Vous êtes sur la page 1sur 142

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


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











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.

This journal has been awarded by the National Science Council, Taiwan. It
can be available from Editorial Office, Biodiversity Research Center, Academia
Monogamous System in the Taiwan Vole Microtus
Sinica, Taipei 115, Taiwan. Printed by Cabin Printing Co., Ltd. 1st Fl., No. 30, kikuchii Inferred from Microsatellite DNA and Home
Lane 210, Sec. 2, Fu-Shin S. Rd., Taipei 100, Taiwan. 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
Graduate Institute of Biotechnology, Tajen Univ., Pingtung 907, Taiwan
National Museum of Marine Biology and Aquarium, Checheng, Pingtung 944, Taiwan
Institute of Marine Biodiversity and Evolution, National Dong Hwa Univ., Checheng, Pingtung 944, Taiwan
Institute of Oceanography, National Taiwan Univ., Taipei 108, Taiwan
Biodiversity Research Center, Academia Sinica, Nangang, Taipei 115, Taiwan
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 symbioses causing coral bleaching was found
mostly found in symbiosis with corals and sea to be closely related to the thermal inhibition of
anemones, are widely considered to underpin the photosystem (PS)II activity of symbiotic algae
ecological success of cnidarian-alga symbioses (Hill et al. 2004, Takahashi et al. 2008). With the
in shallow, nutrient-poor waters (Muscatine and 9 currently described clades (A-I) and numerous
Porter 1977, Falkowski et al. 1993). However, subclades within Symbiodinium (see review in
thermal stress caused by increasing seawater Coffroth and Santos 2005), the algae were also
temperatures results in a breakdown of symbiotic shown to exhibit different extents of tolerance to
associations and seriously threatens coral reefs thermal stress in culture or in hospite (Bhagooli
worldwide (Hoegh-Guldberg et al. 2007, Lesser and Hidaka 2003, Rowan 2004, Tchernov et al.
2007). Thermal breakdown of coral-Symbiodinium 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.

138 Wang et al. – Thermal Tolerance Index of Symbiodinium

2008). Selecting thermally-tolerant Symbiodinium the chlorophyll-a/chlorophyll-c2/peridinin protein

clades or subclades; therefore, was proposed complex, etc. (Takahashi et al. 2008). Thus,
as a way to promote the survival of corals in the the loss of PSII activity is expected to follow the
coming century (Chen et al. 2005a, Berkelmans process of protein denaturation. If the denaturation
and van Oppen 2006). This proposal was based of PSII proteins follows a first-order reaction, then
either on the biogeographic distribution of thermal- the Ea for thermally inhibiting PSII activity could
tolerant Symbiodinium in historically warming be calculated from the Arrhenius equation (a
regions (Chen et al. 2005a b, LaJeunesse et al. kinetic equation for measuring the Ea by linearly
2010) or on thermal-tolerant experiments under regressing the rate constant on the reaction
controlled laboratory conditions (Rowan 2004, temperature in °K). This Ea value could potentially
Tchernov et al. 2004, Sampayo et al. 2008). represent the thermal tolerance of Symbiodinium.
However, conflict occurs when thermal tolerance is Based on this idea, this study was conducted by
determined among different Symbiodinium clades subjecting freshly isolated Symbiodinium (FIS)
or subclades. For example, both biogeographic to elevated temperatures, measuring the rate of
and physiological experiments showed that decline in PSII activity (as indicated by Fv/Fm over
Symbiodinium clade D (specifically subclade D1a) time), and then fitting the rate constants to the
is the most heat-tolerant clade compared to clades Arrhenius equation to calculate the Ea.
A, B, and C (Rowan 2004, Chen et al. 2005b,
LaJeunesse et al. 2010). However, analysis of the
thylakoid membrane integrity showed that there MATERIALS AND METHODS
are also thermal-tolerant subclades within clades A,
B, and C, suggesting that a priori ribosomal DNA Symbiodinium isolation and subclade typing
phylotyping is not diagnostic for thermal sensitivity
of Symbiodinium associations (Tchernov et al. Freshly isolated Symbiodinium (FIS) samples
2004). To resolve this conflict, it is necessary to used in this study were designated Symbiodinium
develop a quantitative comparison with a single C1, C3, C15, D1a, and B1 from 4 hard corals
parameter or index to determine the thermal (Stylophora pistillata, Acropora humilis, Porites
tolerance among Symbiodinium clades and lutea, and Galaxea fasicularis) and a sea anemone
subclades. (Aiptasia pulchella), respectively, based on a
The thermal tolerance between different recent study (Wang et al. 2011). The corals were
Symbiodinium clades or subclades has been collected by scuba diving in Kenting National Park,
compared by estimating the temperature- Taiwan (21°55'54"N, 120°44'45"E), and the sea
dependent performance of the photosynthesis- anemone was obtained from a laboratory culture
irradiation response (Iglesias-Prieto et al. 1992, as described in Wang et al. (2011). Isolation of
Rowan 2004), the degree of decrease in PSII Symbiodinium from each replicate of the animal
activity during heat treatment (Bhagooli and Hidaka host was conducted as previously described
2003, Rowan 2004, Robinson and Warner 2006, (Wang and Douglas 1997, Wang et al. 2011).
Sampayo et al. 2008), or thermal sensitivity to Briefly, coral fragments having about 100 cm2 of
the induction of stress proteins (or enzymes) and live tissue were stripped of tissue using an air
their related genes (Downs et al. 2000, Brown et blast, and tentacles of 10 Aiptasia pulchella were
al. 2002, Souter et al. 2011). However, comparing homogenized with a tissue grinder. After mixing
results between studies has been difficult due to with 2-3 volumes of artificial seawater (Instant
the experimental designs and conditions used. In Ocean, Sarrebourg Cedex, France), the resultant
this study, we attempted to develop a universal slurry was passed through a 15-μm nylon mesh to
index, as indicated by the activation energy (Ea) remove debris. Symbiodinium was then isolated
for thermally inhibiting PSII activity, to represent the by centrifugation at 860 xg for 3 min and washed
thermal tolerance of members of Symbiodinium, with artificial seawater 3 times. Symbiodinium was
since the PSII activity of Symbiodinium is closely preserved in 80% ethanol before conducting the
associated with photosynthesis of the alga and phylotype analysis by resolving the polymerase
its symbiotic stability with corals (Robinson and chain reaction (PCR) product of the ribosomal
Warner 2006, and references therein). Moreover, internal transcribed spacer (ITS) 2 in denatured
the function of the PSII apparatus is determined by gel gradient electrophoresis (DDGE) developed
the natural state of several proteins, such as the by LaJeunesse et al. (2003) and modified as
D1 protein, peridinin-chlorophyll-a-binding proteins, described in Wang et al. (2011). Since the co-
Zoological Studies 51(2): 137-142 (2012) 139

existence of multiple clades or subclades in a Comparisons of Ea values between Symbiodinium

single coral host is well documented (Chen et subclades were made using a one-way analysis
al. 2005a b, Berkelmans and van Oppen 2006), of variance (ANOVA) following by Fisher ’s
FIS phylotyping of the ITS2 gene by PCR- least significance difference (LSD) test, with a
DGGE represented the dominant Symbiodinium significance level of p < 0.05.
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 RESULTS
the possibility of a mixture of clades or subclades
of FIS in the preparation (data not shown). Fv/Fm values of FIS from each preparation,
which ranged 0.618-0.675, were comparable
Fluorescence methodology between Symbiodinium subclades with a 4-h
incubation at 25°C under dim light. When data
FIS samples with about (0.5-0.8) × 10 6 from mixed populations of Symbiodinium subclades
cells/ml, counted with an improved Neubauer were excluded, Fv/Fm values of all subclades tested
hemocytometer (Marienfeld, Germany), were (C1, B1, C3, C15, and D1a) linearly decreased
maintained at 25°C under dim light (< 5 μE/m2/s, with incubation time at an elevated temperature, as
PAR) for 1 h before proceeding with heat treatment. shown by data for Symbiodinium subclade D1a in
Measurement of changes in the maximum figure 1A. Coefficients of the linear regression (r 2)
quantum yield [Fv/Fm = (Fm - Fo)/Fm] of FIS at the of the decrease in Fv/Fm values with incubation time
elevated temperatures began by suspending an at each treated temperature were all significant
algal pellet, collected from centrifugation of 10 ml (p < 0.05), and were 0.982 ± 0.008 (mean ± S.D.,
of an algal suspension at 860 xg for 2 min, in the n = 40) for D1a, 0.979 ± 0.010 (n = 35) for C15,
original volume of artificial seawater which had 0.981 ± 0.015 (n = 30) for C3, 0.980 ± 0.019
been prewarmed to the experimental temperature (n = 30) for B1, and 0.988 ± 0.007 (n = 40) for C1.
(of 31, 33, 35, 37, 39, or 41°C). The value of Fv/Fm When the logarithmically transformed denaturation
of the FIS suspension was directly measured at rate (k) of PSII was plotted against 1/T, coefficients
2-min intervals with a DIVING-PAM fluorometer of the linear regression were also significant
(Walz, Germany) at the DIVING-PAM setting of (p < 0.05), and showed a good fit to the Arrhenius
8 for measuring the light and saturating flash of equation (r 2 = 0.942-0.985), as shown by D1a data
the actinic light. The F v/F m of treated FIS was in figure 1B. The regression coefficients obtained
measured under indoor illumination (< 10 μE/m2/s, were 0.961 ± 0.019 (n = 8) for D1a, 0.966 ± 0.012
PAR), and heat treatment was completed within (n = 7) for C15, 0.965 ± 0.020 (n = 6) for C3, 0.965
12 or 14 min depending on the temperature used. ± 0.017 (n = 6) for B1, and 0.966 ± 0.013 (n = 8)
The F v/F m value of a control FIS that remained for C1. The Ea for PSII denaturation was then
at 25°C under dim light for 4 h was examined to calculated from each Arrhenius equation (Table
evaluate the quality of FIS used in the experiment. 1). The Ea values significantly differed among the
5 different Symbiodinium subclades (F4,30 = 288.3,
Kinetics and statistical analyses p < 0.001). The post-hoc analysis with Fisher’s
LSD test also indicated that Symbiodinium D1a
The rate constant, k (1/min), of PSII pro- displayed the highest Ea, followed in order by C15,
tein denaturation at each temperature was C3, B1, and C1 (Table 1). Ea values for D1a and
obtained from the slope of the linear regression C15 were almost 2-fold higher than those of B1
of Fv/Fm values against incubation times. Then, and C1.
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 DISCUSSION
Arrhenius equation, [ln(k) = ln(A) - (Ea/R)(1/T)],
was examined by a linear regression of ln(k) This study proposes that the activation energy
against 1/T. Therefore, the Ea of each sample was for inhibiting PSII activity under thermal stress
calculated from the Arrhenius equation obtained could be used to represent the thermal tolerance
above with the gas constant, R (= 8.314 J/mol/°K). of Symbiodinium. With such an index, thermal
The coefficient of variation (CV) was used to tolerances among Symbiodinium subclades could
examine the reproducibility between experiments. be compared on a universal scale. In order to
140 Wang et al. – Thermal Tolerance Index of Symbiodinium

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

(A) (B)
0.8 -2


ln (k) (min-1)

Fv /Fm



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) 141

the stability of the PSII apparatus is determined Soc. Lond. B 273: 2305-2312.
by the natural state of a set of proteins, especially Bhggooli R, M Hidaka. 2003. Comparison of stress suscep-
tibility of in hospite and isolated zooxanthellae among five
the D1 protein (Waner et al. 1999, Takahashi et al. coral species. J. Exp. Mar. Biol. Ecol. 291: 191-197.
2008). Fourth, the kinetics of protein denaturation Brown BE, CA Downs, RP Dunne, SW Gibb. 2002. Exploring
under heat treatment were reported to follow a the basis of thermotolerance in reef coral Goniastrea
1st-order reaction and comply with the Arrhenius aspera. Mar. Ecol. Progr. Ser. 242: 119-129.
equation (Weijers et al. 2003), and the Ea for Chen CA, JT Wang, LS Fang, YW Yang. 2005a. Fluctuating
algal symbiont communities in Acropora palifera
the thermal inhibition (or denaturation) of PSII (Scleractinia: Acroporidae) from Taiwan. Mar. Ecol. Progr.
proteins can be easily obtained from the Arrhenius Ser. 295: 113-121.
equation. Consequently, the data obtained in this Chen CA, YW Yang, NV Wei, WS Tsai, LS Fang. 2005b.
study indicated that Ea values for inhibiting PSII Symbiont diversity in the scleractinian corals from tropical
activity of each Symbiodinium subclade were reefs and non-reefal communities in Taiwan. Coral Reefs
24: 11-22.
consistent with previous studies (LaJeunesse Coffroth MA, SR Santos. 2005. Genetic diversity of symbiotic
et al. 2003, Fabricius et al. 2004, Rowan 2004, dinoflagellates in the genus Symbiodinium. Protist 156:
Tchernov et al. 2004, Berkelmans and van Oppen 19-34.
2006, Díaz-Almeyda et al. 2011), i.e., D1a and C15 Dahlhoff E, GN Somero. 1993. Effects of temperature on
were the 2 most thermally tolerant Symbiodinium mitochondria from abalone (genus Haliotis): adaptive
plasticity and its limits. J. Exp. Biol. 185: 151-168.
among the tested subclades. Reproducibility of Díaz-Almeyda E, PE Thomé, M El Hafidi, R Iglesias-Prieto.
the Ea data for each Symbiodinium subclade was 2011. Differential stability of photosynthetic membranes
determined to be acceptable by examining values and fatty acid composition at elevated temperature in
of the CV which ranged 3.2%-9.4%. Symbiodinium. Coral Reefs 30: 217-225.
In summary, a high regression coefficient Downs CA, E Mueller, S Phillips, JE Fauth, CM Woodley.
2000. A biomarker system for assessing the health of
(r 2 > 0.95) obtained from the kinetics data and coral (Montastrea faveolata) during heat stress. Mar.
the low CV between replicates (< 10%) indicated Biotechnol. 2: 533-544.
that the calculated Ea values for PSII denaturation Fabricius KE, JC Mieog, PL Colin, DH Idip, MJH van Oppen.
were stable and reliable. This evidence suggests 2004. Identity and diversity of coral endosymbionts
that the Ea for inhibiting PSII activity during (Zooxanthellae) from three Palauan reefs with contrasting
bleaching, temperature and shading histories. Mol. Ecol.
heat stress can be used to quantify the thermal 13: 2445-2458.
tolerance of Symbiodinium; thus, facilitating Falkowski PG, Z Dubinsky, L Muscatine, L McCloskey. 1993.
ecological, physiological, and evolutionary studies Population control in symbiotic corals. BioScience 43:
of coral symbiosis and bleaching biology. Based 606-611.
on this idea, it is also possible to develop an index Hill R, AWD Larkum, G Frankart, M Kuhl, PJ Ralph. 2004.
Loss of functional photosystem II reaction centres in
to represent bleaching susceptibility of corals by zooxanthellae of corals exposed to bleaching conditions:
selecting proper physiological indicator(s), changes using fluorescence rise kinetics. Photosynth. Res. 82: 59-
in which with an increase in heating temperature or 72.
light intensity follow a 1st-order reaction. 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:
Acknowledgments: The authors would like to 1737-1742.
thank members of the Coral Reef Evolutionary Iglesias-Prieto R, JL Matta, WA Robins, RK Trench. 1992.
Ecology and Genetics (CREEG) Group, Biodi- Photosynthetic response to elevated temperature in the
versity Research Center, Academia Sinica symbiotic dinoflagellates Symbiodinium microadriaticum in
(BRCAS) for field support and D.P. Chamberlin the culture. Proc. Natl. Acad. Sci. USA 89: 10302-10305.
LaJeunesse TC, WK Loh, RV Woesik, O Hoegh-Guldberg,
for English editing. This work was supported by GW Schmidt, WK Fitt. 2003. Low symbiont diversity in
an Academic Sinica Thematic grant (2008-2010) southern Great Barrier Reef corals relative to those of the
to JTW and CAC, and a National Science Council Caribbean. Limnol. Oceanogr. 48: 2046-2054.
grant (NSC96-2628-B-001-004-MY3) to CAC. This LaJeunesse TC, R Smith, M Walther, J Pinzón, DT Pettay,
is CREEG-BRCAS contribution no. 69. 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.
REFERENCES Lesser MP. 2007. Coral reefs bleaching and global climate
change: Can coral survive the next century? Proc. Natl.
Berkelmans R, MJH van Oppen. 2006. The role of zoox- Acad. Sci. USA 104: 5250-5260.
anthellae in the thermal tolerance of corals: a “nugget of Muscatine L, JW Porter. 1977. Reef corals: mutualistic
hope” for coral reefs in an era of climate change. Proc. R. symbiosis adapted to nutrient poor environment.
142 Wang et al. – Thermal Tolerance Index of Symbiodinium

BioScience 27: 154-157. of antenna proteins and photobleaching in cultured

Robinson JD, ME Warner. 2006. Differential impacts of Symbiodinium. Proc. Natl. Acad. Sci. USA 105: 4203-
photoacclimation and thermal stress on the photobiology 4208.
of four different phylotypes of Symbiodinium (Pyrrhophyta). Tchernov D, MY Gorbunov, C de Vargas, SW Yadav, AJ
J. Phycol. 42: 568-579. Milligan, M Häggblom, PG Falkowski. 2004. Membrane
Rowan R. 2004. Thermal adaptation in reef coral symbionts. lipids of symbiotic algae are diagnostic of sensitivity to
Nature 430: 742. thermal bleaching in corals. Proc. Natl. Acad. Sci. USA
Sampayo EM, T Ridgway, P Bongaerts, O Hoegh-Guldberg. 101: 13531-13535.
2008. Bleaching susceptibility and mortality of corals are Wang JT, AE Douglas. 1997. Nutrients, signals, and photo-
determined by fine-scale differences in symbiont type. synthate release by symbiotic algae: the impact of taurine
Proc. Natl. Acad. Sci. USA 105: 1444-1449. on the dinoflagellate alga Symbiodinium from the sea
Souter P, LK Bay, N Anderakis, N Császár, FO Seneca, MJH anemone Aiptasia pulchella. Plant Physiol. 114: 631-636.
van Oppen. 2011. A multilocus, temperature stress- Wang JT, PJ Meng, E Sampayo, SL Tang, CA Chen. 2011.
related gene expression profile assay in Acropora Photosystem II breakdown induced by reactive oxygen
millepora, a dominant reef-building coral. Mol. Ecol. species (ROS) in the freshly-isolated Symbiodinium of
Resources 11: 225-421. Montipora (Scleractinia; Acroporidae). Mar. Ecol. Progr.
Stenseng E, C Braby, GN Somero. 2005. Evolutionary and Ser. 422: 51-62.
acclimation-induced variation in the thermal limits of heart Warner MK, WK Fitt, GW Schmidt. 1999. Damage to photo-
function in congeneric marine snails (genus Tegula): system II in symbiotic dinoflagellates: a determinant of
implications for vertical zonation. Biol. Bull. 208: 138-144. coral bleaching. Proc. Natl. Acad. Sci. USA 96: 8007-
Stillman JH. 2002. Causes and consequences of thermal 8012.
tolerance limits in rocky intertidal porcelain crabs, genus Weijers M, PA Barneveld, MAC Stuart, RW Visschers. 2003.
Petrolisthes. Integr. Compar. Biol. 42: 790-796. Heat-induced denaturation and aggregation of ovalbumin
Takahashi S, S Whitney, S Itoh, T Maruyama, M Badger. 2008. at neutral pH described by irreversible first-order kinetics.
Heat stress causes inhibition of the de novo synthesis 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,*
Biodiversity Research Center, Academia Sinica, Nangang, Taipei 115, Taiwan
Institute of Oceanography, National Taiwan Univ., Taipei 106, Taiwan
Institute of Life Sciences, National Taiwan Normal Univ., Taipei 106, Taiwan
Univ. of Milan-Bicocca, Piazza della Scienza 2, Milan 20126, Italy
Penghu Marine Biological Research Center, Makong 880, Taiwan
Institute of Life Science, National Taitung Univ., Taitung 950, Taiwan
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 and Wallace 1990, Baird et al. 2009, Harrison
are the most important life-history traits in 2011). Three sexual patterns (hermaphroditic,
scleractinian corals, and have been one of the gonochronic, and mixed) and 2 modes of
major research themes over the last 3 decades development (broadcast-spawned gametes and
(reviewed in Richmond and Hunter 1990, Harrison 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

144 Keshavmurthy et al. – Gynodioecy in Scleractinian Corals

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

(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) 145

MATERIALS AND METHODS (2000) using the same terminology. A series of

photographs was taken using an Olympus 5050
Study site and sample collection camera (Tokyo, Japan) attached to the eyepiece
of an Olympus light microscope to obtain images
Coral spawning was observed at CIB of the developmental stages between white and
(23°31'N, 119°33'E), Penghu Is., Taiwan in Apr.- red eggs until the swimming planular larval stage.
June 2011. CIB is a semi-enclosed embayment Galaxea fascicularis white and red eggs inside
where coral communities have developed on top the coral tissues were photographed under 40x
of volcanic rocks, with 75 species of scleractinian magnification (objective lens 4x and eyepiece
corals described (Hsieh 2008, Hsieh et al. 2011). 10x) using an Olympus microscope (model SZ40)
Over 50 colonies of G. fascicularis with a colony fitted with an Olympus C5050 digital camera. The
size of > 10 cm in diameter were collected, gonads were placed in a Petri dish immersed in
deposited in individual buckets, and moved seawater without a cover. Images of white and red
to tanks with a continuous seawater flow and egg were photographed under 100x magnification
aeration system at the joint marine laboratory of (object 10x and eyepiece 10x) using an Olympus
the Biodiversity Research Center, Academia Sinica microscope (model CX31) fitted with an Olympus
(BRCAS)-Penghu Marine Biological Research E510 digital camera. The same gonads were
Center (PMBRC) at CIB. Acropora muricata moved to a glass slide and gently put on the slide
was also collected for reference to compare cover without any pressure. Time-series photos of
developmental stages from fertilized eggs to G. fascicularis were taken under 40x magnification
elongated planular larvae (Miller and Ball 2002). (objective lens 4x and eyepiece 10x) using an
Olympus microscope (model SZ40) fitted with
Observation of spawning and crossing expe- Olympus SP350 and C5050 digital cameras. The
riments cameras were fitted directly to the eyepiece of
the microscope to obtain the photos. Time-series
Observations of spawning behavior at CIB photos of Acropora muricata were taker under 40x
began on 12 Apr. 2011, 5 d before the full moon in magnification with an Olympus C5050 camera.
Apr., based on previous observations (Chen et al. The egg size and scale shown in the photos were
unpubl. data). Throughout the spawning period, obtained by micro-ruler photo of a hemocytometer
seawater flow in the tanks was stopped daily, by obtained at the respective magnifications.
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 RESULTS
time of release of gamete bundles was recorded
once polyps and tentacles were retracted, and Galaxea fascicularis colonies at CIB,
colored bundles, either white or pinkish-red, Penghu Is. were either female (pinkish-red eggs)
were released to the surface of the buckets. On or hermaphroditic (white eggs with sperm sacs)
24 May 2011, 10 colonies of G. fascicularis with (Fig. 1A, B). No spawning was observed for G.
white eggs and 10 colonies with red eggs were fascicularis in Apr. 2011 (normal spawning period
labeled for bundle collection. Gamete bundles in Penghu begins from Apr.). However, on 24
released to the surface of the water in the buckets May 2011, 7 nights after the full moon of May,
were separately scooped up using recycled synchronous spawning of G. fascicularis (> 30
plastic cups, and brought back to the laboratory colonies) was first observed at 19:30, 1 h after
for crossing experiments. Both white and red sunset at the Penghu Is. with a peak of gamete
bundles were filtered through a plankton mesh bundles released at around 20:00 (Fig. 1A, B).
with a 150-μm-mesh size to separate eggs and Continued release of bundles was observed the
sperm. Aliquots of eggs and sperm were collected following 4 nights with a decrease in the number
for size measurements and density counts. of colonies spawned on the 8th night after the full
Sperm density was diluted to 105-106/ml for the moon (Table 1). Another synchronous spawning
crossing experiment (Willis et al. 1997 2006). event of over 30 colonies was observed on 22
White and red eggs were mixed and fertilized June, 6 nights after the full moon of June (Table 1).
with diluted sperm. Developmental stages were Some colonies spawned multiple times either on
observed every hour and categorized based on different nights in May or continuously in June.
stages described for Acropora by Miller and Ball Dissecting gamete bundles suggested
146 Keshavmurthy et al. – Gynodioecy in Scleractinian Corals

that both white and red eggs of G. fascicularis gynodioecious sexual pattern in scleractinian
were mature and had reached the same stage corals.
just before spawning. Some white eggs from
hermaphroditic colonies possessed a clear Egg, embryonic, and larvae development in
nucleus close to the edge of the egg, as seen in Galaxea fascicularis
red eggs (Fig. 1C, D). Spawned white eggs had
a significantly (t-test = -72.1769, p < 0.01) smaller According to a previous study (Harrison
mean diameter (290.20 ± 2.60 μm, n = 171) than 1989) conducted at the GBR, Australia, the
red eggs (438.58 ± 3.13 μm, n = 184) (Fig. 2). reproduction mode in G. fascicularis was reported
Fertilization experiments showed that both to be pseudo-gynodioecious suggesting that
white and red eggs were mature, and embryo white eggs produced by this species cannot be
development was synchronous (Fig. 3). Two- fertilized. This raises the question as to why white
cell cleavage was observed during the 1st hr eggs of G. fascicularis at CIB, Penghu Is., Taiwan
after fertilization (Fig. 3A). The time of the initial were fertile, but those in the GBR, Australia were
development (cell-cleavage stage) cycle in G. not? Two possible scenarios are proposed to
fascicularis embryos was similar to that of A. explain this difference. First, our observed results
muricata before reaching the prawn-chip stage (Fig. may have been due to geographic differentiation
3C-F). Galaxea fascicularis took 8 hr to reach the between G. fascicularis populations in the GBR,
prawn-chip stage, while A. muricata needed at Australia and those at CIB, Penghu Is., Taiwan.
least 12 hr after fertilization (Miller and Ball 2000). In some scleractinian corals, sexual patterns
Also, embryonic development from the donut to the and reproductive modes can vary in different
pear stage in G. fascicularis was significantly faster geographic regions (reviewed in Harrison 2011).
than that of A. muricata (Fig. 3I-P). The swimming For example, histological studies on Pocillopora
ability of planular larvae fertilized from white eggs damicornis colonies in Japan indicated that
did not differ from that of larvae from red eggs. brooded planulae develop from eggs, and may


Our study provides several lines of evi-

Mean diameter of eggs (µm)

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 Red White
a full moon (h after sunset) colonies spawned Egg colour
from a total of n = 50
Fig. 2. Difference in egg sizes between female (red egg color)
May 24 7 1 > 30 and hermaphroditic (white egg color) colonies of Galaxea
25 8 1 > 30 fascicularis. The egg size data were plotted using software R
26 9 1 <5
to generate a box plot. The upper and lower hinge of the box
27 10 1 <5
indicate 75th and 25th percentile of the data set. The line in the
28 11 1 <5
middle of the box represents median for each data set of egg
June 22 6 1 > 30
sizes indicating a skewed data set. Vertical dotted lines with
23 7 1 - whiskers at top and bottom represent maximum and minimum
24 8 1 - values. Circles in the figure are outliers with values outside
the 25%-75% interval. The absence of circles for red eggs
-, not spawning. indicates that there were no outliers.
Zoological Studies 51(2): 143-149 (2012) 147

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


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


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



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


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 (e.g., E. ancora) or hermaphroditic species with
with large quantities of gametes that increased the brooded larvae (e.g., E. glabrescens) (Veron
chances of observing serial embryo development 2000). Gynodieocy in Galaxea might represent a
of white eggs from hermaphroditic colonies. transitional step of sexual pattern evolution in the
Further investigations of the percentage of fertile family Euphyllidae. In addition, gynodieocy also
white eggs in gamete bundles of hermaphroditic suggests a unique inheritance mode of genetics
colonies and of the survival, settlement, recruitment in Galaxea compared to true hermaphroditic or
success, and growth of derived juvenile corals are dioecious species. Further work on ancestral
needed to confirm the contribution of white eggs to reconstruction of life-history traits and genetic
G. fascicularis populations. structuring of populations should provide insights
Overall, results from this study showed that into the evolutionary novelty of gynodioecy in
embryonic development time is much shorter Galaxea among scleractinian corals.
in G. fascicularis compared to that in Acropora.
The length of embryonic development could Acknowledgments: Many thanks go to the
affect dispersal and recruitment among different staff of the Penghu Marine Biological Research
spawning corals (Nakamura and Sakai 2010). For Center (PMBRC), Council of Agriculture for logis-
example, among spawning pocilloporid corals, tical support, and members of the Coral Reef
larvae that develop relatively more rapidly have Evolutionary Ecology and Genetics (CREEG)
higher recruitment at sites where adult coral cover 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,
is high. In contrast, recruitment is not related to Academia Sinica (BRCAS) and 2 anonymous
adult coral cover in acroporid and poritid corals, reviewers for constructive comments. CMH and
the embryonic development times of which are JKL are recipients of a PhD fellowship, and SK
relatively slow (Nakamura and Sakai 2010). is the recipient of a postdoctoral fellowship from
The shorter embryonic development time might Academia Sinica (2010-2012). VD is the recipient
facilitate G. fascicularis settling locally faster, and of a postdoctoral fellowship from the National
helping it become the dominant species after Science Council (NSC), Taiwan. This study was
a series of disturbances and disappearance of made possible by an Academia Sinica Thematic
acroporid corals (Acropora and Montipora) after Grant (AS-100-TP2-A02) and grants from the
a cold shock event in 2008 at CIB (Hsieh et al. NSC (NSC99-2621-B-001-006-MY3) to CAC and
2008 2011). The recruitment of acroporid corals (NSC98-2313-B-056-001-MY3) HJH. This is
may be slower because sources of larvae are from CREEG-BRCAS contribution no. 75, and BRCAS-
neighboring coral communities outside CIB. PMBRC Joint Marine Laboratory contribution no. 1.

Sexual pattern of Galaxea fascicularis

gynodioecy REFERENCES

Completion of embryonic and larval deve- Baird AH, JR Guest, BL Willis. 2009. Systematic and bio-
lopment of white eggs from hermaphroditic colonies geographical patterns in the reproductive biology of
scleractinian corals. Annu. Rev. Ecol. Syst. 40: 551-571.
suggests that the sexual pattern of G. fascicularis Charlesworth D. 2006. Evolution of plant breeding systems.
is gynodioecious, instead of pseudo-gynodioecious Curr. Biol. 16: R726-R735.
as proposed by Harrison (1989). Studies on Chavez-Romo HE, H Reyes-Bonilla. 2007. Sexual repro-
plant reproductive systems have indicated that duction of the coral Pocillopora damicornis in the southern
gynodioecy is a transitional step towards dioecy Gulf of California, Mexico. Cienc. Mar. 33: 495-501.
Colley SB, PW Glynn, AS Mayt, JL Mate. 2006. Species-
(gonochorism) from hermaphroditism (reviewed dependent reproductive responses of eastern Pacific
in Charlesworth 2006). This scenario might be corals to the 1997-1998 ENSO event. In Proceedings
applicable to the evolution of sexual pattern traits of the 10th International Coral Reef Symposium, vol. 1,
in Galaxea. Galaxea is the only coral genus that Okinawa, Japan, 2004, pp. 61-70.
possesses a sexual pattern of gynodieocy, and Dai CF, S Horng. 2010. In, Scleractinia Fauna of Taiwan 1.
The Complex group. National Taiwan University, Taipei
phylogenetic studies have relocated Galaxea from Taiwan, pp. 175.
the family Oculinidae to the Euphyllidae, where it Diah Permata W, RA Kinzie III, M Hidaka. 2000. Histological
forms a sister clade to the genus Euphyllia (Fukami studies on the origin of planulae of the coral Pocillopora
et al. 2008, Dai and Horng 2010). The sexual damicornis. Mar. Ecol. Progr. Ser. 200: 191-200.
patterns of 8 Euphyllia species can be divided into Flot JF, H Magalon, C Cruaud, A Couloux, S Tillier. 2008.
Patterns of genetic structure among Hawaiian corals of
either dioecious species with spawned gametes
Zoological Studies 51(2): 143-149 (2012) 149

the genus Pocillopora yield clusters of individuals that are al. 2011. Establishment of a no-take area (NTA) could
compatible with morphology. Compt. Rend. Biol. 331: not guarantee the preservation of coral communities in
239-247. Chinwan Inner Bay, Penghu, Taiwan. Zool. Stud. 50: 443-
Fukami H, CA Chen, AF Budd, A Collins, C Wallace, YY Chuang 453.
et al. 2008. Mitochondrial and nuclear genes suggest Hsieh HJ, IL Shen, MS Jeng, WS Tsai, WC Su, CA Chen.
that stony corals are monophyletic but most families of 2008. Tropical fish killed by the cold. Coral Reefs. DOI:
stony corals are not (order Scleractinia, class Anthozoa, 10.1007/s00338-008-0378-3.
phylum Cnidaria). PLoS ONE 3: 1-9. Miller DJ, EE Ball. 2000. The coral Acropora: what it can
Glynn PW, SB Colley. 2009. Survival of brooding and contribute to our knowledge of metazoan evolution and
broadcasting reef corals following large-scale the evolution of developmental processes. BioEssays 22:
disturbances: Is there any hope for broadcasting 291-296.
species during global warming? In Proceedings of the Nakamura MS, K Sakai. 2010. Spatiotemporal variability in
11th International Coral Reef Symposium, vol. 1, Ft. recruitment around Iriomote Island, Ryukyu Archipelago,
Lauderdale, FL, USA, 2008, pp. 361-365. Japan: implications for dispersal of spawning corals. Mar.
Glynn PW, NJ Gassman, CM Eakin, J Cortés, DB Smith, HM Biol. 157: 801-810.
Guzmán. 1991. Reef coral reproduction in the eastern Richmond RH, CL Hunter. 1990. Reproduction and recruitment
Pacific: Costa Rica, Panamá, and Galápagos Islands of corals: comparisons among the Caribbean, the tropical
(Ecuador). I. Pocilloporidae. Mar. Biol. 109: 355-368. Pacific, and the Red Sea. Mar. Ecol. Progr. Ser. 60: 185-
Harrison PL. 1989. Pseudo-gynodioecy: an unusual breeding 203.
system in the scleractinian coral Galaxea fascicularis. Souter P. 2010. Hidden genetic diversity in a key model
Proc. 6th. Int. Coral Reef Symp. 2: 699-705. species of coral. Mar. Biol. 157: 875-885.
Harrison PL. 2011. Sexual reproduction of scleractinian Willis BL, RC Babcock, PL Harrison, CC Wallace. 1997.
corals. In Z Dubinsky, N Stambler, eds. Coral reefs: an Experimental hybridization and breeding incompatibilities
ecosystem in transition Part 3. USA: Springer, pp. 59-85. within the mating systems of mass spawning reef corals.
Harrison PL, RC Babcock, GD Bull, JK Oliver, CC Wallace, Coral Reefs 16: S53-S65.
BL Willis. 1984. Mass spawning in tropical reef corals. Willis BL, MJH van Oppen, DJ Mille, SV Vollmer, DJ Ayre.
Science 223: 1187-1188. 2006. The role of hybridization in the evolution of reef
Harrison PL, CC Wallace. 1990. Reproduction, dispersal and corals. Ann. Rev. Ecol. Evol. Syst. 37: 489-517.
recruitment of scleractinian corals. In Z Dubinsky, ed. Veron JEN. 2000. Corals of the world. Townsville, Australia:
Coral reefs. Amsterdam: Elsevier, pp. 133-207. Australian Institute of Marine Science.
Hsieh HJ, KS Chen, YL Lin, YA Huang, AH Baird, WS Tsai et
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
Graduate Institute of Biotechnology, Tajen Univ., Pingtung 907, Taiwan
National Museum of Marine Biology and Aquarium, Pingtung 944, Taiwan
Institute of Marine Biodiversity and Evolution, National Dong Hwa Univ., Checheng, Pingtung 944, Taiwan
Institute of Oceanography, National Taiwan Univ., Taipei 108, Taiwan
Biodiversity Research Center, Academia Sinica, Nangang, Taipei 115, Taiwan
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.

