SEMIDOME BUILDING AS SEXUAL SIGNALING IN THE FIDDLER CRAB
UCA LACTEA (BRACHYURA: OCYPODIDAE)
Tae Won Kim, John H. Christy, and Jae C. Choe
(TWK) Laboratory of Behavior and Ecology, School of Biological Sciences,
Seoul National University, Seoul 151-742, South Korea (ocean74@snu.ac.kr);
(JHC) Smithsonian Tropical Research Institute, Apartado 2072, Balboa, Anco?n,
Repu?blica de Panama? (christyj@naos.si.edu);
(JCC, correspondence) Laboratory of Behavior and Ecology, School of Biological Sciences,
Seoul National University, Seoul 151-742, South Korea (jcchoe@snu.ac.kr)
A B S T R A C T
During the mating season, males of the fiddler crab Uca lactea build semidomes of mud at the entrances
of their burrows to which they attract females for mating in the upper intertidal zone. Related species build
similar structures which either reduce aggression between neighboring males or attract sexually receptive
females. Male U. lactea did not build disproportionately more semidomes as density increased, suggesting
that these structures do not modulate aggression. Larger males built higher and wider semidomes as would
be expected if the semidomes are a courtship signal. When the high tides were too low to cover their
habitat and the sediment dried, males were unable to build new or repair existing semidomes. Towards the
end of the mating season more small males built semidomes perhaps because large males prevented them
from courting earlier when most females mated. We made two experimental enclosures, added males to
one and males and females to the other, and monitored semidome building. Males built significantly more
semidomes in the enclosure with females. Overall, our observations support the hypothesis that the
semidomes of U. lactea are a sexual signal.
Many species of intertidal brachyuran crabs
build mud or sand structures on the surface of
the sediment near or at their burrow entrances,
such as hoods (Zucker, 1981; Clayton, 1988),
pyramids (Linsenmair, 1967), mounds (Wada et
al., 1994), chimneys (Wada and Murata, 2000),
and mudballs (Oliveira et al., 1998). In many
species, only reproductively active males build
structures, which have been shown to reduce
aggression between neighboring males (Zucker,
1974; 1981; Clayton, 1988; Wada et al., 1994)
or to attract females to males? burrows for
mating (Crane, 1975; Christy 1988a, b, 1995;
Oliveira et al., 1998; Christy et al., 2001, 2002).
During the June to August mating season,
males of the fiddler crab Uca lactea construct
low semidomes at the entrances to their burrows
in the upper intertidal zone (Fig. 1). Reproduc-
tively active males of 17 species of fiddler crabs
(approx. 100 species) are known to build similar
structures variously called hoods, pillars,
domes, and semidomes, depending on their size
and shape (Christy, 1988a; Christy et al., 2001).
Semidome building by U. lactea in Japan was
first reported as ??shelter building?? (Yamaguchi,
1971). Recently, Kim and Choe (2003) and Kim
et al. (2004) considered semidomes to be
indicators of courtship activity, although neither
the function of these structures nor their
physical characteristics were described in detail.
Here we describe the semidomes of U. lactea
and aspects of semidome building behavior in
relation to the possible function of these
structures for reducing aggression or attracting
females.
Some other fiddler crabs usually build their
structures in the mid intertidal zone, where they
are destroyed daily by tides (e.g., pillars in Uca
beebei, see Backwell et al., 1995; hoods in Uca
musica, see Christy et al., 2001). In contrast, the
semidomes of U. lactea are built in the upper
intertidal zone in the Korean intertidal mudflat
above the reach of the tide for several days
during the neap tides. We tested whether
sediment moisture content affects structure
building by destroying semidomes on days that
the habitat was, and on days it was not, covered
by the tide. We evaluated whether semidomes
reduce aggression by monitoring semidome
building over the natural range in male density.
