.
J. Trop. Ecol. 6: 421-431.
Francis CM. 1994. Vertical stratification of fruit bats
(Pteropodidae) in the lowland dipterocarp rainforest in
Malaysia. J. Trop. Ecol. 10: 523-530.
Francis CM. 2008. A field guide to the mammals of Southeast
Asia. London: New Holland Publishers, 392 pp.
Freeman PW. 1981. Correspondence of food habits and
morphology in insectivorous bats. J. Mammal. 62: 164-
166.
Fujita MS, MD Tuttle. 1991. Flying foxes (Chiroptera:
Pteropodidae): threatened animals of key ecological and
economic importance. Conserv. Biol. 5: 455.
Fukuda D, OB Tisen, K Momose, S Sakai. 2008. Bat diversity
in the vegetation mosaic around a lowland dipterocarp
forest of Borneo. Raffles Bull. Zool. 57: 213-221.
Funakoshi K, A Zubaid. 1997. Behavioral and reproductive
ecology of the dog-faced fruit bats, Cynopterus brachyotis
and C. horsfieldi, in a Malaysian rainforest. Mammal
Study 22: 95-108.
Gray JE. 1843. List of the specimens of Mammalia in the
collection of the British Museum. London: British Museum
(Natural History) Publications, 216 pp.
Hill JE. 1983. Bats (Mammalia: Chiroptera) from Indo-
Australia. Bull. Br. Mus. Nat. Hist. (Zool.) 45: 103-208.
Hill JE, K Thonglongya. 1972. Bats from Thailand and
Cambodia. Bull. Br. Mus. Nat. Hist. (Zool.) 22: 171-196.
Hodgkison R, ST Balding, A Zubaid, TH Kunz. 2003. Fruit
bats (Chiroptera: Pteropodidae) as seed dispersers and
pollinators in a lowland Malaysian rain forest. Biotropica
35: 491-502.
Ingle NR, LR Heaney. 1992. A key to the bats of the Philippine
islands. Fieldiana Zool. 69: 1-44.
Jayaraj VK. 2009. The phylogenetic relationships of
Megachiroptera in Malaysia inferred from morphological
and DNA analyses. Master’s thesis. Universiti Malaysia
Sarawak, Kota Samarahan.
Jayaraj VK, CJ Laman, MT Abdullah. 2004. Morphological
variation in the genus Cynopterus of Peninsular Malaysia
and Borneo. In AIM Ismail, HL Koh, YA Hasan, eds.
Proceedings of the Regional Conference on Ecological
and Environmental Modelling (ECOMOD 2004), Penang,
Malaysia, pp. 69-81.
Jayaraj VK, CJ Laman, MT Abdullah. 2005. Application of
multivariate techniques in determining morphological
variation in the genus Cynopterus of Peninsular Malaysia
and Borneo. In AA Tuen, I Das, eds. International
Conference on Biogeography and Biodiversity: Wallace in
Sarawak – 150 Years Later: Sarawak. Malaysia, 226 pp.
Jones G, M Morton, PM Hughes, RM Budden. 1993. Echo-
location, flight morphology and foraging strategies of
some West African hipposiderid bats. J. Zool. (Lond.)
230: 285-400.
Joseph FH Jr, RE Anderson, RL Tatham, WC Black. 1992.
Multivariate data analysis with readings, 3rd ed. New
 Zoological Studies 51(2): 259-271 (2012)
York: Macmillan.
Julaihi AM. 2005. A phylogenetic study of the subfamily
Pteropodinae inferred by using partial mitochondrial DNA
cytochrome b gene. BSc thesis, Univ. Malaysia Sarawak
(UNIMAS), Kota Samarahan, Sarawak.
Kalcounis MC, RM Brigham. 1995. Intraspecific variation in
wing loading affects habitat use by little brown bats (Myotis
lucifugus). Can. J. Zool. 73: 89-95.
Kitchener DJ, Maharadatunkamsi. 1991. Description of a new
species of Cynopterus (Chiroptera: Pteropdidae) from
Nusa Tenggara Indonesia. Rec. West. Aust. Mus. 51:
307-363.
Kitchener DJ, Maharadatunkamsi. 1996. Geographic variation
in morphology of Cynopterus nusatenggara (Chiroptera,
Pteropodidae) in southeastern Indonesia, and description
of two new subspecies. Mammalia 60: 255-276.
Kitchener DJ, WC Packer, Maharadatunkamsi. 1995. Morpho-
logical variation in Pteropus lombocensis (Chiroptera:
Pteropodidae) in Nusa Tenggara, Indonesia. Rec. Aust.
Mus. 17: 61-67.
Koopman KF. 1994. Handbook of zoology vol. 8. Chiroptera:
systematics. Berlin: Walter de Gruyter.