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

Wang et al. – Coral-Terpios Interactions 151

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

active radiation, temperature control (25°C), Tokyo, Japan), and coated with platinum in an ion
filtration (EHEIM, Deizisau, Germany), and a sputter (Hitachi E1010). SEM observations were
protein skimmer. The seawater level in the tank made on an SEM (Hitachi S-3500N) at a voltage of
was kept at only 20 cm deep, and 2 underwater 5 kV.
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 RESULTS
debris at the bottom of the tank. Newly growing
sponge on the shell debris was also examined After examining interactions of T. hoshinota
by SEM for comparison of Terpios on a non-coral with 19 species of scleractinian coral and 1
substratum. hydrozoan coral, the results indicated that there
were 4 types of interactions between them, i.e.,
SEM method hairy tips, thick tissue threads, compact edges,
and disintegrated tissues (Table 1). The 2 most
Freshly collected Terpios hoshinota was common features at the coral-facing growth front
persevered in fixative (2.5% glutaraldehyde, of Terpios were hairy tips and compact edges,
2% paraformaldehyde, and 5% sucrose in which were respectively found in 13 (totally 23
0.1 M phosphate buffer at pH 7.0) overnight at specimens) and 12 (totally 23 specimens) species
4°C. Subsequently, specimens were washed in of victim corals. Thick tissue threads were less
phosphate buffer and post-fixed in 2% OsO4/0.1 M often found at the coral-facing growth front of
phosphate buffer (pH 7.3) overnight. Standard the sponge, and were found in only 7 (totally 8
procedures were used to prepare coated samples specimens) species of coral victims, and only that
for SEM observations. Coral-Terpios samples on Por. cylindrical was found in both specimens
were dried in a critical-point dryer (Hitachi HCP-2, 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 displayed no significant changes in the color
sponge, which was only found in 1 specimen morph or Symbiodinium density. Under SEM exa-
of Mil. exaesa. On the coral side, disintegrated mination, hairy tips were found to be occupied
tissues were also rarely observed; only 5 (totally 6 by cyanobacteria, sponge tissues, and spicules
specimens) coral species displayed disintegrated (Fig. 1B-D). Nematocysts obviously from the
tissues at the coral-Terpios border. Eighty percent victim coral were also found on the surface of the
of specimens of the 19 species of coral displayed hairy tips (Fig. 1D). Direct exposure of internal
comparable color morphs and Symbiodinium cyanobacteria and spicules from the hairy tips
densities with nearby corals in the same colony (Fig. 1C, D) was caused by the loss of the fragile
which had not been attacked by Terpios (see pinacoderm during SEM processing.
example photos in Figs. 1A, 2A, 3B). Table 1 also In figure 2, the compact edge and thick
indicates that on I. palifera, Mon. aequituberculata, tissue threads at the Terpios growth front on Pla.
Mon. peltiformis, Poc. verrucosa, Psa. digitata, and ryukyuensis are shown. The thick tissue threads
Eph. aspera, the sponge displayed only 1 feature, were obviously an extension of sponge tissues.
hairy tips, at the coral-facing growth front; but the However, the microscopic image of the compact
sponge on the other 14 species of coral displayed edge of Terpios revealed only spicules but no
more than 1 feature. sponge tissue or cyanobacteria extruding from
Typical examples of detailed interactions the sponge front (Fig. 2B, C). Figure 2B and 2C
between corals and Terpios are shown in figures also indicate that there was no direct contact
1-6. Figure 1 shows lots of hairy tips along between the sponge front and coral tissues, and
the Terpios growth front on I. palifera. Hairy no obvious disintegration was found on the coral
tips of Terpios touch the coral surface when surface. Sometimes, a clear border was also
it moves forward. As shown in figure 1A, the found at the coral-Terpios interface. As shown in
coral surface at the boundary next to Terpios 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 Terpios growth was also found in an extreme case
with disintegrated tissues of S. pistillata. This during our field survey: 1 A. digitifera colony had
interaction was not prevalent throughout the entire maintained the same boundary with Terpios for
Stylophora colony, because thick tissue threads more than a year with no advance or regression.
derived from the sponge were also found to Disintegration of coral tissues at the coral-
interact with the coral on different branches of the Terpios border, even though very seldom, was
same colony (Fig. 3B). Under SEM examination, also found. Taking G. edwardsi as an example,
there was a 200-500-μm wide border without coral the coral tissue displayed disintegration and had
or sponge tissues between the 2 antagonists (Fig. disintegrated into filamentous residues along
3C). When zooming into the Terpios front at the the border contacting Terpios (Fig. 4A). When
coral-interacting side, as shown in figure 3, many examined at high magnification, the growth front
stinging nematocysts were found at the leading of Terpios seemed to move forward by penetrating
edge of Terpios tissues. Opposition against underneath the disintegrated coral tissue (Fig.


(B) (C)




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



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)



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

(A) (B)



(C) (D)



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 157

features of the border between the 2 antagonists (ALS) of Tang et al. (2011). But ALSs of Terpios
were not uniform among coral species or even were less often found in the field. We usually
within the same colony. These observations found ALSs when the growing edge of the sponge
indicate that interactions between corals and ran out of substratum of an invaded coral and
Terpios are dynamic and also not species- tried to climb over an adjacent coral colony or
specific as described in other coral-sponge perhaps faced strong defense by the coral victim
interactions (Averts 1998 2000, McLean and (e.g., Mil. exaesa as seen in the inset of Fig. 1D).
Yoshioka 2008). Averts (2000) also indicated Another feature at the coral-Terpios border was
that the direction of overgrowth by the sponge the smooth and compact growing edge of Terpios,
might be attributed to the level of compactness which is similar to the interaction in other crustose
of the coral, suggesting that the health status of sponges, such as Cliona caribbaea and Cli. lampa,
the coral might also be a determining factor in advancing on coral (Rützler 2002).
Terpios infections. Overgrowth by the sponge The observation of a several-millimeter-wide
when invading a coral was described as occurring band of dead zooids paralleling the growing edge
by elevating the sponge’s growing edge (McLean of the sponge usually indicates that allelochemical
and Yoshioka 2007), but this was not the only interactions are important in spatial competition
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 (A)
tendrils described by Rützler and Muzik (1993),
which are full of sponge tissues, spicules, and
cyanobacteria as found in the arm-like structure



SE WD33.7 mm 5.00 kV x35 1 mm



SE WD33.6 mm 5.00 kV x300 100 μm SE SF WD11.3 mm 5.00 kV x180 250 μm

Fig. 6. Millepora exaesa growing over T. hoshinota. (A, B) Fig. 7. Growth of T. hoshinota on shell debris in an aquarium.
The same specimen at different magnifications; the white (A) Terpios on Isopora palifera maintained in a laboratory
circle indicates the site from where (B) is amplified. CS, coral aquarium; the white arrowhead indicates shell debris used for
surface; NT, disintegrated Terpios tissue with spicules exposed; examination. (B) SEM examination of the shell debris with
S, spicules penetrating out of the coral surface. 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 to findings of spicules protruding from the coral
allelochemical effect was also considered one surface next to the coral-sponge border. This
of the mechanisms by which some aggressive occurred only when the coral grew on un-degraded
sponges invaded corals (Jackson and Buss spicules left behind by disintegrated tissues of
1975, Porter and Targett 1988). In coral-Terpios Terpios. When the soft tissues of coral moved
interactions, both specimens from G. edwardsi over lobed tyrostyle spicules, the spicules were
and one of 2 specimens from 4 other coral lifted up and penetrated through the coral tissue as
species displayed disintegration at the coral- shown in figure 6B.
sponge border. Most of the coral specimens Overgrowth by Terpios has caused sub-
(80%) showed comparable color morphs and stantial losses of coral coverage in Guam (Plucer-
Symbiodinium densities between tissues closely Rosario 1987), the Ryukyus, Japan (Rützler and
contacting the sponge and those remote from the Muzik 1993), and Green I., Taiwan (Liao et al.
coral-Terpios border. Therefore, even though T. 2007). According to the experience in Guam, the
hoshinota was reported to produce chemicals, coral coverage in Guam might recover (Plucer-
such as nakiterpiosin and nakiterpiosinone, Rosario 1987). A similar observation was also
with potent cytotoxicity (Teruya et al. 2004), made in Japan (Reimer et al. 2011). Fortunately,
allelochemicals might not be the major mechanism the coverage of Terpios has not further expanded
by which Terpios kills coral during its competition at Green I. since it was first noted in 2006 (Liao et
for substratum. It was more evident that Terpios al. 2007). The static situation of Terpios coverage
kills corals by overgrowing them. might partly be due to a seasonal typhoon effect,
Scleractinian corals exhibit a wide variety but the dynamic interactive mode between the 2
of offensive and defensive mechanisms for antagonists, as revealed in this study, might be
acquiring and maintaining a living space (Connell a crucial factor promoting the survival of invaded
1973, Wahle 1980, McCook et al. 2001). One of corals. If disturbance decreases, Connell (1978)
the mechanisms is to use stinging warfare such suggested that coral recolonization is possible
as nematocysts or spirocysts. The ‘stinging’ within a period of time. Therefore, it is our hope
mechanism includes the effects of polyps, sweeper that we will see coral recovery from the Terpios
tentacles, and mesenterial filaments. Several invasion, if the environmental conditions of coral
processes were documented in interspecific reefs can be protected from anthropogenic and
competition among corals (reviewed by Lang natural disturbances.
and Chornesky 1990) and also between corals
and a range of other animals such as zoanthids Acknowledgments: We would like to thank
and gorgonians (Karlson 1980, Chornesky 1983, members of the Coral Reef Evolutionary Ecology
Chadwick 1987). During overgrowth by Terpios, and Genetics (CREEG) Group, Biodiversity
some but not all victimized corals were observed Research Center, Academia Sinica (BRCAS) for
to have ejected nematocysts at the contacting field support, and Profs. K. Soong and E. Hirose
border. However, the defenses, including poten- for their valuable comments. This work was made
tial effectors not observable by SEM, such as possible by grants from the National Science
chemicals, did not seem to be very effective in Council, Taiwan (NSC98-2321-B-127-001-MY3) to
most cases. One successful case was found in JTW and (NSC98-2321-B-001-024-MY3) CAC and
the interaction between S. pistillata and Terpios, a BRCAS grant to CAC. This is CREEG-BRCAS
in which nematocysts at the surface of the contribution no. 72.
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 REFERENCES
the coral not only caused disintegration of the
Terpios growing edge but was also overgrowing Averts LAM. 1998. Sponge/coral interactions in Caribbean
the sponge in the reverse direction. Of course, reefs: analysis of overgrowth patterns in relation to
species identity and cover. Mar. Ecol. Progr. Ser. 175:
chemical effects derived from Mil. exaesa cannot 241-249.
be excluded because it is a notorious toxin Averts LAM. 2000. Dynamics behind standoff interactions
producer (Wittle et al. 1971, Shiomi et al. 1989, in three reef sponge species and the coral Montastrea
Radwan and Aboul-Dahab 2004, Iguchi et al. cavernosa. Mar. Ecol. 21: 191-204.
2008). Millepora exaesa was further found to Bryan PG. 1973. Growth rate, toxicity and distribution of the
encrusting sponge Terpios sp. (Hadromerida: Suberitidae)
retrogressively grow on invading Terpios according
Wang et al. – Coral-Terpios Interactions 159

in Guam, Mariana Islands. Micronesica 9: 237-242. anchorata. Carib. J. Sci. 44: 83-89.
Chadwick NE. 1987. Interspecific aggressive behavior of Plucer-Rosario G. 1987. The effect of substratum on the
the corallimorpharian Corynactis californica (Cnidaria: growth of Terpios, an encrusting sponge which kills corals.
Anthozoa): effects on sympatric corals and sea ane- Coral Reefs 5: 197-200.
mones. Biol. Bull. 173: 110-125. Porter JW, NM Targett. 1988. Allelochemical interactions
Chornesky EA. 1983. Induced development of sweeper between sponges and corals. Biol. Bull. 175: 230-239.
tentacles on the reef coral Agaricia agaricites: a response Radwan FY, HM Aboul-Dahab. 2004. Milleporin-1, a new
to direct competition. Biol. Bull. 165: 569-581. phospholipase A2 active protein from the fire coral
Connell JH. 1973. Population ecology of reef-building corals. Millepora platyphylla nematocysts. Comp. Biochem.
In OA Jones, R Endean, eds. Biology and geology of Physiol. Part C Toxicol. Pharmacol. 139: 267-272.
coral reefs, Vol. II: Biology 1. New York: Academic Press, Reimer JD, Y Nozawa, E Hirose. 2011. Domination and
pp. 205-245. disappearance of the black sponge: a quarter century
Connell JH. 1978. Diversity in tropical rain forests and coral after the initial Terpios outbreak in southern Japan. Zool.
reefs. Science 199: 1302-1310. Stud. 50: 395.
Fujii T, S Keshavmurthy, W Zhou, E Hirose, CA Chen, JD Rützler K. 2002. Impact of crustose Clionid sponges on
Reimer. 2011. Coral-killing cyanobacteriosponge (Terpios Caribbean reef corals. Acta Geol. Hispan. 37: 61-72.
hoshinota) on the Great Barrier Reef. Coral Reefs 30: Rützler K, K Muzic. 1993. Terpios hoshinota, a new cyano-
483. bacteriosponge threatening Pacific reefs. Sci. Mar. 57:
Hirose E, A Murakami. 2011. Microscopic anatomy and 395-403.
pigment characterization of coral-encrusting black sponge Shiomi K, M Hosaka, N Yanaike, H Yamanaka, T Kikuchi. 1989.
with cyanobacterial symbiont, Terpios hoshinota. Zool. Partial characterization of venoms from two species of
Sci. 28: 199-205. fire corals Millepora platyphylla and Millepora dichotoma.
Iguchi A, S Iwanaga, H Nagai. 2008. Isolation and charac- Nippon Suisan Gakk 55: 357-362.
terization of a novel protein toxin from fire coral. Biochem. Soong K, SL Yang, CA Chen. 2009. A novel dispersal
Biophys. Res. Comm. 365: 107-112. mechanism of a coral-threatening sponge, Terpios
Jackson JBC, L Buss. 1975. Allelopathy and spatial com- hoshinota (Suberitidae, Porifera). Zool. Stud. 45: 596.
petition among coral reef invertebrates. Proc. Nat. Acad. Suchanek TH, RC Carpenter, JD Witman, CD Harvell.
Sci. USA 72: 5160-5163. 1983. Sponge as important space competitors in deep
Karlson RH. 1980. Alternative competitive strategies in a Caribbean coral reef communities. In ML Reaka, ed. The
periodically disturbed habitat. Bull. Mar. Sci. 30: 894-900. ecology of deep and shallow coral reefs. Symposia Series
Lang JC, EA Chornesky. 1990. Competition between scle- for Undersea Research, NOAA Undersea Research
ractinian reef corals - a review of mechanisms and effects. Program 1(1). Washington DC: National Oceanic and
In Z Dubinsky, ed. Ecosystems of the world, Vol. 25. Atmospheric Administration (NOAA), pp. 55-60.
Coral reefs. Amsterdam: Elsevier, pp. 209-252. Tang SL, MJ Hong, MH Liao, WN Jane, PW Chiang, CB Chen,
Liao MH, SL Tang, GM Hsu, KC Wen, H Wu, WM Chen et al. CA Chen. 2011. Bacteria associated with an encru-
2007. The “black disease” of reef-building corals at Green sting sponge (Terpios hoshinota) and the corals partially
Island, Taiwan - outbreak of a cyanobacteriosponge, covered by the sponge. Environ. Microbiol. 13: 1179-
Terpios hoshinota (Suberitidae: Hadromerida). Zool. 1191.
Stud. 46: 520. Teruya T, S Nakagawa, T Koyama, T Arimoto, M Kita, D
McCook LJ, J Jompa, G Diaz-Pulido. 2001. Competition Uemura. 2004. Nakiterpiosin and nakiterpiosinone,
between corals and algae on coral reefs: a review of novel cytotoxic C-nor-D-homosteroids from the Okinawan
evidence and mechanisms. Coral Reefs 19: 400-417. sponge Terpios hoshinota. Tetrahedron 60: 6989-6993.
McLean EL, PM Yoshioka. 2007. Associations and interactions Wahle CM. 1980. Detection, pursuit, and overgrowth of
between gorgonians and sponges. In MR Custódio, G tropical gorgonians by milleporid hydrocorals: Perseus
Lôbo-Hajdu, E Hajdu, GP Muricy, eds. Porifera research: and Medusa revisited. Science 209: 689-691.
biochemistry, innovation and sustainability. Série Livros Wittle LW, RE Middlebrook, CE Lane. 1971. Isolation and
28. Rio de Janeiro: Museu Nacional, pp. 443-448. partial purification of toxin from Millepora alcicornis.
McLean EL, PM Yoshioka. 2008. Substratum effects on the Toxicon 9: 329-331.
growth and survivorship of the sponge Desmapsamma
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,*
Green Life Science Department, College of Convergence, Sangmyung Univ., 7 Hongij-dong, Jongno-gu, Seoul110-743, South Korea
Institute of Marine Biology, National Taiwan Ocean University, 2 Pei-Ning Road, Keelung 202, Taiwan
South China Sea Institute of Oceanology, Chinese Academy of Science, Guangzhou 510301, China
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- 1992, Carlsson et al. 1995). Coastal tropical
tal and estuarine ecosystems may be smaller, environments are of particular importance in this
in terms of area and volume, but the amount respect (Nittrouer et al. 1995). Increasing attention
of organic carbon exported to the deep ocean has been given to small mountainous rivers with
through the coastal fringe (1.7 × 1015 tons C/yr) drainage areas of < 10,000 km2. These smaller
can reach nearly that of the entire oceanic realm but numerous rivers could collectively be very
(13.2 × 10 15 tons C/yr) (Bienfang and Ziemann 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

Dahms et al. – Lanyang River Copepods 161

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

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

East China Sea
Ga-Yuan Bridge
Taiwan Strait

Ji-Kwan Bridge 24°

Bridge Sh-Gu-Fuh River Pacific Ocean

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 163

in elevation from 1534 to 90 m between the GS Data analysis

Bridge and Tsu-Keng (TK) River stations. The
Sung-Luo (SL) River and TK River stations are Copepod community structures were
located where these streams merge with the analyzed using the Plymouth Routine In
Lanyang River. Whereas 7 upper stations belong Multivariate Ecology Research (PRIMER)
to the freshwater zone, the 2 stations of Lanyang computer package (Version IV; Clarke and
(LY) Bridge and the estuary are influenced by Warwick 1994). In order to reduce the higher
seawater. Waters at the LY Bridge station are less heteroscedasticity observed in the abundance data
affected by seawater as they are somewhat farther for major taxa, a transformation power (λ = 0.983)
from the coast. was generated by regression coefficients, that
Stations were sampled every 2nd month were simultaneously estimated using a method of
during 10 sampling periods from June 2004 to maximizing the log-likelihood function (Box and
Dec. 2005. Tows were performed from a boat Cox 1964). Copepod abundance data were log
at the estuary station and by foot at the river (X + 1)-transformed before clustering, using the
stations. Since the channel of the Lanyang River matrix of abundances composed of samples and
is shallow and narrow, boats cannot navigate in species. Similarity coefficients between samples
the middle and upstream sections. Zooplankton were computed using the Bray-Curtis similarity
samples were obtained by towing a modified North and clustering strategy of flexible links. Three
Pacific (NORPAC) zooplankton net (with a mouth stations (Ji-Kuan (JK) ridge, Ga-Yuan (GY) Bridge,
diameter of 45 cm, a mesh size of 100 µm, and and Niour-Douh (ND) Bridge) were not considered
a length of 180 cm with a Hydrobios flow meter in the cluster analysis since they contained no
(Germany) mounted at the center of the net mouth) copepods. For correlations between abiotic
horizontally for 10 min at the surface at all stations factors and zooplankton abundances, Pearson’s
(Hwang et al. 2007). Zooplankton were preserved product moment correlation coefficients were
in a buffered 5% formalin-seawater solution for calculated with the SPSS computer package
later sorting, identification, and counting in the (Chicago, IL, USA). The Mann-Whitney U-test
laboratory. was applied to compare spatial and seasonal
Water temperature and salinity were mea- differences in surface-water temperatures. A one-
sured with a mercury thermometer and refrac- way analysis of variance (ANOVA) was applied
tometer (S-100, Tanaka Sanjiro Co.,Ltd., Japan), to reveal differences in abundances, numbers of
respectively. Water samples at 1 m in depth were species, and indices of richness, evenness, and
collected with Niskin bottles to determine Chl a diversity among sampling months. The Shannon-
concentrations using the fluorometric method Wiener diversity index (Weaver and Shannon
of Parsons et al. (1984). Other parameters like 1949) together with the richness index and Pielou’s
precipitation, nutrient and organic loadings were evenness index (Pielou 1966) were applied to
not measured but were taken from the literature estimate the copepod community composition.
(Kao and Liu 1996 1997).

Copepod enumeration and identification RESULTS

In the laboratory, zooplankton samples were Overview of the Lanyang River

subsampled with a Folsom splitter. Procedures
for species identification and counting were similar Hydrographic parameters, Chl a, and salinity
to those described by Hwang et al. (1998 2006).
Adult copepods in the subsamples were identified We recorded high surface temperatures
and counted under a stereomicroscope. Species from June to Dec. in both years 2004 and 2005.
were identified according to keys and references Stations along the upper stream showed lower
by Chen and Zhang (1965), Chen et al. (1974), Chl a (0.84 ± 0.25 μg/L) and temperature (19.4
Shih and Young (1995), and Chihara and Murano ± 4.45°C) values than the downstream stations
(1997). Freshwater copepods were identified (2.35 ± 1.90 μg/L for Chl a and 25.5 ± 4.11°C for
according to Dussart and Defaye (2006) and temperature) (Fig. 2). The estuarine station always
Einsle (1996) if not indicated otherwise. 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,
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)

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
hydrographic parameters showed maximum values
35 at the estuarine station (Fig. 2A-C).

30 Copepod abundance and diversity

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

were identified from the marine, estuarine, and
riverine portions of the Lanyang River (Table
2). The Poecilostomatoida was only recorded
at the estuarine and LY Bridge stations. At the
sampling station closest to the Lanyang River
mouth, Apocyclops borneoensis showed the
highest occurrence rate (6.67%). In contrast, 11
copepod species were recorded only once among
(C) all samples (with an occurrence ratio (OR) of
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
river (Table 2).
In terms of both diversity and abundance,
Salinity (PSU)

copepods were the most dominant component

of the zooplankton in all samplings. Among all
zooplankton, cyclopoid copepods were most
prominent, in terms of both diversity and density at
most sampling stations throughout the investigation
period. The number of calanoids was highest in
the estuary. The Poecilostomatoida was only found
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)

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
Sinocalanus sp. I 0.4 0.008 1.11
Sinocalanus tenellus (Kikuchi) 1928 I 0.68 0.014 1.11
Mongolodiaptomus birulai (Rylov) 1922 I 0.68 0.014 1.11
Subeucalanus subcrassus (Giesbrecht) 1888 I 13.14 0.272 1.11
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
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
Temora turbinata (Dana) 1849 I 93.1 ± 128.56 3.854 2.22
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
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
Euterpina acutifrons (Dana) 1847 I 289.15 5.984 1.11
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

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

1000 Ji-Kwan Bridge
Ga-Yuan Bridge
Niour-Douh Bridge
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

Richness index

(D) (E)

0.8 1.4
Evenness index

Diversity index

0.6 1.0


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)


100 2

0 0
(C) (D)
1.2 2.0

Evenness index
Richness index


0.2 0.4

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

Diversity index



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

(A) (B)
1600 30

Number of species (no. station-1)

Copepod abundance (ind. m-3)


100 5

0 0
(C) (D)
1.6 1.0
Evenness index
Richness index

1.0 0.6
0.6 0.4

0.0 0.0
Br r

Br r
Es ge

F e

Y ge

Br e
SL dge

JK ive

TK ive
LY ive

SG ridg

N ridg

G id






Sampling month
Diversity index





F e

B e
D ge

SL dge

Es ge

Br r

TK iver

Br r

LY ive

SG ridg

G idg
JK iv

N rid




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

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

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

Copepod species Group


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
Copepod abundance (ind. m-3)

Number of species

Temperature (°C)

1000 3 18

0 0 15
(C) (D)
40 7

Chlorophyll a (μg/L)
Salinity (PSU)


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 171

(Wetzel 2001). Slack waters are suggested to marine compartment may determine the dynamics
be critical for sustaining a high biomass and of the zooplankton communities in the estuary of
diversity of zooplankton (Thorp and Casper the Lanyang River (Tan et al. 2004). The estuary
2003). Where those are lacking, and the river of the Lanyang River is next to nearshore waters
channel is steep and presents a high current off the northeastern coast of Taiwan. Hwang et
velocity, the zooplankton is affected by turbulence al. (1998) found that copepods represented the
and commonly shows low biomass (Sluss et al. dominant zooplankton group along the northern
2008). This may also hold for the Lanyang River, coast of Taiwan.
particularly the 3 stations (JK Bridge, GY Bridge, River flow and tidal motions respectively
and ND Bridge) in the midstream section where we drive the riverine and marine communities towards
found no copepods in any sampling period. estuaries (Hsieh and Chiu 1997, Waniek et al.
A study by Hsu et al. (2004) further explored 2005, Zhang et al. 2010) and hence shape the
the positive relationship between observed diversity and density of estuarine communities
sediment fluxes and runoff in the Lanyang River (Waniek 2003, Froneman 2004). However, there
which may also have ultimately affected the is the possibility that resident dormant stages
zooplankton communities and caused the lack of contribute to estuarine populations as well, once
copepods at these stations. According to those they emerge from their sedimentary depositions
authors, the annual sediment discharge and (Dahms and Qian 2004, Dahms et al. 2006). The
sediment yield of the Lanyang River are 8.0 Mt biology and mesozooplankton including copepod
and 8154 Mt/km2, respectively. The annual runoff assemblages of the East China Sea and Kuroshio
is 2773 × 10 6 m 3, and transient runoff always Current are little known (Hsiao et al. 2004 2011),
rises sharply from the baseline level of several even though there was an interdisciplinary
tens of m 3 /s (with a mean rate of 62 m 3 /s) to study of the Kuroshio Current (Marr 1970) and
an abnormal level of thousands of m 3/s after a several oceanographic research programs, such
heavy rainfall (with an average annual rainfall of as KEEP (Kuroshio-East China Sea Exchange
3256 mm). Since 1949, the maximal records of Process) in the last decade by several research
daily runoff and suspended sediment concentration institutions and universities in Taiwan (Liu 1997,
are 4580 m3/s and 118 g/L, respectively (Water Hwang et al. 2006). In conclusion, copepod up-
Resources Bureau 1997a b). Sediment discharges stream assemblages were characterized by low
depend on river runoff, and a function relating the abundances and low species diversities, whereas
2 parameters was established (Kao 1996). the estuarine station showed a high abundance
According to Sluss et al. (2008), it is uncertain and a high number of species which were cor-
which abiotic factors control both the relative related with intruding seawater.
abundance of major groups and the relative size
of the zooplankton community of rivers. Biological Acknowledgments: Assistance by laboratory
factors were addressed relatively rarely, but field members of J.S. Hwang at various stages of
surveys and in situ experiments suggest that sample collection and manuscript preparation is
competition and predation play roles in regulating gratefully acknowledged. We are thankful to the
river plankton at least in slack waters (Casper and captain and crew of local ships in the Lanyang
Thorp 2007). River estuary. We acknowledge the initiative of
Our study demonstrated that marine zoo- Drs. K.T. Shao and H.J. Lin to help in getting the
plankton substantially contributed to the present research underway.
estuarine section of the Lanyang ecosystem.
Here, the highest abundance (3410.05 ind./m3)
and species number (12 species/station) cor- REFERENCES
responded with the highest salinity (37.0 psu),
demonstrating the marine role in shaping and Bienfang PK, DA Ziemann. 1992. The role of coastal high
maintaining the estuarine planktonic community. latitude ecosystems in global export production. In PG
Falkowski, AD Woodhead, eds. Primary productivity and
As mentioned, the zooplankton abundance biogeochemical cycles in the sea. New York: Plenum, pp.
of the Lanyang River estuarine station was 285-297.
significantly affected by seawater intrusions, and Bonnet D, C Frid. 2004. Seven copepod species considered
the number of zooplankton groups was affected as indicators of water mass influence and changes:
by water temperature (as affected by the seasonal results from a Northumberland coastal station. S. Afr. J.
Mar. Sci. 61: 485-491.
monsoon; see Hsieh et al. 2011). Hence, the
172 Zoological Studies 51(2): 160-174 (2012)

Box GEP, DR Cox. 1964. An analysis of transformations. J. biomass in a South African temporarily open/closed
Roy. Stat. Soc. Ser. B 26: 211-246. estuary. Estuar. Coast. Shelf Sci. 60: 125-132.
Carlsson P, E Graneli, P Tester, L Boni. 1995. Influences Ho CS. 1975. An introduction to the geology of Taiwan. Taipei,
of riverine humic substances on bacteria, protozoa, Taiwan: Ministry of Economic Affairs, 153 pp.
phytoplankton, and copepods in a coastal plankton Hsiao SH, TH Fang, C Shih, JS Hwang. 2011. Effects of the
community. Mar. Ecol. Prog. Ser. 127: 213-221. Kuroshio Current on copepod assemblages in Taiwan.
Carry AE, CA Nezat, WB Lyons, SJ Kao, DM Hicks, JS Owen. Zool. Stud. 50: 475-490.
2002. Trace metal fluxes to the ocean: the importance of Hsiao SH, CY Lee, CT Shih, JS Hwang. 2004. Calanoid
high-standing oceanic islands. Geophys. Res. Lett. 29: copepods of the Kuroshio Current east of Taiwan, with
2099. notes on the presence of Calanus jashnovi Hulseman,
Casper AF, JH Thorp. 2007. Diel and lateral patterns of 1994. Zool. Stud. 43: 323-331.
zooplankton distribution in the St. Lawrence River. River Hsieh CH, TS Chiu. 1997. Copepod abundance and species
Res. Appl. 23: 73-85. composition of Tanshui River estuary and adjacent waters.
Cearreta A, MJ Irabien, E Leorri, I Yusta, IW Croudace, Acta Zool. Taiwan. 8: 75-83.
AB Cundy. 2000. Recent anthropogenic impacts on Hsieh HY, WT Lo, LJ Wu, DC Liu, WC Su. 2011. Comparison
the Bilbao Estuary, northern Spain: geochemical and of distribution patterns of larval fish assemblages in the
microfaunal evidence. Estuar. Coast. Shelf Sci. 50: 571- Taiwan Strait between the northeasterly and southwesterly
592. monsoons. Zool. Stud. 50: 491-505.
Chang WB, LC Tseng, HU Dahms. 2010. Abundance, Hsu SC, FJ Lin, WL Jeng, YC Chung, LM Shaw, KW Hung.
distribution and community structure of planktonic 2004. Observed sediment fluxes in the southwesternmost
copepods in surface waters of a semi-enclosed Okinawa Trough enhanced by episodic events: flood
embayment of Taiwan during monsoon transition. Zool. runoff from Taiwan rivers and large earthquakes. Deep-
Stud. 49: 735-748. Sea Res. I 51: 979-997.
Chen QC. 1992. Zooplankton of China seas. Beijing: Science Hsu SC, FJ Lin, WL Jeng, TY Tang. 1998. The effect of a
Press, pp. 1-87. cyclonic eddy on the distribution of lithogenic particles in
Chen QC, SZ Zhang. 1965. The planktonic copepods of the the southern East China Sea. J. Mar. Res. 56: 813-832.
Yellow Sea and the East China Sea. I. Calanoida. Stud. Hwang JS, QC Chen, CK Wong. 1998. Taxonomic composition
Mar. Sin. 7: 20-131, 53 plates. (in Chinese, with English and grazing rate of calanoid copepods in coastal waters
summary) of northern Taiwan. Crustaceana 71: 378-389.
Chen QC, SZ Zhang, CS Zhu. 1974. On planktonic copepods Hwang JS, QC Chen, CK Wong. 2003. Taxonomic com-
of the Yellow Sea and the East China Sea. II. Cyclopoida position, density and biomass of free-living copepods in
and Harpacticoida. Stud. Mar. Sin. 9: 27-76, 24 plates. (in the coastal waters of southwestern Taiwan. Crustaceana
Chinese, with English summary) 76: 193-206.
Chihara M, M Murano. 1997. An illustrated guide to marine Hwang JS, HU Dahms, LC Tseng, QC Chen. 2007. Intrusions
plankton in Japan. Tokyo: Tokyo Univ. Press, 1574 pp. of the Kuroshio Current in the northern South China Sea
Chullasorn S, HU Dahms, KW Lee, JS Ki, N Schizas, P Kangtia affect copepod assemblages of the Luzon Strait. J. Exp.
et al. 2011. Upscaled description for Tisbe alaskensis Mar. Biol. Ecol. 352: 12-27.
sp. nov. (Crustacea, Copepoda) combining structural and Hwang JS, R Kumar, HU Dahms, LC Tseng, QC Chen. 2010.
molecular traits. Zool. Stud. 50: 103-117. Interannual, seasonal, and diurnal variation in vertical
Chullasorn S, WX Yang, HU Dahms, P Kangtia, M Holinska, and horizontal distribution patterns of 6 Oithona spp.
W Anansatitporn et al. 2009. Naupliar development of (Copepoda: Cyclopoida) in the South China Sea. Zool.
Eucyclops cf. serrulatus tropicalis, Euc. cf. spatulatus, and Stud. 49: 220-229.
Ectocyclops medius Kiefer, 1930 (Copepoda: Cyclopidae). Hwang JS, R Kumar, CW Hsieh, AY Kuo, S Souissi, MH Hsu et
Zool. Stud. 48: 12-32. al. 2010. Patterns of zooplankton distribution along the
Clarke KR, RM Warwick. 1994. Change in Marine Com- marine, estuarine and riverine portions of the Danshuei
munities: An Approach to Statistical Analysis and ecosystem in northern Taiwan. Zool. Stud. 49: 335-352.
Interpretation. Primer User Manual. (Plymouth Routines Hwang JS, R Kumar, CS Kuo. 2009a. Impact of predation
in Multivariate Ecological Research). Plymouth Marine by the copepod, Mesocyclops pehpeiensis on life table
Laboratory, Bournemouth, UK. Bourne Press Limited. demography and population dynamics of four cladoceran
Dahms HU, X Li, G Zhang, PY Qian. 2006. Resting stages of species: a comparative laboratory study. Zool. Stud. 48:
Tortanus forcipatus (Crustacea, Calanoida) in sediments 738-752.
of Victoria Harbor, Hong Kong. Estuar. Coast. Shelf Sci. Hwang JS, S Souissi, HU Dahms, LC Tseng, FG Schmitt,
67: 562-568. QC Chen. 2009b. Rank-abundance allocations as a
Dahms HU, PY Qian. 2004. Drift pump and drift net. J. Exp. tool to analyze planktonic copepod assemblages off the
Mar. Biol. Ecol. 301: 29-37. Danshuei River estuary (northern Taiwan). Zool. Stud.
Dussart BH, D Defaye. 2006. World directory of Crustacea 48: 49-62.
Copepoda of inland waters. II. Cyclopiformes. Leiden, Hwang JS, S Souissi, LC Tseng, L Seuront, FG Schmitt, LS
the Netherlands: Backhuys Publishers, 354 pp. Fang et al. 2006. A 5-year study of the influence of
Einsle E. 1996. Copepoda: Cyclopoida. Genera Cyclops, the northeast and southwest monsoons on copepod
Megacyclops, Acanthocyclops. In HJF Dumont, ed. assemblages in the boundary coastal waters between the
Guides to the identification of the microinvertebrates of East China Sea and the Taiwan Strait. J. Plankt. Res. 28:
the continental waters of the world. Vol. 10. Leiden, the 943-958.
Netherlands: Backhuys Publishers, 82 pp. Hwang JS, YY Tu, LC Tseng, LS Fang, S Souissi, TH Fang
Froneman PW. 2004. Zooplankton community structure and et al. 2004. Taxonomic composition and seasonal
Dahms et al. – Lanyang River Copepods 173

distribution of copepod assemblages from waters adjacent Freshw. Biol. 28: 217-230.
to nuclear power plant I and II in northern Taiwan. J. Mar. Schminke HK. 2007. Entomology for the copepodologist. J.
Sci. Technol. 12: 380-391. Plankt. Res. 29: 149-162.
Hwang JS, CH Wang, TY Chan, eds. 2000. Proceedings of Shiah FK, SJ Kao, KK Liu. 1996. Spatial variability of
the International Symposium on Marine Biology in Taiwan- phytoplankton production and the implications of its
Crustacean and Zooplankton Taxonomy, Ecology and controlling mechanisms in the coastal zone near the river
Living Resources. Natl. Taiwan Mus. Spec. Publ. Ser. 10: mouth of Lanyang-Hsi. Bot. Bull. Acad. Sin. 37: 9-15.
1-200. Shih CT, SS Young. 1995. A checklist of free-living copepods,
Hwang JS, CK Wong. 2005. The China Coastal Current as a including those associated with invertebrates, reported
driving force for transporting Calanus sinicus (Copepoda: from the adjacent seas of Taiwan. Acta Zool. Taiwan 6:
Calanoida) from its population centers to waters off 64-81.
Taiwan and Hong Kong during the winter NE monsoon Sluss TD, GA Cobbs, JH Thorp. 2008. Impact of turbulence on
period. J. Plankt. Res. 27: 205-210. riverine zooplankton: a mesocosm experiment. Freshw.
Jan S, J Wang, CS Chern, SY Chao. 2002. Seasonal variation Biol. 53: 1999-2010.
of the circulation in the Taiwan Strait. J. Mar. Syst. 35: Tan Y, L Huang, Q Chen, X Huang. 2004. Seasonal variation
249-268. in zooplankton composition and grazing impact on
Jerling HL, TH Wooldridge. 1995. Plankton distribution and phytoplankton standing stocks in the Pearl River Estuary,
abundance in the Sundays River estuary, South Africa, China. Cont. Shelf Res. 24: 1949-1968.
with comments on potential feeding interactions. S. Afr. J. Thor P, TG Nielson, P Tiselius, T Juul-Pederson, C Michel,
Mar. Sci. 15: 169-184. EF Møller et al. 2005. Post-spring bloom community
Kâ S, JS Hwang. 2011. Mesozooplankton distribution and structure of pelagic copepods in the Disko Bay, western
composition on the northeastern coast of Taiwan during Greenland. J. Plankt. Res. 27: 341-356.
autumn: effects of the Kuroshio Current and hydrothermal Thorp JH, AR Black, KH Haag, JD Wehr. 1994. Zooplankton
vents. Zool. Stud. 50: 155-163. assemblages in the Ohio River: seasonal, tributary, and
Kao SJ. 1996. Total organic carbon freight of a subtropical navigation dam effects. Can. J. Fish. Aquat. Sci. 51:
mountainous river (Lanyang-Hsi) in Taiwan. Acad. Bull. 1634-1643.
Chin. Naval Acad. 6: 19-70. Thorp JH, AF Casper. 2003. Importance of biotic interactions
Kao SJ, KK Liu. 1996. Particulate organic carbon export from in large rivers: an experiment with planktivorous fish,
a subtropical mountainous river (Lanyang-Hsi) in Taiwan. dreissenid mussels and zooplankton in the St. Lawrence
Limnol. Oceanogr. 41: 1749-1757. River. River Res. Appl. 19: 265-279.
Kao SJ, KK Liu. 1997. Fluxes of DOC and fossil POC Thorp JH, S Mantovani. 2005. Zooplankton in turbid and
from an Oceania small river (Lanyang-Hsi) in Taiwan. hydrologically dynamic, prairie rivers. Freshw. Biol. 50:
Biogeochemistry 39: 255-269. 1474-1491.
Kao SJ, KK Liu. 2002. Exacerbation of erosion induced by Tseng LC, HU Dahms, QC Chen, JS Hwang. 2009. Copepod
human perturbation in a typical Oceania watershed: feeding study in the upper layer of the tropical South
insight from 45 years of hydrological records from China Sea. Helg. Mar. Res. 63: 327-337.
the Lanyang-Hsi River, northeastern Taiwan. Global Tseng LC, R Kumar, HU Dahms, QC Chen, JS Hwang. 2008a.
Biogeochem. Cycles 1: 1-7. Copepod gut contents, ingestion rates and feeding impact
Lee HJ, SY Chao. 2003. A climatological description of in relation to their size structure in the southeastern
circulation in and around the East China Sea. Deep-Sea Taiwan Strait. Zool. Stud. 47: 402-416.
Res. II 50: 1065-1084. Tseng LC, R Kumar, HU Dahms, CH Wu, QC Chen, JS Hwang.
Liang WD, TY Tang, YJ Yang, MT Ko, WS Chuang. 2003. 2008b. Monsoon-driven seasonal succession of copepod
Upper-ocean currents around Taiwan. Deep-Sea Res. II assemblages in the coastal waters of the northeastern
50: 1085-1106. Taiwan Strait. Zool. Stud. 47: 46-60.
Liu KK, ed. 1997. Kuroshio edge exchange processes Vandromme P, FG Schmitt, S Souissi, EJ Buskey, JR Strickler,
collected papers. Vol. 4. Taipei, Taiwan: Joint Global CH Wu et al. 2010. Symbolic analysis of plankton
Ocean Flux Study (China-Taipei) Committee. swimming trajectories: case study of Strobilidium sp.
Liu KK, TH Peng, PT Shaw, FK Shiah. 2003. Circulation and (Protista) helical walking under various food conditions.
biogeochemical processes in the East China Sea and Zool. Stud. 49: 289-303.
the vicinity of Taiwan: an overview and a brief synthesis. Waniek JJ. 2003. The role of physical forcing in the initiation of
Deep-Sea Res. II 50: 1055-1064. spring blooms in the Northeast Atlantic. J. Mar. Syst. 39:
Marr J, ed. 1970. Symposium on the Cooperative Study of 57-82.
the Kuroshio and Adjacent Regions (1968: East-West Waniek JJ, NP Holliday, R Davidson, L Brown, SA Henson.
Center). Honolulu, HI: East-West Center Press. 2005. Freshwater control of onset and species
Nittrouer CA, GJ Brunskill, AG Figueiredo. 1995. Importance composition of Greenland shelf spring bloom. Mar. Ecol.
of tropical coastal environments. Geo-Mar. Lett. 15: 121- Progr. Ser. 288: 45-57.
126. Water Resources Bureau. 1997a. Hydrological year book of
Parsons TR, Y Maita, CM Lalli. 1984. A manual of chemical Taiwan, Republic of China, 1995. Taipei, Taiwan: Ministry
and biological methods for seawater analysis. Oxford, of Economic Affairs, 399 pp.
UK: Pergamon Press, 1-173. Water Resources Bureau. 1997b. Hydrological year book of
Pielou EC. 1966. The measurement of diversity in different Taiwan, Republic of China, 1995. Taipei, Taiwan: Ministry
types of biological collections. J. Theor. Biol. 13: 131-144. of Economic Affairs, 400 pp.
Richardson WB. 1992. Microcrustacea in flowing water: Weaver W, CE Shannon. 1949. The mathematical theory of
experimental analysis of washout times and a field test. communication. Urbana, IL: Univ. of Illinois.
174 Zoological Studies 51(2): 160-174 (2012)

Wetzel RG. 2001. Limnology: lake and river ecosystems, 3rd Zhang G, S Sun, ZL Xu, QL Zhang. 2010. Unexpected
ed. New York: Academic Press/Elsevier, 1006 pp. dominance of the subtropical copepod Temora turbinata
Wu CH, HU Dahms, EJ Buskey, JR Strickler, JS Hwang. 2010. in the temperate Changjiang river estuary and its possible
Behavioral interactions of Temora turbinata with potential causes. Zool. Stud. 49: 492-503.
ciliate prey. Zool. Stud. 49: 157-168.
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
Mohammed Reza Zargaran1,*, Nadir Erbilgin2, and Youbert Ghosta1
Plant Protection Department- Sero Road- Agricultural Faculty, Urmia Univ., PO Box 165, Urmia, Iran
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.