If males build semidomes in response to
declining nearest-neighbor distances, then their
numbers should increase disproportionately
with density (Zucker, 1981). For example,
given a random dispersion of burrows, the
mean nearest-neighbor distance ? 1/2
p
density
673
JOURNAL OF CRUSTACEAN BIOLOGY, 24(4): 673?679, 2004
(Clark and Evans, 1954). If semidomes are
a sexual signal to females, then males should
build them more often when females are present
than when they are absent. We tested this by
manipulating male and female density in
enclosures and monitoring hood building.
MATERIALS AND METHODS
Study Species
Uca lactea live in the upper intertidal zone on mudflats in the
temperate, tropical and subtropical Indo-Pacific (Kim, 1973;
Crane, 1975). In South Korea, during the cold months from
November to April, this species remains underground in
Fig. 1. (A) Male Uca lactea with a semidome, (B) lateral view of a semidome.
674 JOURNAL OF CRUSTACEAN BIOLOGY, VOL. 24, NO. 4, 2004
burrows. On the warmer days in early May, as the tide recedes
and exposes their habitat, they emerge and feed almost
continuously on organic matter on the sediment surface. When
the tide covers their habitat, at night, and during heavy
rains, crabs remain in their burrows which they plug with
mud. Except when mating, crabs occupy burrows alone.
At the beginning of the breeding season in early June,
males dig their burrows, which have one entrance, to a depth
of about 40 to 50 cm. During 3- to 5-day periods of intense
courtship twice each lunar month, many males construct
semidomes out of mud at the entrance to their burrow. Males
court passing females by waving claws toward them and
females respond by approaching and briefly entering males?
burrows. Females sequentially visit several males before
they choose a mate by staying in a male?s burrow. In some
cases, about 15 to 30 minutes later, the male emerges and
walks away leaving the burrow to the female, which she then
plugs with mud from below. In other cases, the male may
plug the burrow and stay with the female for a day or more
before leaving. Finally, males sometimes approach neigh-
boring females, follow them to their burrows, and then tap
for several minutes on the edge of the burrow opening with
their walking legs while the female is below. If the female
emerges, the crabs mate on the surface, after which the male
returns to his burrow.
Study Site
The study was conducted at the ??Choji?? intertidal mudflat in
Kilsang-myon, Kanghwa Island (378359N, 1268329E) from
June to August in 2000?2001. This period included the
monsoon season in South Korea. The study site is along the
intracoastal waterway, downstream of the Han River, about
1 km from Daemyong Port in the Kimpo Area. The sediment
was primarily silt (sand: 6.9?17.2%; silt: 72.9?82.8%; clay:
5.9?10.3%; mean grain diameter: 4.4 phi) and contained
0.5% total organic carbon. Maximum tidal height ranged
from 10 to 930 cm. Uca lactea lives in a 400?500 m2 area
on the upper intertidal mudflat at from 700 to 850 cm
elevation. This area is not flooded by the tides and dries out
for 5?6 days each semilunar cycle.
Semidome Building Behavior
In an area with high male density from July 1?August 3,
2000, we delimited a 2 3 2 m plot with nylon strapping
supported by four 50-cm-long stakes of PVC tubing. Each
male?s burrow inside the plot was marked with a small
numbered flag. We sat in a chair near the plot and observed
crabs directly or through binoculars. For the three days when
most males courted during each semilunar period, we made
notes on each crab?s behavior and recorded the number of
semidomes on the surface every hour.
The Relationship Between Male Size and Burrow Size
We determined the relationship between male size and the
diameter of the burrow entrance by measuring the entrance
diameter of males? burrows with vernier calipers (0.1 mm
precision). We then captured the males at these burrows by
placing the chisel-shaped end of a 1.2 m bamboo stick next
to the burrow opening. When the male emerged and walked
about 20 cm away, we quickly blocked the burrow entrance
with the stick and enclosed the male with a 10 cm high
plastic ring. Using the vernier calipers we measured the
carapace width and the propodus length of the large claw of
each male.
Seasonal Variation in the Sizes of Semidome Builders
To determine whether the size of semidome builders shows
seasonal variation, we randomly chose burrows with semi-
domes and measured the entrance diameter and the height
and width of the semidome using vernier calipers. We made
these measurements on June 11 (n? 11), June 18 (n? 18),
July 5 (n ? 13), July 26 (n ? 20), and August 7 (n ? 19),
2000. In addition, we captured active males separately and
measured the carapace width of active crabs on June 4 (n?