Lekagul B, JA McNeely. 1977. Mammals of Thailand.
Sahankarnbhat, Bangkok, 177 p.
Lim BL. 1970. Food habits and breeding cycle of the
Malaysian fruit-eating bat, Cynopterus brachyotis. J.
Mammal. 51: 174-177.
Manly BFJ. 1994. Multivariate statistical methods. A primer,
2nd ed. London: Chapman and Hall.
Medway L. 1978. Mammals of Borneo: field keys and an
annotated checklist. Kuala Lumpur: Monographs of the
Malayan Branch of the Royal Asiatic Society no. 7.
Menon V. 2007. A study on the wing-loading and aspect-ratio
for fruit bats of family Pteropodidae. BSc thesis, Universiti
Malaysia Sarawak (UNIMAS), Kota Samarahan, Sarawak.
Mickleburgh SP, AM Hutson, PA Racey. 1992. Old world fruit
bats: an action plan for their conservation. Sultanate of
Oman and Gland, Switzerland: IUCN/FFPS/ZSL/WWF/
JWPT/NWF.
Miller GS Jr. 1906. Seven new Malayan bats. Proc. Biol. Soc.
Washington 19: 63.
Mohd Azlan J, RSK Sharma, M Zakaria. 2000. Species
diversity and relative abundance of understorey bats at Air
Hitam Forest Reserve, Selangor, Malaysia. Malay. Nat. J.
54: 69-75.
Müller S. 1838. Over eenige nieuwe zoogdieren van Borneo.
Tijdschrift voor Natuurlijke Geschiedenis en Physiologie, 5:
134-150.
Nargorsen DW, RL Peterson. 1980. Mammal collector’s
271
manual. Life Science Miscellaneous Publications.
Toronto: Royal Ontario Museum.
Nor SM. 1996. The mammalian fauna on the islands at the
northern tip of Sabah, Borneo. Fieldiana Zool. 83: 1-51.
Payne J, CM Francis, K Phillipps. 1985. A field guide to the
mammals of Borneo. Kota Kinabalu, Malaysia: The
Sabah Society and World Wildlife Fund Malaysia.
Peters HW. 1861. Monatsberichte der Koniglich Preussischen
Akademie der Wissenschaften zu Berlin 1861: 708.
Peterson AT, LR Heaney. 1993. Genetic differentiation
in Philippines bats of the genera Cynopterus and
Haplonycteris. Biol. J. Linn. Soc. 49: 203-218.
Phua PB, RT Corlett. 1989. Seed dispersal by the lesser short-
nosed fruit bat (Cynopterus brachyotis, Pteropodidae,
Megachiroptera). Malay. Nat. J. 42: 251-256.
Rhodes MP. 2002. Assessment of source of variance and
patterns of overlap in microchiropteran wing morphology
in southeast Queensland, Australia. Can. J. Zool. 80:
450-460.
Schluter D. 1993. Adaptive radiation in sticklebacks: size,
shape, and habitat use efficiency. Ecology 74: 699-709.
Schmitt LH, DJ Kitchener, RA How. 1995. A genetic per-
spective of mammalian variation and evolution in the
Indonesian archipelago: biogeographic correlates in the
fruit bat genus Cynopterus. Evolution 49: 399-414.
Simmons NB. 2005. Order Chiroptera. In DE Wilson, DM
Reeder, eds. Mammal species of the world: a taxonomic
and geographical reference, 3rd ed. Washington DC:
Smithsonian Institution Press.
Storz JF, HR Bhat, TH Kunz. 2001. Genetic consequences
of polygyny and social structure in an Indian fruit bat,
Cynopterus sphinx. Inbreeding, outbreeding, and
population subdivision. Evolution 55: 1215-1223.
Tan KH, A Zubaid, TH Kunz. 1998. Food habits of Cynopterus
brachyotis (Muller) (Chiroptera: Pteropodidae) in
Peninsular Malaysia. J. Trop. Ecol. 14: 299-307.
Temminck CJ. 1825. Monographes de Mammalogie. G.
Dufour & E. D’Ocagne, Paris, 1: 198.
Timoh JR. 2006. Diversity and abundance of bats in an oil
palm plantation and neighboring forest in Bratak oil
palm estate, Bau, Sarawak. BSc thesis, Univ. Malaysia
Sarawak (UNIMAS), Kota Samarahan, Sarawak.
Vahl M. 1979. Beskrivelse på tre nye arter flagermuse. Skrifter
Naturhistorie-Selskabet Copenhagen, 4: 121-138.
Wain-Wright PC. 1996. Ecological explanation through
functional morphology: the feeding biology of sunfishes.
Ecology 77: 1339-1343.
Zar JH. 1984. Biostatistical analysis. 2nd ed. Englewood
Cliffs, NJ: Prentice-Hall.
272