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

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

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

176 Zargaran et al. – Powdery Mildew Reduced Diversity of Cynipid Gall Wasps

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

species were reported for the 1st time (Sadeghi

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


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 RESULTS

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

Table 2. Conversion of common indices to true diversities

Index x Diversity in terms of x Diversity in terms of pi

Shannon entropy x ≡ - Σ p ln p
i = 1
i i exp (x) exp (- Σ p ln p )
i = 1
i i

Gini-Simpson index x ≡ 1- Σp
i = 1
i 1/(1-x) 1/ Σp
i = 1
Zoological Studies 51(2): 175-184 (2012) 179

baccarum, the later was more abundant. Several uninfected trees. Gini-Simpson indices were lower
stem gall wasps and 1 leaf gall wasp, C. cornifex, in 2010 compared to values in 2009. Among
were found equally on infected and uninfected sites, uninfected trees at Ghabre-hossein had
trees. None of the leaf gall wasps were found the highest Gini-Simpson indices in both years,
together on the same leaf in the current study. followed by Dare-ghabr and Rabat. Mirabad
Overall, oak trees with oak powdery mildew had the lowest Gini-Simpson indices. Likewise,
infection had reduced richness and diversity of Shannon’s H index indicated that uninfected trees
oak gall wasps (Table 4). The highest species had the highest oak gall wasp diversity compared
richness was found on uninfected trees in all study to infected trees at all sites (Table 4). Differences
sites in the 2 consecutive years. Gini-Simpson among sites were similar to the Gini-Simpson
indices were 1.75 in 2009 and 1.71 in 2010 for index, with Ghabre-hossein having the highest
infected and 2.33 in 2009 and 2.13 in 2010 for 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 + (-) + (-) - (-) - (-) + (-) + (-) + (-) + (-)
Species commonly found on uninfected trees, rarely found on lightly infected trees, and virtually absent from highly infected trees.
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. respectively. The mean number of oak galls was
An increase in either the Gini-Simpson index or generally higher on uninfected trees than infected
Shannon’s H index reduced the evenness of gall trees at all sites in 2 the consecutive years (Table
wasps (Table 4). The highest and lowest species 5). Trees infected with powdery mildews showed
evenness values were found at Mirabad and the lowest mean number of cynipid galls. Among
Ghabre-hossein, respectively. uninfected trees, the mean number of galls on
Cluster analysis dendrogram are shown uninfected trees ranged from14.8 at Rabat in
in figures 1 and 2. Dendrograms cluster sites 2010 to 61.2 at Ghabre-hossein in 2009, while the
according to how strongly correlated the sites are, same means for infected trees ranged from 6.2
and if sites are highly correlated, they will have a at Rabat to 25.4 at Ghabre-hossein in 2009. The
correlation value of 1 or close to 1. In the current maximum and minimum uninfected: infected ratios
study, the highest value of the Sorensen similarity were 2.8 and 2.1 in Dare-ghabr in 2009 and 2010,
between Dare-ghabr and Ghabre-hossein was 0.72 respectively.
for uninfected trees in 2009, while the similarity Pearson’s correlation coefficients between
between these 2 sites measured from infected the number of galls and percent infection showed
trees was 0.41. The lowest index of similarity significant negative correlations in 2009 (r = -0.714,
was recorded between Rabat and Dare-ghabr n = 20, p < 0.01) and 2010 (r = -0.581, n = 20,
on infected trees in 2009. In 2010, Sorensen p < 0.01) (Fig. 3). On infected oak trees, leaf gall
similarity indices of infected and uninfected trees abundances declined with increasing levels of
at Ghabre-hossein and Rabat were 0.27 and 0.40 powdery mildew.

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

Diversity indices

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

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

Rescaled Distance Chister Combine (A)
80 60 40 20 0
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
Uninfected Infected

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- infection either increases tannin contents in all
formers in 2 possible ways. First, E. alphitoides- tissues such that high tannin contents in the
infection of oak trees most likely systemically inner layers of galls might not be suitable for the
changes the host plant suitability, particularly host developing larvae, or significantly reduces tannin
nutrients and host secondary compounds, as contents such that developing larvae are not
chemical interactions between gall-formers and protected from endophytic fungi, or a combination
their host plants are important for both the success of both, as tannin content are very likely to vary
and avoidance of gall formation. For example, with the severity of pathogen infection (Bonello et
nutrients of plant tissues play critical roles in the al. 2006).
selection of oviposition sites and subsequent gall A 2nd possible alternative to explain the plant-
development (Hartley 1998, Stone and Schönrogge mediated interaction between E. alphitoides and
2003). Female cynipid gall wasps prefer host cynipid gall wasps is that the presence of a fungus
tissues with high nutritional quality (Stone et al. may prohibit oviposition by female wasps, as we
2002), and it is likely that an E. alphitoides infection observed that hyphae of E. alphitoides covered
may diminish nutritional substances in oak tissues the surface of host leaves such that females could
(Stone and Schönrogge 2003). Decreasing not lay eggs. An E. alphitoides infestation on
nitrogen or increasing carbon, due to increased leaves initially appears as light-green to yellow
metabolism of carbon-based metabolites such as spots. As the disease severity progresses, spidery
tannins and lignin (Scriber and Slansky 1981, Wold or threadlike white patches typically develop
and Marquis 1997) or increases in photosynthesis with scattered small, black fruiting bodies. The
(Bagatto et al. 1996) may alter carbon: nitrogen presence of an infestation of plant tissues by
(C:N) ratio of plant tissues. Scriber and Slansky E. alphitoides could also indirectly increase
(1981) suggested that tissues with high C: N ratios competition for suitable oviposition sites among
provide low-quality food for developing immature leaf gall wasps (Gilbert et al. 1994). However,
wasps inside galls. Additional investigations in this this avoidance mechanism might only explain the
system should focus on changes in plant nutritional reduction in leaf gall wasp diversity and species
quality due to E. alphitoides infestations to fully richness, not stem galls, as E. alphitoides is only
understand interactions between the disease and present on oak leaves.
cynipid oak gall wasps. Plant-mediated interactions between patho-
Further, our study cannot rule out the role gens and herbivorous insects were commonly
of secondary metabolites, particularly tannin reported in other systems (Krause and Raffa
levels, in reducing the abundance and richness of 1992, Felton and Korth 2000, Eyles et al. 2007).
cynipid gall wasps in diseased oak trees. Tannin For example, Krause and Raffa (1992) found that
is a phenolic compound used for defense against infection of larch Larix decidua with the fungal
a variety of organisms and is also induced by pathogen Mycosphaerella laricina induced a
pathogen infestation (Stone et al. 2002). Tannins systemic reduction in host quality for the larch
in cynipid galls are known to be concentrated in sawfly Pristiphora erichsonii. Likewise, Eyles et
the outer layers, where they may protect the gall al. (2007) reported that an infection by Diplodia
from endophytic fungi (Taper et al. 1986, Taper pinea elicited resistance against the defoliating
and Case 1987, Wilson and Carroll 1997). For European pine sawfly Neodiprion sertifer in
example, the endophytic fungus Discula quercina Austrian pine Pinus nigra. In our system, plant
(Coelomycetes) was shown to cause almost 100% defensive responses apparently seem to be
cynipid gall wasp mortality in artificial infection operating against only the gall wasp community
experiments (Wilson and Carroll 1997). Tannins and not E. alphitoides because we observed
may also protect gall-formers against parasitoids continuous colonization of the same oak trees
(Cornell 1983, Taper and Case 1987). This close by E. alphitoides after an initial infection. This
association of cynipid gall wasp and tannin levels suggests a possible “cross-talk” between defensive
could explain the observed positive relationships pathways against E. alphitoides (associated with
of oak tannin levels with cynipid diversity and salicylic acid) and cynipid gall wasps (associated
abundance (Taper and Case 1987, Wold and with jasmonic acid) in oaks (Bostock 2005, Heil
Marquis 1997, Stone et al. 2002). and Ton 2008). Even though we do not have a
Although we do not know how different complete understanding of the defensive pathways
severities of Erysiphe alphitoides infestation induced in oaks by either oak gall wasps or E.
affect tannin contents, we suspect that pathogen alphitoides, further studies of oak systems should
Zoological Studies 51(2): 175-184 (2012) 183

identify these pathways and determine whether Physiol. Mol. Plant. Pathol. 68: 95-104.
induction of 1 pathway prevents synthesis of the Bostock RM. 2005. Signal crosstalk and induced resistance:
straddling the line between cost and benefit. Annu. Rev.
other, thereby leading to an impaired capacity of Phytopathol. 43: 545-580.
a plant to respond to either pathogen infection or Braun U. 1995. The powdery mildews (Erysiphales) of Europe.
insect damage (Bostock 2005). New York: Gustav Fischer Verlag Press.
We currently do not know why the abundance Chodjai M. 1980. Study of the gall forming cynipid
and species diversity of gall wasps were higher Hymenoptera of Iranian oak forests. J. Entomol. Soc.
Iran. 3: 1-44.
at the Ghabre-hossein and Dare-ghabr sites Colgan LJ, N Erbilgin. 2011. Tree-mediated interactions
compared to the others. Despite differences in between the jack pine budworm and a mountain pine
climate, both sites have similar vegetative cover beetle fungal associate. Ecol. Entomol. (in press)
and similar species abundance and richness levels Cornell H. 1983. The secondary chemistry and com-
of cynipid gall wasps. Despite their similarities plex morphology of galls formed by the Cynipinae
(Hymenoptera): why and how? Am. Midl. Nat. 110: 225-
in climate, Ghabre-hossein and Mirabad had 236.
different vegetative cover, and the former had a Cuevas-Reyes P, M Quesada, P Hanson, R Dirzo, K Oyama.
larger species complex. This suggests that the 2004. Diversity of gall-inducing insects in a Mexican
distribution of host plant species may be highly tropical dry forest: the importance of plant species
critical for determining patterns of herbivore richness, life-forms, host plant age and plant density.
Ecology 92: 707-716.
abundances (Starzomski et al. 2008) along with Egan SP, JR Ott. 2007. Host plant quality and local adaptation
factors like climate and phenological synchrony of determine the distribution of a gall-forming herbivore.
herbivores with host plants. This is not surprising Ecology 88: 2868-2879.
considering the fact that the abundance and Eyles A, P Bonello, R Ganley, C Mohammed. 2010. Induced
richness of gall wasps are related to the richness resistance to pests and pathogens in trees. New Phytol.
185: 893-908.
and abundance of host plant species (Starzomski Eyles A, R Chorbadjian, C Wallis, R Hansen, D Cipollini, D
et al. 2008). Likewise, the species richness of oak Herms et al. 2007. Cross-induction of systemic induced
gall wasps in Mexico was highly correlated to their resistance between an insect and a fungal pathogen in
host plants (Cuevas-Reyes et al. 2004). Further, Austrian pine over a fertility gradient. Oecologia 153:
Stone et al. (2002) suggested that geographical 365-374.
Farr DF, AY Rossman. 2010. Fungal databases. Available at
differences in the oak gall wasp fauna were related http://nt.ars-grin.gov/ fungal databases Accessed 10 Apr.
to oak distribution patterns in different regions. The 2011.
current study also added further complexity to host Felton GW, KL Korth. 2000. Trade-offs between pathogen and
plant-gall wasp interactions; i.e., the role of plant herbivore resistance. Curr. Opin. Plant. Biol. 3: 309-314.
pathogens in the spatial distribution of herbivorous Foss LK, LK Rieske. 2004. Stem galls affect oak foliage with
potential consequences for herbivory. Ecol. Entomol. 29:
insects. Specifically, as indicated by the cluster 273-280.
analysis dendrogram, although the Sorensen Genimar-Reboucas J, V Eduardo-Martins, F Wilson. 2003.
similarity value for uninfected trees was fairly high Richness and abundance of gall-forming insects in the
between Dare-ghabr and Ghabre-hossein (0.72), Mamiraua Varzea, a flooded Amazonian forest. Uakari 1:
E. alphitoides infestations dramatically reduced the 39-42.
Gilbert F, C Astbury, J Bedingfield, B Ennis, S Lawson, T Sitch.
similarity index between these 2 sites to 0.41. 1994. The ecology of the pea galls of Cynips divisa. In
MAJ Williams, ed. Plants galls: organisms, interactions,
Acknowledgments: The authors are grateful to populations. Oxford, UK: Clarendon, pp. 331-349.
the head of the Agricultural and Natural Resources Griffon E, A Maublanc. 1912. Les Microsphaera des chênes.
Research Center of West Azerbaijan (Dr. R. Bull. Soc. Mycol. 28: 88-104.
Hartley SE. 1998. The chemical composition of plant galls: Are
Sokouti Oskouii) for his scientific and financial levels of nutrients and secondary compounds controlled
support. by the gall-former? Oecologia 113: 492-501.
Heil M, J Ton. 2008. Long-distance signalling in plant defence.
Plant. Sci. 13: 264-272.
REFERENCES 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
Bagatto G, LC Paquette, D Shorthouse. 1996. Influence of biota. J. Linn. Soc. 68: 87-112.
galls of Phanacis taraxaci on carbon partitioning within Jost L. 2006. Entropy and diversity. Oikos 113: 363-375.
common dandelion, Taraxacum sp. Entomol. Exp. Appl. Karban R, IT Baldwin. 1997. Induced responses to herbivory.
79: 111-117. Chicago, IL: Univ. of Chicago Press.
Bonello P, TR Gordon, DA Herms, DL Wood, N Erbilgin. 2006. Koleff P, K Gaston, J Lennon. 2003. Measuring beta diversity
Nature and ecological implications of pathogen-induced for presence/absence data. J. Anim. Ecol. 72: 367-382.
systemic resistance in conifers: a novel hypothesis.
184 Zargaran et al. – Powdery Mildew Reduced Diversity of Cynipid Gall Wasps

Kosaka H, T Aikawa, N Ogura, K Tabata, T Kiyohara. 2001. relationship between regional and local species richness:
Pine wilt disease caused by the pine wood nematode: the a test of saturation theory. J. Ecol. 89: 1921-1930.
induced resistance of pine trees by a virulent isolates of Stone GN, K Schönrogge. 2003. The adaptive significance of
nematode. Eur. J. Plant. Pathol. 107: 667-675. insect gall morphology. Trends in Ecol. Evol. 18: 512-522.
Krause SC, KF Raffa. 1992. Comparison of insect, fungal, and Stone GN, K Schönrogge, R Atkinson, D Bellido, J Pujade-
mechanically induced defoliation of larch-effects on plant Villar. 2002. The population biology of oak gall wasps
productivity, host susceptibility. Oecologia 90: 411-416. (Hymenoptera: Cynipidae). Annu. Rev. Entomol. 47: 633-
Långström B, H Solheim, C Hellqvist, P Krokene. 2001. Host 668.
resistance in defoliated pine: effects of single and mass Stout MJ, JS Thaler, BPHJ Thomma. 2006. Plant-mediated
inoculations using bark beetle-associated blue-stain fungi. interactions between pathogenic microorganisms and
Agric. For. Entomol. 3: 211-216. herbivorous arthropods. Annu. Rev. Entomol. 51: 663-
Magurran AE. 2004. Measuring biological diversity. Oxford, 689.
UK: Blackwell Publishing Press. Taper ML, TJ Case. 1987. Interactions between oak tannins
Mougou-Hamdane A, X Giresse, C Dutech, ML Desprez- and parasite community structure: Unexpected benefits of
Loustau. 2010. Spatial distribution of lineages of oak tannins to cynipid gall-wasps. Oecologia 71: 254-261.
powdery mildew fungi in France, using quick molecular Taper ML, EM Zimerman, TJ Case. 1986. Sources of
detection methods. Ann. For. Sci. 67: 449-457. mortality for a cynipid gall-wasp Dryocosmus dubiosus
Nakamura N, Y Miyamoto, T Ohgushi. 2003. Gall initiation (Hymenoptera: Cynipidae): the importance of tannin/
enhances the availability of food resources for herbivorous fungus interaction. Oecologia 68: 437-495.
insects. Funct. Ecol. 17: 851-857. Tavakoli M, G Melika, SE Sedeghi, ZS Penzes, M Assareh, R
Prior KM, JJ Hellmann. 2010. Impact of an invasive oak Atkinson et al. 2008. New species of Cynipid gall wasps
gall wasp on a native butterfly: a test of plant-mediated from Iran and Turkey (Hymenoptera: Cynipini). Zootaxa
competition. Ecology 91: 3284-3293. 1699: 1-64.
Raffa KF, SC Krause, PB Reich. 1998. Long-term effects of Tavanaei GH. 2006. Report of two fungi causing powdery
defoliation on red pine suitability to insects feeding on mildew on oak trees in forests of west Azarbaijan and
diverse plant tissues. Ecology 79: 2353-2364. Kordistan Provinces. Iran. J. For. Range Prot. Res. 4: 46-
Ronquist F, J Liljeblad. 2001. Evolution of the gall wasp-host 47.
plant association. Evolution 55: 2503-2522. Veldtman R, MA McGeoch. 2003. Gall-forming insect
Sadeghi SE, G Melika, GN Stone, M Assareh, H Askary, M species richness along a non-scleromorphic vegetation
Tavakoli et al. 2010. Oak gall wasps fauna of Iran rainfall gradient in South Africa: the importance of plant
(Hymenoptera: Cynipidae: Cynipini). 7th International community composition. Aust. Ecol. 28: 1-13.
Congress of Hymenoptroists 112, Hungary: The inter- Wallin KF, KF Raffa. 2001. Effects of folivory on subcortical
national society of Hymenopterists. plant defences: Can defence theories predict interguild
Scriber JM, F Slansky. 1981. The nutritional ecology of processes? Ecology 82: 1387-1400.
immature insects. Annu. Rev. Entomol. 26: 183-211. Wilson D, GC Carroll. 1997. Avoidance of high-endophyte
Southwood TR, PA Henderson. 2000. Ecological methods. space by gall-forming insects. Ecology 78: 2153-2163.
Oxford, UK: Blackwell. Wold EN, RJ Marquis. 1997. Induced defence in white oak:
Staley JT, A Stewart-Jones, TW Pope, DJ Wright, SR Leather, effects on herbivores and consequences for the plant.
P Hadley et al. 2010. Varying responses of insect Ecology 78: 1356-1369.
herbivores to altered plant chemistry under organic and Zargaran MR, SE Sadeghi, S Hanifeh. 2008. Identification of
conventional treatments. Proc. R. Soc. Lond. Biol. 277: oak gall wasps and their parasitoids in west Azerbaijan
779-786. Province of Iran, Urmia, Iran. Institute of Forests and
Starzomski BM, RL Parker, DS Srivastava. 2008. On the 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.

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

186 Huo et al. – Age and Growth of Oxygymnocypris stewartii

IUCN’s Red List of Threatened Species as a near- MATERIALS AND METHODS

threatened fish (Ng 2010). However, attempts
to develop effective population management In total, 712 O. stewartii individuals were
strategies have been obstructed by a lack of basic obtained from the Yarlung Tsangpo River and a
biological information. The available information tributary (Xiang Qu) monthly from Aug. 2008 to
mainly focuses on its taxonomy (Lloyd 1908, Aug. 2009 by means of floating gillnets, bottom
Cao and Deng 1962), its origin and evolution gillnets, and trap nets (Fig. 1). More than 30 fish
(Cao et al. 1981), and aspects of its phylogenetic were collected each month. The standard length
development and biogeography (Chen 1998, Chen (SL) of each fresh specimen was measured to the
2000, He and Chen 2007). There have been few nearest 1 mm using a tapeline, and body weight
studies on the otolith microstructure, age and (BW) was measured to the nearest 0.1 g with an
growth of O. stewartii (Jia and Chen 2009 2011). electronic balance. Lapillus otoliths were extracted
Accurate age determination and estimates of from each fish, washed with 95% ethanol, air-
growth and mortality parameters are fundamental dried, and then stored in labeled tubes.
requirements for understanding population Both right and left lapillus otoliths were
dynamics and provide essential data needed to removed from each individual, but usually just the
maintain sustainable yields by fisheries (Campana left otolith was used for the analysis. Otoliths were
and Thorrold 2001). Results from the only 2 mounted with the ventral face on a glass slide
studies of age and growth of O. stewartii (Jia and using thermoplastic glue, with the dorsoventral
Chen 2009 2011) validated the periodicity of otolith axis perpendicular to the slide plane. The otoliths
increment formation. However, no studies have were then ground from the dorsal face using wet
provided evidence to validate the 1st annual ring sandpaper (600-1500 grit) and polished with
or the precision of aging methods. This is often an alumina paste (3 μm) until the core was visible
overlooked but necessary component of any aging under a compound microscope. The otoliths were
study (Campana 2001). The objectives of this removed by dissolving the glue with xylene, and
study were first to describe annulus characteristics then the otoliths were re-affixed to the glass slides
of otoliths, second to validate annuli and verify using nail polish, with the polished face down. The
annual periodicity in otoliths, and finally to estimate ventral face was then ground and polished until the
the age and growth of O. stewartii. core was exposed again.

88°W 90°W

Xiang Qu River
Lhasa River

Xaitongmoin Nyemo Quxu

Yarlung Tsangpo River Xigaze Rinbung

Yarlung Tsangpo River
Nyang Qu River


0 25 50 100 km


Fig. 1. Sampling locations of Oxygymnocypris stewartii in the Yarlung Tsangpo River.

Zoological Studies 51(2): 185-194 (2012) 187

The presumed daily growth increments Lt = L∞ {1 - exp[-k (t - t0)]};

(DGIs) of otoliths of 6 young fish of 45-63 mm
SL (5 individuals caught on 4 Jan. 2009 and 1 where L ∞ is the asymptotic body length-at-age,
individual caught on 23 Feb. 2009) were counted, which represents the average body length-at-age
and the radius of the otoliths was measured along individuals would attain if they grew indefinitely, k
the posterior axis (Sequeira et al. 2009). The DGI is the growth coefficient, and t0 is the age at length
periodicity was validated to be daily (Jia and Chen 0.
2009). The growth performance index, Ø (Ø =
A marginal increment ratio (MIR) analysis was log 10 k + 2log 10 L ∞), was calculated to compare
used to verify the period of opaque zone formation the growth parameters obtained in the present
in the otoliths. Monthly variations of the MIR paper with values reported by other authors for
(1-8 annuli) were established using the following schizothoracine fishes (Munro and Pauly 1983).
equation: The index of the average percentage error
(IAPE) and coefficient of variation (CV) were
MIR = (R - Rn) / (Rn - Rn-1); calculated to measure the ageing precision
between the 2 readings. The equations (Campana
where R is the otolith radius, Rn is the distance 2001) are expressed as follows:
from the focus to the outer edge of the last annulus

Σ Σ |X X- X | ) × 100%, and
formed, and Rn-1 is the distance from the focus to 1
ij j
the outer edge of the penultimate complete annulus IAPE = N
j = 1 i = 1 j
(Haas and Recksiek 1995). Measurements were
conducted along the posterior growth axis using an
image analysis system (Leica Application Suite EZ, R

Σ (XR--X1)
N 2

ij j
Heerbrugg, Switzerland) with a direct data feed CV =
) × 100%;
N i = 1
between the dissecting microscope (Leica EZ 4D) j = 1

and a computer. Xj
Each fish was assigned to an age class where N is the number of fish aged, R is the
assuming 1 Mar. as the birth date, which appro- number of times each fish is aged, Xij is the ith age
ximately corresponds to the peak spawning determination of the jth fish, and Xj is the mean
season. A new ring mark found on the otolith of age calculated for the jth fish.
a fish captured before 1 Mar. was not considered The BW-SL relationship and von Bertalanffy
to be an annulus in the age assignment, whereas function were calculated by a non-linear regression
when a fish sampled after the assumed birth analysis (the Levenberg-Marquardt method;
date had no new ring mark, an annulus that was Levenberg 1944). The difference in the BW-SL
supposed to have formed was considered in the relationship between the sexes was compared by
age estimation (Granada et al. 2004). an analysis of covariance (ANCOVA). Deviation
Otolith readings were made along the of the allometric coefficient, b, from the theoretical
posterior growth axis. The reader had no prior value of isometric growth (b = 3) was tested by
knowledge of the length, sex, or time of capture a t-test (Pauly 1984). A residual sum of squares
before the age estimation. All ages were deter- analysis (ARSS) was used to determine whether
mined twice by the same interpreter after a any significant difference existed in the von
considerable time (3 wk). Readings were only Bertalanffy equations for males and females (Chen
accepted if both counts by the same examiner et al. 1992).
were in agreement. If the 2 readings differed, The analysis was carried out using SPSS
then the otolith was recounted, and the final count 16.0 (Chicago, IL, USA), Originpro 8.0 (Originlab,
was then accepted as the agreed age. If the Northampton, USA), Microsoft Excel 2003
3rd reading had no consensus with either of the (Redmon, WA, USA), and Photoshop CS4
previous 2 readings, the sample was discarded. Extended (Adobe, San Jose, USA). Statistical
SL-BW relationgships were calculated by the significance was accepted when p < 0.05.
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:
188 Huo et al. – Age and Growth of Oxygymnocypris stewartii

Length-frequency distributions Undetermined

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

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

n = 206); and

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

(R = 0.981, n = 133).

0.6 The allometric index value (b) obtained

MIR (%)

from the function significantly differed from 3 for

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).

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

(A) (B)

11 12
13 19

2 4 5 6

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,
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.

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
24 2 496-536 516 ± 28.3
25 1 502 502
Zoological Studies 51(2): 185-194 (2012) 191

Scizothoracinae fishes. Numbers of DGIs within counts can often be used to confirm the identity
the 1st annulus were 137-154 in Ptychobarbus of the 1st annulus (Waldron 1994, Lehodey and
dipogon (Li et al. 2009), 121-184 in O. stewartii Grandperrin 1996). DGIs can be used to estimate
(Jia and Chen 2009), and 130-168 in Schizothorax the expected radius of the 1st annulus in otoliths
o’connori (Ma et al. 2011). The daily width (Campana 2001). In this study, the mean radius
increment declined from 4.6 to 0.5 µm between of older fish approached that of the 1st annulus in
the core and the otolith margin for YOY fishes. YOY fish, indicating that the 1st annulus of lapilli
The variability in the daily width increment may was validated in O. stewartii.
be related to several factors, especially the water Mechanisms of growth increment formation
temperature (Marshall and Parker 1982, Neilson are poorly understood. There are several possible
and Geen 1982, Campana 1984). The variation in explanations for band formation; temperature,
the daily width increment within the annulus was feeding regimes, and the reproductive cycle may
consistent with that of water temperature. The be factors affecting growth increment formation
daily width increment declined near the translucent (Beckman and Wilson 1995, Morales-Nin 2000,
zone when water temperatures decreased Tserpes and Tsimenides 2001, Grandcourt et al.
indicating that O. stewartii growth became slower at 2006, Liu et al. 2009). In this study, deposition
that time (Jia and Chen 2009). Translucent zones of the opaque zone in otoliths of O. stewartii
representing slow or no growth periods in the year occurred in the spring and early summer, whereas
may be composed of many fine increments, but the hyaline zone was formed in winter months. A
these increments were too fine to be seen under a similar phenomenon for Scizothoracinae fishes
light microscope. Therefore, the above-described was also demonstrated for O. stewartii (Jia and
phenomenon could be attributed to the theoretical Chen 2009), P. dipogon (Li et al. 2009), S. waltoni
resolution limit of the light microscope (Campana (Qiu and Chen 2009), and S. o’connori (Ma et
and Neilson 1985). Perhaps, a scanning electron al. 2011). Deposition of the opaque zone occurs
microscope would be a more-accurate method. from Mar. to June (Fig. 4), corresponding to the
Identification of the 1st or innermost growth reproductive activity and water temperature
increment is an important component of any age variations, and deposition of the hyaline zone
validation study. Validation of the 1st increment occurs during winter months, when there is
is a mandatory adjunct to an age determination; reduced metabolic activity. Formation of the
without a correctly defined starting point, age opaque zone in otoliths of O. stewartii appeared to
determinations will be consistently wrong by be partially associated with reproductive activity.
a constant amount. In species with a clearly Peak reproductive activity (which occurs in Mar.,
interpretable otolith microstructure, daily increment unpubl. data) may promote a redirection of energy
to reproduction, with a consequent reduction in
700 somatic growth, which could possibly affect the
physiology of otolith growth. Moreover, formation
600 of the opaque zone in otoliths of O. stewartii may
be partially related to water temperature. The
Standard length (mm)

average water temperature around the Xigaze is
400 2°C in Feb., rising abruptly to 15°C in June (Li et
al. 2010). This abrupt rise in water temperature
300 could cause changes in metablolic activities of
Female fish and could result in opaque zone deposition.
Male Similar influences by reproducion and water
100 temperature on other fishes were reported in past
studies (Morales-Nin and Ralston 1990, Mann
0 and Buxton 1997, Pajuelo et al. 2003, Bustos
0 5 10 15 20 25 30
et al. 2009). Also, in a review of otolith studies
Age (yr)
including 94 species from 36 families, Beckman
and Wilson (1995) found that the formation of
Fig. 7. The von Bertalanffy growth curve of O. stewartii with
the observed standard length at age estimated from otoliths.
opaque and hyaline zones might be related to
The length-at-age data of the undetermined specimens were water temperatures and spawning activity. Thus,
included in models fitting both sexes (except for two 6 yr old formation of the annulus in otoliths of O. stewartii
individuals) 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 its food could contain more energy than that of
in this study were 25 yr for females and 17 yr other Schizothoracinae fishes mentioned above.
for males, which were comparable to values The von Bertalanffy growth coefficient (k)
obtained by Jia and Chen (2011) for both sexes, is a useful index for addressing the potential
indicating that females live longer than males. vulnerability of stocks to excessive mortality
Li and Chen (2009) recorded 45 yr for females (Musick 1999). Comparing the parameters of
and 24 yr for males of P. dipogon based on an some Schizothoracinae fishes, Li and Chen (2009)
interpretation of sectioned otoliths. Chen et al. suggested that they are slow-growing species
(2009) found 18 yr for females and 16 yr for males with k values of around 0.1. Slow-growing, long-
of S. younghusbandi younghusbandi by means lived fishes tend to be particularly vulnerable to
of otoliths. Yao et al. (2009) obtained 24 yr for excessive exploitation and exhibit rapid stock
females and 18 yr for males of S. o’connori using collapse, after which population turnover may
otoliths. Ma et al. (2010) also reported 50 yr be lower than expected, and their responses to
for females and 40 yr for males of S. o’connori rehabilitation measures slower than predicted
based on otolith observations. Those studies (Musick 1999). In this study, the estimated
revealed that great longevity is a common trait in maximum age was 25 yr, and the k value was
schizothoracine fishes. around 0.1, indicating that O. stewartii is a slow-
Comparing results of the growth of O. growing, long-lived species. Therefore, it is
stewartii with those of Jia and Chen (2011), the k essential to establish reasonable management
value obtained in this study was smaller (Table 2). practices for this species to allow for its sustain-
Differences among all of the estimated parameters able use. First, more scientific work such as
could be attributed to several factors: (1) different biological studies, resource investigations, and
sampled locations, (2) different age groups life history studies should be vigorously carried
employed to fit the VBGF function (the previous out in the near future to accumulate fundamental
study used 1-6 age groups), and (3) different size biological data in order to properly manage this
distributions. species; and 2nd, based on these data, new
The growth performance indices (Ø) of fishery regulations should be established, and the
O. stewartii were larger than those of other effectiveness of these regulations assessed by
schizothoracines (Table 2), indicating that the the continuous monitoring of stocks. These new
growth of O. stewartii is relatively greater than fishery regulations should focus on a sustainable
other fishes of the Schizothoracinae, which inhabit fishing intensity, a minimum catch size, and proper
the same elevatons and region. These growth fishing methods to prevent overfishing. Prohibiting
differences might be related to feeding (Jia and fishing and marketing during the spawning season
Chen 2011). Carnivorous fishes can obtain more may be 1 way to protect the older classes of O.
energy than that gained by other feeding habits 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) 193

guide local customs like “Releasing Day” to protect growth, mortality and reproduction of the blackspot
spawning populations and recruitment. snapper, Lutjanus fulviflamma (Forsskål, 1775), in the
southern Arabian Gulf. Fish. Res. 78: 203-210.
Haas RE, CW Recksiek. 1995. Age verification of winter
Acknowledgments: The authors thank the flounder in Narragansett Bay. Trans. Am. Fish. Soc. 124:
Institute of Hydrobiology, Chinese Academy of 103-111.
Sciences, Wuhan, China for providing the otolith He DK, YF Chen. 2007. The molecular phylogeny and bio-
image analysis system. 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.
REFERENCES 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
Beckman DW, CA Wilson. 1995. Seasonal timing of opaque Oxygymnocypris stewartii (Cypriniformes, Cyprinidae,
zone formation in fish otoliths. In DH Secor, JM Dean, Schizothoracinae) in the Lhasa River in Tibet, China.
SE Campana, eds. Recent developments in fish otolith Environ. Biol. Fish. 28: 45-52.
research. Columbia, SC: Univ. of South Carolina Press, Jia YT, YF Chen. 2011. Age structure and growth chara-
pp. 27-43. cteristics of the endemic fish Oxygymnocypris stewartii
Bustos R, Á Luque, JG Pajuelo. 2009. Age estimation and (Cypriniformes: Cyprinidae: Schizothoracinae) in the
growth pattern of the island grouper, Mycteroperca fusca Yarlung Tsangpo River, Tibet. Zool. Stud. 50: 69-75.
(Serranidae) in an island population on the northwest Lehodey P, R Grandperrin. 1996. Age and growth of the
coast of Africa. Sci. Mar. 73: 319-328. alfonsino Beryx splendens over the seamounts off New
Campana SE. 1984. Lunar cycles of otolith growth in the Caledonia. Mar. Biol. 125: 249-258.
juvenile starry flounder Platichthys stellatus. Mar. Biol. Levenberg K. 1944. A method for the solution of certain non-
80: 239-246. linear problems in least squares. Q. Appl. Math. 2: 164-
Campana SE. 2001. Accuracy, precision and quality control 168.
in age determination, including a review of the use and Li HJ, N Zhang, XT Lin. 2010. Spatio-temporal characteristics
abuse of age validation methods. J. Fish Biol. 59: 197- of Yarlung Zangbo River in Tibet. J. Henan Norm. Univ.
242. 38: 126-130. (in Chinese)
Campana SE, J Neilson. 1985. Microstructure of fish otoliths. Li XQ, YF Chen. 2009. Age structure, growth and mortality
Can. J. Fish Aquat. Sci. 42: 1014-1032. estimates of an endemic Ptychobarbus dipogon (Regan,
Campana SE, SR Thorrold. 2001. Otoliths, increments, and 1905) (Cyprinidae: Schizothoracinae) in the Lhasa River,
elements: keys to a comprehensive understanding of fish Tibet. Environ. Biol. Fish. 28: 97-105.
populations? Can. J. Fish Aquat. Sci. 58: 30-38. Li XQ, YF Chen, DK He, F Chen. 2009. Otolith characteristics
Cao WX, YY Chen, YF Wu, SQ Zhu. 1981. Origin and and age determination of an endemic Ptychobarbus
evolution of schizothoracine fishes in relation to the dipogon (Regan, 1905) (Cyprinidae: Schizothoracinae) in
upheaval of the Xizang Plateau. In CAS, ed. The the Yarlung Tsangpo River, Tibet. Environ. Biol. Fish. 86:
comprehensive scientific expedition to the Qinghai-Xizang 53-61.
Plateau, studies on the period, amplitude and type of the Liu KM, ML Lee, SJ Joung, YC Chang. 2009. Age and growth
uplift of the Qinghai-Xizang Plateau. Beijing: Science estimates of the sharptail mola, Masturus lanceolatus, in
Press, pp. 118-130. (in Chinese) waters of eastern Taiwan. Fish. Res. 95: 154-160.
Cao WX, ZL Deng. 1962. The schizothoracine fishes in the Lloyd RE. 1908. Report on the fish collected in Tibet by Capt
region and adjacent area of western Szechwan Province. FH and Stewart IMS. Rec. Indian Mus. 2: 341-344.
Acta Hydrobiol. Sin. 2: 27-53. (in Chinese) Ma BS, CX Xie, B Huo, XF Yang, HP Huang. 2010. Age and
Chen F, YF Chen, DK He. 2009. Age and growth of Growth of a Long-Lived Fish Schizothorax o’connori in the
Schizopygopsis younghusbandi younghusbandi in the Yarlung Tsangpo River, Tibet. Zool. Stud. 49: 749-759.
Yarlung Zangbo River in Tibet, China. Environ. Biol. Fish. Ma BS, CX Xie, B Huo, XF Yang, P Li. 2011. Age validation,
28: 155-162. and comparison of otolith, vertebra and opercular bone for
Chen Y, DA Jackson, HH Harvey. 1992. A comparison of von estimating age of Schizothorax o’connori in the Yarlung
Bertalanffy and polynomial functions in moldelling fish Tsangpo River, Tibet. Environ. Biol. Fish. 90: 159-169.
growth data. Can. J. Fish Aquat. Sci. 49: 1228-1235. Mann B, C Buxton. 1997. Age and growth of Diplodus sargus
Chen YF. 1998. Study on phylogenetic and distribution pattern capensis and D. cervinus hottentotus (Sparidae) on the
of the schizothoracine fishes. I Phylogenetic. Acta Tsitsikamma coast, South Africa. Cybium 21: 135-147.
Zootaxon. Sin. 23: 17-25. (in Chinese) Marshall SL, SS Parker. 1982. Pattern identification in the
Chen YF. 2000. Systermatics and resource biology of the microstructure of sockeye salmon (Oncorhynchus nerka)
schizothoracine fishes. PhD dissertation, Chinese otoliths. Can. J. Fish Aquat. Sci. 39: 542-547.
Academy of Science, Wuhan, China. (in Chinese) Morales-Nin B, S Ralston. 1990. Age and growth of Lutjanus
Chen YF, WX Cao. 2000. Schizothoracinae. In PQ Yue, kasmira (Forskål) in Hawaiian waters. J. Fish Biol. 36:
ed. Fauna sinica osteicthtyes cypriniformes III. Beijing: 191-203.
Scince Press, pp. 273-388. (in Chinese) Morales-Nin B. 2000. Review of the growth regulation
Granada VP, Y Masuda, T Matsuoka. 2004. Age and growth of processes of otolith daily increment formation. Fish. Res.
the yellowbelly threadfin bream Nemipterus bathybius in 46: 53-67.
Kagoshima Bay, southern Japan. Fish. Sci. 70: 497-506. Munro JL, D Pauly. 1983. A simple method for comparing the
Grandcourt EM, TZA Abdessalaam, F Francis. 2006. Age, growth of fishes and invertebrates. ICLARM Fishbyte 1:
194 Huo et al. – Age and Growth of Oxygymnocypris stewartii

5-6. Ichthyol. Res. 56: 260-265.