6), June 18 (n? 30), July 5 (n?23), July 17 (n? 6), July 29
(n ? 37), and August 1 (n ? 19) to see whether the size of
active crabs changes with the season.
Semidome Rebuilding
To determine whether sediment moisture content affects the
ability of males to build semidomes, we destroyed 36
randomly selected semidomes during low tide, shortly after
they were built, on days when the previous tide had covered
the habitat. We also did this on days that the habitat had not
been inundated and the sediment was relatively dry. Thirty
minutes after we destroyed these semidomes we recorded
how many were rebuilt.
Social Context and Semidome Building
If males build semidomes to reduce the frequency with
which they interact aggressively with other males, then they
should build them whether females are present or not. In
contrast, if males build semidomes to attract females, they
should build them only when females are present. We tested
these predictions by putting males in two enclosures in the
field, one with males only and the other with both sexes, and
we recorded the frequency of semidome building. The
enclosures were 1 3 1 m2 with walls of stainless steel mesh
(, 0.5 mm mesh width) 30 cm tall buried to 15 cm in the
mud. The corners were fastened with wire and the top edge
was capped with split plastic pipe (3 cm diameter) and elbow
fittings forming a barrier that the crabs could not climb. We
selected the area where crabs are relatively rare, and then set
up the enclosures. We removed all burrow-holding crabs
from inside the enclosures by capturing them using the
method described above. This process was done for 10 days
before the experiment started. We measured each male?s
carapace width and length and the length of the propodus
and individually marked each with paint and adhesive labels.
We made burrows for males inside the enclosures by forcing
bamboo sticks (1 cm diameter) into the mud in an evenly
spaced array. We added 20 males to each plot and 10 adult
females to one plot only. Male size did not differ between
plots (Mann-Whitney U-test, carapace width: U?163.5, Z?
0.987, P? 0.32; carapace length, U? 162.5, Z?
1.01, P?
0.31; major claw length: U? 159, Z?
1.10, P? 0.27).
From June 1 to 30, 2001, a period spanning two
semilunar cycles, we observed the enclosures daily from 3
h before to 3 h after low tide. On June 7, males began
semidome building. We observed the crabs from about 1 m
away while sitting on a chair. We recorded the number of
males active on the surface, the number of semidome
builders, the times that semidomes were built, and the
number of semidomes. We excluded from the analysis
observations on days that few crabs were active either
because low tide occurred at dusk (June 15?18) or because it
rained heavily (June 24, 29?30). On June 11?14 and 27?28,
the high tide did not reach the enclosures, and the sediment
became too dry for crabs to build semidomes. On those days,
we destroyed any semidomes remaining from the previous
day, and we poured 10 L of seawater into each enclosure to
675KIM ET AL.: SEMIDOME BUILDING AS COURTSHIP SIGNAL
extend the days possible for semidome building. We did this
because we wanted to know the exact number of semidome
builders on each day regardless of sediment condition and
the number of semidomes built previously.
RESULTS
Semidome Building Behavior
Males began building semidomes in early June
at the start of the breeding season. They built
semidomes by scraping up bits of sediment from
the surface with their four walking legs on the
side with the major claw, and stacking the
material at the edge of their burrow entrance
(Fig. 2). Males did not finish their semidomes at
once but rather added material to them in
repeated trips over many minutes. Over 80
percent of males completed their semidomes by
the time of low tide (Fig. 3A) when the number
of waving males peaked (Fig. 3B).
The Relationship Between Male and
Semidome Size
Semidomes averaged 17.6 6 5.0 mm (n ? 38)
high and 26.7 6 5.6 mm (n? 37) wide. Burrow
entrance diameter was significantly correlated
with male carapace width (r ? 0.729, n ? 34,
P , 0.0001) and major claw length (r? 0.705,
n?33, P , 0.0001). Males with larger diameter
burrows (larger males) built higher and wider
semidomes (height ? 5.699 ? 1.161 3 entrance
diameter, r ? 0.344, F1,36 ? 4.839, P ? 0.03;
width? 6.359? 1.999 3 entrance diameter, r?
