Southeastern Naturalist
523
K.J. Farrell, R.P. Creed, and B.L. Brown
22001144 SOUTHEASTERN NATURALIST 1V3o(3l.) :1532,3 N–5o2. 93
Reduced Densities of Ectosymbiotic Worms (Annelida:
Branchiobdellida) on Reproducing Female Crayfish
Kaitlin J. Farrell1,3,*, Robert P. Creed1, and Bryan L. Brown2
Abstract - Cleaning symbioses provide net benefits by improving each partner’s fitness.
Ectosymbiotic Cambarincola spp. (branchiobdellidans) can increase growth and survival
of Cambarus chasmodactylus (New River Crayfish), but the nature of the symbiosis might
change with female reproductive state because brooding offspring (eggs, young) and worms
inhabit the same surfaces. Here, we present the results of field surveys that examined
whether the number and location of branchiobdellidans on New River Crayfish varies as a
function of female crayfish reproductive state. Reproducing female New River Crayfish had
fewer total worms, an absence of cocoons, and a relatively greater proportion of worms on
lateral body surfaces than non-reproducing crayfish. The altered distribution and reduced
abundance of worms suggest that the symbiosis changes with female reproductive status,
but additional experiments will be needed to identify the mecha nism responsible.
Introduction
Cleaning symbioses are interspecific interactions in which a cleaner species
removes detritus, parasites, or other epibionts from the client organism, and have
been considered textbook examples of mutualism (Bshary et al. 2007, Losey 1979,
Poulin and Grutter 1996). While much of the literature on cleaning symbioses has
focused on interactions between cleaner and client fishes on coral reefs (Cheney
and Côté 2003, Limbaugh 1961, Poulin and Grutter 1996), some species of crayfish
and branchiobdellidan worms (Annelida, Branchiobdellida) also appear to engage
in cleaning symbioses (Brown et al. 2002, 2012; Lee et al. 2009 ).
The nature of the crayfish–branchiobdellidan relationship can change depending
on the density of branchiobdellidans on a crayfish (Brown et al. 2002, 2012)
and environmental conditions (Lee et al. 2009). Cambarus chasmodactylus James
(New River Crayfish) hosts an ectosymbiotic branchiobdellidan, Cambarincola
ingens Hoffman, which removes particulate matter and fouling epibionts from the
crayfish exoskeleton including the gills (Brown et al. 2002). Lab and field experiments
have indicated that intermediate densities of worms can have direct positive
effects on the crayfish by increasing growth and survival compared to crayfish
without C. ingens (Brown et al. 2002, 2012). Lee et al. (2009) also found positive
effects of a mixed community of four species of branchiobdellidans on growth rates
of a host crayfish, Cambaroides similis Koelbel, in high-fouling environments but
no effects in low-fouling environments. The nature of these relationships may also
1Department of Biology, Appalachian State University, Boone, NC 28608. 2Department of
Biological Sciences, Virginia Tech, Blacksburg, VA 24061. 3Present address - Odum School
of Ecology, University of Georgia, Athens, GA 30602. *Corresponding author - kfarrell@
uga.edu.
Manuscript Editor: Nathan Dorn
Southeastern Naturalist
K.J. Farrell, R.P. Creed, and B.L. Brown
2014 Vol. 13, No. 3
524
change with the reproductive status of female crayfish. Branchio bdellidans inhabit
the same abdominal surfaces as brooding offspring (eggs, young), and may affect
crayfish offspring and/or be removed by female cleaning behaviors. Altered worm
counts and distributions on reproducing female crayfish could indicate that the nature
of the symbiosis depends on host reproductive state. To determine whether the
relationship is stable through reproductive cycles, we quantified Cambarincola spp.
abundance and distribution on reproducing and non-reproducing female crayfish;
here, we present the results of our field survey .
Methods
Field surveys
We conducted field surveys of Cambarus chasmodactylus and Cambarincola
spp. at 4 locations in the New River watershed in Watauga County, NC (36°12'N,
81°42'W). We collected crayfish and worms during the summer of 2011 from early
June to early August in three 3rd-order tributaries of the South Fork of the New
River: Meat Camp Creek, Howard’s Creek, and the Middle Fork of the New River.
New River Crayfish is a dominant crayfish in these tributaries (Fortino and Creed
2007). We also collected crayfish and worms in the South Fork of the New River, a
4th-order stream, where adult New River Crayfish are co-dominant with Orconectes
cristavarius Taylor (Spiny Stream Crayfish) in the sampled areas (Fortino and
Creed 2007, Helms and Creed 2005). Reaches surveyed in each stream were short
(~100 m), and the predominant substrate in all reaches was cobb le/boulder.