Musick JA. 1999. Ecology and Conservation of Long-lived Sequeira V, A Neves, AR Vieira, I Figueiredo, LS Gordo. 2009.
Marine Animals. Am. Fish. Soc. Symp. 23: 1-10. Age and growth of bluemouth, Helicolenus dactylopterus,
Neilson JD, GH Geen. 1982. Otoliths of chinook salmon from the Portuguese continental slope. ICES J. Mar. Sci.
(Oncorhynchus tshawytscha): daily growth increments 66: 524-531.
and factors influencing their production. Can. J. Fish Tserpes G, N Tsimenides. 2001. Age, growth and mortality of
Aquat. Sci. 39: 1340-1347. Serranus cabrilla (Linnaeus, 1758) on the Cretan shelf.
Ng HH. 2010. Oxygymnocypris stewartii. IUCN Publishing Fish. Res. 51: 27-34.
Red List of Threatened Species. Available at http://www. von Bertalanffy L. 1938. A quantitative theory of organic
iucnredlist.org/apps/redlist/details/168408/0 Accessed growth (inquiries on growth laws II). Hum. Biol. 10: 181-
Apr. 2010. 213.
Pajuelo JG, JM Lorenzo, R Domínguez-Seoane. 2003. Age Waldron ME. 1994. Validation of annuli of the South African
estimation and growth of the zebra seabream Diplodus anchovy, Engraulis capensis, using daily otolith growth
cervinus cervinus (Lowe, 1838) on the Canary Islands increments. ICES J. Mar. Sci. 51: 233-234.
shelf (central-East Atlantic). Fish. Res. 62: 97-103. Wu YF, CZ Wu, eds. 1992. The fishes of the Qinghai-Xizang
Pauly D. 1984. Fish population dynamics in tropical waters: a Plateau. Chengdu, China: Sichuan Publingshing House
manual for use with programmable calculators. Manila, of Science and Technology. (in Chinese)
the Philippines: International Center for Living Aquatic Yao JL, YF Chen, F Chen, DK He. 2009. Age and growth of
Resources Management. an endemic Tibetan fish, Schizothorax o’connori, in the
Qiu H, YF Chen. 2009. Age and growth of Schizothorax Yarlung Tsangpo River. J. Freshw. Ecol. 24: 343-345.
waltoni in the Yarlung Tsangpo River in Tibet, China.
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,*
Departamento de Biologia, Univ. Estadual do Centro-Oeste, Rua Presidente Zacarias 875, CEP: 85010-990, Guarapuava (PR), Brasil
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.

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

196 Gonçalves et al. – Pollen Sources of Centris tarsata

1983, Michener 2000). 2006). In southern Brazil, C. tarsata is the most

Some bee species belonging to the tribes abundant bee species (Buschini 2006). It prefers
Tapnotaspidini and Centridini exhibit reproductive open habitats and shows greater nesting activity
cycles and nesting activities that are synchronized during the hot season, especially in Dec. and Jan.
with the flowering periods of certain species (Buschini and Wolff 2006).
of plants (Rocha-Filho et al. 2008, Aguiar and Several studies were conducted in Brazil
Melo 2009, Bezerra et al. 2009, Gaglianone et to identify sources of pollen used by different
al. 2011). These bees visit flowers to obtain oil, species of bees and to understand the degree
pollen, nectar, and resin (resources needed to of association between bees and the plants that
build parts of their nests) to feed the larvae and they visit. Through an analysis of pollen grains,
maintain adults and their reproductive activities it is possible to identify the main floral resources
(Vogel 1974, Buchmann 1987, Roubik 1989, used by bees. This information allows the asse-
Vinson et al. 1996). Some studies showed the ssment of resource availability in the field and
importance of these bees as pollinators of various the identification of times of resource scarcity
species of Neotropical plants (Frankie et al. (Salgado-Labouriau 1961, Ortiz 1994, Bastos et al.
1976, Gottsberger et al. 1988, Freitas 1997), 2003).
including those producing oil, such as species of An analysis of the pollen spectrum of C.
the Malpighiaceae (Rêgo and Albuquerque 1989, tarsata based on samples from nests in the
Freitas et al. 1999) and Scrophulariaceae (Vogel northeastern micro-region of Bahia State, Brazil
and Machado 1991). indicated the presence of 17 pollen types from 7
The genus Centris is typically tropical, and plant families. These samples, representing an
its species belong to 12 subgenera. The species assemblage of 5-11 pollen types, identified plants
richness of Centris is high in South America. In used by the bees to feed their offspring (Dórea et
Brazil, these bees are found in various eco- al. 2009). In Maranhão State, also in northeastern
systems, such as dunes and sandbanks (Silva Brazil, pollen analyses of C. tarsata showed
and Martins 1999, Silva et al. 2001, Viana and relatively high quantities of pollen grains from
Alves-dos-Santos 2002), caatinga (Martins 1994, Banisteriopsis sp. (Malpighiaceae) and Cassia sp.
Zanella 2000, Aguiar and Almeida 2002, Aguiar et (Caesalpiniaceae).
al. 2003), grasslands, and savannas (Silveira and Centris tarsata is endemic to Brazil. No
Campos 1995, Albuquerque and Mendonça 1996). previous studies of the interactions of plants with
Centris tarsata has only been recorded from this bee species have been conducted in southern
Brazil. The distribution of C. tarsata in Brazil is Brazil, where it is the most abundant trap-nesting
based on Aguiar and Garófalo (2004), information bee. The goals of this study were to investigate
from specimens deposited in entomological the plants that constitute the larval food supply for
collections (J.M.F. Camargo, pers. commun.), C. tarsata and determine whether this bee has a
samples of females and/or males collected on polylectic or an oligolectic tendency in this region.
flowers (Camargo and Mazucato 1984, Vogel
and Machado 1991, Martins 1994, Silveira and
Campos 1995, Albuquerque and Mendonça 1996, MATERIALS AND METHODS
Freitas 1997, Schlindwein 1998, Zanella 2000),
and the location of nests (Chandler et al. 1985, This study was carried out in the Parque
Camilo et al. 1995, Silva et al. 2001, Viana et al. Municipal das Araucárias, located in the
2001, Aguiar and Martins 2002). This information municipality of Guarapuava, Paraná State,
indicates that C. tarsata occurs in the states of PA, southern Brazil (25°21'06"S, 51°28'08"W). The
MA PI, CE, PB, PE, BA, MG, SP, PR, RS, MS, MT, area of the park is approximately 104 ha. The
and GO. vegetation is composed of mixed ombrophilous
In the savanna area of Uberlândia (Minas forest (42.75%), gallery forest (10.09%), fields
Gerais State, Brazil), C. tarsata was recorded (6.8%), swamps (7.13%), and altered areas
as one of the principal pollinators of West Indian (33.23%). The grasslands are physionomically
cherry Malpighia emarginata DC (Malpighiaceae) characterized by areas of low grasses and no
(Vilhena and Augusto 2007). This bee is solitary bushes. Species of the families Cyperaceae,
and tends to nest in preexisting cavities. Its Leguminosae, Verbenaceae, Compositae, and
nests can be built in trap-nests (Silva et al. 2001, Umbelliferae are the principal plants in this habitat.
Aguiar and Garófalo 2004, Buschini and Wolff The grasslands are surrounded by Araucaria
Zoological Studies 51(2): 195-203 (2012) 197

forests, dominated by Araucaria angustifolia Phytolaccaceae. Phytollaca dioica L. supplied

(Coniferae: Araucariaceae). The swamps are 15.4% of the pollen in the samples. The family
located in the lowest-elevation regions of the Malpighiaceae represented 4.5% of the pollen
park and are primarily composed of grasses and in the samples, whereas the families Lauraceae
members of the Compositae (Buschini and Fajardo (3.2%), Myrthaceae (2.8%), Melastomataceae
2010). According to the Köppen classification, the (1.01%), Lythraceae (0.9%), Campanulaceae
climate is humid mesothermic, with no dry season (0.4%), Convolvulaceae (0.2%), Caesalpiniaceae
and mild summers because of the elevation. The (0.16%), Asteraceae (0.1%), Amaranthaceae
winter is moderate, with the frequent occurrence (0.08%), and Polygalaceae (0.07%) occurred at
of frost. The annual mean temperature is appro- low percentages. Erythroxylum deciduum A. St.
ximately 16°C. Hil. (0.03%), of the family Erythroxilaceae, and
In this study, pollen grains were removed another species not yet identified (Undetermined-1)
from provisioned cells of 11 nests of C. tarsata (0.01%) appeared in more than 1 sample but at
from a total of 128 trap-nests installed in the low occurrence percentages. Although the pollen
swamp and grassland areas. Centris tarsata has types of Styrax leprosum Hook and Am. (0.09%)
a seasonal pattern of nesting activity from Nov. and another unidentified species (Undetermined-2)
to May (Buschini and Wolff 2006), so the pollen (0.04%) were recorded in only 1 sample, their
grains used in this study were collected from Mar. percentages were higher than those of Ery.
2002 to Dec. 2003. Pollen collected from the deciduum and Undetermined-1.
nests was preserved on permanent slides with the The frequencies of occurrence of pollen
acetolysis method (Erdtman 1960). Five pollen types in the 11 samples analyzed showed that Sol.
grain slides were made for each nest to produce a americanum and Sol. variabile (100%) were the
total of 55 slides. Pollen grains were also collected most consistent, followed by Janusia guaranitica
from flowering plants from May 2006 to Apr. 2007. and Cinnamomum amoenum (Ness and Mart.)
Pollen was collected throughout the area in which Kosterm (60%), Gomphrena elegans Mart., and
nests were built. The pollen was removed from Ipomoea grandifolia Lam. (40%). The 14 other
the flowers and/or buttons of each plant to obtain pollen types occurred in 10%-30% of samples:
2 slides per plant. All pollen grain slides from Phytollaca dioica, Vernonia sp. Schreb, Senna
both nests and plants were examined using light multijuga (Rich.) H. S., Cuphea sp. P. Browne
microscopy to identify plants used by the bee. The (30%), Ery. deciduum, Janusia sp. A. Juss,
pollen was quantified by consecutively counting Tibouchina cerastifolia Cong, and Undetermined-1
300 pollen grains per slide. Total numbers of (20%), and Baccharis sp. L., Lobelia sp. Pohl,
pollen grains counted were 1500 grains per Ipomoea purpurea (l.) Roth, Campomanesia
nest and 16,500 grains in all. Subsequently, we adamantium O. Berg, Polygala sp. L., Sty.
determined the percentages of occurrence of each leprosum Hook. and Arn, and Undetermined-2
species and botanical family in C. tarsata nests (10%).
according to the classification of Barth (1970) and
Louveaux et al. (1970 1978). Thus, pollen types
were classified as dominant (> 45% of total grain DISCUSSION
on the slides), accessory (15%-45%), important
isolates (3%-14%), and occasional isolates (< 3%). Although C. tarsata used 20 types of pollen
grains, pollen of Sol. americanum, Sol. variabile,
and Phy. dioica were most common in the larval
RESULTS diet. The importance of the family Solanaceae as
a source of pollen for C. tarsata was also reported
We collected 99 flowering plant species in the by Aguiar et al. (2003) and Dórea et al. (2009) in
study area during the activity period of C. tarsata. the Caatinga, xerophytic vegetation predominant
Overall, 20 pollen types from 17 plant families in semi-arid northeastern Brazil. According to
were collected by this bee (Fig. 1, Table 1). Buchmann (1983), the presence of poricidal
The family Solanaceae was visited most anthers in flowers of the Solanaceae establishes
often (71%). Solanum americanum Mill. (28.6%) a close relationship with females of Centris. In
and Sol. variabile Mart. (42.4%) were the primary this plant-pollinator relationship, pollination by
pollen sources for C. tarsata in the study area. vibration (buzz-pollination) is an effective method
The 2nd most frequently visited family was the 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 Rego 2007) as sources of pollen for C. tarsata.
also occurs on flowers of the Melastomataceae Aguiar et al. (2003), in Itatim (BA), northeastern
(Buchmann and Hurley 1978) and Caesalpiniaceae Brazil, recorded the presence of pollen of the
(Moure and Castro 2001). This method of polli- Caesalpiniaceae in the diet of C. tarsata offspring.
nation is associated with small pollen grains, as Senna Mill. was also found to represent a frequent
in Solanum species. These grains have a high source of pollen and nectar for this bee. Plants of
amount of protein, which is important for larval this genus are also associated with a mechanism
development (Roulston et al. 2000). of pollination by vibration (Santos et al. 2004,
Studies in different Brazilian biomes also Anacleto and Marchini 2005, Andena et al. 2005).
highlighted the importance of the families Moreover, Aguiar et al. (2003) stated that solitary
Solanaceae (Dórea et al. 2009), Malpighiaceae, bees, such as species of Centris, are more likely to
Caesalpiniaceae, and Myrtaceae (Mendes and 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

Gomphrena elegans Mart. - Herb Swamp Mar. 0.08% POI
Baccharis sp. L. Pollen, nectar Shrub Grassland Feb., Mar. 0.04% POI
Vernonia sp. Schreb. Pollen, nectar Tree Forest Oct. 0.06% POI
Lobelia sp. Pohl. Herb Swamp Feb. 0.4% POI
Ipomoea grandifolia Lam. Pollen Liana Swamp Mar. 0.13% POI
Ipomoea purpúrea (l.) Roth. Pollen Liana Swamp Mar. 0.07% POI
Erythroxylum deciduum A. St. Hil. Pollen, nectar Shrub Grassland Sept. 0.03% POI
Senna multijuga (Rich.) H. S. Pollen, nectar Tree Grassland Feb. 0.16% POI
Cinnamomum amoenum (Nees) Kosterm. Pollen Tree Forest Oct. 3.2% PII
Cuphea sp. P. Browne. Pollen, nectar Herb Swamp Apr. 0.9% POI
Janusia guaranítica (A. St.-Hil.) A. Juss. Pollen, oil Herb Grassland Dec. 3.2% PII
Janusia sp. A. Juss. Pollen, oil - - - 1.3% POI
Tibouchina cerastifolia Cong. Pollen, oil Herb Grassland Jan., Feb. 1.01% POI
Campomanesia adamantium O. Berg. Pollen Tree Grassland Oct. 2.8% POI
Phytolacca dioica L. Pollen Tree Grassland, forest Oct. 15.4% PA
Polygala sp. L. Pollen, nectar - - - 0.07% POI
Solanum americanum Mill. Pollen Herb Grassland, Mar. 28.6% PA
Solanum variabile Mart. Pollen Tree Grassland Nov. 42.4% PA
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 those found for Sol. americanum, Sol. variabile,
also stated that the exploitation of resources from and Phy. dioica. According to Severson and
the families Caesalpiniaceae and Malpighiaceae is Parry (1981), measurements of a pollen sample
frequently found in different biomes. should be representative of the mass of pollen by
In Salinas (MG), southeastern Brazil, including the average number of grains counted
Guimarães (2006) found that the family Myrtaceae and should also reflect the estimated volumetric
was visited by several species of Centris. Similar contribution of the grain type. Thus, the degree
results were obtained in the São Francisco Valley of importance of 1 type of pollen grain should
of Brazil by Siqueira et al. (2005), who reported a not be based solely on its percentage but should
high frequency of Centris and Xylocopa visitation also include both its numeric and volumetric
to flowers of this family. Pollen is the primary representation in the sample.
resource provided by this family for bees, which The sporadic presence of pollen of the
are probably its most efficient pollinators (Gressler Melastomataceae in nests of C. tarsata in
et al. 2006). In these plants, pollination also Guarapuava may reflect the tendency of the bees
occurs by vibration, although the anthers exhibit to seek the oil of these plants to build their nests.
longitudinal dehiscence and are not poricidal When collecting the oil, they place their ventral
(Proença 1992). abdomen and thorax on the stigma and anthers of
Although the percentage of pollen from plants the flowers. This behavior facilitates the transfer
of the family Malpighiaceae in the diet of C. tarsata of pollen to the stigma (Gimenes and Lobão
in Guarapuava was low (4.5%), this finding does 2006) and also results in the transport of small
not mean that these plants have little importance amounts of pollen from the plants to the bees’
as resource suppliers for these bees. According to nests. In studies in Camaçari (BA), northeastern
Anderson (1979), Vogel (1990), and Ramalho and Brazil, Oliveira-Rebouças and Gimenes (2004)
Silva (2002), a close relationship between bees observed that medium- and large-sized species
of the tribe Centridini and plants of this family can of Centris were highly efficient at collecting pollen
be interpreted as a product of a long evolutionary from flowers of the Melastomataceae. In the study
history of interactions between the 2 groups. This region, the use of pollen of the Convolvulaceae
history could even explain the high reproductive (e.g., Ipomoea) by Centris may be related to the
success of these plants in the Americas. The morphology of the pollen grains. These grains
plants provide both oil and pollen to feed the larvae are large-sized, are porate and colpate with a
of these bees. They bloom almost year-round, but perforated exine, and are spiculated and hairy.
the flowers are more highly abundant during the The spine characteristic of this genus assists in
warm and rainy period (Silberbauer-Gottsberger the attachment of pollen grains to the hair of bees,
and Gottsberger 1988). In the Brazilian savanna thereby optimizing the transport process (Machado
(i.e., the cerrado), the nesting and foraging and Melhem 1987, Sengupta 1972, Leite et al.
activities of the Centridini are generally more 2005).
frequent during the period of peak flowering of The occurrence of pollen from the
the Malpighiaceae (Rocha-Filho et al. 2008). The Phytolaccaceae, Lauraceae, and Styracaceae
Centridini is considered to be key pollinators of in the diet of C. tarsata may be related to the
this plant family (Michener 2000, Machado 2002 bees’ search for resources in plants located in
2004, Machado et al. 2002, Alves dos Santos et al. transitional areas between the grassland and
2007). The system of oil production in these plants Araucaria forest. These areas are close to sites
and collection of the oil by the bees require a where the bees nest. Frankie et al. (1983) also
series of morphological adaptations in both groups observed many species of Centris foraging in the
and behavioral adaptations by the bees (Simpson canopy of mass-flowering tree species.
and Neff 1977). The oil, the primary resource Although C. tarsata was found to visit 20
that attracts the bees to the plants, is secreted by species of plants, it preferred Solanum species
glands called elaiophores (Vogel 1974, Simpson with poricidal anthers and pollen grains with high
and Neff 1981) and is included in the diet of larval amounts of protein. This selective behavior by
bees. females of C. tarsata indicates that this bee is
In the Malpighiaceae, pollen grain sizes oligolectic in its larval provisioning in this region
usually range from medium to large. Pollen of of southern Brazil. Because C. tarsata occurs in
Janusia occurred in small quantities in bee nests, areas of natural grasslands and collects pollen from
but these quantities were considerably higher than plants in transition zones between these areas and
Zoological Studies 51(2): 195-203 (2012) 201

Araucaria forest, these bees undoubtedly promote Bezerra ES, AV Lopes, IC Machado. 2009. Biologia
the pollination of various plant species in these reprodutiva de Byrsonima gardnerana A. Juss.
(Malpighiaceae) e interações com abelhas Centris
areas that are currently suffering from severe (Centridini) no Nordeste do Brasil. Rev. Bras. Bot. 32: 95-
exploitation and fragmentation in southern Brazil. 108.
Further studies should be conducted to investigate Boucher DH, S James, KH Keeler. 1982. The ecology of
the ability of these bees to explore different mutualism. Ann. Rev. Ecol. Syst. 13: 315-347.
grassland fragments in this region and transport Buchmann SL. 1983. Buzz pollination in angiosperms. In
CE Jones, RJ Little, eds. Handbook of experimental
pollen grains between them, thereby increasing pollination biology. New York: Scientific and Academic
the genetic variability of the region’s plants. Editions, pp. 73-113.
Buchmann SL. 1987. The ecology of oil flowers and their
Acknowledgments: Partial support was provided bees. Ann. Rev. Ecol. Syst. 18: 343-369.
by Fundação Araucária (The State of Paraná Buchmann SL, JP Hurley. 1978. A biophysical model for buzz
pollination in angiosperms. J. Theor. Biol. 72: 639-657.
Research Foundation) and UNICENTRO (Univ. Buschini MLT. 2006. Species diversity and community
Estadual do Centro-Oeste). We thank Prof. Dr. J. structure in trap-nesting bees in southern Brazil.
Cordeiro for plant identification. 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-
REFERENCES Stockh. 91: 426-432.
Buschini MLT, LL Wolff. 2006. Biologia de nidificação de
Aguiar AJC, CF Martins. 2002. Abelhas e vespas solitárias Centris (Hemisiella) tarsata Smith no Sul do Brasil
em ninhos-armadilha na Reserva Biológica Guaribas (Hymenoptera, Apidae, Centridini). Braz. J. Biol. 66:
(Mamanguape, Paraíba, Brasil). Rev. Bras. Zool. 19: 1091-1101.
101-116. Camargo JMF, M Mazucato. 1984. Inventário da apifauna e
Aguiar AJC, GAR Melo. 2009. Notes on oil sources for flora apícola de Ribeirão Preto, SP - Brasil. Dusenia 14:
the bee genus Caenonomada (Hymenoptera, Apidae, 55-87.
Tapinotaspidini). Rev. Bras. Entomol. 53: 154-156. Camillo E, CA Garófalo, JC Serrano, G Muccillo. 1995. Diver-
Aguiar CML, GF Almeida. 2002. Atividade diária de coleta de sidade e abundância sazonal de abelhas e vespas
óleo por espécies de Centris (Hymenoptera: Apidae) em solitárias em ninhos-armadilha (Hymenoptera, Apocrita,
Mcvaughia bahiana W.R. Anderson (Malpighiaceae) na Aculeata). Rev. Bras. Entomol. 39: 459-470.
caatinga. Acta Biol. Leop. 24: 131-140. Chandler L, JAF Barrigosi, EBS Diaz. 1985. The first defi-
Aguiar CML, CA Garófalo. 2004. Nesting biology of Centris nitive host record for Leucospis cayennensis Westwood
(Hemisiella) tarsata Smith (Hymenoptera, Apidae, (Hymenoptera: Leucospidae). Rev. Ceres 32: 170-174.
Centridini). Rev. Bras. Zool. 21: 477-486. Del-Claro K. 2004. Multitrophic relationships, conditional
Aguiar CML, FCV Zanella, CF Martins, CAL Carvalho. 2003. mutualisms, and the study of interaction biodiversity in
Plantas visitadas por Centris spp. (Hymenoptera: Apidae) tropical savannas. Neotrop. Entomol. 33: 665-672.
na caatinga para obtenção de recursos florais. Neotrop. Dórea MC, FAR Santos, LCL Lima, LER Figueroa. 2009.
Entomol. 32: 247-259. Análise Polínica do Resíduo Pós-Emergência de
Albuquerque PMC, JAC Mendonça. 1996. Anthophoridae Ninhos de Centris tarsata Smith (Hymenoptera: Apidae,
(Hymenoptera, Apoidea) e Flora associada em uma Centridini). Neotrop. Entomol. 38: 197-202.
formação de cerrado no município de Barreirinhas, Ma, Erdtman G. 1960. The acetolized method. A revised
Brasil. Acta Amazôn. 26: 45-54. description. Svensk Bot. Tidskrift 54: 561-564.
Alves-dos-Santos I, IC Machado, MC Gaglione. 2007. História Faegri K, L Van der Pijl. 1979. The principles of pollination
natural das abelhas coletoras de óleo. Oecol. Bras. 11: ecology. 3rd ed. Oxford: Pergamon Press, 244 pp.
242-255. Frankie GW, PA Opler, KS Bawa. 1976. Foraging behaviour of
Anacleto DA, LC Marchini. 2005. Análise faunística de abelhas solitary bees: implications for outcrossing of a Neotropical
(Hymenoptera, Apoidea) coletadas cerrado do Estado de forest tree species. J. Ecol. 64: 1049-1057.
São Paulo. Acta Sci. Biol. Sci. 27: 277-284. Frankie GW, SB Vinson, PA Opler. 1983. Centris aethyctera
Andena SR, LR Bego, MR Mechi. 2005. A Comunidade de and Centris fuscata in the lowland dry forest of Costa Rica
abelhas (Hymenoptera, Apoidea) de uma área de cerrado (Hymenoptera: Anthophoridae). In DH Janzen, ed. Costa
(Corumbataí, SP) e suas visitas às flores. Rev. Bras. Rica Natural History. Chicago, Univ. Press, pp. 707-708.
Zoociênc. 7: 55-91. Freitas BM. 1997. Number and distribution of cashew
Anderson WR. 1979. Floral conservatism in Neotropical (Anacardium occidentale) pollen grains on the bodies
Malpighiaceae. Biotropica 11: 219-223. of its pollinators, Apis mellifera and Centris tarsata. J.
Barth OM. 1970. Análise microscópica de algumas amostras Apicult. Res. 36: 15-22.
de mel. Pólen dominante. Anais Acad. Bras. Ciênc. 42: Freitas BM, JE Alves, GF Brandão, ZB Araujo. 1999.
351-366. Pollination requirements of West Indian cherry (Malpighia
Bastos EMAF, VM Silveira, AEE Soares. 2003. Pollen emarginata) and its putative pollinators, Centris bees. J.
spectrum of honey produced in cerrado areas of Minas Agric. Sci. 133: 303-311.
Gerais State (Brazil). Braz. J. Biol. 63: 599-615. Gaglianone MC, AJC Aguiar, F Vivallo, I Alves-dos-Santos.
Bawa KS. 1990. Plant-pollinator interactions in tropical rain 2011. Checklist das Abelhas Coletoras de Óleos do
forests. Ann. Rev. Ecol. Syst. 21: 399-422.
202 Gonçalves et al. – Pollen Sources of Centris tarsata

Estado de São Paulo, Brasil. Biot. Neotrop., 11: 1-10. pollination. London: Harper Collins Publishers, 496 pp.
(Online Edição em Inglês) Proença CEB. 1992. Buzz pollination older and more
Gimenes M, CS Lobão. 2006. A polinização de Krameria widespread than we think? J. Trop. Ecol. 8: 115-120.
bahiana B.B. Simpson (Krameriaceae) por abelhas Ramalho M, M Silva. 2002. Flora oleífera e sua guilda de
(Apidae) na restinga, BA. Neotrop. Entomol. 35: 440-445. abelhas em uma comunidade de restinga tropical. Rev.
Gottsberger G, JMF Camargo, I Silberbauer-Gottsberger. Univ. Estadual Feira Santana Bras. 2: 34-43.
1988. A bee-pollinated tropical community: the beach Rêgo MMC, PMC Albuquerque. 1989. Comportamento das
dune vegetation of Ilha de São Luís, Maranhão. Braz. abelhas visitantes de murici, Byrsonima crassifolia (L.).
Bot. Jahrb. Syst. 109: 469-500. Kunth, Malpighiaceae. Bol. Mus. Para. Emílio Goeldi.
Gressler E, MA Pizo, L Patrícia, C Morellato. 2006. Polinização Sér. Zool. 5: 179-193.
e dispersão de sementes em Myrtaceae do Brasil. Rev. Rocha-Filho LC, CI Silva, MC Gaglianone, SC Augusto. 2008.
Bras. Bot. 29: 509-530. Nesting behavior and natural enemies of Epicharis
Guimarães RA. 2006. Abelhas (Hymenoptera: Apoidea) (Epicharis) bicolor Smith, 1854 (Hymenoptera: Apidae).
visitantes das flores da goiaba (Psidium guajava L.), Trop. Zool. 21: 227-242.
laranja (Citrus sinensis L.) e tangerina (Citrus reticulata Roubik DW. 1989. Ecology and natural history of tropical
B.) em pomares comerciais em Salinas - MG. Vitória da bees. Cambridge Tropical Biology Series. New York:
Conquista, BA: Univ. Estadual do Sudoeste da Bahia. Cambridge Univ. Press, 514 pp.
Kevan PG, HG Baker. 1983. Insects as flower visitors and Roubik DW. 1993. Tropical pollinators in the canopy and
pollinators. Ann. Rev. Entomol. 28: 407-453. understory - field data and theory for stratum preferences.
Leite KRB, R Simão-Bianchini, FAR Santos. 2005. Morfologia J. Insect Behav. 6: 659-673.
polínica do gênero Merremia Dennst. (Convolvulaceae) Roulston TH, JH Cane, SL Buchmann. 2000. What governs
ocorrentes no Estado da Bahia, Brasil. Acta Bot. Bras. protein content of pollen: pollinator preferences, pollen-
19: 313-321. pistil interactions, or phylogeny? Ecol. Monogr. 70: 617-
Louveaux J, A Maurizio, G Vorwohl. 1970. Methods of 643.
melissopalynology. Bee World 51: 25-138. Salgado-Labouriau ML. 1961. Pollen grains of plants of the
Louveaux J, A Maurizio, G Vorwohl. 1978. Methods of “Cerrado” I. Ann. Acad. Bras. Cienc. 33: 119-130.
melissopalynology. Bee World 59: 139-157. Santos FM, CAL Carvalho, RF Silva. 2004. Diversidade
Machado IC. 2002. Flores de óleo e abelhas coletoras de óleo de abelhas (Hymenoptera: Apoidea) em uma área de
floral: conceitos, distribuição e revisão. Ribeirão Preto, transição Cerrado-Amazônia. Acta Amazon. 34: 319-328.
SP, BR.: Anais do V Encontro Sobre Abelhas, pp. 129- Schlindwein C. 1998. Frequent oligolecty characterizing a
135. diverse bee-plant community in a xerophytic bushland of
Machado IC. 2004. Oil-collecting bees and related plants: a subtropical Brazil. Stud. Neotrop. Fauna E 33: 46-59.
review of the studies in the last twenty years and case Sengupta S. 1972. On the pollen morphology of Convol-
histories of plants occurring in NE Brazil. In BM Freitas, vulaceae with special reference to taxonomy. Rev.
JOP Pereira, eds. Solitary bees. Conservation, rearing Palaeobot. Palynol. 13: 157-212.
and management for pollination. Fortaleza, CE, BR: Severson DW, JE Parry. 1981. A chronology of pollen
Imprensa Univ., pp. 255-281. collection by honeybees. J. Apic. Res. 20: 97-103.
Machado IC, TS Melhem. 1987. Morfologia polínica de Silberbauer-Gottsberger I, G Gottsberger. 1988. A polinização
Ipomoea hederifolia L. e I. quamoclit L. (Convolvulaceae). de plantas do cerrado. Rev. Bras. Biol. 48: 651-663.
Hoehnea 14: 25-30. Silva FO, BF Viana, EL Neves. 2001. Biologia e Arquitetura
Machado IC, S Vogel, AV Lopes. 2002. Pollination of de ninhos de Centris (Hemisiella) tarsata Smith
Angelonia cornigera Hook. (Scrophulariaceae) by long- (Hymenoptera: Apidae: Centridini). Neotrop. Entomol. 30:
legged oil-collecting bees in NE Brazil. Plant. Biol. 4: 352- 541-545.
359. Silva MCM, C Martins. 1999. Flora apícola e relações tróficas
Martins CF. 1994. Comunidade de abelhas (Hym., Apoidea) de abelhas (Hymenoptera: Apoidea) em uma área
da caatinga e do cerrado com elementos de campo de duna (praia de Interlagos, Cabedelo - PB, Brasil).
rupestre do Estado da Bahia, Brasil. Rev. Nord. Biol. 9: Principia 7: 40-51.
225-257. Silveira FA, MJO Campos. 1995. A mellissofauna de
Mendes FN, MMC Rêgo. 2007. Nidificação de Centris Corumbataí (SP) e Paraopeba (MG) e uma análise
(Hemisiella) tarsata Smith (Hymenoptera, Apidae, da biogeografia das abelhas do cerrado brasileiro
Centridini) em ninhos-armadilha no Nordeste do (Hymenoptera, Apoidea). Rev. Bras. Entomol. 39: 371-
Maranhão, Brasil. Rev. Bras. Entomol. 51: 382-388. 401.
Michener CD. 2000. The bees of the World. Baltimore, MD: Simpson BB, JL Neff. 1977. Krameria, free-fatty acids and oil-
Johns Hopkins Univ. Press, 913 pp. collecting bees. Nature 267: 150-151.
Moure JS, MS Castro. 2001. Uma nova espécie de Centris Simpson BB, JL Neff. 1981. Floral rewards: alternatives to
Fabricius (Hymenoptera, Apoidea, Anthophoridae) do pollen and nectar. Ann. MO Bot. Gard. 68: 301-322.
Nordeste do Brasil. Rev. Bras. Zool. 18: 329-333. Siqueira KMM, CF Martin, C Schlindwein, LHP Kill. 2005.
Oliveira-Rebouças P, M Gimenes. 2004. Abelhas (Apoidea) Polinizadores efetivos da goiaba (Psidium guajava) em
Visitantes de Flores de Comolia ovalifolia DC Triana área irrigada do submédio do vale do São Francisco.
(Melastomataceae) em uma Área de Restinga na Bahia. Ann. Enc. Zool. Nord. 15: 260.
Neotrop. Entomol. 33: 315-320. Viana BF, I Alves-dos-Santos. 2002. Bee diversity of the
Ortiz PL. 1994. The Cistaceae as food resources for honey coastal sand dunes of Brazil. In PG Kevan, VL Imperatriz-
bees in SW Spain. J. Apic. Res. 33: 136-144. Fonseca, eds. Pollinating bees: the conservation link
Proctor M, P Yeo, A Lack. 1996. The natural history of between agriculture and nature. Brasília, BR: Ministério
Zoological Studies 51(2): 195-203 (2012) 203

do Meio Ambiente, pp. 135-153. Vogel S. 1974. Ölblumen und ölsammelnde Bienen. Trop.
Viana BF, FO Silva, AMP Kleinert. 2001. Diversidade e Subtrop. Pflanz. 7: 285-369.
sazonalidade de abelhas solitárias (Hymenoptera: Vogel S. 1990. History of the Malpighiaceae in the light of
Apoidea) em dunas litorâneas no nordeste do Brasil. pollination ecology. Memoir. NY Bot. Gard. 55: 130-142.
Neotrop. Entomol. 30: 245-251. Vogel S, ICS Machado. 1991. Pollination of four sympatric
Vilhena AMGF, SC Augusto. 2007. Polinizadores da aceroleira species of Angelonia (Scrophulariaceae) by oil-collecting
Malpighia emarginata DC (Malpighiaceae) em área de bees in NE, Brasil. Plant. Syst. Evol. 178: 153-178.
Cerrado no Triângulo Mineiro. Biosci. J. 23: 14-23. Zanella FCV. 2000. The bees of the Caatinga (Hymenoptera,
Vinson SB, GW Frankie, HJ Williams. 1996. Chemical ecology Apoidea, Apiformes): a species list and comparative notes
of bees of the genus Centris (Hymenoptera: Apidae). Fla. regarding their distribution. Apidologie 31: 579-592.
Entomol. 79: 109-129.
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 monogamy include 1) pair bonds between males
as monogamous, polygynous, polyandrous, or and females in reproductive and non-reproductive
promiscuous based on the number of partners seasons (Carter et al. 1995), 2) aggressive
each individual has (Wittenberger 1979, Clutton- behaviors toward strange individuals (Carter et
Brock 1989). In the past, a species’ mating system al. 1995, Back et al. 2002), 3) bi-parental care
was indirectly determined by sexual dimorphism, (Solomon 1993a, Patris and Baudoin 2000), 4)
space use, and mating behaviors (Emlen and regulation of social factors (e.g., estrus induction)
Oring 1977, Getz and Hofmann 1986, Carter et (Carter et al. 1995), 5) the same home range sizes
al. 1995). Monogamy occurs in < 3% of mammal for males and females (Gaulin and FitzGerald
species (Kleiman 1977) and has attracted much 1988), and 6) shared use of a territory (e.g.,
research interest. Traditional ways to determine 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