0.529, F1,35 ? 13.57, P , 0.001; Fig. 4a).
Semidome height and width were correlated
significantly (r ? 0.383, n ? 37, P , 0.0001,
Fig. 4b; width? 2.678? 0.56 3 height; F1,35 ?
13.0, P , 0.0001).
Seasonal Variation in Body Size of
Semidome Builders
There was significant variation in the burrow
entrance diameter of semidome builders across
the season (One-way ANOVA, F? 4.86, d.f.?
4, 76, P , 0.002; Fig. 5). Pair-wise compar-
isons of dates (Fisher?s PLSD) showed signif-
icant differences in entrance diameter between
June 18 and July 26 (P ? 0.039), June 18 and
August 7 (P? 0.0003), and July 5 and August 7
(P , 0.001). The sizes of semidome builders
Fig. 2. Semidome building: (A) dragging mud, (B)
stacking mud, and (C) modifying the structure.
Fig. 3. Mean (6 SD) numbers of (A) semidomes built and
(B) waving males on the surface of the 2 3 2 m plot in
relation to the time of low tide.
676 JOURNAL OF CRUSTACEAN BIOLOGY, VOL. 24, NO. 4, 2004
decreased toward the end of the breeding
season. However, the carapace width of ac-
tive males did not differ between any sampling
dates (mean range: 14.03?15.03 mm; ANOVA,
F5,115 ? 0.590, P? 0.70).
The Density of Waving Males and
Semidome Building
The number of semidomes did not increase
disproportionately with the number of waving
males as expected if a density-dependent
mechanism affects the proportion of dome
builders. Semidomes increased linearly with
waving-male density from 1 to 25 males/4m2
(r? 0.939, n ? 16, P , 0.0001; Fig. 6).
Semidome Rebuilding
On the days when the habitat was covered by
the previous tide, 84.0% (n? 137 of 163) of the
courting males built semidomes. On these days,
17 of 36 males (47.2%) rebuilt semidomes
within 30 min after we destroyed them at low
tide. On the days when the tides did not flood
the habitat, 82.1% (n? 55 of 67) of the wavers
had semidomes at their burrow entrance. None
of the 36 males rebuilt their semidomes from the
dry mud after we destroyed them.
Social Context
In the enclosure with females, each male built
more semidomes than did the males that had no
contact with females (t? 2.933, n? 19, 20, P?
0.0057), whereas there was no significant
Fig. 4. (A) Semidome width (open squares) and height
(filled circles) in relation to burrow entrance diameter. (B)
Semidome width in relation to height. The lines are from
least squares regression.
Fig. 5. Seasonal changes in burrow entrance diameter
(mean 6 SD) of semidome builders. See Materials and
Methods for sample sizes.
Fig. 6. The relationship between the numbers of waving
males and semidomes in a 2 3 2 m plot. Observations were
made on 16 clear days. The line is from least squares
regression (Y ? 
0.531 ? 0.872 X; r2adj ? 0.874, P ,
0.0001).
677KIM ET AL.: SEMIDOME BUILDING AS COURTSHIP SIGNAL
difference in the number of days each male was
active in the two enclosures (t?
1.377, n? 19,
20, P ? 0.17).
DISCUSSION
The results of this study suggest that the primary
function of semidomes in Uca lactea is related to
courtship signaling but not to aggression between
males. During the breeding season in June
through August, males built semidomes on the
days of each semilunar cycle that they courted
females with claw waving. The number of semi-
domes increased linearly with the number of
waving males, which would not be expected if
males built semidomes in response to the non-
linear decrease in interburrow distance with
density. Finally, males built significantly more
semidomes when females were present in the
experimental enclosure than they did when they
had only male neighbors. Because we added
females to only one enclosure, the densities in the
two enclosures were not equal, and this density
difference could have affected our results. How-
ever, usually only two to three females were
active each day in the female-added enclosure,
and we doubt this relatively small difference in
density is sufficient to explain the highly
significant difference in semidome building
between the two enclosures.