ZOOLOGICAL STUDIES
INSTRUCTIONS TO CONTRIBUTORS

Zoological Studies publishes original research papers in six major fields, including Behavioral biology,
Ecology, Evolution, Systematics and Biogeography, Comparative physiology and Animals/plants
interaction. Manuscripts are welcome from around the world, but must be written in English. Authors who
submit exceptionally long manuscripts may be asked to defray a portion of related printing cost. Color
,
photographs can be printed at the author s expense. The journal will provide 25 free reprints of accepted
,
articles per publication; additional reprints are available for order at the author s expense.
With the exception of invited review papers, submissions must include a cover letter containing the
basic information and stating that the manuscript is based on previously unpublished original research
and has not been submitted to another journal for publication. If the paper concerns the use of animals or
specimens in research, a statement to the effect that the author(s) has adhered to the legal requirements of
the country in which the work was carried out or to any institutional guidelines should be included. Authors
are encouraged to provide the names and e-mail addresses of four possible reviewers.
The Editorial Board has final authority concerning acceptance or rejection of any manuscript. As a
,
condition of publication, the authors copyright automatically belongs to Zoological Studies. If the author(s)
does not have clear title to the copyright of any part of the manuscript, it is the sole responsibility of the
author(s) to obtain written permission from the copyright holder and present it to the editor of Zoological
Studies.
The following format guidelines should be followed for all papers submitted.

I. Submission procedure
Manuscripts must be submitted via e-mail as electronic files to the Editorial Office: zoolstud@gate.
sinica.edu.tw. The text should be submitted as a Microsoft Word file and a PDF file. The MS Word file will
allow a timely review process by allowing reviewers to insert comments on the electronic copy. Figures
should be included at the end of the PDF file containing the text, but for publication of accepted manuscripts,
separate text and figure files are requested as described below. To reduce the PDF file size for more-efficient
transmission, embed fonts, use the“optimize”function in Adobe Acrobat (or other program), and use no
more than 300 dpi resolution for figures. To aid the Editor in file management, please begin all filenames with
the surname of the first author; it would also be useful to include the date: e.g., Randall_et_al_4Sep02.doc
(spell out month to avoid confusion). Important: Please place the date of submission in the top right corner
of the title page and change the date on subsequent revisions. All corresponding authors will receive a
confirmation of receipt of the manuscript via e-mail. Authors who have not received a confirmation within
one week should check with the editorial office. After an electronic submission has been examined and
determined to be appropriate for Zoological Studies, it will be sent electronically to reviewers, who will return
it the same way.