We captured crayfish by lifting boulders from the stream bed and washing crayfish
into dip nets; exposed individuals on the stream bed were also captured with
dip nets. After capture, we placed crayfish in individual, covered plastic containers
filled with stream water to prevent possible transfer of worms between crayfish. We
examined the exterior of each female crayfish on site by holding the crayfish in a
shallow dish of water and visually examining ventral, lateral, and dorsal body surfaces
using a 10x OptiVisor binocular headband magnifier (Donegan Optical Company,
Lenexa, KS). Cambarincola spp. attached to the exterior carapace are easily
located on submerged crayfish by the swaying of the anterior portion of their bodies
(Brown and Creed 2004). For each crayfish, we recorded total carapace length (CL,
from the tip of the rostrum to the posterior margin of the cephalothorax), the location
(dorsal, lateral, or ventral crayfish surfaces) and number of large, potentially
reproductive (≥6 mm; Brown et al. 2002, 2012) and small (<6 mm) Cambarincola
spp., and the location and number of branchiobdellidan cocoons. Other researchers
have found that most visible branchiobdellidans on the external surfaces of
New River Crayfish in these tributaries are C. ingens; thus, we are confident that
many of the field-identified worms were C. ingens, though a few may have been
Cambarincola philadelphicus Leidy. For this reason, we will refer to the worms as
Cambarincola. The only other branchiobdellidan species observed on the external
surfaces of New River Crayfish during these surveys was Pterodrilus alcicornis
Moore, which we recovered from a few crayfish in Howard’s Creek but did not include
in this analysis. We also examined female crayfish to determine if there were
Southeastern Naturalist
525
K.J. Farrell, R.P. Creed, and B.L. Brown
2014 Vol. 13, No. 3
attached eggs, i.e., they were ovigerous, or if they were carrying recently hatched
young; crayfish were then released.
Data analysis
We compared the number and location of large, small, and total Cambarincola
as well as the number of Cambarincola cocoons found on ovigerous, young-bearing,
and non-reproducing females. We excluded surveyed female crayfish from this
analysis if their carapace length was less than that of the smallest reproducing female
(32 mm CL) to control for reduced worm numbers on smaller crayfish (Brown
and Creed 2004). In addition, we excluded recently molted crayfish (identified by
incomplete hardening of the carapace) from this analysis because these individuals
usually have fewer worms and cocoons than intermolt individuals (R.P. Creed, unpubl.
data). The analysis thus included 18 females without eggs or attached young,
7 ovigerous females, and 3 bearing recently hatched young crayfish. The carapace
lengths of these crayfish ranged from 32 mm to 45 mm.
When assessing differences in worm number between reproducing and nonreproducing
crayfish, we used general linear models fitted to negative binomial
distributions because these models fit our overdispersed count data better than models
fitted to a Poisson distribution (Chatterjee and Simonoff 2013). We also tested
for effects of crayfish size (CL) on worm counts through a model that included crayfish
CL as a covariate. We used a chi-square goodness-of-fit test to assess whether
branchiobdellidans were distributed on crayfish exoskeletons in proportion to the
available area (sensu Brown and Creed 2004). We used Fisher’s exact test to compare
the proportion of worms found on each body surface (dorsal, lateral, ventral) between
reproducing and non-reproducing female crayfish. We used R v3.0.2 (R Core
Team 2013) to conduct our analyses. General linear models were fitted to a negative
binomial distribution using the MASS package (Venables and Ripley 2002).
Results
We captured a total of 18 non-reproducing female crayfish (32–38 mm CL), 7
ovigerous females (CL 32–45 mm), and 3 carrying recently hatched young (34–35
mm CL). There were no significant differences between the numbers of Cambarincola
nor worm cocoons present on ovigerous female crayfish and those bearing
recently hatched young (large worms: Z = -0.612, P = 0.540; small worms: Z =
-1.115, P = 0.265; cocoons: not present on any individuals), so we pooled these 2
groups for subsequent analyses and hereafter, we refer to them as reproducing females.
Worm number did not differ as a function of crayfish size because both CL
and CL × reproductive status were non-significant factors (P-values > 0.10) in the
models; CL was removed from subsequent analyses.
There were significantly more total (large and small) worms on non-reproducing
females (6.72 ± 0.87) than on reproducing females (3.10 ± 0.66; Z = -3.161, P =
0.002). Although the number of small worms did not differ between reproducing
and non-reproducing females (Fig. 1, Z = -1.440, P = 0.150), the number
of large worms was significantly lower on reproducing female crayfish, with
Southeastern Naturalist
K.J. Farrell, R.P. Creed, and B.L. Brown
2014 Vol. 13, No. 3
526
non-reproducing females hosting over two times as many large worms as reproducing
females (Fig. 1; Z = -2.976, P = 0.003).