Zucker (1981) suggested that male U. musica
build hoods to reduce territorial overlap and rates
of aggression with neighbors, giving each male
more time to court during limited mating periods.
Christy (1988a) tested this idea in Uca beebei and
found that the pillars males of this species build
do not affect space use or aggression between
neighbors. Instead, Christy (1988b) found that
males with pillars were more attractive to females
perhaps because they elicit landmark orientation,
a behavior that reduces the risk of predation on
crabs that move away from the immediate safety
of a burrow, as do mate-sampling females
(Christy, 1995). This idea subsequently received
strong support from an experimental study of
the function of hoods in Uca musica (Christy
et al., 2001, 2002, 2003a, b).
The semidome destruction test showed that
almost half of the semidome builders would
rebuild them but not when the previous neap
high tide was too low to have covered the
habitat. Semidome building probably is limited
by the water content of the sediment. Alterna-
tively, it is possible that the tendency for males
to attempt to rebuild is lower during the neap
tides. However, this does not seem to be the
case because nearly the same proportion of
active males built semidomes during each
period. When we poured water on the dry
sediment during the neap tides, males were
stimulated to build semidomes. The general
level of semidome building was the same
because of the almost same proportion of
semidomes built by courting males irrespective
of tidal periods. If female U. lactea orient to
structures when they move between males as
they search for a mate as do female U. musica
and U. beebei, then males that have their
semidomes destroyed during neap tides will be
less attractive to females.
The entrance diameter of burrows with semi-
domes, and hence, the sizes of semidome
builders, decreased towards the end of the mating
season. Zucker (1984) found that smaller court-
ing males from a low-density population failed to
court when they were placed in a high-density
population with larger males. She suggested that
smaller males delay courtship in response to
competition from larger males. In U. annulipes,
females tended to visit smaller males more often
later in the semilunar cycle, suggesting that the
benefits of investing in structure building vary
with male size and the timing of the investment
(Jennions and Backwell, 1998). It is unclear
whether competitive inhibition of structure
building by smaller males or variation in the
allocation of time to this behavior according to
potential benefits explains why smaller U. lactea
built more structures later in the season.
In a separate study, Kim and Choe (2003)
found that food availability influenced not only
the intensity of semidome building but also the
onset of courtship behavior. This suggests that
semidome building may signal male condition as
affected by food availability. In most Uca species
in which males build structures, females visit
several males before they choose a mate, and the
presence of a structure has been shown to
increase male attractiveness. This should create
directional selection for more frequent structure
building (Christy et al., 2001) and perhaps for
building large structures that might be seen at
greater distances. We showed that larger males
built wider and higher semidomes, as expected if
they are a signal under directional selection, but
we do not know whether male attractiveness
increases with semidome size. Indeed, there
currently is no strong evidence that female fiddler
crabs prefer large males, males with relatively
large claws, or males with other large signals.
678 JOURNAL OF CRUSTACEAN BIOLOGY, VOL. 24, NO. 4, 2004
ACKNOWLEDGEMENTS
We thank Sanha Kim for reading the manuscripts. The
Benthic Ecology Laboratory of SNU analyzed the sediment
characteristics of the study site. This study was supported by
the Brain Korea 21 Research Fellowship from the Korean
Ministry of Education and Human Resource Development.
LITERATURE CITED
Backwell, P. R. Y., M. D. Jennions, J. H. Christy, and
U. Schober. 1995. Pillar building in the fiddler crab Uca
beebei: evidence for a condition-dependent ornament.?
Behavioral Ecology and Sociobiology 36: 185?192.
Christy, J. H. 1988a. Pillar function in the fiddler crab Uca
beebei (I): effects on male spacing and aggression.?
Ethology 78: 53?71.
???. 1988b. Pillar function in the fiddler crab Uca beebei
(II): competitive courtship signalling.?Ethology 78:
113?128.