Wu et al. – Monogamy of Taiwan Voles 205

These traditional methods provide clues for social (1998) studied the home ranges of M. kikuchii
monogamy, but not for genetic monogamy. Social using radio-tracking and field trapping. He found
monogamy indicates that 1 male and 1 female that home range sizes did not differ between males
show social living behavior, but it infers no sexual and females, only opposite sexes had overlapping
or reproductive patterns (Reichard 2003). Genetic ranges, and each range overlapped with only
monogamy is when 1 male and 1 female have an 1 individual of the opposite sex. Those results
exclusive reproductive relationship, and there are suggest social monogamy. To date; however, there
no extra-pair copulations (Reichard 2003). has been no study of the genetic components of M.
With the development of molecular genetic kikuchii mating systems.
techniques, genetic data are being used to exa- Parental care by both parents and pairing
mine mating systems (Queller et al. 1993, Avise exclusivity are supporting behavioral components
1994). Social mating systems may differ from of monogamy (Solomon 1993b, Carter et al.
genetic mating systems (Clutton-Brock and Isvaran 1995, Patris and Baudoin 2000). These home-
2006). In small mammals, for example, Neotoma range and behavioral observational studies led
cinerea in North America was thought to be socially us to hypothesize that M. kikuchii is socially and
polygynous based on sexual dimorphism and genetically monogamous. Since microsatellite
female clustering (Finley 1958, Escherich 1981). It DNA can be sensitive enough to identify parental
was identified as genetically monogamous using relationships, relatedness, and dispersal rates
DNA fingerprinting techniques (Topping and Millar (Scribner and Pearce 2000), we used microsatellite
1998). In contrast, socially monogamous species DNA and capture-recapture methods to identify
were found to engage in extra-pair copulations consistencies between the social and genetic
suggesting they are not genetically monogamous. mating systems of M. kikuchii. We tested the
These include the genetically promiscuous following predictions: 1) adult home-range sizes
Apodemus sylvaticus in the UK (Baker et al. 1999) do not significantly differ between sexes, 2) home-
and the genetically polygynous A. argenteus range overlaps among adults during reproductive
in Japan (Ohnishi et al. 2000). Peromyscus seasons are extensive or exclusive to a single
polionotus (Foltz 1981) and P. californicus (Ribble individual of the opposite sex, and 3) there is a
1991); however, are both socially and genetically lack of evidence of males mating with more than 1
monogamous. To distinguish between social and female at the same time (both females alive) or of
genetic components of a mating system (Hughes litters sired by multiple males.
1998), monogamous mating systems should be
examined with field observations and genetic
analyses that indicate parentage of offspring and MATERIALS AND METHODS
rule out extra-pair copulations (Kraaijeveld-Smit
2004). Trapping
In Microtus species occurring in the New
and Old Worlds (Hoffmann and Koeppl 1985), Trapping was carried out from June 2004 to
promiscuousness and polygyny are common, Sept. 2005 in the Hehuan Mt. area (121°17'17.4"E,
but monogamy is rare (Wolff 1985). Microtus 24°08'36.4"N) of Taroko National Park, central
kikuchii is the only Microtus species endemic to Taiwan. The elevation is about 3000 m. The
Taiwan. It is the southernmost Old World Microtus dominant vegetation is Yushan cane grassland.
species (Hoffmann and Koeppl 1985). It lives in Sherman traps baited with sweet potato were
diverse habitats, such as grasslands, scrub, and set up in a 4-ha (200 × 200-m) grid. To reduce
forests, including coniferous, broadleaf, and mixed trapping mortality and help retain warmth in cold
coniferous and broadleaf forests. It is mainly months, balled-up wads of shredded paper were
found at elevations of > 2000 m in alpine habitats put in front of the trigger of each Sherman trap.
of coniferous forests and Yushan cane (Yushania Previous trapping results with 10-m trap spacing in
niitakayamensis) grasslands. The reproductive the same Hehuan Mt. area revealed mean home
season is from Mar. to Aug. (Lu 1991). Chen et range sizes of 447.9 m 2 in spring, 423.4 m 2 in
al. (2006) studied the behavior of M. kikuchii in summer, 258.3 m2 in fall, and 210.6 m2 in winter
captivity, and found that when given the freedom with no statistical differences among seasons or
to choose, it spent significantly more time with its between sexes (Wu 1998). Radio-tracking of 8
mated partner than with strange individuals. They individuals for at least 24 h of continuous tracking
also observed paternal care of offspring. Wu 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 compare home-range sizes between adult males
between radio locations in a single day (average and females and between reproductive and non-
47.5 m, range 22-89 m) (Wu 1998). To maximize reproductive seasons. Home-range overlap was
the number of individuals trapped, we chose a examined for each 3-mo period in the reproductive
20-m trap spacing to form a 40,000-m2 grid with seasons to ensure that home ranges overlapped
121 Sherman traps. In an attempt to catch all at least part of the time. For example, examination
members of each family group, we trapped 5 of home-range overlap from Mar. to May could
successive nights each month using the capture- guarantee temporal overlap because individuals in
recapture method. Because activity of M. kikuchii the home-range analysis were captured in at least
peaks at 04:00-10:00 and 16:00-21:00 (Wu 1998), one of the following scenarios: Mar.-Apr., Apr.-May,
we checked traps 3 times a day at 05:00-07:00, or Mar.-May. Thus, ranges of individuals trapped in
09:00-11:00, and 19:00-22:00. two of these 3 sessions would overlap temporally.
We used toe-clipping to mark each vole The percentage of home-range overlap was only
the 1st time it was caught. Clipped toes and calculated between resident females and males.
additionally clipped pieces of the left ear and tail
were preserved in 99.9% alcohol for the genetic Selection of microsatellite primers
analysis. For each capture, we recorded the trap
site, individual number, sex, age, body weight, There are no primers designed for Microtus
and reproductive status (e.g., whether the female kikuchii. Microtus montebelli and M. oeconomus
nipple size indicated it was pregnant or lactating are the most closely related species to M. kikuchii
and whether a male had descended testes). (Conroy and Cook 2000). Therefore, we tested
Adults were distinguished from immature voles by primers for loci MSMM-1-8 designed for M.
the reproductive condition or weight (adult > 30 g). montebelli (Ishibashi et al. 1999) and for loci Moe1-
8 designed for M. oeconomus (Van de Zande et al.
Home-range size and overlap 2000) to determine their suitability for analyzing M.
kikuchii microsatellite DNA sequences.
Adult individuals captured more than 10 times Genomic DNA was extracted from left-
(Swilling and Wooten 2002) and residing over a ear tissue with a DNA purification kit (Epicentre,
month within the trapping grid (i.e., trapped in at Madison, WI, USA). The above primers amplified
least 2 different but not necessarily consecutive specific sequences. A polymerase chain reaction
monthly trap sessions) were considered residents (PCR) used a total volume of 50 μl with 1 μl of a
and used for the home-range analysis. Although fluorescence-labeled forward primer (25 mM), 1 μl
Seaman et al. (1999) suggested ≥ 30 relocations of an unlabeled reverse primer (25 mM), 5 μl PCR
for a home range to reduce bias, it was almost buffer (10x), 0.6 μl DNA, 0.6 μl DNTP (10 mM),
impossible to achieve this number of trapping 0.6 μl Taq, and 41.2 μl water. Because the lengths
locations because of our regime of 4 trap nights of these PCR products were too short for direct
per month and the short life of voles. Therefore, sequencing, they were excised from the agarose
we calculated home ranges for individuals gel for TA cloning. Each specific sequence was
captured in at least 2 monthly trap sessions and ligated with a vector (Invitrogen, Grand I., NY,
with ≥ 10 locations. Individuals caught in only USA) and put into competent cells for TA cloning.
1 monthly trapping session were not included All clones were further re-amplified with M13
in the home-range analysis. Home-range sizes primers (forward and reverse) which were supplied
were estimated for the reproductive and non- with TA cloning kit (Invitrogen) for length check.
reproductive seasons. ArcView GIS 3.3 (ESRI PCR protocol of the TA cloning check started
1996) with animal movement analysis (Hooge from denaturation at 94°C for 10 min. Twenty-five
and Eichenlaub 2000) was used to estimate the cycles were performed at the following conditions:
minimum convex polygon (MCP) home-range 1 min at 94°C, 1 min at 55°C for annealing, and
size (m2) from trapping locations. The percentage 1 min at 72°C for extension. Horizontal elec-
of home-range overlap between males and trophoresis with a 2% agarose gel was used to
females was subsequently calculated. Because check the sequence lengths of the clones.
of the small sample sizes and dependency of Clones containing sequences of < 500 bp
some individuals on home ranges in both the (Schlotterer and Harr 2001) were sent to Mission
reproductive and non-reproductive seasons, the Biotech Company (Taichung, Taiwan) to be
nonparametric Mann-Whitney U-test was used to sequenced on an ABI PRISM TM 3730xl DNA
Wu et al. – Monogamy of Taiwan Voles 207

Analyzer (Applied Biosystems, Carlsbad, CA, RESULTS

USA). Usable loci were determined using BioEdit
6.0.5 (Hall 1999) to check each sequence for In total, 169 voles (79 males and 90 females)
repeating units and whether both sides of the were caught in 1615 captures. One vole was
sequence were conserved and stable. Primers of excluded from the parentage analysis due to
usable microsatellite loci were used to synthesize failure to amplify its DNA.
fluorescent primers for the microsatellite analysis.
Polymorphism of microsatellite loci
Genetic data analysis
In total, 11 microsatellite loci were chosen.
For the microsatellite analysis, protocols Seven loci (MSMM-2, -3, -4, -5, -6, -7, and -8)
for DNA extraction and the PCR were the same used primers designed from Microtus montebelli
as those described above. PCR products were and 4 loci (Moe1, -2, -5, and -6) used primers
separated on an ABI 310 genetic analyzer (Applied designed from M. oeconomus (Table 1). Except
Biosystems). Individuals were genotyped using for MSMS-7, numbers of alleles were > 10;
Genotyper vers. 2.0 (Applied Biosystems). averaging 14.3 (range 8-19). All amplified loci
Tests of pairwise linkage disequilibrium were polymorphic. The observed heterozygosity
between loci were conducted using GenePop (Ho) value of each locus was close to the expected
(Raymond and Rousset 1995). Allele diversity, heterozygosity (He) value. Average values of Ho
heterozygosity (observed Ho and expected He), and He were both 0.88. As a result, the mean
and Hardy-Weinberg equilibrium of each loci were inbreeding coefficient, F, was essentially 0 at
calculated and tested using GENALEX 6 (Peakall -0.002. There were no departures from Hardy-
and Smouse 2006). Weinberg equilibrium (Table 1), indicating that
C E RV U S 2 . 0 ( S l a t e e t a l . 2 0 0 0 ) a n d the study population was under Hardy-Weinberg
GENALEX 6 (Peakall and Smouse 2006) were equilibrium. Locus pairs MSMM-4/MSMM-7
used to estimate parentage. Since the real and MSMM-4/Moe5 showed significant linkage
parentage of any individual could not be assured disequilibrium. Most loci showed no significant
based on the capture data, we randomly compared linkage disequilibrium. Therefore, locus MSMM-4
genotypes of all individuals to all males to identify was not used for the genetic analysis.
the most likely fathers. These offspring-father pairs
were randomly compared to all females to identify Parentage analysis
the most likely mothers. An error rate of 1% was
incorporated into the simulation with 80% relaxed The critical ΔLOD with a 95% level of
and 95% strict confidence intervals. Parentage certainty was 0.08 (with 95% of the parentage
was confirmed on the basis of mismatching resolved) if neither parent was known. Of the total
putative parentage at 0 loci or 1 locus, the LOD 168 voles used for the parentage analysis, 20
score (log-likelihood of each candidate parent), and mated pairs were found (18 males and 20 females)
the confidence of ΔLOD (the difference between to have 31 offspring. In total, 69 voles were
the 2 most likely parents). A ΔLOD score of > 3.0 assigned parentage (Table 2). In other words,
confirmed parentage, while a ΔLOD score of < -3.0 41.1% (69/168) of the 168 voles, including adult
rejected parentage (Slate et al. 2000). A ΔLOD and immature voles, could be assigned parentage
score was calculated by the difference in LOD with both parents identified. Except for 2 males
scores between the most likely and the 2nd most (49M and 50M), each male mated with only 1
likely candidate parents (either of which might be female during the study period. The 2 males who
the true parent). The most likely candidate parent mated with more than 1 female did so in different
was the one with a ΔLOD score exceeding the breeding seasons in different years and after the
critical ΔLOD score with a 95% confidence inter- 1st female was no longer present.
val. Relationships of individuals with overlapping
home ranges were checked with results of the Home-range size and overlap and their relation-
parentage analysis to see whether they were ships
mates or family members.
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- pairwise combinations between male and female
Whitney U-test, p = 0.31). Average home-range ranges were detected as sexually paired partners.
sizes in the reproductive season were 1826.1 ± Six (75%) of these 8 reproductive pairs had
1463.9 m2 (n = 23) for adult males and 1684.2 ± exclusive, overlapping home ranges. The average
1256.7 m2 (n = 19) for adult females. The average overlap of a male’s range with a female’s range
home-range sizes in the non-reproductive season did not statistically differ from the average overlap
were 2072.7 ± 1637.7 m2 (n = 11) for adult males of a female’s range by a male’s range (Wilcoxon
and 1125.0 ± 874.6 m2 (n = 8) for adult females. rank-sum test, p = 0.126). Average percentages
No significant difference in home-range sizes of a female’s home range overlapped by a male’s
was found between sexes (Mann-Whitney U-test, were significantly larger in mated pairs (77.1%,
p = 0.74 for the reproductive season and p = 0.16 n = 8) than in pairs not found to produce offspring
for the non-reproductive season). The sample size (41.8%, n = 11) (Mann-Whitney U-test, p = 0.0372).
for the non-reproductive season was small. Average percentages of a female’s home range
We separated the reproductive season into overlapped by a male’s were also significantly
3 periods of 3 mo each and compared the home- larger than a male’s home range overlapped by his
range overlap among resident individuals (Fig. 1). sexual female partner’s (42.4%, Wilcoxon rank-
Only those with ranges that overlapped ranges sum test, p = 0.0499, n = 8). In other words, a
of the opposite sex are shown. There were 21 female’s home range tended to overlap more with
pairwise combinations of home ranges overlapping that of her sexual partner, but males did not show
those of the opposite sex. Eleven (52.4%) showed this trend.
exclusive home-range overlap between 1 male In total, 14 social units were recognized
and 1 female (Fig. 1). Eight (38.1%) of these 21 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)

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

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

MSMM-3a (CA)15 52 TACGCCCTTCAAACTCATGTG 102-136 14 0.833 0.827 -0.008 NS 0.279
MSMM-4a (CA)19 52 TGTTTCAAGGCAATAAGGTGG 143-187 18 0.913 0.872 -0.047 NS 0.512
MSMM-5a (CA)17 52 TCTAATACCCTCTTCCTTGGG 69-111 19 0.900 0.910 0.011 NS 0.247
MSMM-6a (CA)20 52 TCCTATCAAGGGGCATTCATCT 145-167 12 0.873 0.857 -0.019 NS 0.987
MSMM-7a (CA)18 56 TAAGAAGGGCCACTAAGACCC 105-123 8 0.820 0.830 0.012 NS 0.915
MSMM-8a (CA)17 50 TGCTTAGTTCACTGCTGAACC 170-196 13 0.927 0.886 -0.046 NS 0.246
Moe1b (GT)18 60 TGGTTGTTCTGTGGTGAATACAG 93-133 15 0.847 0.890 0.049 NS 0.770
Moe2b (GT)17 60 CATCTGATGAGTCCCTGAGG 145-185 15 0.887 0.889 0.002 NS 0.480
Moe5b (TC)25 60 GGTCATGCTCCAAGAAGCTC 108-138 14 0.833 0.866 0.037 NS 0.663
Moe6b (GT)25 60 GGTTTTCTGATTCAGGCAGG 210-246 14 0.913 0.998 -0.017 NS 0.420
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

41M 102M
41M 60M
60M 138F
80F 135F
42F 25F 80F
8F 136F 41M

45M 109M 36F

36F 108M
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*

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
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 Solomon 2000, Lucia et al. 2008). We maintain
exclusively between 1 male and 1 female. Six of that M. kikuchii is socially monogamous because
these 11 social units (54.5%) were paired sexually of similar home-range sizes between sexes, the
and monogamously produced offspring, and 5 very high proportion of social units comprised of
social units were not detected to have produced male-female pairs, and the relatively low overlap
any offspring. For the 3 social units with > 2 in home ranges of individuals without reproductive
individuals, only the social unit with 2 males and relationships (e.g., not sexual partners or family
2 females (June-Aug. 2004) was not detected to members).
have reproduced. The 2nd social unit (Mar.-May Only six of the 11 male-female pairs iden-
2005) had a mated pair (44M-80F). The 3rd (June- tified by overlapping home ranges were found
Aug. 2005) had a family group (parents 44M-80F to be paired partners that had successfully
and 2 daughters: 135F and 136F). No sexual produced young. Male-female pairs determined
pairing or family groups were found in June-Aug. by overlapping home ranges might not truly
2004. In 2005, all social units found to reproduce (8 be breeding pairs. In M. ochrogaster, socially
of 10, 80%) practiced monogamy and were either monogamous, multiple paternity in five of 9 litters
mated pairs or family groups. In total, 8 (57.1%) was also identified by a genetic study (Solomon
social units were found to have produced offspring. et al. 2004). Thus, data from home-range overlap
Six (75%) of these had exclusive home-range cannot reflect true pairing conditions or whether M.
overlap between 1 male and 1 female. kikuchii is genetically monogamous. To determine
the mating system of a species, field data and
genetic data are both necessary (Hughes 1998,
DISCUSSION Kraaijeveld-Smit 2004). As we found no litters
sired by multiple fathers, genetic monogamy of
Our results strongly support the hypotheses M. kikuchii should be assured. The parentage
that Microtus kikuchii is socially and genetically analysis showed survival of 1 or 2 offspring in each
monogamous. We identified litters sired by 18 litter. This is consistent with Lu’s (1991) data from
males and 20 females. The 2 males that fathered field trapping (range 1-3, average litter size 2.1)
more than 1 litter did so in different years and with and our own observations from captive breeding
a different mate because their 1st mate was no (litter size 1-3 with 2 most frequent) (pers. unpubl.
longer present. None of the 9 litters with multiple data). With an average litter size of 2.1, it may
offspring had more than 1 father (Table 2). We be more difficult to detect multiple paternity in M.
found no significant differences in home-range kikuchii than for species with larger litter sizes,
sizes between sexes. Most (78.6%) social units such as M. ochrogaster. Low detectability of
consisted of only 1 male and 1 female. Home- multiple paternity due to small litter size is unlikely
range overlaps between male and female pairs not for M. kikuchii because Lu (1991) found a very
found to produce offspring were relatively small low post-implantation mortality rate (2 resorbed
(44M-58F, 41M-80F, and 120M-136F), except for embryos of 64 embryos). Moreover, we are con-
60M and 80F. Therefore, overlap in home range fident that we trapped most of the population
was mostly by a single male, a single female, because our extensive trapping effort spanned
and their offspring, strongly suggesting social 2 breeding seasons, and we had high recapture
monogamy. rates (71.5%-96.5%). Even so, we still found no
Microtus kikuchii was suspected of being litters sired by multiple fathers.
socially monogamous based on home ranges Some studies are beginning to show that
(Wu 1998) and observations of captive individuals in a number of species considered to be mono-
(Chen et al. 2006). Combining our home-range gamous, females mate with multiple males. In
data with a genetic analysis of parentage provided mammals, extra-pair copulation was found in
more in-depth information on the relationships of some socially monogamous species (Richardson
voles observed to have overlapping home ranges. 1987, Agren et al. 1989, Solomon et al. 2004,
Microtus ochrogaster is considered a socially Mabry et al. 2011). Previously, only 2 known
monogamous species (Hofmann et al. 1984, Carter rodent species simultaneously showed social and
et al. 1995) even though not all adults live as male- genetic monogamy, i.e., Peromyscus polionotus
female pairs. Adults live in groups of single males (Foltz 1981) and P. californicus (Ribble 1991).
and females, and groups of 3 or more adults were Genetic promiscuity and polygyny are common in
also documented (Getz et al. 1993, Cochran and Microtus (e.g., M. pennsylvanicus, M. richardsoni,
Wu et al. – Monogamy of Taiwan Voles 211

M. xanthognathus, and M. californicus), but ESRI. 1996. Arcview GIS Software. Redlands, CA: Enviro-
monogamy is rare (Wolff 1985). Our study has nmental Systems Research Institute (ESRI).
Finley RB. 1958. The woodrats of Colorado: distribution and
added M. kikuchii to the list of rodent species ecology. Univ. KS Publ. Mus. Nat. Hist. 10: 213-552.
simultaneously showing social and genetic Foltz DW. 1981. Genetic evidence for long-term monogamy
monogamy. in a small rodent, Peromyscus polionotus. Am. Nat. 117:
Acknowledgments: The authors thank the Gaulin SJC, RW FitzGerald. 1988. Home-range size as a
predictor of mating systems in Microtus. J. Mammal. 69:
anonymous reviewers for their critical comments 311-319.
and many valuable remarks on the original Getz LL, JE Hofmann. 1986. Social organization in free-
manuscript. We thank Q.W. Zhu, J.K. Lin, G.H. living prairie voles, Microtus ochrogaster. Behav. Ecol.
Zhen, S.L. Yuan, and many other volunteers for Sociobiol. 18: 275-282.
help with fieldwork. We thank M.Y. Zhen, S.F. Getz LL, B Mcguire, T Pizzuto, JE Hofmann, B Frase.
1993. Social-organization of the prairie vole (Microtus
Zhen, Z.X. Zhang, L.Y. Liu, R.P. Huang, W.Y. ochrogaster). J. Mammal. 74: 44-58.
Zhen, and Z.Y. Guo for help with genetic work. We Hall TA. 1999. BioEdit: a user-friendly biological sequence
thank the high-elevation experimental station of alignment editor and analysis program for Window 95/998/
the Taiwan Endemic Species Research Institute NT. Nucleic Acids Res. Symp. Ser. 41: 95-98.
for accommodations during fieldwork. Trapping Hoffmann RS, JW Koeppl. 1985. Zoogeography. In RH
Tamarin, ed. Biology of New World Microtus. Boston,
protocols complied with government regulations. MA: American Society of Mammalogist, pp. 84-115.
Permits were obtained from Taroko National Park. Hofmann JE, LL Getz, L Gavish. 1984. Home range overlap
This research was funded by the National Science and nest cohabitation of male and female prairie voles.
Council of Taiwan (NSC96-2621-B-029-001-MY3) Am. Midl. Nat. 112: 314-319.
and Tunghai Univ. 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,
REFERENCES Hughes C. 1998. Integrating molecular techniques with field
methods in studies of social behavior: a revolution results.
Agren G, Q Zhou, W Zhong. 1989. Ecology and social beha- Ecology 79: 383-399.
viour of Mongolian gerbils, Meriones unguiculatus, at Ishibashi Y, Y Yoshinaga, T Saitoh, S Abe, H Iida, MC Yoshida.
Xilinhot, Inner Mongolia, China. Anim. Behav. 37: 11-27. 1999. Polymorphic microsatellite DNA markers in the field
Avise JC. 1994. Molecular markers, natural history and vole Microtus montebelli. Mol. Ecol. 8: 163-164.
evolution. New York: Chapman and Hall. Kleiman DG. 1977. Monogamy in mammals. Q. Rev. Biol. 52:
Back SR, LA Beeler, RL Schaefer, NG Solomon. 2002. 39-69.
Testing functional hypotheses for the behavior of resident Kraaijeveld-Smit F. 2004. Behavioural ecology. In TJC
pine voles, Microtus pinetorum, toward non-residents. Beebee, G Rowe, eds. An introduction to molecular
Ethology 108: 1023-1039. ecology. Oxford, UK: Oxford Univ. Press, pp. 81-109.
Baker RJ, KD Makova, RK Chesser. 1999. Microsatellites Lu MS. 1991. Ecology and reproduction of Microtus kikuchii
indicate a high frequency of multiple paternity in in the Hohuanshan area. Master’s thesis, Tunghai Univ.,
Apodemus (Rodentia). Mol. Ecol. 8: 107-111. Taichung, Taiwan. (in Chinese with English abstract)
Carter CS, AC De Vries, LL Getz. 1995. Physiological Lucia KE, B Keane, LD Hayes, YK Lin, RL Schaefer, NG
substrates of mammalian monogamy: the prairie vole Solomon. 2008. Philopatry in prairie voles: an evaluation
model. Neurosc. Biobehav. Rev. 19: 303-314. of the habitat saturation hypothesis. Behav. Ecol. 19:
Chen Y-C, Z-Y Shieh, L-K Lin. 2006. Mating system of Taiwan 774-783.
vole: evidence of behavioral studying. Chin. Biosci. 49: Mabry KE, CA Streatfeild, B Keane, NG Solomon. 2011.
51-73. (in Chinese) avpr1a length polymorphism is not associated with either
Clutton-Brock TH. 1989. Mammalian mating systems. Proc. R. social or genetic monogamy in free-living prairie voles.
Soc. Lond. Ser. B Biol. Sci. 236: 339-372. Anim. Behav. 81: 11-18.
Clutton-Brock TH, K Isvaran. 2006. Paternity loss in con- Ohnishi N, T Saitoh, Y Ishibashi. 2000. Spatial genetic
trasting mammalian societies. Biol. Lett. 2: 513-516. relationships in a population of the Japanese wood mouse
Cochran GR, NG Solomon. 2000. Effects of food supple- Apodemus argenteus. Ecol. Res. 15: 285-292.
mentation on the social organization of prairie voles Patris B, C Baudoin. 2000. A comparative study of parental
(Microtus ochrogaster). J. Mammal. 81: 746-757. care between two rodent species: implications for
Conroy CJ, JA Cook. 2000. Molecular systematics of a the mating system of the mound-building mouse Mus
Holarctic rodent (Microtus: Muridae). J. Mammal. 81: spicilegus. Behav. Proc. 51: 35-43.
344-359. Peakall R, PE Smouse. 2006. GENALEX 6: genetic analysis
Emlen ST, LW Oring. 1977. Ecology, sexual selection, and the in Excel. Population genetic software for teaching and
evolution of mating systems. Science 197: 215-223. research. Mol. Ecol. Notes 6: 288-295.
Escherich PC. 1981. Social biology of the bushy-tailed Queller DC, JE Strassmann, CR Hughes. 1993. Microsatellite
woodrat, Neotoma cinerea. Univ. CA Publ. Zoology 110: and kinship. Trends Ecol. Evol. 8: 285-288.
1-132. Raymond M, F Rousset. 1995. An exact test for population
212 Zoological Studies 51(2): 204-212 (2012)

differentiation. Evolution 49: 1280-1283. and female prairie voles (Microtus ochrogaster). Can. J.
Reichard UH. 2003. Monogamy: past and present. In UH Zool. Rev. Can. Zool. 71: 434-437.
Reichard, C Boesch, eds. Monogamy. Cambridge, UK: Solomon NG, B Keane, LR Knoch, PJ Hogan. 2004. Multiple
Cambridge Univ. Press, pp. 3-25. paternity in socially monogamous prairie voles (Microtus
Ribble DO. 1991. The monogamous mating system of ochrogaster). Can. J. Zool. 82: 1667-1671.
Peromyscus californicus as revealed by DNA finger- Swilling WR, MC Wooten. 2002. Subadult dispersal in a
printing. Behav. Ecol. Sociobiol. 29: 161-166. monogamous species: the Alabama beach mouse
Richardson PRK. 1987. Aardwolf mating system: overt (Peromyscus polionotus ammobates). J. Mammal. 83:
cuckoldry in an apparently monogamous mammal. S. Afr. 252-259.
J. Sci. 83: 405-410. Topping MG, JS Millar. 1998. Mating patterns and reproductive
Schlotterer C, B Harr. 2001. Microsatellite instability. In success in the bushy-tailed woodrat (Neotoma cinerea) as
Encyclopedia of life sciences (eLS). NJ: J. Wiley & Sons, revealed by DNA fingerprinting. Behav. Ecol. Sociobiol.
Ltd., pp. 1-4. (http://www.els.net) 43: 115-124.
Scribner KT, JM Pearce. 2000. Microsatellite: evolutionary and Van de Zande L, RC van Apeldoorn, AF Blijdenstein, D de
methodological background and empirical applications Jong, W van Delden, R Bijlsma. 2000. Microsatellite
at individual, population and phylogenetic levels. In analysis of population structure and genetic differentiation
AJ Baker, ed. Molecular methods in ecology. Berlin: within and between populations of the root vole, Microtus
Blackwell Science, pp. 235-273. oeconomus in the Netherlands. Mol. Ecol. 9: 1651-1656.
Seaman DE, JJ Millspaugh, BJ Kernohan, GC Brundige, KJ Wittenberger JF. 1979. The evolution of mating systems in
Raedeke, RA Gitzen. 1999. Effects of sample size on birds and mammals. In P Marler, JG Vandenbergh, eds.
kernel home range estimates. J. Wildl. Manage. 63: 739- Handbook of behavioral neurobiology. New York: Plenum,
747. pp. 271-349.
Slate J, T Marshall, J Pemberton. 2000. A retrospective Wolff JO. 1985. Behavior. In RH Tamarin, ed. Biology of
assessment of the accuracy of the paternity inference New World Microtus. Boston, MA: American Society of
program CERVUS. Mol. Ecol. 9: 801-808. Mammalogists, pp. 340-372.
Solomon NG. 1993a. Body size and social preferences of Wu PJ. 1998. Population ecology and activity pattern of
male and female prairie voles, Microtus ochrogaster. Kikuchii’s field vole (Microtus kikuchii) in Hohuanshan
Anim. Behav. 45: 1031-1033. area. Master’s thesis, Tunghai Univ., Taichung, Taiwan. (in
Solomon NG. 1993b. Comparison of parental behavior in male 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,*
Biodiversity Research Center, Academia Sinica, Nangang, Taipei 115, Taiwan
Comparative Genomic Centre and ARC Centre of Excellence for Coral Reef Studies, James Cook Univ., Townsville 4810, Australia
Natural History Museum and Institute, Chiba 955-2, Japan
Institute of Oceanography, National Taiwan Univ., Taipei 106, Taiwan
Department of Life Science, National Taitung Univ., Taitung 950, Taiwan
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.

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

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

*To whom correspondence and reprint requests should be addressed. Tel: 886-2-27899549. Fax: 886-2-28958059.

214 Lin et al. – New Scleractinian Coral From Taiwan

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




30°S 11 - 50 m
> 100 m

150°W 90°W °W 0°E 60°E 120°E

Fig. 1. Worldwide distribution and depths of Polycyathus spp.

Zoological Studies 51(2): 213-221 (2012) 215

N (A) (B)

Kaohsiung City TAIWAN


(C) (D)

120°E 122°E

(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

(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. remaining ones were used to calculate posterior

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


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 species are known from Pacific waters (P. palifera
elements. Septal and palar faces bearing several Verrill 1869 [reef depth], P. hondaensis (Durham
pointed granules aligned perpendicular to septal/ & Barnard 1952) [55-64 m], P. fulvus [30-50 m], P.
palar edges. Fossa moderately deep, containing isabela Wells, 1982 [14-23 m], P. hodgsoni [> 35 m];
elongate papillose columella. Columella composed P. marigondoni [35 m]; P. furanaensis Verheij &
of 5-7 slender, irregularly shaped rods. Best 1987 [6-52 m], P. norfolkensis [10-20 m], and
Remarks: Polycyathus chaishanensis sp. nov. P. octuplus [90-441 m]).
differs from all other known species of this genus Among Pacific and Indian congeners that
by having a much smaller corallite. Twenty-one have small corallites, P. chaishanensis sp. nov. is
corallites examined from the holotype colony had most similar to P. difficilis (Mergui Archipelago).
a mean CD of 3.05 ± 0.26 mm (Fig. 5), whereas Both species have an exserted S1, indistinct
corallites among the other 18 valid Polycyathus P1, and S2 and S3 with dentate/laciniate axial
species are significantly larger (mean CD of edges. However, P. chaishanensis sp. nov. differs
4.38 ± 1.10 mm). In addition, P. chaishanensis in having 4 incomplete cycles of septa, while P.
sp. nov. has one of the shallowest bathymetric difficilis has 3 cycles of septa.
ranges known from representatives of this genus Interestingly, DGGE from the ITS-2 confirmed
(≤ 3 m) (Fig. 1), and all colonies were found to the presence of Symbiodinium subclade C1
inhabit tidal pools. Of the 18 extant Polycyathus associated with P. chaishanensis sp. nov. Although
species, 3 were described from the Atlantic Ocean Wijsman-Best (1970) described the association
(P. atlanticus Duncan, 1876 [depth unknown], P. of zooxanthellae with P. fulvus, to date, all other
senegalensis Chevalier 1966 [12-143 m], and representatives of this genus are considered
P. mayae Cairns 2000 [110-309 m]; 5 from the azooxanthellate (Cairns et al. 1999). However, to
Indian Ocean (P. persicus Duncan 1876 [depth reinvestigate this important ecological aspect of
unknown], P. fuscomarginatus Klunzinger 1879 shallow-water Polycyathus, new samples enabling
[depth unknown], P. verrilli [depth unknown], P. the examination of their tissue must be collected.
difficilis Duncan 1889 [depth unknown], and P. Etymology: This species is named for the
andamanensis [depth unknown]); 1 species from uplifted reef in southern Taiwan (Chaishan) from
the Mediterranean Sea (P. muellerae Abel 1959 which it was collected and to which it is possibly
[10-32 m]); and according to Cairns (1999), 9 endemic.
Distribution: Known only from the sublittoral
zone (< 3 m deep) near Chaishan, Kaohsiung,
8.00 Taiwan (22°37'13"N, 120°15'56"E to 22°38'18"N,
Calicular Diameter (mm)


5.00 Phylogeny of Polycyathus

To test the hypothesis that Polycyathus
is a natural genus, a 16S rRNA sequence
was extracted from the P. chaishanensis sp.
2.00 nov. mt genome (accession no.: NC 015642;
Lin et al. 2011) and aligned with previously
1.00 published sequences from 8 representatives
P. chaishanensis Polycyathus spp. of morphologically related caryophylliid genera
and 13 representatives of non-caryophylliid
Fig. 5. Measurement of the calicular diameter (CD) of P. families. Results of the phylogenetic analysis are
chaishanensis sp. nov. (21 corallites) and extant Polycyathus summarized in figure 6, and following Romano
species (18 species). The CD of each P. chaishanensis
and Cairns (2000), Le-Goff Vitry et al. (2004),
corallite and its congeners are indicated by black circles in
the box plot. The non-parametric Wilcoxon-Mann-Whitney and Fukami et al. (2008), sequences from 4
rank sum test showed no significant difference (p = 0.1135) scleractinian species in the “complex” coral clade
in calicular diameters between P. chaishanensis sp. nov. and were used as an outgroup. Despite the fact that
extant Polycyathus species. only 2 Polycyathus species were represented in
Zoological Studies 51(2): 213-221 (2012) 219