Visible large worms were not distributed evenly on the external body surfaces
on either reproducing or non-reproducing crayfish. We found more large worms and
Cambarincola cocoons on the ventral surfaces of non-reproducing crayfish than
would be expected given an area-weighted distribution (χ2
2 df = 17.52, P = 0.0002
and χ2
2 df = 72.04, P < 0.0001, respectively). In contrast, reproducing crayfish had
fewer large worms than expected on the ventral surfaces (χ2
2 df = 7.21, P = 0.027)
and lacked any attached Cambarincola cocoons. Small worms on reproducing
crayfish were not evenly distributed, with more small worms found on the lateral
surfaces than would be expected (χ2
2 df = 6.57, P = 0.023). Small worms were distributed
evenly on non-reproducing crayfish.
Figure 1. Counts (bars) and spatial distribution (pie charts) of Cambarincola spp. and worm
cocoons on reproducing (n = 10) and non-reproducing (n = 18) female New River Crayfish.
Pie charts show mean proportion of worms and cocoons on lateral (white), ventral (medium
gray), and dorsal (dark gray) surfaces of female crayfish. Bars represent mean counts ± 1
SE. Statistical results reported on the figure are for negative binomial general linear models
comparing counts of worms on reproducing and non-reproducing females. Two asterisks
(**) denotes P < 0.01.
Southeastern Naturalist
527
K.J. Farrell, R.P. Creed, and B.L. Brown
2014 Vol. 13, No. 3
When we compared worm distributions between crayfish groups, non-reproducing
females had a greater proportion of visible large worms on the ventral surfaces
(0.52 ± 0.08) than did reproducing females (0.08 ± 0.06), where large worms were
most often found on lateral surfaces (Fisher’s exact test: two-sided P = 0.0005;
Fig. 1). Distributions of small worms also differed between crayfish groups, with
small worms found most often on the lateral surfaces of reproducing crayfish; their
distribution was more even on non-reproducing crayfish (Fig. 1, Fisher’s exact test:
two-sided P = 0.002).
Discussion
Cambarincola worms and cocoons on reproducing female New River Crayfish
were significantly reduced compared to the number on non-reproducing female
crayfish. The difference was mainly due to a significant reduction in the number
of large worms. Worms on reproducing crayfish were also located on different
surfaces of the exoskeleton than on non-reproducing crayfish, with large worms
found predominantly on the lateral surfaces of reproducing females, versus on the
ventral surfaces of non-reproducing females. The data suggest that the nature of the
symbiosis between female New River Crayfish and Cambarincola spp. may vary
depending on the reproductive state of the host, a condition that warrants experimentation
to pinpoint potential mechanisms driving the dif ferences we observed.
Several mechanisms could explain the reduced worm numbers we documented
on reproducing female crayfish. Crayfish grooming behaviors prior to egg extrusion
may contribute to the reduction in large branchiobdellidans and their cocoons. Previous
studies on the natural history of various crayfish taxa reported that females
thoroughly clean the ventral surfaces of their exoskeletons 4–5 days prior to egg
extrusion (Andrews 1904, Tack 1941). It is possible that such intensive grooming
behaviors could dislodge branchiobdellidans and their cocoons from the ventral
surface of the crayfish abdomen prior to egg-extrusion. Indeed, laboratory observations
of crayfish grooming have demonstrated that crayfish can dislodge attached
branchiobdellidans (Farrell et al. 2014, Skelton et al. 2014). Crayfish behaviors
following egg extrusion may also limit worm re-colonization of the ventral surface
of the abdomen. After the eggs are extruded and attached, females beat the pleopods
to ensure water circulation around the eggs and, after hatching, the young crayfish,
which prevents embryo fouling and death (Bauer 1989). Such frequent disturbances
could contribute to the continued reduction in large worms and cocoons found on
the ventral abdomens of female crayfish carrying recently hatche d young.
Emigration could also contribute to the observed reduction of large worms on
reproducing females. Branchiobdellidans are known to disperse between crayfish
when the hosts come into direct contact (Young 1966), thus, large worms could
leave reproducing females if the crayfish come into contact with other crayfish. Our
observed pattern of large worm reductions only on reproducing females suggests
that the worms emigrate prior to, or soon after, egg extrusion. Further investigation
is needed to determine why large worms remaining on reproducing crayfish
were not laying cocoons because cocoons were absent from the dorsal and lateral
surfaces on reproducing female crayfish.
Southeastern Naturalist
K.J. Farrell, R.P. Creed, and B.L. Brown
2014 Vol. 13, No. 3
528
The number of small worms did not differ between reproducing and non-reproducing
females, though the distribution of small worms was different between
crayfish groups. We predict that when small worms mature into large worms, they
recolonize the ventral surfaces after the crayfish reproductive period is over, though
more extensive investigations are needed to evaluate this hypothesis, as well as assess
other patterns of small-worm distribution, such as the lack of small worms on
the dorsal surfaces of reproducing crayfish.