Peer review involves the following steps:

1. Author submits manuscript; recommend 4 potential overseas referees;
2. Editorial Office checks formatting;
3. Chief Editor assigns a Subject Editor;
4. Subject Editor assigns Reviewers;
5. Reviewers submit review reports;
6. Subject Editor makes recommendation;
7. Chief Editor makes final decision; and
8. Author is informed of decision.

Manuscript Types

Reports: Reports are full-length papers and should not exceed 8000 words (including tables and figure
legends).
Invited Reviews: The Chief Editor invites an author to write a review. Invited reviews should normally be
the same length as a report.
 273

II. Title Page

The title page should include the manuscript title; names of all authors; address(es) of where the
research was conducted and, if different, all current addresses of the authors including fax and e-mail if
available; a short running title of less than 40 characters; name, address, telephone, and fax numbers where
correspondence should be sent; and no more than five keywords preferably not in the title.

III. Abstract
The abstract should be a factual condensation of the entire paper, including a statement of purpose, a
clear description of observations and findings, and a concise presentation of the conclusions. It should not
exceed 300 words. Literature citations should be avoid.

IV. Text
Manuscripts should include the following sections: Abstract, Materials and Methods, Results,
Discussion, Acknowledgments, References, Figures, and Tables. Begin each section on a separate sheet.
The full text of the Abstract to the References should be double-spaced with a minimum of 1.5-inch margins.
Numbered lines should be marked through the text to make it easier to refer to corrections in the review
process. The font of the entire manuscript should be set to 12 point Times New Roman. Scientific binomials
should be italicized.

V. References
References should be cited in the text using the following formats: (Smith 1992), (Smith et al. 1992),
or (Smith 1978a b 1983 1992, Jones 1990). Bibliographic citations should be arranged alphabetically
according to the surname of the primary author, and formatted as in the following examples.

Aranishi F. 2005b. Rapid PCR-RFLP method for discrimination of imported mackerel and domestic
mackerel. Mar. Biotechnol. (in press)
Chen W. 1974. Butterflies of Taiwan in colour. Taipei: Chinese Culture Press. (in Chinese)
Elzinga A, N Alonzo. 1983. Analysis for methylated amino acids in proteins. In CHW Hirs, SN Timasheff,
eds. Methods in enzymology. Vol. 91, Part I. New York: Academic Press, pp. 8-13.
Fishbase. 2005. A global information system on fishes. Available at http://fishbase.sinica.edu.tw/home.htm
Fisher CR, JJ Childress. 1986. Translocation of fixed carbon from symbiotic bacteria to host tissues in the
gutless bivalve Solemya reidi. Mar. Biol. 93: 59-68.
Fujioka T, H Chiba. 1988. Notes on distributions of some Japanese butterflies. Spec. Bull. Lep. Soc. Jap. 6:
141-149. (in Japanese with English summary)
Mills SC, JD Reynolds. 2003. The bitterling-mussel interaction as a test case for co-evolution. J. Fish Biol.
63 (Supplement A): 84-104.
Munday PL, PJ Eyre, GP Jones. 2003. Ecological mechanisms for coexistence of colour polymorphism in a
coral-reef fish: an experimental evaluation. Oecologia 442: 519-526.
Lee CL. 1998. A study on the feasibility of the aquaculture of the southern bluefin tuna in Australia.
Department of Agriculture, Fisheries and Forestry (AFFA), Canberra, ACT 1998, 92 pp.
Summerfelt RC, GE Hall, eds. 1987. Age and growth in fish. Ames, IA: Iowa State University Press.

VI. Tables
Tables should not duplicate material found in the text or in accompanying illustrations. Tables must be
numbered consecutively in the order of mention in the text, and be described in brief but complete legends.
All tables must be typed double-spaced without vertical lines, one table per page. All symbols (a, b, c, etc.)
and abbreviations used must be briefly and clearly explained in the table footnotes. Asterisks should be
used to indicate levels of significance: a single asterisk (*) for p ≤ 0.05, double asterisks (**) for p ≤ 0.01, and
triple asterisks (***) for p ≤ 0.001).