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


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 Psammocora sp. and Porites sp., were found on
that this genus is not monophyletic. Polycyathus limestone or among fleshy algae. However, most
chaishanensis sp. nov. was not grouped with of these scleractinian species were found in tidal
any other congener (P. muellerae) or any other pools of < 5 m deep. Reefs in shallow water with
caryophylliid representative. Instead, our results less light are usually dominated by zoanthus and
show that P. chaishanensis sp. nov. has a genetic sea anemones, probably including Anthopleura sp.
immediacy to some representatives of the (Fig. 2). In addition, overhangs and overhanging
Siderastreidae (Coscinaraea and Psammocora), surfaces with less light are primarily dominated by
Fungiidae (Zoopilus and Fungia), and Faviidae encrusting sponges. Polycyathus chaishanensis
(Leptastrea) (Fig. 6). In addition, our results sp. nov. was only found on well-lit reefs dominated
support P. muellerae having a close relationship by green and encrusting calcareous red algae, and
with Paracyathus pulchellus (Kitahara et al. 2010b) was generally rare (Fig. 2). However, this area
but not with Rhizosmilia maculata. These results is dominated by green algae and turbid waters
were also supported by the COI sequence data caused by erosion, which may have inhibited the
(data not shown). occurrence of most other scleractinians. The small
In previous molecular studies, many of the population of this new species raises concerns as
morphologically defined families, especially those to its vulnerability to natural and anthropogenic
composed of zooxanthellate species, showed threats.
extensive polyphyly (Romano and Cairns 2000,
Le Goff-Vitry et al. 2004, Fukami et al. 2008, Acknowledgments: We thank Dr. K. Soong
Kitahara et al. 2010a). In an attempt to clarify (National Sun-Yat Sen Univ., Kaohsiung, Taiwan)
the validity of morphology-based taxonomy, for ecological information on the Chaishan area,
additional taxon sampling, more-comprehensive Mr. L.C. Wang (National Taiwan Univ., Taipei,
morphological analyses, and additional molecular Taiwan) for assistance with the SEM technology,
data are required (Fukami et al. 2008). Therefore, and Dr. Y. Nozawa for help with photography.
molecular data from other Polycyathus species are Constructive comments from the members of the
needed to clarify the phylogenetic status of this Coral Reef Evolutionary and Ecological Genetics
genus. (CREEG) Laboratory, Biodiversity Research
Center, Academia Sinica (BRCAS) and 3 anony-
Ecology of Polycyathus chaishanensis sp. nov. mous reviewers are especially appreciated. SK
is the recipient of a postdoctoral fellowship from
The rare distribution and the small-sized Academia Sinica (2010-2012). This study was
population of this new species raise several supported by a BRCAS Thematic grant (2006-
concerns as to its vulnerability to natural and 2008) and one from the National Science Council,
anthropogenic threats, in a period of intense urban Taiwan (NSC94-2621-B-001-005) to CAC. This is
development at Chaishan. the CREEG Laboratory contribution no. 73.
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 REFERENCES
Taiwan occurs in the Gutingkeng Formation near
Kaohsiung (Gong et al. 1998). A debris avalanche Cairns SD. 1995. The marine fauna of New Zealand:
and sandy substrate form the main characteristics Scleractinia (Cnidaria: Anthozoa). NZ Oceanogr. Inst.
Mem. 103: 1-216.
of the Chaishan area and have contributed to Cairns SD. 1999. Cnidaria, Anthozoa: deep-water
the benthic communities of this region. Among azooxanthellate Scleractinia from Vanuatu, and Wallis and
hermatypic organisms reported from Chaishan’s Futuna Islands. Mém. Mus. natl. d'Hist. nat. 180: 31-167.
formation, the most important are scleractinian Cairns SD, BW Hoeksema, J van der Land. 1999. Species
corals (such as Acropora, Porites, Favia, and richness of recent Scleractinia. Atoll Res. Bull. (National
Museum of Natural History Smithsonian Institution
Favites), mollusks, and encrusting calcareous Washington, DC, USA) 459: 1-46.
red algae (Gong et al. 1998). Hard surfaces CPAMI. 2008. The program of sustainable coastal deve-
exposed to light in the Chaishan area were found lopment. Taipei, Taiwan: Construction and Planning
to be heavily dominated by algae, primarily the Agency Ministry of the Interior (CPAMI) Press. (in
green algae Ulva fasciata and U. lactuca, and Chinese)
Duncan PM. 1876. Notices of some deep-sea and littoral
some turf algae such as Chaetomorpha antennina corals from the Atlantic Ocean, Caribbean, Indian, New
(Huang 2003). Colonial zooxanthellate corals,
Zoological Studies 51(2): 213-221 (2012) 221

Zealand, Persian Gulf, and Japanese etc. seas. Proc. Biol. 141: 387-400.
Zool. Soc. Lond. 44: 428-442. LaJeunesse TC, RK Trench. 2000. The biogeography of two
Fukami H, AF Budd, DR Levitan, J Jara, R Kersanach, N species of Symbiodinium (Freudenthal) inhabiting the
Knowlton. 2004. Geographic differences in species intertidal anemone, Anthopleura elegantissima (Brandt).
boundaries among members of the Montastraea annularis Biol. Bull. 199: 126-134.
complex based on molecular and morphological markers. Le Goff-Vitry MC, AD Rogers, D Baglow. 2004. A deep-sea
Evolution 58: 324-337. slant on the molecular phylogeny of the Scleractinia. Mol.
Fukami H, CA Chen, AF Budd, A Collins, C Wallace, YY Chuang Phylogen. Evol. 30: 167-177.
et al. 2008. Mitochondrial and nuclear genes suggest Lin MF, KS Luzon, WY Licuana, MC Ablan-Lagman, CA Chen.
that stony corals are monophyletic but most families of 2011. Seventy-four universal primers for characterizing
stony corals are not (order Scleractinia, class Anthozoa, the complete mitochondrial genomes of scleractinian
phylum Cnidaria). PLoS One 3: e3222. corals (Cnidaria; Anthozoa). Zool. Stud. 50: 513-524.
Gong SY, SW Wang, TY Lee. 1998. Pleistocene coral reefs Nylander JAA. 2004. MrModeltest v2. Program distributed by
associated with claystones, southwestern Taiwan. Coral the author. Evolutionary Biology Centre, Uppsala Univ.
Reefs 17: 215-222. Romano SL, SD Cairns. 2000. Molecular phylogenetic
Huang SF. 2003. Seaweeds of northeastern Taiwan. Taipei, hypotheses for the evolution of scleractinian corals. Bull.
Taiwan: National Taiwan Museum Press. (in Chinese) Mar. Sci. 67: 1043-1068.
Huelsenbeck JP, F Ronquist. 2001. MrBayes: Bayesian Tamura K, M Dudley, M Nei, S Kumar. 2007. MEGA4:
inference of phylogeny. Bioinformatics 17: 754-755. Molecular Evolutionary Genetics Analysis (MEGA) soft-
Kitahara MV, SD Cairns, DJ Miller. 2010a. Monophyletic ware vers. 4.0. Mol. Biol. Evol. 24: 1596-1599.
origin of Caryophyllia (Scleractinia, Caryophylliidae), with Verheij E, MB Best. 1987. Notes on the genus Polycyathus
descriptions of six new species. Syst. Biodivers. 8: 91- Duncan, 1876 and a description of three new scleractinian
118. corals from the Indo-Pacific. Zool. Mededelingen 61: 147-
Kitahara MV, SD Cairns, J Stolarski, D Blair, DJ Miller. 2010b. 154.
A comprehensive phylogenetic analysis of the Scleractinia Wijsman-Best M. 1970. A new species of Polycyathus
(Cnidaria, Anthozoa) based on mitochondrial CO1 Duncan, 1876, from New Caledonia and a new record of
sequence data. PLoS One 5: e11490. Polycyathus senegalensis Chevalier, 1966 (Madreporaria).
LaJeunesse TC. 2002. Diversity and community structure of Beaufortia 227: 79-84.
symbiotic dinoflagellates from Caribbean coral reefs. Mar.
Zoological Studies 51(2): 222-231 (2012)

Systematic Study of the Simocephalus Sensu Stricto Species Group

(Cladocera: Daphniidae) from Taiwan by Morphometric and Molecular
Shuh-Sen Young1,*, Mei-Hui Ni2, and Min-Yun Liu3
Department of Applied Science, National Hsinchu University of Education, Hsinchu 300, Taiwan
Hsinchu Municipal Hsinchu Elementary School, Hsinchu 300, Taiwan
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.

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

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

Young et al. – Study of Simocephalus from Taiwan 223

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

with eggs) were collected in 2003 and 2004 and Hudec, 1991 (Daphniidae) and Diaphanosoma
used for the morphometric analysis: 45 individuals dubium Manuilova, 1964 (Sididae) from Taiwan
of S. vetulus from 8 sites, 72 individuals of S. were analyzed in order to obtain outgroup
vetuloides from 11 sites, and 70 individuals of sequences.
S. mixtus from 10 sites. From this set of 187
individuals, 72 individuals, including 22 individuals Morphometric analysis
of S. vetulus from 8 sites, 28 individuals of S.
vetuloides from 11 sites, and 22 individuals of S. Lateral-view images of S. vetulus, S.
mixtus from 10 sites, were selected for the DNA vetuloides, and S. mixtus were taken using a
analysis. Additionally, 7 individuals of S. serrulatus digital camera under a stereomicroscope for the
(Fig. 1) from 3 sites and 19 individuals of S. morphometric study. Morphometric characters
heilongjiangensis (Fig. 1) from 5 sites were also were extracted from the photographic images, and
included in the DNA analysis. Daphnia similoides 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 225

24 length measurements, each of which was animals. Each animal was taken from 95% EtOH
divided by body length to obtain size-free ratios and placed into pure water for 1 h for cleaning.
(Fig. 2). The body length and dorsal valve angle After that, each animal was placed at the bottom
(Fig. 2) were also measured on the photographic of a 0.5-ml centrifuge tube for 30 min to dry in a
images, and the clutch size of each individual speed vacuum-drying system. Dried samples
was assessed under a microscope. SPSS vers. were then ground up by needles, and 50 μl of a 5%
10.0.1 (Chicago, IL, USA) was used to analyze Chelex solution was used to extract the DNA by
the numerical data. The data matrix was tested incubation at 56°C for 2-3 h, followed by incubation
using the Kaiser-Meyer-Olkin (KMO) measure at 90°C for 8 min. For each polymerase chain
of sampling adequacy and by the Bartlett X2 test reaction (PCR), 5 μl of upper cleaning was used as
prior to the principle component analysis (PCA). the DNA template after centrifugation at 104 rpm
For individuals with eggs, Pearson’s correlation (9168g) for 3 min.
analyses and non-linear regressions among the We employed the universal primers, LCO
dorsal angle, body length, and clutch size were 1490 (5'-GGTCAACAAATCATAAAGATATTGG-3')
carried out. and HCO2918 (5'-TAAACTTCAGGGTGACCAA
AAAATCA-3') (Folmer et al. 1994), to amplify the
DNA extraction, amplification, and sequencing mitochondrial COI gene by a PCR. Each PCR
sample had a total volume of 50 μl and consisted
Total genomic DNA was extracted using of pH 9.2 buffer solution (50 mM Tris-HCl, 16 mM
Chelex (InstaGene Matrix BIO-RAD 7326030, ammonium sulfate, 2.5 mM MgCl 2 , and 0.1%
Bio-Rad Laboratories, Hercules USA) from single 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
B 45 s, and extension at 72°C for 45 s; followed by
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
G H C an ultraviolet light box. DNA fragments were
60° 60°
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
E D with our sequences. The download sequences
included S. vetulus from the UK (accession no.,
Fig. 2. Morphometry of each specimen extracted from 8
data points (A-H), from which we constructed 24 length
DQ889172: Costa et al. 2007), S. cf. punctatus
measurements; each length measurement was then divided by from Mexico and Guatemala (EU702310 and
body length (AE) to obtain size-free ratios. The angle between EU702282, Elias-Gutierrez et al. 2008), S.
lines AE and ED was taken as the dorsal valve angle. cf. exspinosus from Mexico and Guatemala
226 Zoological Studies 51(2): 222-231 (2012)

(EU702296 and EU702279, Elias-Gutierrez et al. positively correlated with the dorsal valve angle
2008), S. cf. mixtus from Mexico and Guatemala and body length. The valve shape was not a
(EU702305 and EU702281, Elias-Gutierrez et species-specific character, but rather it depended
al. 2008), and S. serrulatus from Mexico and on the clutch size.
Guatemala (EU702312, Elias-Gutierrez et al.
2008). COI gene sequences were aligned by eye Molecular analysis of COI sequences
using the BioEdit program vers. 7.0.2 (Hall 1999).
We calculated the haplotype diversity (Hd, Nei We used 110 COI sequences from S. vetulus
1987), nucleotide diversity (π, Nei 1987), genetic (n = 22), S. vetuloides (n = 28), S. mixtus (n = 10),
distance (D xy, Nei 1987), and average genetic S. serrulatus (n = 7), S. heilongjiangensis (n = 19),
distances between each pair of species using Daphnia similoides (n = 5), and Diaphanosoma
MEGA 3 vers. 3.0 (Kumar et al. 2004). Daphnia dubium (n = 7) for the phylogenetic analysis. Each
similoides and Diaphanosoma dubium were used sequence was 641 bp long. Twelve haplotypes
as outgroups, and the phylogenetic tree was were detected for the 5 species of Simocephalus
derived using all sequences by the Neighbor- with 151 segregation sites; the genetic diversity,
joining (NJ) and maximum-parsimony (MP) Hd, was 0.891, and the nucleotide diversity, π, was
methods (Saitou and Nei 1987) based on Kimura 0.07049. Simocephalus vetulus had 4 haplotypes
2-parameter (K2P) distances with 1000 bootstraps from 8 sites (Hd = 0.576), S. vetuloides had 6
using MEGA 3. 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
RESULTS (Hd = 0.571), and S. heilongjiangensis had 3
haplotypes from 6 sites (Hd = 0.374) (Table 1).
Morphometric comparisons of Simocephalus Genetic distances (Dxy) between each pair of
vetulus, S. vetuloides, and S. mixtus species based on the COI gene ranged 0.00671-
0.1604 (Table 2). Genetic distances among S.
The KMO value for the morphometric data vetulus, S. vetuloides, and S. mixtus were all
matrix was 0.81, and Bartlett’s X 2 was 2583.96 < 0.01, while those between S. serrulatus and the
(d.f. = 276; p = 0.000), demonstrating the other species were > 0.15, and those between
suitability of the PCA. After the PCA, 91% of
the variance was explained by the 1st, 2nd, and 3 S. vetulus S. mixtus S. vetuloides
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 0
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 -1
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 -2
size was r = 0.70 (p = 0.000). The relationship -4 -3 -2 -1 0 1 2 3 4
between clutch size (Y) and dorsal valve angle (X) PCA 2
fit a quadratic function Y = 0.0088X2 - 0.9091X +
Fig. 3. Results of the principal component analysis of the
25.3361 (r 2 = 0.53), and the one between clutch morphometric dataset: 1st and 2nd principle component plot.
size (Y) and body length (X) also fit a quadratic Simocephalus vetulus and S. mixtus were well separated with
function Y = 9.81X2 - 20.48X + 12.00 (r 2 = 0.49). a distribution gap, while S. vetuloides filled the gap and mixed
Hence, irrespective of the species, clutch size was with those 2 species.
Young et al. – Study of Simocephalus from Taiwan 227

S. heilongjiangensis and the other species were h-j ) were well separated, with each group being
> 0.14. supported by a 99% bootstrap value. The dorsal
In the phylogenetic NJ tree (Fig. 4), S. valve shape variation was not associated with
vetulus, S. vetuloides, and S. mixtus (hap a-g) genetic differences based on the COI gene. The
were mixed together as a well-supported group most protruding valve shape (S. mixtus) was
with a bootstrap value of 99%. Simocephalus common in haplotypes a and b. Valve shapes of
serrulatus (hap k-l) and S. heilongjiangensis (hap 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 of the scale of intercontinental divergence for the
by all 3 morphospecies (Fig. 5, Table 1). We cladoceran genus Daphnia, Adamowicz et al.
reconstructed the phylogenetic trees by including (2009) observed a pairwise sequence divergence
both our sequences and downloaded sequences, within the D. obtusa complex of up to a maximum
and obtained NJ and MP phylogenetic trees with of 16.9%, with divergences of up to 19% within
similar tree structures (Fig. 6). Haplotypes a-g the D. longispina complex. In our study, S.
from Taiwan were all placed in the same group. serrulatus and S. heilongjiangensis showed 14%-
16% COI divergence from each other and from
Simocephalus sensu stricto. These interspecific
DISCUSSION differences were similar to most crustaceans
(Hebert et al. 2003).
DNA barcoding can be helpful in species Based on the morphological differences
identification within cryptic species groups (Hebert described by Orlova-Bienkowskaja (2001), 3
et al. 2004, Belyaeva and Taylor 2009). In general, species - S. vetulus, S. vetuloides, and S. mixtus
sequence divergences are much lower among - were previously recorded in Taiwan. Indeed, our
individuals of a species than between closely morphometric analysis of the valve shape revealed
related species. For example, congeneric species a significant difference between S. vetulus
of moths exhibit an average sequence divergence and S. mixtus from Taiwan, which appeared
of 6.5% in the mitochondrial COI gene, whereas to support their taxonomic status as different
divergences among conspecific individuals species. However, when all 3 putative species
average only 0.25% (Moore 1995, Hebert et al. were included in the analysis, the PCA did not
2004). Similar values were obtained in birds, separate S. vetulus, S. vetuloides, and S. mixtus
with intraspecific divergences of COI averaging from one another, as they formed a morphological
0.27%, whereas congener divergences averaged continuum. This is consistent with a single
7.93% (Hebert and Stoeckle et al. 2004). Among morphologically variable species. Furthermore,
1781 congeneric species pairs of crustaceans, differences in valve shape among S. vetulus, S.
only 1.3% had COI gene divergences of < 2%, vetuloides, and S. mixtus collected in Taiwan were
13.4% had COI gene divergences ranging 4%- not associated with genetic variations. The genetic
8%, and 81.8% had COI gene divergences distances in COI among them were very small
ranging 8%-32% (Hebert et al. 2003). In a study (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
99 Daphnia similoides
Diaphanosoma dubium


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.
Young et al. – Study of Simocephalus from Taiwan 229


Hap d

sf-1 ha-5
Hap a
scf-1 dgA-1
hs3-1 dgB-1
Hap e

Hap f
mn-2 dd-1
tt-2 Hap b gA-1

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 (*)

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
Simocephalus cf. punctatus (#) 85 Simocephalus cf. punctatus (#)
Simocephalus punctatus (#) Simocephalus punctatus (#)
Simocephalus cf. exspinosus (#) 63 Simocephalus cf. exspinosus (#)
Simocephalus cf. exspinosus (#) Simocephalus cf. exspinosus (#)
Simocephalus cf. mixtus (#) Simocephalus cf. mixtus (#)
Simocephalus cf. mixtus (#) Simocephalus cf. mixtus (#)
Hap k (*) 99 Hap k (*)
62 Hap l (*)
Hap l (*)

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 Belyaeva and Taylor 2009, Abreu et al. 2010), our
to treat all morphotypes of Simocephalus sensu results indicate that S. cf. vetulus from Taiwan
stricto from Taiwan as a single species, S. cf. is probably not the same species as S. vetulus
vetulus, as the publication time of S. vetulus was from the UK, and S. serrulatus from Taiwan is not
earlier than those of the other 2 species. conspecific with S. cf. serrulatus from Mexico.
COI sequence comparison of S. cf. vetulus Simocephalus cf. vetulus from Taiwan appears to
from Taiwan with the European S. vetulus showed be geographically isolated from populations on
that these were not conspecific (Fig. 6). As no other continents. Future studies should collect
sequences of S. mixtus or S. vetuloides from the barcodes of all morphospecies of Simocephalus
areas of their primary distribution were available from different locations around the world in order
for comparison, it remains unclear whether the to reconstruct their systematic relationships.
species found in Taiwan are conspecific with those
species. It is possible that Simocephalus found Acknowledgments: We thank the National Sci-
in Taiwan is either S. mixtus or S. vetuloides or a ence Council of Taiwan for their grant (NSC87-
new undescribed species. Future studies should 2311-B-134-001) to support part of this work. We
compare sequences of S. vetulus, S. mixtus, and are very grateful to the anonymous reviewers for
S. vetuloides collected from the type locations with their critical and constructive comments on our
sequences of S. cf. vetulus from Taiwan to verify manuscript.
its taxonomic status.
According to allozymic studies by Hann
(1995), intraspecific differentiation within S. cf. REFERENCES
vetulus in North America was very slight. North
American and European populations were gene- Abreu MJ, MJ Santos-Wisniewski, O Rocha, TC Orlando.
tically distinct according to the allozyme data, but 2010. The use of PCR-RFLP to genetically distinguish
the morphologically close species: Ceriodaphnia dubia
no morphological distinctiveness was identified. Richard, 1894 and Ceriodaphnia silvestrii Daday, 1902
In the past, conspecific populations from different (Crustacean Cladocera). Braz. J. Biol. 70: 121-124.
continents were believed to be widespread Adamowicz SJ, A Petrusek, JK Colbourne, PDN Hebert, JDS
within the Cladocera based on morphological Witt. 2009. The scale of divergence: a phylogenetic
identifications. An intercontinental distribution of appraisal of intercontinental allopatric speciation in a
passively dispersed freshwater zooplankton genus. Mol.
a species is generally presumed to be a result Phylogen. Evol. 50: 423-436.
of passive transport by migratory birds or other Belyaeva M, DJ Taylor. 2009. Cryptic species within the
dispersal mechanisms (Dumont and Negrea 2002, Chydorus sphaericus species complex (Crustacea:
Adamowicz et al. 2009). The alternative hypothesis Cladocera) revealed by molecular markers and sexual
of geographical isolation assumes that gene flow stage morphology. Mol. Phylogen. Evol. 50: 534-546.
Blaxter ML. 2004. The promise of a DNA taxonomy. Phil.
among populations of cosmopolitan species on Trans. R. Soc. B Biol. Sci. 359: 669-679.
different continents is interrupted, and therefore Boonsom J. 1984. The freshwater zooplankton of Thailand
the question is how large their genetic divergence (Rotifera and Crustacea). Hydrobiologia 113: 223-229.
is relative to the geographical dis-continuum Chen CS, CH Tzeng, TS Chiu. 2010. Morphological and
scale. For example, Xu et al. (2009) explored the molecular analyses reveal separations among
spatiotemporal populations of anchovy (Engraulis
global phylogeography of the non-cosmopolitan japonicus) in the southern East China Sea. Zool. Stud.
freshwater cladoceran Polyphemus pediculus 49: 270-282.
(Linnaeus, 1761) (Crustacea, Onychopoda) using Chiang SC, NS Du. 1979. Freshwater Cladocera. Fauna
2 mitochondrial genes, COI and 16s ribosomal (r) Sinica. Crustacea. Beijing, China: Science Press,
RNA, and 1 nuclear marker, 18s rRNA. The P. Academia Sinica, 297 pp.
Costa FO, JR deWarrd, J Boutillier, S Ratnasingham, RT
pediculus complex represents an assemblage of at Dooh, M Hajibabaei, PDN Hebert. 2007. Biological
least 9 largely allopatric, cryptic species. The Far identifications through DNA barcodes: the case of the
East harbors exceptionally high levels of genetic Crustacea. Can. J. Fish. Aquat. Sci. 64: 272-295.
diversity at both the regional and local scales. Du NS. 1993. Crustacean biology. Beijing, China: Science
In contrast, little genetic subdivision is apparent Press, 1003 pp. (in Chinese)
Elias-Gutierrez M, FM Jeronimo, NV Ivanova, M Valdez-
across the formerly glaciated regions of Europe Moreno, PDN Hebert. 2008. DNA barcodes for
and North America. Cladocera and Copepoda from Mexico and Guatemala:
Similar to Xu et al. (2009) and many other highlights and new discoveries. Zootaxa 1893: 1-42.
previous studies on cosmopolitan cladoceran Dumont HJ, SV Negrea. 2002. Introduction to the class
species (Ishida et al. 2006, Rowe et al. 2007, Branchiopoda. (No. 19) Guides to the identification of the
Young et al. – Study of Simocephalus from Taiwan 231

microinveryebrates of the continental waters of the world. zooplankton in Japan. Hiroshibacho, Suita-shi, Osaka,
Leiden, the Netherlands: Backhuys Publishers, 398 pp. Japan: Hoikusha Publishing, 532 pp.
Flössner D. 1972. Krebstiere, Crustacea. Kiemen- und Moore WS. 1995. Inferring phylogenies from mtDNA variation:
Blattfüsser, Branchiopoda. Fischläuse, Branchiura. mitochondrial-gene trees versus nuclear-gene trees.
Tierw. Deutschl. 60. Gustav Fischer Verlag, Jena. 501 p. Evolution 49: 718-726.
(in German) Negrea S. 1983. Cladocera. Fauna Republicii Socialiste
Folmer O, M Black, W Hoeh, R Lutz, R Vrijenhoek. 1994. DNA Romania. Volume IV. Fascicula 12. Cladocera. Editura
primers for amplification of mitochondrial cytochrome c Academiei Republicii Romania. Bucuresti, 399p.
oxidase subunit I from diverse metazoan invertebrates. Nei M. 1987. Molecular evolutionary genetics. New York:
Mol. Mar. Biol. Biotechnol. 3: 294-299. Columbia Univ. Press.
Fryer G. 1957. Freeliving freshwater Crustacea from Lake Ni MH. 2005. Systematic studies of Simocephalus (Cladocera:
Nyassa and adjoining waters. Part II. Cladocera and Daphnnidae) from Taiwan. Master’s thesis, Department
Conchostraca. Arch. Hydrobiol. 53: 223-239. of Applied Science, National Hsinchu Univ. of Education,
Hall TA. 1999. BioEdit: a user-friendly biological sequence Hsinchu, Taiwan, 69 pp.
alignment editor and analysis program for Windows 95/98/ Orlova-Bienkowskaja MY. 1998. A revision of the clarocen
NT. Nucleic Acids. Symp. Ser. 41: 95-98. Genus Simocephalus (Crustacea: Daphniidae). Bull. Nat.
Hann BJ. 1995. Genetic variation in Simocephalus Hist. Mus. Lond. (Zool.) 64: 1-62.
(Anomopoda: Daphniidae) in North America: patterns and Orlova-Bienkowskaja MY. 2001. Cladocera: Anomopoda
consequences. Hydrobiologia 307: 9-14. Daphniidae: genus Simocephalus. Leiden, the Nether-
Harding JP. 1961. Some South African Cladocera collected by lands: Backhuys, 130 pp.
Dr. A. D. Harrison. Ann. S. Afr. Mus. 46: 35-46. Rajapaksa R, CH Fernando. 1982. The Cladocera of Sri Lanka
Hebert PDN, CA Ball, SL deWaard Jr. 2003. Biological identi- (Ceylon), with remarks on some species. Hydrobiologia
fications through DNA barcodes. Proc. R. Soc. Lond. B 94: 49-69.
Biol. Sci. 270: 313-321. Rowe CL, SJ Adamowicz, PDN Hebert. 2007. Three new
Hebert PDN, EH Penton, JM Burns, DH Janzen, W Hallwachs. cryptic species of the freshwater zooplankton genus
2004. Ten species in one: DNA barcoding reveals cryptic Holopedium (Crustacea: Branchiopoda: Ctenopoda),
species in the Neotropical skipper butterfly Astraptes revealed by genetic methods. Zootaxa 1656: 1-49.
fulgerator. PNAS 101: 14812-14817. Rylov WM. 1930. Cladocera et Copepoda in Abhandlungen
Hebert PDN, S Ratnasingham, SL deWarrd JR. 2003. Bar- der Pamir-Expedition 1928. II. Zoologie: 105-133.
coding animal life: cytochrome c oxidase subunit 1 Saitou N, M Nei. 1987. Neighbor-joining method. Mol. Biol.
divergences among closely related species. Proc. R. Soc. Evol. 4: 406-425.
Lond. B 270(Supplement): 96-99. Sharam BK. 1978. A note on freshwater cladocerans from
Hebert PDN, MY Stoeckle, TS Zemlak, CM Francis. 2004. West Bengal. Bangladesh J. Zool. 6: 139-151.
Identification of birds through DNA barcodes. Plos Biol. 2: Sars GO. 1903. On the crustacean fauna of central Asia. Part
1657-1663. II. Appendix. Cladocera. Ann. Mus. Zool. Acad. Sci. St.
Hemsen J. 1952. Ergebnisse Österreichischen Iran Expedition Petersburg 8: 157-264.
1949/50. Cladoceren und freilebende Copepoden, usw. Sars GO. 1916. The freshwater Entomostraca of Cape
Österr. Akad. Wiss. Mathem.-nat. Kl. Abt. I 161: 585-644. Province (Union of South Africa) Part I. Cladocera. Ann.
(in German) S. Afr. Mus. 15: 303-351.
Ishida S, AA Kotov, DJ Taylor. 2006. A new divergent lineage of Shi XL, XB Shi. 1996. On the species and distribution of
Daphnia (Cladocera: Anomopoda) and its morphological Simocephalus in Heilongjiang Province, China. Acta
and genetical differentiation from Daphnia curcirostris Zootax. Sin. 21: 263-276. (in Chinese)
Eylmann, 1887. Zool. J. Linn. Soc. 146: 385-405. Tautz D, P Arctander, A Minelli, RH Thomas, AP Vogler. 2003.
Johnson DS. 1953. On some Cladocera from South African A plea for DNA taxonomy. Trends Ecol. Evol. 18: 70-74.
muds. Ann. Mag. Nat. Hist. 12: 923-928. Tuo YS. 2002. Freshwater Cladocera of Taiwan. Master’s
Kim IH. 1998. Key to the Korean freshwater Cladocera. Kor. J. thesis, Institute of Mathematics and Science Education,
Syst. Zool. Spec. Issue 2: 43-65. (in Korean) Hsinchu Education College, Hsinchu, Taiwan, 60 pp.
Kumar S, K Tamura, M Nei. 2004. MEGA 3: integrated Uéno M. 1927. On some freshwater branchiopods from China.
software for molecular evolutionary genetic analysis Annot. Zool. Jpn. 11: 157-163.
sequence alignment. Bioinformatics 5: 150-163. Uéno M. 1966. Cladocera and Copepoda from Nepal. Jpn. J.
Lefébure T, CJ Douady, M Gouy, J Gibert. 2006. Relationships Zool. 26: 95-100.
between morphological taxonomy and molecular Werestschagin G. 1923. Notiz über die Süsswasserfauna
divergence within Crustacea: proposal of a molecular des Pamirs. Bull. I`Inst. Hydrobiol. Russie 6: 1-40. (in
threshold to help species delimitation. Mol. Phylogen. German)
Evol. 40: 435-447. Xu S, PDN Hebert, AA Kotov, ME Cristescu. 2009. The non-
Manuilova EF. 1964. Vetvistousye rakoobraznye fauny SSSR. cosmopolitanism paradigm of freshwater zooplankton:
Opredeliteli po faune SSSR 88. Moscow and Leningrad: insights from the global phylogeography of the predatory
Nauka, 327 pp. (in Russian) cladoceran Polyphemus pediculus (Crustacea,
Mayer E, PD Ashlock. 1991. Principles of systematic zoology, Onychopoda). Mol. Ecol. 18: 5161-5179.
2nd ed. New York: McGraw-Hill, pp. 42-51. Yoon SM, HS Kim. 1987. A systematic study on the freshwater
Michael RG, BK Sharma. 1998. Indian Cladocera: fauna of Cladocera from Korea. Kor. J. Syst. Zool. 3: 175-207. (in
Indian and adjacent countries. Shillong, Meghalaya, Korean)
India: Department of Zoology, North-Eastern Hill Univ., Yoon SM, W Kim. 2000. Taxonomic review of the cladoceran
262 pp. genus Simocephalus (Branchiopoda, Anomopoda,
Mizuno T, E Takahashi. 1991. An illustrated guide to freshwater 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,

(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.

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

Azman and Othman – A New Genus and Species of Aoroid Amphipod 233

The following abbreviations are used: A, setae and a few long, slender setae. Gnathopod
antenna; ABD, abdomen; G, gnathopod; HD, head; 1 enlarged in both sexes, or only in males; coxa 1
l, left; LL, lower lip; MD, mandible; MX, maxilla; enlarged, larger than coxa 2. Merus of gnathopod
MP, maxilliped; P, pereopod; PL, pleopod; r, right; 2 not enlarged. Pereopods 5-7 without accessory
T, telson; U, uropod; UR, urosome; UL, upper spines on anterior margin. Pereopod 7 longer or
lip; ♂ , male; ♀, female. The type materials of much longer than pereopod 6. Urosomites not
the new species are deposited at the Universiti coalesced. Uropods 1 and 2 without a dense array
Kebangsaan Malaysia Muzium Zoologi (UKMMZ), of robust setae. Peduncle of uropod 3 relatively
Bangi, Malaysia. short, length usually ≤ 2 times breadth; with 2, 1,
or no rami. Telson without hooks or denticles.

RESULTS Aoridae Stebbing, 1899

Corophiida Leach, 1814 Diagnosis: Anteroventral margin of head

moderately excavate. Pereopod 7 very elongate,
Aoroidea Stebbing, 1899 entire propodus extending beyond pereopod 6.

Diagnosis (description based on Myers Grandidierella Coutière, 1904

and Lowry 2003): Head rectangular, anterodistal
margin recessed, lateral cephalic lobe weakly Diagnosis: Eyes small to medium. Accessory
extended, eye, if present, situated proximal to flagellum of antenna 1 minute, 1-segmented.
lobe; anteroventral margin weakly recessed, Inner plate of maxilla 1 vestigial. Coxae very
moderately excavate. Mandible palp 3-articulated small, relatively short, of various sizes and shapes.
or absent, article 3, when present, asymmetrical, Gnathopod 1 (male) complexly subchelate and
distally rounded, with setae extending along most much larger than gnathopod 2. Gnathopod 2
of posterodistal margin, or approximately parallel- subchelate. Dactylus of pereopods 6 and 7
sided with distal setae only; posterior margin with elongate, falcate. Uropods 1 and 2 biramous;
setae of variable length, or with comb of short rami slightly subequal; peduncle with ventrodistal
process. Uropod 3 uniramous. Telson entire.
Species composition: Grandidierella contains
40 species of G. africana Schellenberg, 1936; G.
N 1 km bispinosa Schellenberg, 1938; G. bonnieroides
Stephensen, 1948; G. cabindae (Schellenberg,
THAILAND KLEBANG 1925); G. chelata K.H. Barnard, 1951; G.
chaohuensis Hou and Li, 2002; G. dentimera
Port of Sungai Udang 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
Strait of Malacca
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;
SUMATERA 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
Fig. 1. Map showing the sampling area. Morino and Dai, 1990; G. teres Myers, 1981; G.
234 Zoological Studies 51(2): 232-247 (2012)

trispinosa Bano and Kazmi, 2010; G. unidentata 2 (MX2, Fig. 3) broad medially, pointed distally,
Ren, 2006; and G. vietnamica Dang, 1968. outer margin naked; outer plate extending just
beyond inner one, both outer and inner margins
Grandidierella melakaensis sp. nov. naked. Inner plate of maxilliped (MP, Fig. 3)
(Figs. 2-5) elongate, extending well beyond proximal article
of palp, medially narrow, apically truncate; outer
Material examined: Holotype. ♀, Malaysia, plate almost reaching end of palp article 2, inner
Strait of Malacca, Melaka, Port of Sungai margin straight and outer margin evenly convex,
Udang, (2°14'3"N, 102°7'43"E), 17 m, muddy dense bristles on outer margin; palp consisting of
bottom, 26 May 1995, C. Zaidi, M. Soed, S. 4 articles, article 4 small, subtriangular, tapering to
Zuhaimi (Smith-McIntyre grab). UKM I.D. 3611 truncate tip and ending in stout spine. Pereonites
(UKMMZ-1273). Allotype. ♂ , data same as for 1-5 subequal to each other in length, 6 and 4 of
holotype. (UKMMZ-1274). Paratypes. Data same equal length, and 5-7 deeper than preceding ones,
as for holotype, UKMMZ-1275 (2 ♂♂ , 2 ♀♀); pereonite 1 anteroventrally roundly produced.
UKMMZ-1276 (3 ♂♂ , 6 ♀♀); UKMMZ-1277 Coxal plates small, shallow, separated.
(5 ♂♂ , 3 ♀♀). Gnathopod 1 (G1 ♀, Fig. 2) subequal in
Description: Female (holotype). Total body size with gnathopod 2, length ratio of articles from
length 2.6 mm (from tip of rostrum to apex of basis to dactylus approximately 16: 3: 4: 15: 9: 7;
telson). Head (HD, Fig. 3) with short, pointed basis stout, anterior margin straight; ischium short,
rostrum, about as long as pereonites 1 and 2 subrectangular, anterior margin distally weakly
combined, with triangular-shaped anterior head produced and naked; merus slightly longer than
lobe, inferior antennal sinus deep, beyond middle ischium, distally tapering to become subtriangular,
of head. Eye small, oval, placed just behind posterior margin and submargin throughout with
anterior head lobe. Antenna 1 (A1, Fig. 3) much numerous setae which are peculiarly very long and
longer than antenna 2, with peduncle longer than bristly; carpus about as long as basis, elongate,
flagellum, length ratio of 9: 11: 4; flagellum shorter posterior margin weakly convex but crenulate and
than peduncle, composed of 15 articles, distal one both its margin and submargin throughout densely
of which vestigial, each article distally provided covered with very long bristly setae; propodus
with tuft of long and short setae. Antenna 2 (A2, narrower and slightly longer than 1/2 of carpus,
Fig. 3) short and stout, 4-segmented in ratio of 5: slightly curved but with uniform width, densely
7: 16: 14; 1st and 2nd peduncular articles very covered with very long setae both anteriorly
short, their combined length subequal to article and posteriorly; dactylus shorter than propodus,
3, broader than those of antenna 1; flagellum stout, falcate, tapering to pointed tip, grasping
very short, slightly longer than 1/2 length of margin minutely serrated medially. Length ratio
peduncular article 4, 3-articulate, all articles of articles of gnathopod 2 (G2 ♀, Fig. 2) from
setiferus, distalmost article apically armed with 2 basis to dactylus approximately 14: 3: 4: 9: 12: 3;
stout spines surrounded by a tuft of setae. Apical brood plate narrow and elongate, about as broad
margin of upper lip (UL, Fig. 3) broad, slightly as basis and about 1/2 as long as gnathopod 2;
concave medially, bearing minute bristle. Inner basis elongate and parallel-sided; ischium short,
plate of lower lip (LL, Fig. 3) developed, broad with distally slightly produced anterior margin and
and angular, minutely pubescent, outer plate with naked posterior margin; merus slightly longer than
rounded shoulder, densely pubescent, and with ischium, subcircular, as long as broad; carpus
strongly developed, rounded mandibular process. shorter than propodus, naked along its length;
Incisor of mandible (MD, Fig. 3) well-developed, propodus elongate, as broad as and subequal
with 4 teeth on left mandible and 5 teeth on right to basis in length, palm transverse, defined by 3
one; lancinia mobilis armed with 4 teeth on both stout spines, palm margin possessing some robust
left and right mandibles; accessory blades 8 setae; dactylus stout, short, as long as palm, claw-
on left mandible and 7 on right one; right molar like, grasping margin with a hump near proximal
process developed, with circular apex, fringed with end. Pereopod 3 (P3, Fig. 4) longer than pereopod
apically branched processes; palp triarticulate. 4; brood plate elongate and lanceolate; length ratio
Inner plate of maxilla 1 (MX1, Fig. 3) small and of articles from basis to dactylus approximately
short, with setae; outer plate distally truncate; 13: 3: 6: 4: 5: 8; basis linear; ischium short,
palp biarticulate, extending slightly beyond outer subrectangular, anterior margin medially concave;
plate, with rounded apex. Inner plate of maxilla 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)