???. 1995. Mimicry, mate choice, and the sensory trap
hypothesis.?American Naturalist 146: 171?181.
???, J. Baum, and P. R. Y. Backwell. 2003a.
Attractiveness of sand hoods built by courting male
fiddler crabs Uca musica: test of a sensory trap
hypothesis.?Animal Behaviour 66: 89?94.
???, P. R. Y. Blackwell, and S. Goshima. 2001. The
design and production of a sexual signal: hoods and hood
building by male fiddler crabs Uca musica.?Behaviour
138: 1065?1083.
???, ???, ???, and T. Kreuter. 2002. Sexual
selection for structure building by courting male fiddler
crabs: an experimental study of behavioral mecha-
nisms.?Behavioral Ecology 13: 366?374.
???, ???, and U. Schober. 2003b. Interspecific
attractiveness of structures built by courting male fiddler
crabs: experimental evidence of a sensory trap.?
Behavioral Ecology and Sociobiology 53: 84?91.
Clark, P. J., and F. C. Evans. 1954. Distance to nearest
neighbor as a measure of spatial relationships in
populations.?Ecology 35: 445?453.
Clayton, D. A. 1988. Hood construction as a spacing
mechanism in Cleistostoma kuwaitense (Crustacea: Ocy-
podidae).?Marine Biology 99: 57?61.
Crane, J. 1975. Fiddler Crabs of the World. Princeton
University Press, Princeton.
Jennions, M. D., and P. R. Y. Backwell. 1998. Variation in
courtship rate in the fiddler crab Uca annulipes: is
it related to male attractiveness??Behavioral Ecology 6:
605?611.
Kim, H. S. 1973. Illustrated Encyclopedia of Fauna & Flora
of Korea. Vol. 14. Anomura Brachyura. Korean Ministry
of Education.
Kim, T. W., and J. C. Choe. 2003. The effect of food
availability on the semilunar courtship rhythm in the
fiddler crab Uca lactea (De Haan) (Brachyura:
Ocypodidae).?Behavioral Ecology and Sociobiology
54: 210?217.
???, K. W. Kim, R. B. Srygley, and J. C. Choe. 2004.
Semilunar courtship rhythm of the fiddler crab Uca lactea
in a habitat with great tidal variation.?Journal of
Ethology 22: 63?68.
Linsenmair, K. E. 1967. Konstucktion und Signalfunktion
der Sandpyramide der Reiterkrabbe Ocypode saratan
Forsk. (Decapoda Brachyura Ocypodidae).?Zeitschrift
fu?r Tierpsychologie 24: 403?456.
Oliveira, R. F., P. K. McGregor, F. R. L. Burford, M. R.
Custo?dio, and C. Latruffe. 1998. Function of mudballing
behaviour in the European fiddler crab Uca tangeri.?
Animal Behaviour 55: 1299?1309.
Wada, K., and I. Murata. 2000. Chimney building in the
fiddler crab Uca arcuata.?Journal of Crustacean Biology
20: 505?509.
???, S. Yum, and J. K. Park. 1994. Mound building in
Ilyoplax pingi (Crustacea: Brachyura: Ocypodidae).?
Marine Biology 121: 61?65.
Yamaguchi, T. 1971. The courtship behavior of a fiddler
crab, Uca lactea.?Kumamoto Journal of Science Bi-
ology 10: 13?37.
Zucker, N. 1974. Shelter building as a means of reducing
territory size in the fiddler crab, Uca terpsichore
(Crustacea: Ocypodidae).?American Midland Naturalist
91: 224?236.
???. 1981. The role of hood-building in defining
territories and limiting combat in fiddler crabs.?Animal
Behaviour 29: 387?395.
???. 1984. Delayed courtship in the fiddler crab Uca
musica terpsichores.?Animal Behaviour 32: 735?742.
RECEIVED: 4 November 2003.
ACCEPTED: 14 May 2004.
679KIM ET AL.: SEMIDOME BUILDING AS COURTSHIP SIGNAL