VII. Figures
Figures should be in the following format.
1. Figures must be in finished form and ready for reproduction.
2. Number the figures using Arabic numerals according to the order of mention in the text.
3. Appropriate lettering and labeling should be used with letters and numbers which will be at least
1.5 mm high in the final reproduction.
274

4. The Font of the lettering should be Arial. All figures should be one or two column widths (either 8 or
17 cm) in size. The maximum printed page height is 23 cm. Include scale bars where appropriate. Color and
grayscale photograph should be saved in EPS format.
5. Color photographs should be at a resolution of 300 pixels/inch. Grayscale photographs should be
saved in 8 bits/channel. Photographs should be saved in CMYK which is suitable for printing. Do not save
the format in indexed color.
6. Line drawings should be prepared in TIFF format at a resolution of 1200 pixels/inch. Figures are
edited using EXCEL, so please provide the original files.
7. Authors should prepare any TIFF- or EPS-formatted figures at the intended final size which is
suitable for editing, and also prepare figures with no labels or words after the manuscript is accepted.
8. If all parts of a figure can be clearly seen in the printed version, then this is a good indication that the
figure will be acceptable.
9. The maximum size for all originals should not exceed the size of a printed page. High-quality original
artwork or glossy prints should be submitted for reproduction mounted on appropriate mounting cards.
10. Authors may indicate their size preferences of each figure (i.e., two-column width,“do not
reduce,”etc.). All lines must be dark and sharply drawn. Reproductions may be used for review copies of a
manuscript.

VIII. Figure Legends

Each figure should be accompanied by a title and explanatory figure legend. All associated descriptive
legends should be typed (double-spaced) on a separate sheet; sufficient detail should be given in each
legend to understand the figure independent of the text.

IX. Nucleotide and Protein Sequences

Newly reported nucleotide and protein sequences must be deposited in the DDBJ/EMBC/GenBank
databases. Accession numbers must be included in the final version of the manuscript.