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






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

U1 U3





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; exilis also differ.

propodus much shorter and narrower than carpus, Etymology: The new species of Grandidierella
proximally and distally broader than medial part, is named after its type locality, Melaka as
anterior margin uneven, posterior concave margin melakaensis.
medially produced forming strong and apically
blunt process, throughout its length covered with Unciolidae Myers and Lowry, 2003
several plumose setae; dactylus rather stout,
somewhat straight, proximally wider and tapering Diagnosis (description from Myers and
to blunt tip, grasping margin proximally bearing Lowry 2003): Anteroventral margin of head
single small tooth and subapically with 2 pairs of moderately excavate, or strongly excavate for
small teeth, anterior margin with pair of setae near receiving enlarged antenna 2. Pereopod 7 not
proximal end. Gnathopod 2 (G2 ♂ , Fig. 2) in both very elongate, entire propodus not extending
female and male rather similar except for length beyond pereopod 6. Included subfamilies/genera.
of basis which in male is distinctly longer than Acuminodeutopinae: Acuminodeutopus J.L.
propodus (1.4 times as long as propodus). Barnard, 1959; Klebang gen. nov.; Rudilemboides
Remarks: The genus Grandidierella Coutière, J.L. Barnard, 1959; and Wombalana Thomas
1904, a member of amphipods of the family and Barnard, 1991. Unciolinae: Dryopoides
Aoridae, is characterized by a subcylindrical Stebbing, 1888; Janice Griffiths, 1973; Liocuna
body, small- to medium-sized eyes, small Myers, 1981a; Neohela Smith, 1881; Orstomia
coxae, gnathopod 1 larger than gnathopod 2, Myers 1998; Pedicorophium Karaman, 1981;
male gnathopod 1 carpochelate, and with a Pseudunciola Bousfield, 1973; Pterunciola Just,
uniramous uropod 3. According to Myers (1970), 1977; Ritaumius Ledoyer, 1978; Rildardanus J.L.
Grandidierella presumably originated from the old Barnard, 1969; Uncinotarsus L’Hardy and Truchot,
Tethys Sea and is considered to have a tropical 1964; Unciola Say, 1818; Unciolella Chevreux,
affinity. Records of this genus appear scattered 1911; and Zoedeutopus J.L. Barnard, 1979.
throughout the Caribbean Sea to Madagascar, Remarks: Myers and Lowry (2003) esta-
Tanzania, and India (Myers 1970). To date, the blished the family Unciolidae and included it
genus Grandidierella is known to contain 40 together with the existing Aoridae Stebbing
species, with recent additions by Ren (2006), in the superfamily Aoroidea. It can easily be
Myers (2009), Bochert and Zettler (2010), and distinguished by a combination of characters that
Bano and Kazmi (2010). The present work includes a moderate or strong excavation along
reports on the 1st record of this genus occurring in the anteroventral margin of the head for receiving
Malaysian waters, along the coast of the state of the enlarged antenna 2; antenna 1 article 3 short,
Melaka, Peninsular Malaysia. ≤ 1/2 the length of article 2; an enlarged gnathopod
Grandidierella melakaensis sp. nov. can be 1; pereopods 5, 6, and 7 in a regular length
easily distinguished from all other species in the progression; and all urosomites free. Currently,
genus by a set of characters known only in this the Unciolidae is composed of the 2 subfamilies
species: (1) an obviously projecting rostrum, (2) of the Acuminodeutopinae with 3 genera and
pereopod 5 having a merus and ischium of equal the Unciolinae with 14 genera and is distributed
lengths, and (3) epimerons 1 and 2 with long worldwide in both cold and warm waters.
plumose setae posteroventrally. Nonetheless,
the specimens examined resemble G. elongata
in having a triangular ocular lobe; and G. exilis, Klebang gen. nov.
G. gilesi, G. mahafalensis, G. palama, and G.
indentata in bearing several very long plumose Type species: Klebang barnardi sp. nov., present designation.
setae on the propodus, carpus, and merus of Included species: K. barnardi sp. nov.
gnathopod 2 of both the male and female and
possessing a single posterodistal spine on male Diagnosis: Rostrum short, ocular lobes
gnathopod 1, but clearly differ in many other moderate, produced to front, pointed. Eyes
respects, especially in the form of gnathopod 1. moderate. Antenna 1 slightly longer than antenna
Furthermore, the inflated uropod 3 peduncle and 2, both slender; peduncular article 3 slightly shorter
very short mandibular palp article 1 in G. elongata, than article 1, article 2 longest, accessory flagellum
the much inflated uropod 3 peduncle in G. gilesi, present. Peduncular article 3 of antenna 2 short,
and the ventral pereon process on pereonite 1 in G. flagellum with only 3 or 4 articles. Mandibular palp
Azman and Othman – A New Genus and Species of Aoroid Amphipod 241

setose; article 2 longest. Male gnathopods 1 and Pantai Klebang, Melaka, Malaysia the general area
2 subequal, subchelate, and carpochelate. Outer in Melaka where this genus was discovered.
ramus of uropod 1 with brush setae. Uropod 3
uniramus; peduncle short; ramus elongate with Klebang barnardi sp. nov.
robust setae on both margins. Telson semicircular (Figs. 6-8)
and lobed.
Remarks: The diagnosis of the new genus Material examined: Holotype. ♂ , Malaysia,
is based on the type-species described be- Strait of Malacca, Melaka, Port of Sungai Udang,
low. Klebang gen. nov. is closely related to St. CS, Petronas (2°14'43"N, 102°6'53"E), 20 m,
Grandidierella Coutière, from which it shares muddy bottom, 22 Oct. 2003, C. Zaidi, M. Soed,
several generic characters in having a sub- S. Zuhaimi (Smith-McIntyre grab). UKM I.D.
cylindrical body, an enlarged carpochelate 7187 (ref: UKMMZ-1350). Paratypes. From the
gnathopod 1, free urosomites, and a uniramus same sample as holotype, UKMMZ-1352 (7 ♂♂ );
uropod 3. A careful examination of the newly UKMMZ-1353 (4 ♂♂ ); UKMMZ-1354 (8 ♂♂ ).
acquired material on the other hand, although Description: Male (holotype). Total body
closely similar morphologically to Grandidierella, length 6.7 mm (from tip of rostrum to apex of
suggests that it represents a new genus in the telson). Body rather slender. Head (HD, Fig. 6)
Aoroidea. Myers and Lowry (2003) provided broader and deeper than pereonite 1; rostrum
a valuable updated key to the families and not developed, anterior lateral head lobe (ocular
subfamilies of the Corophiidea. Some key chara- lobe) extending forward and anteriorly pointed in
cters show that our material naturally fits into the triangular shape; inferior antennal sinus deep and
Acuminodeutopinae, like the short article 3 of straight vertically; eye distinct and located behind
antenna 1 at ≤ 1/2 the length of article 2, uropod anterior head lobe. Antenna 1 (A1, Fig. 6) slightly
3 lacking recurved robust setae, gnathopods 1 longer than antenna 2, ratio of peduncular articles
and 2 not together forming a sieving basket, free 1-3 as 1.1: 1.5: 1; article 1 with 4 postero-marginal
urosomites, an enlarged gnathopod 1, pereopods setae; flagellum with 5 articles, 2 times as long
5, 6, and 7 in a regular length progression, and as peduncle; accessory flagellum uni-articulate,
most importantly the acute lateral cephalic lobes short. Peduncular article 3 of antenna 2 (A2, Fig.
of the head. As shown by the excellent series of 6) with 3 long and 1 short setae posterodistally;
head drawings of selected genera in Myers and article 4 slightly shorter than article 5 with row of
Lowry (2003), the acute head cephalic lobes are long setae along posterior margin; flagellum short,
of special importance in the classification of this composed of 4 articles. Labrum of upper lip (UL,
group (Fig. 4 in Myers and Lowry 2003). Currently, Fig. 7) broad, its apical margin weakly concave
the acuminodeutopine clade includes only the 3 mid-ventrally and pubescent on each lobe. Inner
genera of Acuminodeutopus, Rudilemboides, and plates of lower lip (LL, Fig. 7) highly developed
Wombalano, and all 3 share the characteristic and subtriangular, mandibular process narrow but
of having the acute, triangular, lateral cephalic well-developed; outer plates with bristly shoulders.
lobes. Clearly within this clade only Wombalano Both mandibles (MD, Fig. 7) similar to each other
possesses the same distinctive generic characters except for number of accessory blades with 4 on
shown in the Klebang gen. nov. material in right and 5 on left; incisor produced to interior,
having a uniramus uropod 3. However, the broad, with 5 teeth; lacinia mobilis on both sides
unique formation of the male gnathopod 2 (with 4-toothed, followed by 4 or 5 accessory blades;
an expanded basis and carpus) in Wombalano molar process medium, ridged distally and serrate
is an advanced character that separates it from marginally, with a single seta; palp triarticulate.
the Klebang gen. nov. material. At the same Inner plate of maxilla 1 (MX1, Fig. 7) reduced;
time, Klebang gen. nov. is highly distinctive in outer plate with truncate apical margin; palp
having this combination of characters: (1) the extending beyond outer plate, biarticulate. Inner
unique carpal configuration of gnathopod 2, (2) a plate of maxilla 2 (MX2, Fig. 7) slightly shorter than
largely expanded posterior margin of the carpus outer one; outer plate larger than inner one, distally
of gnathopod 1, and (3) the densely setose broadest and with rounded apical margin. Inner
mandibular palp that has not yet been formulated. plate of maxilliped (MP, Fig. 7) short, not extending
Therefore, we consider the current species to be beyond tip of palmer proximal article; outer plate
representative of a new genus. extending beyond 1/2 of palmer article 2, outer
Etymology: The name Klebang refers to margin naked, evenly convex; palp 4-articulated,
242 Zoological Studies 51(2): 232-247 (2012)

terminal article distally tapering and ending in a shorter, coxa plate with ventral margins medially
long nail-like spine-tooth. produced into triangular expansion; length ratio of
Gnathopod 1 (G1, Fig. 6) subchelate and articles from basis to dactylus approximately 15:
carpochelate, subequal to gnathopod 2 in size; 3: 7: 7: 8: 5; and with less setation than pereopod
coxa plate shallow, rhomboidal, smaller than 3. Pereopod 5 (P5, Fig. 8) slightly longer than
others, anteroventral angle markedly produced; pereopod 4; coxa posteroventrally expanded into
length ratio of articles from basis to dactylus long and narrowly obtuse angle; length ratio of
approximately 14: 3: 4: 12: 10: 7; basis linear, articles from basis to dactylus approximately 20:
almost parallel-sided; ischium short, posterior 3: 11: 9: 9: 3; basis slightly expanded in proximal
margin 3 times as long as anterior margin; merus part, about 1/3 as long as pereopod 5; ischium
subrectangular, longer than wide, anterior margin short, anterior margin longer than posterior one;
naked; carpus robust, mainly subrectangular merus longer than carpus, uniform in width, apical
except proximally small and short, with triangular margin anteriorly produced into triangular process;
ending, greatly wider and longer than propodus, carpus subequal to merus in width, parallel-sided
posterodistally 3/4 margin throughout equally except near proximal end; propodus about as long
produced into widely expanded plate which is as carpus; dactylus short, 1/3 as long as propodus,
proximoventrally oblique and distoventrally pointing weakly curved, tapering to pointed tip, bearing 1
forwards and ending in a small tooth; propodus plumose seta at anterior proximal end and 1 thin
shorter and narrower than carpus, anterior evenly spine in middle of grasping margin. Pereopod 6
convex and posterior margin barely concave; (P6, Fig. 8) rather long, extending well beyond
dactylus stout, fairly curved, about 1/2 as long end of telson and uropods; coxa with posteriorly
as carpus, tapering to pointed tip. Gnathopod 2 produced fairly narrow and rounded lobe, anteriorly
(G2, Fig. 6) longer than gnathopod 1, subchelate and anteroventrally rounded; length ratio of
and carpochelate; coxa plate shallow with articles from basis to dactylus approximately 12:
ventral margins medially produced into triangular 2: 10: 5: 7: 4; basis almost linear and uniform in
expansion; length ratio of articles from basis to width, about 1/3 as long as pereopod 6; ischium
dactylus 18: 3: 5: 10: 14: 6; basis linear; ischium short, posterodistally slightly produced; merus 2
shortest of all articles, subrectangular; merus times longer than carpus, distinctly narrower than
slightly longer than ischium, posterodistal angle basis, twisted near distal end; carpus shorter than
with triangular spine-tooth; carpus at mid-length propodus, anterior and posterior margins curved
twice as long as merus but shorter than propodus, forward forming a groove along its length; dactylus
proximally narrowing into triangular end and distally about 1/2 as long as propodus, tapering to sharply
widening with truncate apical margin, distal 1/2 of pointed tip, posterior margin with long slender
posterior margin produced into large elongated spine at 2/3 from proximal end. Pereopod 7 (P7,
expansion which is proximally wider, distally Fig. 8) extremely long, extending well beyond end
tapering to rounded tip, outwardly deflected, of pereopod 6; coxa comparatively shallower;
and reaching near distal margin of propodus; length ratio of articles from basis to dactylus 7: 1:
propodus rather long and stout, proximally narrow 8: 4: 6: 3; basis weakly expanded, 1/4 as long as
and distally obviously broader, palm strongly pereopod 7; ischium short, posterior margin distally
transverse, with minutely serrated marginal spines; slightly produced; merus longer but narrower
dactylus fitting on palm, stout, tapering to pointed than basis; anterior and posterior margins of
tip, anteroproximally with long setae. Pereopod 3 carpus curved to rear forming a groove; propodus
(P3, Fig. 8) thin and elongate; coxa plate shallow elongate and rather slender; dactylus 1/2 as long
with ventral margins medially produced into as propodus, weakly curved, tapering to pointed
triangular expansion; length ratio of articles from tip, with 1 long thin spine at 2/3 from proximal end.
basis to dactylus approximately 20: 3: 8: 9: 10: 8; Epimeron 1 (ABD, Fig. 6) subrectangular,
basis linear, almost uniform in width, 1/3 as long 2 posteriorly evenly rounded, and 3 with roundly
as pereopod 3; ischium short, subrectangular; produced antero- and posteroventral angles.
merus shorter than carpus, anterodistally weakly Urosomites 1-3 in combined length as long as
produced; carpus rectangular; propodus narrower epimeron 3.
than carpus; dactylus rather long, 4/5 as long as Pleopods 1-3 (PL1, PL2, PL3, Fig. 8)
propodus, gently curved, tapering to pointed tip, similar to each other; peduncles cylindrical and
anteroproximally armed with a seta. Pereopod 4 anterodistally with plumose setae, each one
(P4, Fig. 8) rather similar to pereopod 3 but slightly distinctly shorter than inner ramus but equal to
Azman and Othman – A New Genus and Species of Aoroid Amphipod 243







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)










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








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

Schellenberg A. 1936. Zwei neue Amphipoden des Stillen Stebbing TRR. 1908. South African Crustacea (Part IV). Ann.
Ozeans und zwei Berichtungen. Zool. Anzeiger 116: 153- South Afr. Mus. 6: 1-96.
156. Stephensen K. 1938. Grandidierella japonica n. sp. A new
Schellenberg A. 1938. Littoral Amphipoden des Topischen amphipod with stridulating organ from brackish water in
Pazifiks. K. svenska Vetensk Akad. Handl. 16: 1-105. Japan. Annot. Zool. Jpn. 17: 179-184
Shoemaker CR. 1935. A new species of amphipod of the Stephensen K. 1948. Amphipods from Curaçao, Bonaire,
genus Grandidierella and a new record for Melita nitida Aruba and Margarita. Stud. Fauna Curaçao, Aruba,
from Sinaloa, Mexico. J. Wash. Acad. Sci. 25: 65-71. 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
Research Institute and Natural Museum of Senckenberg, Senckenberganlage 25, 60325 Frankfurt am Main, Germany
School of Biology, College of Science, Univ. of Tehran, Tehran, Iran
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 Philyra Leach, 1817. Therefore, the actual number
faunal elements of littoral and sublittoral soft- of leucosiid species recorded from the Persian Gulf
sediment habitats in the Persian Gulf and Gulf of is currently 37. However, the leucosiid fauna of
Oman. They are the most diverse of all brachyuran the Gulf of Oman remains poorly known, and only
families (Stephensen 1946, Titgen 1982, Apel 11 species of this group have been recorded there
2001). Apel (2001) listed 30 leucosiid species from (Nobili 1906, Stephensen 1946, Tan and Ng 1995,
the Persian Gulf plus 2 additional species only Apel 2001). By adding 2 recorded species herein,
known from the Gulf of Oman and commented that the number of known leucosiid species from the
records of 4 more species reported from the region Gulf of Oman rises to 13.
are doubtful. Thus almost 1/6 of all brachyuran Galil (2009), in her recent treatment of
crab species of the Persian Gulf belong to the Philyra, divided the genus into 8 genera. Among
Leucosiidae (Apel 2001). Some additional new those, Hiplyra Galil, 2009, comprises 6 species
species were also described and recorded from the distributed in the Indo-West Pacific: H. elegans
Persian Gulf in recent years, raising the number Gravier, 1920; H. longimana A. Milne Edwards,
of leucosiid crabs in the region to 35 species 1874; H. michellinae Galil, 2009; H. platycheir
(Naderloo and Sari 2005). Recently, Galil (2009) De Haan, 1841; H. sagitta Galil, 2009; and H.
added 2 more species (Hiplyra sagitta Galil, 2009, variegata (Rüppell, 1830). The genus is chara-
and Lyphira perplexa Galil, 2009) when she revised cterized by elongate adult chelipeds with the

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

Naderloo and Apel – Hiplyra in Northern Indian Ocean 249

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

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


1 mm

(C) (D)

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 251

dorsal surface. For the male abdomen, the telson in H. elegans

Female gonopore (Fig. 1E) on inner anterior has 2 distinct elevations proximally on the lateral
edge of sternite 5, nearly round; large membranous portions (Fig. 1D), while that of H. variegata is
oval process directed anteroposteriorly. First smooth, with no elevation (Fig. 7C). In addition,
somite of female abdomen not distinctly trilobate, these 2 species have distinct morphologies of
with granular distal margin. the female gonopore allowing females of these
Remarks: Stephensen (1946) listed subs- congeners to readily be distinguished. While H.
tantial material from the Persian Gulf and Gulf of elegans has a distinct oval membranous process
Oman under the name Philyra variegata (Rüppell, which is directed anteroposteriorly (Fig. 1E), the
1830). We had the opportunity to reexamine most female gonopore of H. variegata has a large
of Stephensen’s (1946) material, compared it with opening (Fig. 7D) on the inner side of a prominent
Rüppell’s type material of H. variegata from the elevation. As Galil (2009) discussed, the 1st
Red Sea, and found that some specimens were somite of the female abdomen of H. variegata
not H. variegata but H. elegans instead. Hiplyra is distinctly trilobate, while this somite is simple
elegans is distinguished from H. variegata by the in females of H. elegans. It must be noted that
carapace shape, morphology of the G1, the form drawings provided by Stephensen (1946: 88, fig.
of the male abdomen, and the gonopore structure 15F-K) clearly depict H. variegata, in particular,
of females. The carapace of H. elegans is slightly the male abdomen which shows a smooth telson
longer than wide (mean CL/CB = 1.05), while the lacking any elevation.
carapace of H. variegata is as long as wide, and Distribution: Madagascar, Gulf of Oman, Sri
even in large specimens is only slightly wider than Lanka.
long. The apical process of G1 in H. elegans is
very small, subdistal, and directed laterally (Fig. Hiplyra ramli sp. nov.
1D), while in H. variegata, the small apical process (Figs. 3, 4)
is completely distal and directed ventrally (Fig. 7A).
Type locality: Al Aqah, Fujairah, east coast of
UAE, Gulf of Oman.
Material examined: Holotype 1 ♂ (SMF
(A) 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
(B) 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
Fig. 2. Hiplyra elegans (Gravier, 1920). Male holotype, CL = immediately behind frontal ridge. Front slightly
9.90 mm, CB = 9.33 mm (ZMK CRU929). (A) dorsal surface; (B) shorter than posterior margin of carapace,
ventral surface. 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 Ischium of 3rd maxilliped slightly longer than
laterally. merus, outer surface nearly smooth, distal margin
Epibranchial margin with small granules; minutely granular; merus long-triangular, outer
anterolateral margin with large granules anteriorly, surface finely granular, granules larger distally
becoming smaller posteriorly; posterolateral margin on outer surface; exopod large, wider distally,
finely granular, granules continuing to posterior outer surface smooth, margins minutely serrate.
margin, small granules below posterior margin. Abdominal sternums granular, granules larger
Anterior margin of efferent channel nearly straight, anteriorly; anterior margin of abdominal sulcus
separated from lateral granulated margin by deep granular.
U-shaped incision. Subhepatic and pterygostomial Male chelipeds (Figs. 3A, 4A) with moderately


1 mm


(C) (D)

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 253

long merus, distinctly shorter than carapace proximally with small granules; somite 6 firmly
breadth (mean ML/CB = 0.72); upper surface merged to fused somites 2-5, not freely movable;
granular proximally; anterior margin with large lateral margin sharply diverging proximally, gently
granules, granules becoming larger medially; converging distally along most of its length,
posterior margin with small granules, proximally prominent elevated arrow-shaped process distally
moderately larger. Anterior lower and upper on outer surface; telson elongate-triangular, slightly
margins of carpus minutely granular. Upper shorter than somite 6, with 2 processes at basis of
surface of manus smooth, faint row of very small lateral portion, margins smooth.
granules on lower portion, extending parallel G1 (Fig. 3C, D) slightly curved laterally,
to lower margin in proximal 1/2; lower margin narrowing medially; apical portion expanded, with
granular; upper margin faintly serrate. Movable long setae on lateral margin, relatively short setae
finger slightly shorter than manus, arched on mesial margin; small apical process directed
medially, cutting edge blade-shaped; immovable dorsally; sperm channel curved on dorsal surface.
finger shorter than movable finger, curved gently Female gonopore (Fig. 3E) on inner ante-
downward; cutting edge with small triangular teeth rior edge of sternite 5, obliquely directed antero-
distally, 2 or 3 distal ones larger; short dense setae dorsally; small membranous oval process on outer
along cutting edge of immovable finger, shorter margin of opening.
distally; short setae along round process of distal Remarks: Hiplyra ramli sp. nov. is a relatively
margin of manus on articulation to movable finger. small-sized species, which is morphologically
Male abdomen (Figs. 3B, 4B) long-triangular, closest to H. sagitta and H. elegans. With regard
scarcely granular; somites 2-5 completely fused, to the lengths of the carapace and male chelipeds,
fused somites proximally with large granules, the new species; however, is clearly distinct from
medially with depression, lateral margins 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
(A) 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
(B) 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
Fig. 4. Hiplyra ramli sp. nov. male holotype, CL = 7.4. CB = 6.8 females, with the ratio of ML/CB in males ranging
(SMF 38466). (A) dorsal surface; (B) ventral surface. 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. Type locality: Near Bushehr, Persian Gulf,

Distribution: Presently only known from the Iran.
Gulf of Oman coast of the UAE (Fujairah). Material examined: Persian Gulf: 2 ♂♂
Etymology: The species is named after the (CL = 9.68, 16.04 mm, CB = 9.45, 14.56 mm)
Arabic word “raml” for sand, since it was collected (ZMK CRU880), St. 32, 7.5 m, N of Kharg I., G.
in sandy substrate. The name is used as a noun Thorson, 23 Mar. 1937; 1 ♂ (CL = 16.67 mm,
in apposition. CB = 15.12 mm), 3 ♀♀ (CL = 15.42-19.07 mm,
CB = 14.09-17.66 mm) (SMF 38392), 22 m,
Hiplyra sagitta Galil, 2009 Kuwait, 28°53'N, 48°24'E, trawl, 24 Apr. 1995, F.
(Figs. 5, 6) Krupp; 2 ♀♀ (CL = 15.55, 17.00 mm, CB = 14.23,
16.13 mm) (SMF 38393), 13-17 m, Kuwait,
Philyra platychira Alcock 1896: 242. 29°10'N, 48°28'E, trawl, 23 Apr. 1995, F. Krupp.
Philyra platycheir Tirmizi and Kazmi 1986: 100, fig. 29.
Philyra variegata Stephensen 1946: 89-93 (not Hiplyra
Diagnosis: Carapace (Fig. 6A) slightly
variegata (Rüppell 1830) (part of the material from st. 27, longer than wide (CL/CB = 1.1); dorsal surface
near Bushehr, Persian Gulf)). finely punctate medially, laterally, and posteriorly.
Hiplyra sagitta Galil 2009: 296-297, 315 (in key), figs. 11, 12A. Anterior margin of efferent channel straight,

(A) (B) (C)

(D) 1 mm

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).
Naderloo and Apel – Hiplyra in Northern Indian Ocean 255

separated from lateral granulated margin by is absent in all other known congeners except
somewhat wide U-shaped incision. Ischium of for the newly described H. ramli sp. nov. In the
3rd maxilliped slightly longer than merus, about latter, the process; however, is distinct from that
1.2-times merus length. Male abdomen (Figs. of H. sagitta in its short and triangular form (Fig.
5C, 6B) elongate-triangular; segment 6 with long 3B). Apart from the 2 discriminative characters
arrow-shaped process, telson narrow and smooth. presented by Galil (2009), we add the feature of
Male chelipeds (Fig. 6) moderately long; movable the distinctive apical part of G1 and the structure
finger about as long as manus, or slightly shorter. of the female gonopore, which separate H. sagitta
Immovable finger with small denticles along cutting from other congeners treated here (discussed
edge, 2 large triangular teeth subdistally, cutting under “Remarks” of the new species, H. ramli sp.
edge with dense short setae. G1 (Fig. 5A, B) nov.).
long, narrow, with moderately large apical process, The holotype and paratypes described by
directed laterally. Female gonopore (Fig. 5D) Galil (2009) are from Stephensen’s (1946) material
small, on inner side of large elevation. examined under H. variegata from Bushehr in the
Remarks: Hiplyra sagitta was recently Persian Gulf.
described from Bushehr in the Persian Gulf by Hiplyra sagitta is one of the largest species in
Galil (2009). She mentioned that H. sagitta differs the genus with the largest male found at Kharg I. in
from its congeners by having a triangular incision the Persian Gulf (CL = 16.04 mm, CB = 14.56 mm)
which separates the anterior margin of the efferent and the largest female recorded from Kuwait in the
channel from the lateral granulated margin (Galil Persian Gulf (CL = 19.07 mm, CB = 17.66 mm).
2009: 297), while this incision is narrow and Distribution: Persian Gulf, India, Andaman
U-shaped in the 3 other species she examined. Sea.
Another distinct character mentioned by Galil
(2009) is the particular arrow-shaped process on Hiplyra variegata (Rüppell, 1830)
the distal part of male abdominal somite 6 which (Figs. 7, 8)

Myra variegata Rüppell 1830: 17, pl. 4-4.

(A) 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
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),
(B) 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.
Fig. 6. Hiplyra sagitta Galil, 2009, male, CL = 16.79, CB Diagnosis: Carapace (Fig. 8A) as long as
= 15.13 mm (SMF 38392). (A) dorsal surface; (B) ventral wide, slightly wider than long; dorsal surface
256 Zoological Studies 51(2): 248-258 (2012)

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

(A) (B) (C)

1 mm

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 257

Persian Gulf and by Nobili (1906) who studied the DISCUSSION

type material of H. variegata (Rüppell 1830). Such
an indentation was seen in the material examined The genus Hiplyra Galil, 2009, was recently
in the present paper. Galil (2009) recorded 2 other separated from Philyra Leach, 1817, using the
characters as discriminative for distinguishing the following characters: apical process of G1 minute,
species from its congeners, including a marbled the presence of a thick fringe of setae on the inner
color pattern of the carapace and 2 triangular margin of the movable finger, and the cheliped
teeth distally on the cutting edge of the immovable merus being longer than the carapace in males
finger. (Galil 2009: 314). The 1st 2 characters clearly
We found 3 additional morphological chara- distinguish Hiplyra from Philyra. The 3rd character
cters which we believe are even more significant is rather confusing as in all known species of
and readily distinguish it from all other congeners Hiplyra, even the type species, H. platycheir (De
treated here. Hiplyra variegata has a unique Haan, 1841), the merus is actually clearly shorter
G1 structure and female gonopore, and its male than the carapace.
abdomen is completely smooth with no elevation Hiplyra currently includes 7 species, which are
or process on the 6th male abdominal somite primarily distinguished from each other using the
or telson (Fig. 7A-D). Detailed discussions on morphology of the carapace and male abdomen
the differences between H. variegata and its (Galil 2009). Apart from these 2 characters, 2
congeners are presented under “Remarks” of the more-important discriminative characters including
former species. G1 and the female gonopore were found to be
Distribution: Kenya, Red Sea, Gulf of Aden, useful here to separate closely related species.
Persian Gulf, Gulf of Oman. 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
(A) 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

(B) 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
Fig. 8. Hiplyra variegata (Rüppell, 1830), male lectotype, - Male telson completely smooth; G1 narrowing distally,
CL = 7.66, CB = 7.73 mm (SMF 11121). (A) dorsal surface; (B) apical process located distally, directed ventrally; 1st somite
ventral surface. of female abdomen trilobate .................... Hiplyra variegata
258 Zoological Studies 51(2): 248-258 (2012)

Acknowledgments: We are grateful to J. Olesen by Professor Herdman at Ceylon, in 1902. In WA

(ZMK) for kindly providing us with the valuable Henderman ed. Report to the Government of Ceylon on
the Pearl Oyster Fisheries of the Gulf of Manaar. Part v.
brachyuran material collected from the “Danish Supplementary Report 40: 349-432, pls. 1, 2.
Scientific Expedition in Iran” conducted in 1937/38. Laurie RD. 1915. Reports on the marine biology of the
We are indebted to Prof. M. Türkay (SMF) for Sudanese Red Sea. XXI. On the Brachyura. J. Linn.
his support as supervisor of both authors, and Soc. Lond. 31: 407-475, figs. 1-5, pls. 42-45.
to Deutscher Akademischer Austausch Dienst Naderloo R, A Sari. 2005. Iranian subtidal leucosiid crabs
(Crustacea: Decapoda: Brachyura) of the Persian Gulf:
(DAAD) for financial support in the form of a PhD taxonomy and zoogeography. Iran. J. Anim. Biosys. 1:
scholarship to R. Naderloo. Furthermore, we are 31-46.
grateful to J.A. Khan and the team of the Arabian Nobili G. 1906. Faune carcinologique de la Mer Rouge
Seas Expedition who gave great support to one of décapodes et stomatopodes. Ann. Sci. Nat. (Zool.) 4:
the authors (M. Apel) during a survey of the UAE 1-347, figs. 1-12, pls. 1-11.
Paulson OM. 1875. Izsledovaniya rakoobraznykh Krasnago
coastline in 1995. Morya s zametkami otnositel’no rakoobraznykh drugikh
morei. Chast‘ I. Podophthalmata i Edriophthalmata
(Cumacea). Kiew, Kul’zhenko 1875: I-XIV + 1-144, pls.
Rüppell E. 1830. Beschreibung und Abbildung von 24
Arten kurzschwanzigen Krabben, als Beitrag zur
Alcock A. 1896. Materials for carcinological fauna of India. N.
Naturgeschichte des rothen Meers. Frankfurt, Germany:
2: The Brachyura Oxystomata. J. Assoc. Soc. Beng. 65:
H.L. Brönner, 28 pp.
Serène R. 1968. Prodromus for a check list of the non-
Apel M. 2001. Taxonomie und Zoogeographie der Brachyura,
planctonic marine fauna of South East Asia. Sing. Nat.
Paguridea und Porcellanidae (Crustacea: Decapoda) des
Acad. Sci. Spec. Pub. 1: 1-122.
Persisch-Arabischen Golfes: 1-268. PhD dissertation,
Stephensen K. 1946. The Brachyura of the Iranian
Johann Wolfgang Goethe-Univ., Frankfurt am Main,
Gulf. Danish scientific investigations in Iran, part IV.
Copenhagen: E. Munksgaard, pp. 57-237.
Balss H. 1915. Anomuren, Dromiaceen und Oxystomen.
Tan CGS, PKL Ng. 1995. A revision of the Indo-Pacific genus
XXXI. Die Decapoden des Roten Meeres. Expeditionen
Oreophorus Rüppel, 1830 (Crustacea: Decapoda:
S.M. Schiff “Pola” in das Rote Meer nordliche und sudliche
Brachyura: Leucosiidae). In B Richer De Forges, ed.
halfte 1895/96-1897/98. Berichte der Kommission für
Les fonds meubles des lagons de Nouvelle-Calédonie
ozeanographische Forschungen, 18 pp.
(Sédimentologie, benthos). Etudes & Thèses, Vol. 2.
Galil B. 2009. An examination of the genus Philyra Leach,
Paris: Orstom, pp. 101-189.
1817 (Crustacea, Decapoda, Leucosiidae) with description
Tirmizi NM, QB Kazmi. 1986. Marine fauna of Pakistan. 4.
of seven new genera and six new species. Zoosystema
Crustacea: Brachyura (Dromiacea, Archaebrachyura,
31: 279-320.
Oxystomata, Oxyrhyncha). Publication I, BCCI Foun-
Gravier C. 1920. Sur une collection de crustacés recueillis à
dation Chair, Institute of Marine Sciences, Univ. of
Madagascar par M. le Lieutenant Decary. Bull. Mus. Hist.
Karachi, Pakistan, pp. 1-244.
Nat. Paris 26: 376-383.
Titgen RH. 1982. The systematics and ecology of the
Guinot D. 1967. La faune carcinologique (Crustacea,
decapods of Dubai, and their zoogeographic relationships
Brachyura) de l’Ocean Indien occidental et de la Mer
to the Persian Gulf and the western Indian Ocean. PhD
Rouge. Catalogue remarques biogéographiques et
dissertation, Texas A&M Univ., Texas, USA.
bibliographie. Mém. Inst. Fond. Afr. 77: 235-352.
Laurie RD. 1915. 1906. Report on the Brachyura collected
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
Faculty of Agro Industry and Natural Resources, Universiti Malaysia Kelantan, Locked bag 36, Pengkalan Chepa, Kelantan 16100,
Department of Zoology, Faculty of Resource Science and Technology, Universiti Malaysia Sarawak, Kota Samarahan 94300, Sarawak,

(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.

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

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

(1992) listed 19 synonyms of C. brachyotis, but data on samples from Borneo and Peninsular
Simmons (2005) recognized only seven of them, Malaysia to the southern tip of Thailand. Results of
with most of them lacking data on their status and those studies showed that 2 forms of C. brachyotis
current distribution. Abdullah (2003) compared inhabited 2 contrasting habitats (in Peninsular
morphological measurements of Cynopterus Malaysia and Borneo). The larger form was found
from various sources (Andersen 1912, Hill and to inhabit open areas, whereas the smaller form
Thonglongya 1972, Lekagul and McNeely 1977, was confined to primary forests. Abdullah et al.
Medway 1978, Hill 1983, Payne et al. 1985, (2000) postulated that these differences found in
Kitchener and Maharadatunkamsi 1991 1996, C. brachyotis are based on ecological differences
Ingle and Heaney 1992, Nor 1996) and found in the habitats they occupy. Later Campbell
that a lot of morphological measurements overlap et al. (2004) reexamined the species complex
within and between species across its distribution. using different genetic markers and discovered
This species is widely distributed throughout 4 additional distinct lineages in the C. brachyotis
Southeast Asia (Fig. 1) and can be found at areas complex scattered in the Indo-Malayan region.
up to 1600 m in elevation in Borneo (Lekagul and These 4 lineages are respectively found in India,
McNeely 1977, Medway 1978, Bergmans and Myanmar, Sulawesi, and the Philippines. Abdullah
Rozendall 1988, Corbet and Hill 1992, Peterson and Jayaraj (2006) later performed a cluster
and Heaney 1993, Abdullah 2003). It can be found analysis on the type specimen of C. brachyotis
in many habitats (but most frequently in disturbed using morphological measurements described
forest) including lower montane forest, dipterocarp by Müller (1938), and the results showed that the
forest, gardens, mangroves, and strand vegetation. nominate C. brachyotis was clustered with the
Francis (1990) found that there were forearm larger form of C. brachyotis.
length differences in C. brachyotis caught in A recent study using microsatellites and 2
primary forests and that from secondary habitats mitochondrial (mt)DNA genes by Fong (2011)
in Sepilok, Sabah. This observation was later showed congruent findings with Abdullah et al.
investigated by Abdullah et al. (2000) and Abdullah (2000), Abdullah (2003), Campbell et al. (2004
(2003) using molecular and external morphometric 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
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).
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
Fig. 1. Distribution of 8 subspecies of C. brachyotis in the (2006) study showed that the larger form was
Indo-Malayan region (Mickelburgh et al. 1992, Simmons indeed the assigned C. brachyotis, it is apparent
2005). (a) C. b. altitudinis found in highlands of the Main that the small form may be a new species of
Range, Peninsular Malaysia; (b) C. b. brachysoma found in Cynopterus yet to be described. However, a recent
the Andaman Is.; (c) C. b. ceylonensis found in Sri Lanka; (d) taxonomy of the Cynopterus by Simmons (2005)
C. b. concolor found on Enganno I.; (e) C. b. hoffeti found in
did not include this new form, and Francis (2008)
Vietnam; (f) C. b. insularum found in the Kangean Is. and Laut
Kecil Is.; (g) C. b. javanicus found in Bali, Java, Madura, and assigned C. cf. brachyotis Sunda to the large form
Penidah; and (h) C. b. brachyotis found in Bangka, Belitung, of C. brachyotis commonly found in open areas
Borneo, Lombok, the Nicobar Is., Peninsular Malaysia, the and C. cf. brachyotis Forest to the small form of C.
Philippines, Singapore, Sulawesi, Sumatra and Thailand. brachyotis commonly found in primary forests. For
Zoological Studies 51(2): 259-271 (2012) 261

the easy interpretation of this paper, we assign C. confirmed by Abdullah (2003) and Fong (2011)
brachyotis as the known large form as verified by using DNA sequences of the partial Cytochrome
Abdullah and Jayaraj (2006), and C. cf. brachyotis b (700 bps) and Cytochrome Oxidase 1 (486 bps)
Forest as the new undescribed form found in were used in this study.
primary forests. Twenty-eight morphological measurements
In terms of forearm length differences, (skull, dental, and external morphological
Francis (1990) showed that C. brachyotis has a measurements; Fig. 2) were recorded following
mean forearm length of 62.1 mm (n = 22) and C. Kitchner et al. (1995) and Jayaraj et al. (2004
cf. brachyotis Forest has a mean forearm length 2005). Abbreviations for the characters measured
of 58.4 mm (n = 21). Abdullah (2003) reported are as follow: BL, bulla length; C1BW, canine tooth
that the forearm length of C. cf. brachyotis Forest basal width; C1C1B, breadth across both canine
was 59.43 (± standard deviation (SD) 2.70) mm, outside surfaces; C1M3L, canine molar length or
and C. brachyotis had a mean forearm length of maxillary tooth row length; CW, cranial width; DBC,
63.87 (± 5.02) mm. Campbell et al. (2006) used distance between cochleae; DL, dentary length;
a forearm length of 63.8 (± 1.6) mm to identify C. D3MCL, 3rd digit metacarpal length; D4MCL,
brachyotis and a forearm length of 59.5 (± 1.7) mm 4th digit metacarpal length; D5MCL, 5th digit
to identify C. cf. brachyotis Forest. metacarpal length; D3P1L, 3rd digit 1st phalanx
The high reliance on forearm length to length; D3P2L, 3rd digit 2nd phalanx length; EL,
distinguish these 2 forms is problematic as ear length; GBPL, greatest basial pit length; GSL,
various authors have reported different forearm great skull length; IOW, interorbital width; M3L,
length measurements used for differentiation. 3rd molar tooth crown length; M3W, 3rd molar
The development of additional characters to tooth crown width; M3M3B, breadth across outside
differentiate these 2 forms would be useful in the surfaces of both 3rd molar teeth; MW, mastoid
field, as this will aid field ecologists in accurately width; PES, pes length; PL, palatal length; POW,
identifying both C. brachyotis and C. cf. brachyotis postorbital width; PPL, postpalatal length; RL,
Forest in southern Thailand, Peninsular Malaysia, radius length; TL, tibia length; TVL, tail to ventral
and Borneo. Thus, in this study, we attempted to length; and ZW, zygomatic width. Bat skulls were
further describe detailed morphometric variations extracted after morphological data were collected
that exist in the genus Cynopterus from Peninsular following Nargorsen and Peterson (1980).
Malaysia and Borneo using multivariate statistics. A cluster analysis using Euclidean distances
The approach was to develop a classification with the unweighted pair-groups method
function that can be used to differentiate C. average (UPGMA) was performed to construct
brachyotis and C. cf. brachyotis Forest in the field a morphometrics-based phylogeny and confirm
and verify museum specimens. the initial grouping of samples (Everitt 1993).
The E. major measurements were used as the
outgroup for this analysis. Data of confirmed
MATERIALS AND METHODS groupings were then subjected to a t-test to check
for sexual dimorphism. Levene’s test for equality
In total, 74 specimens (10 individuals of of variances was used as a selection criterion for
C. horsfieldii, 34 individuals of C. brachyotis, the assumption of equal or unequal variances
29 individuals of C. cf. brachyotis Forest, and 1 prior to the t-test (Zar 1984). The normality of
individual of Eonycteris major) were used in this the data was checked using a normal probability
study. These specimens were either collections plot and the Shapiro-Wilk test. The assumption
from field sampling done in various localities within of homoscedasticity was tested using Box’s M
Borneo and Peninsular Malaysia or museum test, and the assumption of multicolinearity was
samples from the zoological museum at Universiti checked by observing the tolerance value for all
Malaysia Sarawak (Sarawak, Malaysia) and independent variables (Joseph et al. 1992). Next,
the Department of Wildlife and National Parks the data were subjected to a stepwise discriminant
(PERHILITAN) Museum (Pahang, Malaysia). Due function analysis following Joseph et al. (1992)
to a limited number of samples of Cynopterus, we and Manly (1994). Two separate analyses
opted to focus on the problem of differentiating were performed: 1) using a combination of all
C. brachyotis and C. cf. brachyotis Forest using available characters and 2) using only external
multivariate statistics. Only specimens of C. morphological characters. Data were analyzed
brachyotis and C. cf. brachyotis Forest previously 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 standard deviations (SDs) of all characters are
(SPSS, Chicago, IL, USA). 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
RESULTS not follow a normal distribution either prior to or
after transformation to achieve normality. Box’s
The UPGMA cluster analysis (Fig. 3) shows M statistics had a value of 23.406 (probability of
th groupings of Cynopterus spp. based on p = 0.483, p > 0.001) indicating homoscedasticity.
morphological measurements. Based on the Thus, the data were analyzed using a pooled
phylogram, there are 3 major clades consisting covariance matrix for classification. Multicolinearity
of C. horsfieldii, C. cf. brachyotis Forest, and C. among the independent variables was not present,
brachyotis. Of the 28 characters examined, 3 as tolerance values for all variables were > 0.10.
characters (IOW for C. horsfieldii and D3MCL For analysis of all remaining characters, the
and D5MCL for C. brachyotis) were found to be stepwise method identified 1 discriminant function
sexually dimorphic (Table 1). The means and (Function 1) that was statistically significant based











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

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 discriminant function (Function 1) that was stati-
(GBPL, M3L, M3W, TVL, D3P2L, and RL; Table stically significant (Table 6) with 4 characters (TVL,
4) were generated from the stepwise procedure. D3P1L, D3MCL, and RL; Table 7) generated from
The characters with the highest weight on function the stepwise procedure. The character with the
1 were RL (0.957) and M3L (0.506), whereas highest weight and loading was RL (weight = 1.240;
M3L (3.912) and M3W (-3.454) had the highest loading = 0.706), while D3MCL (-0.731) had the
discriminant loadings. All 6 characters determined 2nd-highest weight. All 4 characters determined
by the stepwise procedure produced a discriminant by the stepwise procedure produced a discriminant
function with an accuracy rate of 100% (see function with an accuracy rate of 96.8% (see
accuracy rates, Table 5). accuracy rates, Table 8).
For analysis of only external morphological A histogram of the discriminant scores
characters, the stepwise method identified a 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 have a forearm of > 60 mm. Adults of C. cf.
C. brachyotis or C. cf. brachyotis Forest. The brachyotis Forest have a smaller body size with
predictive models are as follows: duller coloration and usually have a forearm length
for all remaining characters of < 60 mm.
Comparison with previous bat surveys
ŷ = -0.405a + 3.912b - 3.454c - 0.149d + (Timoh 2006, Fukuda et al. 2008) and personal
0.148e + 0.546f - 37.326 (constant) (a) observations indicate that C. brachyotis was
sampled across a wide variety of vegetation
and for only external morphological characters types with different capture rates, whereas C.
cf. brachyotis Forest was confined to primary
ŷ = -0.216d + 0.442 - 0.433h - 0.706f - 34.507 forests. Capture rates of C. brachyotis were 44%
(constant); (b) in secondary forests, 41% in orchards, and 72%
in oil palm plantations (Fukuda et al. 2008). The
where ŷ is the discriminant score (a negative score high capture rate in oil palm plantations is probably
indicates C. cf. brachyotis Forest and a positive associated with the abundance of oil palm fruit,
score indicates C. brachyotis), a is the GBPL, b i.e., a food source (Fukuda et al. 2008). We also
is the M3L, c is the M3W, d is the TVL, e is the speculated that this abundant food source would
D3P2L, f is the RL, g is the D3P1L, and h is the also likely increase the life expectancy of C.
D3MCL. brachyotis in oil palm plantations as many older
individuals were captured (with distinct reddish-
brown fur on their shoulders and worn out or
DISCUSSION missing teeth in most individuals) in Timoh’s (2006)
General discussion of statistical results In this study, the analyses revealed that the
RL, M3L, and M3W had the highest discriminant
Based on the cluster analysis, a clear division loading and weight, and this was reflected by
(approximately 15.48% distance, based on the importance of these characters during the
estimates from the graph) was observed between identification process. The RL or forearm length
C. brachyotis and C. cf. brachyotis Forest. Visual is one of the characters useful in identifying bats,
observations of samples during field sampling especially fruit bats of the family Pteropodidae.
indicated that adult C. brachyotis can be vaguely This character was also previously used to diffe-
identified due to the brown fur with a pronounced rentiate C. brachyotis, C. cf. brachyotis Forest,
yellowish or reddish tinge, and these bats usually 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


4 6


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
Zoological Studies 51(2): 259-271 (2012) 267

al. 2000, Abdullah 2003, Campbell et al. 2004 mance of bats as the total body mass is negatively
2006 2007, Jayaraj et al. 2004 2005, Fong 2011). correlated with wing loading, a measure of the
Although M3L and M3W are not generally used ability to navigate around obstacles (Aldridge
for species identification, molar differences in C. 1986, Aldridge and Rautenbach 1987, Jones et
horsfieldii, C sphinx, and C. brachyotis are a key al. 1993, Rhodes 2002) and maneuverability in
character which can be used to differentiate these cluttered areas (Aldridge and Rautenbach 1987,
3 species of Cynopterus in Malaysia. The D3MCL Jones et al. 1993, Kalcounis and Brigham 1995,
and D3P1L both contribute to the length and size Brigham et al. 1997, Rhodes 2002). This can be
of the wings, and this may reflect the habitats that directly linked to the habitats of both species, with
both species occupy. C. brachyotis occupying less-cluttered habitats and
Th e cl uste r a n d d iscri mi n a nt fu ncti o n C. cf. brachyotis Forest occupying dense areas
analyses showed that C. brachyotis and C. cf. (Abdullah et al. 2000, Abdullah 2003, Jayaraj
brachyotis Forest populations are morphologically et al. 2004 2005). Body size seems to be the
distinct, congruent with previous results using discriminating factor in the cluster analysis for
molecular methods (Abdullah 2003, Campbell et effectively discriminating these 2 species, which
al. 2004 2006, Julaihi 2005, Fong 2011). Although explains why both species can be separated, but
the topology of the dendogram generated by body size per se does not depict the entire picture
the cluster analysis was not similar to previous of the divergence of these bats. In terms of the
molecular studies, it was able to differentiate flight apparatus and dimensions, both species
C. brachyotis from C. cf. brachyotis Forest. apparently did not undergo the change in wing
The different topologies might be a reflection shape indicated in a recent study by Campbell et
of the morphological appearances of these al. (2007), but rather a change in body size which
bats. Morphologically both C. brachyotis and might have been due to selective pressures for C.
C. cf. brachyotis Forest look similar, whereas C. brachyotis and C. cf. brachyotis Forest to fit into
horsfieldii is very much larger with distinct cusps their respective habitats. Similarly, Menon (2007)
on the lower premolar and 1st lower molar; these revealed that the aspect-ratio and wing-loading
characteristics are not present in C. brachyotis or C. indices cannot be used to differentiate these 2
cf. brachyotis Forest. species in Borneo.
The prediction models developed will Previous studies (Freeman 1981, Schluter
be particularly useful in accurately identifying 1993, Wain-Wright 1996) noted that there was a
C. brachyotis and C. cf. brachyotis Forest in relationship between the structure of the feeding
Malaysia. Specifically function (a) can be used apparatus and diet in bats. As M3W and M3L
to verify museum specimens, and function (b) are associated with feeding and foraging, it was
will be more appropriate for field identification. speculated that the shape and dimension of the
Function (a) requires cranial, dental, and external dentition are associated with the diet. Current
morphological measurements; thus, unverified knowledge of the diet and foraging behavior of C.
museum specimens can be identified once the brachyotis in Malaysia was previously documented
skull is extracted, reducing the cost of validating by Lim (1970), Phua and Corlett (1989), Fujita
the species using molecular tools. Although having and Tuttle (1991), Francis (1994), Funakoshi and
a lower accuracy rate, function (b) can be used in Zubaid (1997), Tan et al. (1998), Mohd Azlan et al.
the field as only external morphological characters (2000), and Hodgkison et al. (2003), but none of
are needed to identify the species. If needed, those authors focused on differences in the diets
however, a tissue sample via skin scraping or a and foraging behaviors of C. brachyotis and C.
wing punch can be taken for species verification cf. brachyotis Forest. Thus a more-detailed study
in the lab. An accurate identification method will of the diets and foraging behaviors would shed
definitely aid ecologists, conservationists, and law more light on ecological differences between C.
enforcement officials in studying and conserving brachyotis and C. cf. brachyotis Forest.
this species complex.
Implications of recent studies for the taxono-
Body sizes and relation to habitat types of mic status of the Cynopterus brachyotis
Cynopterus brachyotis and C. cf. brachyotis complex
It was proven by various studies that C.
Body size can be related to the flight perfor- 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 of Cynopterus across their distribution. Although
geographically distinct lineages respectively from C. brachyotis is widely distributed, information on
India, Myanmar, Sulawesi, and the Philippines, the current status of the 6 lineages of C. brachyotis
and 2 sympatric forms (recognized as C. cf. especially is not clear. Confounded by the non-
brachyotis Forest and C. brachyotis in this study) recognition of these new C. brachyotis lineages
in southern Thailand, Peninsular Malaysia, and (Forest, India, Myanmar, Sulawesi, and the
Borneo. These 2 sympatric forms are found in Philippines) as distinct species (see Abdullah and
distinct habitats: C. brachyotis is found in open Jayaraj 2006), the survival of these rare species
areas, and C. cf. brachyotis Forest is found in may be threatened if no clear and proper planning
the primary and old secondary forests (Abdullah for conservation is put in place.
2003, Campbell et al. 2004 2006, Jayaraj et al. In terms of biogeography, the existing reco-
2004 2005, Julaihi 2005, Fukuda et al. 2008, Fong gnized Cynopterus species of C. brachyotis,
2011). Cynopterus brachyotis is the ancestral C. horsfieldii, C. luzoniensis, C. minutus, C.
lineage of all Cynopterus in Peninsular Malaysia nusatenggara, C. sphinx, and C. tithaecheilus are
and Borneo with nucleotide divergence ranging distributed in the Indo-Malayan region and their
8%-9%, whereas C. cf. brachyotis Forest is closely distributions overlap. Simmons (2005) listed their
related to C. horsfieldii, differing by only a genetic distributions as follow: C. brachyotis is distributed
divergence of 3.5% (Abdullah 2003). in Sri Lanka, India, Nepal, Burma, Thailand,
This scenario is not new to the taxonomy Cambodia, Vietnam, South China, Malaysia, the
of Cynopterus as C. nusatenggara described by Nicobar and Andaman Is., Borneo, Sumatra,
Kitchner and Maharadatunkamsi (1991 1996) Sulawesi, Magnole, Sanana, Sangihe I., and
was also found within Cynopterus populations Talaud I. with possible occurrence in the Palawan
during field sampling on islands of Nusa Tenggara, region of the Philippines; C. horsfieldii is limited to
Indonesia. Cynopterus bats are currently repre- Thailand, Cambodia, Peninsular Malaysia, Borneo,
sented by 7 species (Simmons 2005), but there Java, Sumatra, the Lesser Sunda Is., and adjacent
are a lot of variations in terms of body size and small islands; C. luzoniensis is found in Sulawesi,
coloration between and within species. These the Philippines, and adjacent small islands; C.
variations were observed in island populations and minutus is found in Sumatra, Java, Borneo, and
highland populations, and are due to differences Sulawesi; C. nusatenggara is found in Lombok,
in vegetation and other ecological factors (see Moyo, Sumbawa, Sangeang, Komodo, Flores,
Hill and Thonglongya 1972, Lekagul and McNeely Sumba, Adonara, Lembata, Pantar, Alor, and the
1977, Medway 1978, Payne et al. 1985, Kitchner Wetar Is.; C. sphinx is found in Sri Lanka, Pakistan,
and Maharadatunkamsi 1991 1996, Ingle and Bangladesh, India, South China, Southeast Asia
Heaney 1992, Schmitt et al. 1995, Nor 1996, including Burma, Vietnam, Cambodia, Peninsular
Abdullah et al. 2000, Storz et al. 2001, Abdullah Malaysia, Sumatra, and possibly in Borneo; and
2003, Campbell et al. 2004 2006 2007, Jayaraj et C. tithaecheilus is found in Sumatra, Java, Bali,
al. 2004, Menon 2007, Fukuda et al. 2008, Fong Lombok, Timor, and adjacent small islands.
2011). In Malaysia; however, only 5 species of
Menon (2007) collected an unidentified Cynopterus coexist together, i.e., C. horsfieldii,
Cynopterus specimen from Satang I., Borneo, C. sphinx, C. brachyotis, C. minutus, and C. cf.
Malaysia, and this specimen was later identified brachyotis Forest. Cynopterus horsfieldii, C.
using DNA techniques. The forearm length of brachyotis, and C. cf. brachyotis Forest have a
this Cynopterus specimen was 69 mm indicating high geographic distributional overlap (Abdullah
it was C. sphinx, but DNA identification indicated 2003, Campbell et al. 2004 2006 2007), but
that it was C. brachyotis (unpubl. data). A molar C. cf. brachyotis Forest’s distribution extends
examination of the specimen did not show a clear farther north into Thailand, Vietnam, and probably
distinction between C. brachyotis and C. sphinx. Cambodia and Laos (Campbell et al. 2004
Such an observation is the norm when individuals 2006). Ecologically, C. sphinx and C. brachyotis
from this genus are collected in a wide range of are common in open habitats, orchards, and
vegetative types, which indicates that there are agricultural areas, whereas C. horsfieldii and C.
high intra- and interspecific variations among cf. brachyotis Forest are found in primary and old
Cynopterus representatives. The lack of such secondary forests in Peninsular Malaysia and
knowledge indicates the necessity for a current southern Thailand. Cynopterus cf. brachyotis
large-scale study on inter- and intraspecific forms Forest is also rare in Peninsular Malaysia, as its
Zoological Studies 51(2): 259-271 (2012) 269

occurrence is dictated by the existence of primary rate of of identifying both C. brachyotis and C. cf.
and old secondary forests. Cynopterus sphinx brachyotis Forest. Based on the 1st prediction
is found in both habitat types, but declines in model (function a), 6 chara-cters are needed
number near forest edges (Campbell et al. 2006). to accurately differentiate C. brachyotis from
Similar observations were found in Borneo, but C. cf. brachyotis Forest in southern Thailand,
with the exclusion of C. sphinx as records of this Peninsular Malaysia, and Borneo. This model
species occurring in Borneo are only from Central would be more appropriate for use on museum
Kalimantan (Payne et al. 1985, Abdullah et al. specimens as skull and dental characters are
1997), and to date, there are no recent records of needed for the calculation. The 2nd prediction
this species in Malaysian Borneo. The occurrence model (function b) can be used during field
of C. minutus in Borneo is still in question, as sampling, as only external morphological
there is little information on it, but a recent survey measurements are needed for identification.
by Benda (2010) did record C. minutus in Sabah. These prediction models can subsequently be
The forearm length of C. minutus captured in his used by bat biologists to correctly identify adult C.
study was 54.3-58.1 (mean, 55.69, SD, 1.644) mm brachyotis forms in southern Thailand, Peninsular
(n = 5), which is slightly smaller but overlaps with Malaysia, and Borneo, thus aiding in research and
forearm length measurements of C. cf. brachyotis conservation efforts of both C. brachyotis and C.
Forest in Abdullah (2003), Campbell et al. (2004 cf. brachyotis Forest in this region. Further sug-
2006 2007), Jayaraj et al. (2004 2005), Jayaraj gestions on taxonomic research of this species
(2009), and Fong (2011). complex should include verification of multiple
As Abdullah and Jayaraj’s (2006) preliminary genetic markers, examination of detailed morpho-
investigation of the nominate specimen of C. metrics, and a review of the taxonomic status of
brachyotis revealed that the type specimen of C. the 6 existing C. brachyotis forms in the Indo-
brachyotis described by Müller (1838) is the larger Malayan region. Conservation of this species
form, it is apparent that the remaining C. brachyotis complex needs to be carefully planned in order to
lineages (Forest, India, Myanmar, Sulawesi, and ensure that all 6 divergent forms do not go extinct,
the Philippines) require further study to clarify their as these are suspected of being undescribed
phylogenetic positioning and taxonomic status. To species in the Indo-Malayan region.
date, there are more than 10 studies (see Abdullah
et al. 2000, Abdullah 2003, Campbell et al. 2004 Acknowledgments: We would like to thank Dr. L.
2006 2007, Jayaraj et al. 2004 2005, Julaihi 2005, S. Hall for the inspiration to go further and realize
Abdullah and Jayaraj 2006, Jayaraj 2009, Fong our potential, to F.A.A. Khan, A.K.H. Guan, B.
2011) that have validated the existence of C. cf. Ketol, F.P. Har and all staffs in Molecular Ecology
brachyotis Forest, but there are no published Laboratory Universiti Malaysia Sarawak (UNIMAS)
studies on the remaining C. brachyotis lineages for all their support and comments to improve our
in the Indo-Malayan region. Thus, a complete work and being great companions during field trips
phylogenetic tree of all 7 recognized species and and laboratory sessions. We would also like to
recorded divergent forms of Cynopterus (including thank the Sarawak Forest Department for issuing
the 6 divergent forms of C. brachyotis) should permit no. 04608 and the Sarawak Forestry
be generated to clarify the taxonomic status of Corporation for their hospitality during our visit
all Cynopterus spp. in the Indo-Malayan region. to protected areas in Sarawak. We extend our
Clarification of C. luzoniensis from Sulawesi and gratitude to the Faculty of Resource Science and
Palawan is also needed, as there is the possibility Technology, UNIMAS, Department of Wildlife and
that the Sulawesi and Philippine forms of C. National Parks (Kuala Lumpur), Sabah Parks,
brachyotis previously described by Campbell et al. Sabah Wildlife Department, and many other
(2004) could possibly be C. luzoniensis, or these 2 individuals for various administrative and logistical
C. brachyotis forms may differ from C. luzoniensis support throughout the course of the study. We
altogether. Finally, because C. minutus is thank 2 critical anonymous reviewers who tre-
recognized as a distinct species (Simmons 2005), mendously helped us improve the taxonomic
there is a need to check the status of this species perspectives of an initial draft of this paper. The
in Borneo as little information is available. main author, JVK, would also like to thank M.
Two models to differentiate C. brachyotis Muhamad for her comments on the statistical
and C. cf. brachyotis Forest were developed analysis and the overall manuscript. This study
using multivariate statistics with a high accuracy 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 Press.

colleagues, a UMK Short Term Grant (R/SGJP/ Cuvier F. 1824. Des dents des mammiferes. F. G. Strasbourg,
France: Levrault, 256 pp.
A03.00/00481A/001/2010/000037) awarded to JVK Everitt BS. 1993. Cluster analysis, 3rd ed. New York: Halsted
and YCW and ARA, and a UNIMAS ZAMALAH Press.
scholarship awarded to JVK. 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
REFERENCES understorey of lowland dipterocarp rain forest in Malaysia.
J. Trop. Ecol. 6: 421-431.
Abdullah MT. 2003. Biogeography and variation of Cynopterus Francis CM. 1994. Vertical stratification of fruit bats
brachyotis in Southeast Asia. PhD dissertation, Univ. of (Pteropodidae) in the lowland dipterocarp rainforest in
Queensland, St. Lucia, Australia. Malaysia. J. Trop. Ecol. 10: 523-530.
Abdullah MT, VK Jayaraj. 2006. Preliminary investigation on Francis CM. 2008. A field guide to the mammals of Southeast
the relationship of the nominate C. brachyotis with the Asia. London: New Holland Publishers, 392 pp.
small-sized and large-sized C. brachyotis using clustering Freeman PW. 1981. Correspondence of food habits and
analysis. Sarawak Mus. J. 62: 223-236. morphology in insectivorous bats. J. Mammal. 62: 164-
Abdullah MT, C Moritz, GC Grigg, LS Hall. 2000. Evidence 166.
of cryptic species within Cynopterus brachyotis by using Fujita MS, MD Tuttle. 1991. Flying foxes (Chiroptera:
mtDNA sequence. In Z Yaacob, S Moo-Tan, S Yorath, Pteropodidae): threatened animals of key ecological and
eds. Proceedings of the International Conference on In- economic importance. Conserv. Biol. 5: 455.
Situ And Ex-Situ Biodiversity Conservation in the New Fukuda D, OB Tisen, K Momose, S Sakai. 2008. Bat diversity
Millennium. Kota Kinabalu, Malaysia: Yayasan Sabah. in the vegetation mosaic around a lowland dipterocarp
Abdullah MT, MA Rahman, LS Hall. 1997. New records for forest of Borneo. Raffles Bull. Zool. 57: 213-221.
bats in Sarawak, Malaysia. Malay. Nat. J. 50: 365-367. Funakoshi K, A Zubaid. 1997. Behavioral and reproductive
Aldridge HDJN. 1986. Manoeuvrability and ecological ecology of the dog-faced fruit bats, Cynopterus brachyotis
segregation in the little brown (Myotis lucifugus) and Yuma and C. horsfieldi, in a Malaysian rainforest. Mammal
(M. yumamensis) bats (Chiroptera: Vespertilionidae). Study 22: 95-108.
Can. J. Zool. 64: 370-382. Gray JE. 1843. List of the specimens of Mammalia in the
Aldridge HDJN, IL Rautenbach. 1987. Morphology, echo- collection of the British Museum. London: British Museum
location and resource partitioning in insectivorous bats. J. (Natural History) Publications, 216 pp.
Anim. Ecol. 56: 763-778. Hill JE. 1983. Bats (Mammalia: Chiroptera) from Indo-
Andersen K. 1912. Catalogue of the Chiroptera in the Australia. Bull. Br. Mus. Nat. Hist. (Zool.) 45: 103-208.
collections of the British Museum. Vol. I. Megachiroptera. Hill JE, K Thonglongya. 1972. Bats from Thailand and
London: British Museum of Natural History. Cambodia. Bull. Br. Mus. Nat. Hist. (Zool.) 22: 171-196.
Benda P. 2010. On a small collection of bats (Chiroptera) from Hodgkison R, ST Balding, A Zubaid, TH Kunz. 2003. Fruit
western Sabah (North Borneo, East Malaysia). Vespertilio bats (Chiroptera: Pteropodidae) as seed dispersers and
13-14: 45-76. pollinators in a lowland Malaysian rain forest. Biotropica
Bergmans W, FG Rozendaal. 1988. Notes on collections 35: 491-502.
of fruit bats from Sulawesi and some off-lying islands Ingle NR, LR Heaney. 1992. A key to the bats of the Philippine
(Mammalia, Megachiroptera). Zoologische Verhan- islands. Fieldiana Zool. 69: 1-44.
delingen 248: 1-74. Jayaraj VK. 2009. The phylogenetic relationships of
Brigham RM, SD Grindal, MC Firman, JL Morissette. 1997. Megachiroptera in Malaysia inferred from morphological
The influence of structural clutter on activity patterns of and DNA analyses. Master’s thesis. Universiti Malaysia
insectivorous bats. Can. J. Zool. 75: 131-136. Sarawak, Kota Samarahan.
Bumrungsri S, PA Racey. 2005. Field discrimination between Jayaraj VK, CJ Laman, MT Abdullah. 2004. Morphological
lesser short-nosed fruit bat (Cynopterus brachyotis Müller, variation in the genus Cynopterus of Peninsular Malaysia
1838) and greater short-nosed fruit bat (C. sphinx Vahl, and Borneo. In AIM Ismail, HL Koh, YA Hasan, eds.
1797) (Chiroptera: Pterpodidae) in southern Thailand. Proceedings of the Regional Conference on Ecological
Nat. Hist. Bull. Siam Soc. 53: 111-121. and Environmental Modelling (ECOMOD 2004), Penang,
Campbell P, CJ Schneider, AM Adnan, A Zubaid, TH Kunz. Malaysia, pp. 69-81.
2004. Phylogeny and phylogeography of Old World Jayaraj VK, CJ Laman, MT Abdullah. 2005. Application of
fruit bats in the Cynopterus brachyotis complex. Mol. multivariate techniques in determining morphological
Phylogen. Evol. 33: 764-781. variation in the genus Cynopterus of Peninsular Malaysia
Campbell P, CJ Schneider, AM Adnan, A Zubaid, TH Kunz. and Borneo. In AA Tuen, I Das, eds. International
2006. Comparative population structure of Cynopterus Conference on Biogeography and Biodiversity: Wallace in
fruit bats in Peninsular Malaysia and southern Thailand. Sarawak – 150 Years Later: Sarawak. Malaysia, 226 pp.
Mol. Ecol. 15: 29-47. Jones G, M Morton, PM Hughes, RM Budden. 1993. Echo-
Campbell P, CJ Schneider, AM Adnan, A Zubaid, TH location, flight morphology and foraging strategies of
Kunz. 2007. Morphological and ecological correlates some West African hipposiderid bats. J. Zool. (Lond.)
of coexistence in Malaysian fruit bats (Chiroptera: 230: 285-400.
Pteropodidae). J. Mammal. 88: 105-118. Joseph FH Jr, RE Anderson, RL Tatham, WC Black. 1992.
Corbet GB, JE Hill. 1992. The mammals of the Indomalayan Multivariate data analysis with readings, 3rd ed. New
region: a systematic review. Oxford, UK: Oxford Univ.
Zoological Studies 51(2): 259-271 (2012) 271

York: Macmillan. manual. Life Science Miscellaneous Publications.

Julaihi AM. 2005. A phylogenetic study of the subfamily Toronto: Royal Ontario Museum.
Pteropodinae inferred by using partial mitochondrial DNA Nor SM. 1996. The mammalian fauna on the islands at the
cytochrome b gene. BSc thesis, Univ. Malaysia Sarawak northern tip of Sabah, Borneo. Fieldiana Zool. 83: 1-51.
(UNIMAS), Kota Samarahan, Sarawak. Payne J, CM Francis, K Phillipps. 1985. A field guide to the
Kalcounis MC, RM Brigham. 1995. Intraspecific variation in mammals of Borneo. Kota Kinabalu, Malaysia: The
wing loading affects habitat use by little brown bats (Myotis Sabah Society and World Wildlife Fund Malaysia.
lucifugus). Can. J. Zool. 73: 89-95. Peters HW. 1861. Monatsberichte der Koniglich Preussischen
Kitchener DJ, Maharadatunkamsi. 1991. Description of a new Akademie der Wissenschaften zu Berlin 1861: 708.
species of Cynopterus (Chiroptera: Pteropdidae) from Peterson AT, LR Heaney. 1993. Genetic differentiation
Nusa Tenggara Indonesia. Rec. West. Aust. Mus. 51: in Philippines bats of the genera Cynopterus and
307-363. Haplonycteris. Biol. J. Linn. Soc. 49: 203-218.
Kitchener DJ, Maharadatunkamsi. 1996. Geographic variation Phua PB, RT Corlett. 1989. Seed dispersal by the lesser short-
in morphology of Cynopterus nusatenggara (Chiroptera, nosed fruit bat (Cynopterus brachyotis, Pteropodidae,
Pteropodidae) in southeastern Indonesia, and description Megachiroptera). Malay. Nat. J. 42: 251-256.
of two new subspecies. Mammalia 60: 255-276. Rhodes MP. 2002. Assessment of source of variance and
Kitchener DJ, WC Packer, Maharadatunkamsi. 1995. Morpho- patterns of overlap in microchiropteran wing morphology
logical variation in Pteropus lombocensis (Chiroptera: in southeast Queensland, Australia. Can. J. Zool. 80:
Pteropodidae) in Nusa Tenggara, Indonesia. Rec. Aust. 450-460.
Mus. 17: 61-67. Schluter D. 1993. Adaptive radiation in sticklebacks: size,
Koopman KF. 1994. Handbook of zoology vol. 8. Chiroptera: shape, and habitat use efficiency. Ecology 74: 699-709.
systematics. Berlin: Walter de Gruyter. Schmitt LH, DJ Kitchener, RA How. 1995. A genetic per-
Lekagul B, JA McNeely. 1977. Mammals of Thailand. spective of mammalian variation and evolution in the
Sahankarnbhat, Bangkok, 177 p. Indonesian archipelago: biogeographic correlates in the
Lim BL. 1970. Food habits and breeding cycle of the fruit bat genus Cynopterus. Evolution 49: 399-414.
Malaysian fruit-eating bat, Cynopterus brachyotis. J. Simmons NB. 2005. Order Chiroptera. In DE Wilson, DM
Mammal. 51: 174-177. Reeder, eds. Mammal species of the world: a taxonomic
Manly BFJ. 1994. Multivariate statistical methods. A primer, and geographical reference, 3rd ed. Washington DC:
2nd ed. London: Chapman and Hall. Smithsonian Institution Press.
Medway L. 1978. Mammals of Borneo: field keys and an Storz JF, HR Bhat, TH Kunz. 2001. Genetic consequences
annotated checklist. Kuala Lumpur: Monographs of the of polygyny and social structure in an Indian fruit bat,
Malayan Branch of the Royal Asiatic Society no. 7. Cynopterus sphinx. Inbreeding, outbreeding, and
Menon V. 2007. A study on the wing-loading and aspect-ratio population subdivision. Evolution 55: 1215-1223.
for fruit bats of family Pteropodidae. BSc thesis, Universiti Tan KH, A Zubaid, TH Kunz. 1998. Food habits of Cynopterus
Malaysia Sarawak (UNIMAS), Kota Samarahan, Sarawak. brachyotis (Muller) (Chiroptera: Pteropodidae) in
Mickleburgh SP, AM Hutson, PA Racey. 1992. Old world fruit Peninsular Malaysia. J. Trop. Ecol. 14: 299-307.
bats: an action plan for their conservation. Sultanate of Temminck CJ. 1825. Monographes de Mammalogie. G.
Oman and Gland, Switzerland: IUCN/FFPS/ZSL/WWF/ Dufour & E. D’Ocagne, Paris, 1: 198.
JWPT/NWF. Timoh JR. 2006. Diversity and abundance of bats in an oil
Miller GS Jr. 1906. Seven new Malayan bats. Proc. Biol. Soc. palm plantation and neighboring forest in Bratak oil
Washington 19: 63. palm estate, Bau, Sarawak. BSc thesis, Univ. Malaysia
Mohd Azlan J, RSK Sharma, M Zakaria. 2000. Species Sarawak (UNIMAS), Kota Samarahan, Sarawak.
diversity and relative abundance of understorey bats at Air Vahl M. 1979. Beskrivelse på tre nye arter flagermuse. Skrifter
Hitam Forest Reserve, Selangor, Malaysia. Malay. Nat. J. Naturhistorie-Selskabet Copenhagen, 4: 121-138.
54: 69-75. Wain-Wright PC. 1996. Ecological explanation through
Müller S. 1838. Over eenige nieuwe zoogdieren van Borneo. functional morphology: the feeding biology of sunfishes.
Tijdschrift voor Natuurlijke Geschiedenis en Physiologie, 5: Ecology 77: 1339-1343.
134-150. Zar JH. 1984. Biostatistical analysis. 2nd ed. Englewood
Nargorsen DW, RL Peterson. 1980. Mammal collector’s Cliffs, NJ: Prentice-Hall.


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
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
Invited Reviews: The Chief Editor invites an author to write a review. Invited reviews should normally be
the same length as a report.

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.

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

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
Indexed/Abstracted in:
Biological Abstracts
Chemical Abstracts
Current Awareness in Biological Sciences
Current Contents
Entomology Abstracts
Life Sciences
Vol. 51, No. 2
Zoological Record March, 2012

J.T. Wang, P.J. Meng, Y.Y. Chen, 137 Determination of the Thermal Tolerance of Symbiodinium Using the Activation
and C.A. Chen Energy for Inhibiting Photosystem II Activity

S. Keshavmurthy, C.M. Hsu, C.Y. 143 Larval Development of Fertilized “Pseudo-Gynodioecious” Eggs Suggests
Kuo, V. Denis, J.K. Leung, S. a Sexual Pattern of Gynodioecy in Galaxea fascicularis (Scleractinia:
Fontana, H.J. Hsieh, W.S. Tsai, Euphyllidae)
W.C. Su, and C.A. Chen

J.T. Wang, Y.Y. Chen, P.J. Meng, 150 Diverse Interactions between Corals and the Coral-Killing Sponge, Terpios
Y.H. Sune, C.M. Hsu, K.Y. Wei, and hoshinota (Suberitidae: Hadromerida)
C.A. Chen

H.U. Dahms, L.C. Tseng, S.H. 160 Biodiversity of Planktonic Copepods in the Lanyang River (Northeastern
Hsiao, Q.C. Chen, B.R. Kim, and Taiwan), a Typical Watershed of Oceania
J.S. Hwang

M.R. Zargaran, N. Erbilgin, and Y. 175 Changes in Oak Gall Wasps Species Diversity (Hymenoptera: Cynipidae) in
Ghosta Relation to the Presence of Oak Powdery Mildew (Erysiphe alphitoides)

B. Huo, C.X. Xie, B.S. Ma, X.F. 185 Age and Growth of Oxygymnocypris stewartii (Cyprinidae: Schizothoracinae)
Yang, and H.P. Huang in the Yarlung Tsangpo River, Tibet, China

L. Gonçalves, C.I. da Silva, and 195 Collection of Pollen Grains by Centris (Hemisiella) tarsata Smith (Apidae:
M.L.T. Buschini Centridini): Is C. tarsata an Oligolectic or Polylectic Species?

J.S. Wu, P.J. Chiang, and L.K. Lin 204 Monogamous System in the Taiwan Vole Microtus kikuchii Inferred from
Microsatellite DNA and Home Ranges


M.F. Lin, M.V. Kitahara, H. 213 A New Shallow-Water Species, Polycyathus chaishanensis sp. nov.
Tachikawa, S. Keshavmurthy, and (Scleractinia: Caryophylliidae), from Chaishan, Kaohsiung, Taiwan
C.A. Chen

S.S. Young, M.H. Ni, and M.Y. Liu 222 Systematic Study of the Simocephalus Sensu Stricto Species Group
(Cladocera: Daphniidae) from Taiwan by Morphometric and Molecular

B.A.R. Azman and B.H.R. Othman 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

R. Naderloo and M. Apel 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

V.K. Jayaraj, C.J. Laman, and M.T. 259 A Predictive Model to Differentiate the Fruit Bats Cynopterus brachyotis and
Abdullah C. cf. brachyotis Forest (Chiroptera: Pteropodidae) from Malaysia Using
Multivariate Analysis