X. Special Notes on Taxonomic Papers

Taxonomic papers submitted to Zoological Studies will be considered by the uniqueness of the taxa
under study (e.g., a poorly described taxonomic group). Authors describing a new species are encouraged
to incorporate a revision of that particular group or relationships to existing species. Simple taxonomic
descriptions are no longer considered for publication in Zoological Studies. Those papers submitted to
Zoological Studies should follow the following style conventions.
1. Upon the first mention of a species or infra-familial in both the abstract and text, the author of the
animal taxon must be cited referring to the International Code of Zoological Nomenclature. Do not abbreviate
,
the generic name of a taxon upon first mention or at the beginning of a sentence. Author s names of a taxon
must not be abbreviated except for Linnaeus (as L.) and Fabricius (as Fabr.). When multiple authorships are
involved, authors, names should be separated by“et”or“and”. When citing authors of a taxon, citation of
the year is optional. If used, however, the year must be enclosed within parentheses or square brackets, and
the citation must be considered a reference citation within the article and be listed in the references.
2. New taxa or synonymies that are erected should be clearly and appropriately marked as: comb.
nov., com. rev., nom. nov., sp. nov., stat. nov., stat. rev., syn. nov., etc. A new taxon must list the name of the
describing author(s) after the binomial or trinomial, even if it is the same as the manuscript author(s).
3. Types: Descriptions and revisions also require comments on the types involved. Comments on types
should be in a separate paragraph, and should include collection data and deposition information.
4. Keys: Keys are not essential in taxonomic work, but are highly recommended. Keys must be concise,
clear, easy to follow, and have reversibility provisions. Keys must also be in adjacent couplet style, and each
couplet should preferably contain more than a single, non-overlapping attribute.
5. Materials examined: Holotype and paratype(s) must be designated if a new taxon is being published.
Designation of an allotype is not necessary. The collecting site, number of specimens examined, sex, date,
and collector should be stated.
6. The result section of the systematic papers should be in the order of scientific name,
synonyms, Material examined (inc. holotype and paratype), Etymology, Diagnosis, Description (inc.
Measurements), then a Distribution. The Discussion section should be included at the end of main
text.
Indexed/Abstracted in:
Biological Abstracts
Chemical Abstracts
Current Awareness in Biological Sciences
Current Contents
Entomology Abstracts
Life Sciences
Zoological Record
ORIGINAL PAPERS
J.T. Wang, P.J. Meng, Y.Y. Chen,
and C.A. Chen
S. Keshavmurthy, C.M. Hsu, C.Y.
Kuo, V. Denis, J.K. Leung, S.
Fontana, H.J. Hsieh, W.S. Tsai,
W.C. Su, and C.A. Chen
J.T. Wang, Y.Y. Chen, P.J. Meng,
Y.H. Sune, C.M. Hsu, K.Y. Wei, and
C.A. Chen
H.U. Dahms, L.C. Tseng, S.H.
Hsiao, Q.C. Chen, B.R. Kim, and
J.S. Hwang
M.R. Zargaran, N. Erbilgin, and Y.
Ghosta
B. Huo, C.X. Xie, B.S. Ma, X.F.
Yang, and H.P. Huang
L. Gonçalves, C.I. da Silva, and
M.L.T. Buschini
J.S. Wu, P.J. Chiang, and L.K. Lin
M.F. Lin, M.V. Kitahara, H.
Tachikawa, S. Keshavmurthy, and
C.A. Chen
S.S. Young, M.H. Ni, and M.Y. Liu
B.A.R. Azman and B.H.R. Othman
R. Naderloo and M. Apel
V.K. Jayaraj, C.J. Laman, and M.T.
Abdullah
Vol. 51, No. 2
March, 2012
COMPARATIVE PHYSIOLOGY
137 Determination of the Thermal Tolerance of Symbiodinium Using the Activation
Energy for Inhibiting Photosystem II Activity
ECOLOGY
143 Larval Development of Fertilized “Pseudo-Gynodioecious” Eggs Suggests
a Sexual Pattern of Gynodioecy in Galaxea fascicularis (Scleractinia:
Euphyllidae)
150 Diverse Interactions between Corals and the Coral-Killing Sponge, Terpios
hoshinota (Suberitidae: Hadromerida)
160 Biodiversity of Planktonic Copepods in the Lanyang River (Northeastern
Taiwan), a Typical Watershed of Oceania
175 Changes in Oak Gall Wasps Species Diversity (Hymenoptera: Cynipidae) in
Relation to the Presence of Oak Powdery Mildew (Erysiphe alphitoides)
185 Age and Growth of Oxygymnocypris stewartii (Cyprinidae: Schizothoracinae)
in the Yarlung Tsangpo River, Tibet, China
195 Collection of Pollen Grains by Centris (Hemisiella) tarsata Smith (Apidae:
Centridini): Is C. tarsata an Oligolectic or Polylectic Species?
204 Monogamous System in the Taiwan Vole Microtus kikuchii Inferred from
Microsatellite DNA and Home Ranges
SYSTEMATICS AND BIOGEOGRAPHY
213 A New Shallow-Water Species, Polycyathus chaishanensis sp. nov.
(Scleractinia: Caryophylliidae), from Chaishan, Kaohsiung, Taiwan
222 Systematic Study of the Simocephalus Sensu Stricto Species Group
(Cladocera: Daphniidae) from Taiwan by Morphometric and Molecular
Analyses
232 Two New Species of Amphipods of the Superfamily Aoroidea (Crustacea:
Corophiidea) from the Strait of Malacca, Malaysia, with a Description of a
New Genus
248 Leucosiid Crabs of the Genus Hiplyra Galil, 2009 (Crustacea: Brachyura:
Leucosiidae) from the Persian Gulf and Gulf of Oman, with Description of a
New Species
259 A Predictive Model to Differentiate the Fruit Bats Cynopterus brachyotis and
C. cf. brachyotis Forest (Chiroptera: Pteropodidae) from Malaysia Using
Multivariate Analysis