The observed differences in worm location between crayfish groups suggest that
worms may be moving from the ventral surfaces to other locations on the crayfish
exoskeleton. Previous surveys in this system found that nearly 50% of C. ingens
were located on the ventral surface of the abdomen or on the underside of the cephalothorax
(Brown et al. 2002). Similarly, more than 80% of Cambarincola cocoons
are located on the ventral surface of the abdomen (R.P. Creed, unpubl. data). In
our surveys, this pattern held true for non-reproducing females, but on reproducing
females, the majority of large and small worms were found on the lateral surfaces,
specifically the margins of the carapace. Worms may relocate when the host extrudes
eggs to avoid reductions in their own fitness due to smothering. However,
such relocation alone would not result in a reduction in worm number as was observed
in our surveys unless some worms move into the gill chamber. Because we
only quantified worms on the exterior of the crayfish, additional investigations are
needed to test these hypotheses.
While it is possible that our counts failed to account for branchiobdellidans in
the gill chamber, extensive observations of Cambarincola in lab and field experiments
suggest that few worms, if any, spend significant amounts of time in the gill
chambers. Large worms tend to be found in the same places on the external surface
of the crayfish for extended periods (Thomas et al. 2013). Therefore, we doubt that
our counts missed a significant proportion of the worms on these crayfish. If some
large worms moved from the ventral surface to the gill chamber, it is immaterial to
the distributional result (reduced density on ventral surfaces) .
The results of our field surveys suggest that reproduction by female crayfish
may alter the nature of the symbiosis between New River Crayfish and
Cambarincola spp. such that the mutual benefits temporarily wane for one or
both partners. Further experiments will be necessary to assess the mechanisms
responsible and determine whether branchiobdellidans have measurable direct
effects on crayfish reproduction.
Acknowledgments
We are grateful to Michael J. Thomas and April L. Meeks for field assistance. Comments
from Nathan Dorn and two reviewers improved this manuscript. Funding was
provided by National Science Foundation grants to R.P. Creed (DEB-0949823) and B.L.
Brown (DEB-0949780).
Literature Cited
Andrews, E.A. 1904. Breeding habits of crayfish. American Naturalist 38:165–206.
Southeastern Naturalist
529
K.J. Farrell, R.P. Creed, and B.L. Brown
2014 Vol. 13, No. 3
Bauer, R.T. 1989. Decapod crustacean grooming: Functional morphology, adaptive value,
and phylogenetic significance. Pp. 49–73, In B.E. Felgenhauer, L. Watling, and A.B.
Thistle (Eds.). Functional Morphology of Feeding and Grooming in Crustacea. CRC
Press, Boca Raton, FL. 320 pp.
Brown, B.L., and R.P. Creed. 2004. Host preference by an aquatic ectosymbiotic annelid
on 2 sympatric species of host crayfishes. Journal of the North American Benthological
Society 23:90–100.
Brown, B.L., R.P. Creed, and W.E. Dobson. 2002. Branchiobdellid annelids and their crayfish
hosts: Are they engaged in a cleaning symbiosis? Oecologia 132:250–255 .
Brown, B.L., R.P. Creed, J. Skelton, M.A. Rollins, and K.J. Farrell. 2012. The fine line between
mutualism and parasitism: Complex effects in a cleaning symbiosis demonstrated
by multiple field experiments. Oecologia 170:199–207.
Bshary, R., R.F. Oliveria, T.S.F. Oliveira, and A.V.M. Canário. 2007. Do cleaning organisms
reduce the stress response of client reef fish? Frontiers i n Zoology 4:21–29.
Chatterjee, S., and J.S. Simonoff. 2013. Handbook of Regression Analysis. John Wiley and
Sons, Inc., Hoboken, NJ. 236 pp.
Cheney, K.L., and I.M. Côté. 2003. The ultimate effect of being cleaned: Does ectoparasite
removal have reproductive consequences for damselfish clients? Behavioral Ecology
14:892–896.
Farrell, K.J., R.P. Creed, and B.L. Brown. 2014. Preventing overexploitation in a mutualism:
Partner regulation in the crayfish–branchiobdellid symbiosis. Oe cologia 174:501–510.
Fortino, K., and R.P. Creed. 2007. Abiotic factors, competition, or predation: What determines
the distribution of young crayfish in a watershed? Hydrobi ologia 575:301–314.
Helms, B.S., and R.P. Creed. 2005. The effects of 2 coexisting crayfish on an Appalachian
river community. Journal of the North American Benthological Society 24:113–122.
Lee, J.H., T.W. Kim, and J.C. Choe. 2009. Commensalism or mutualism: Conditional outcomes
in a branchiobdellid-crayfish symbiosis. Oecologia 159:217 –224.
Limbaugh, C. 1961. Cleaning symbiosis. Scientific American 205:42–49.
Losey, G.S. 1979. Fish cleaning symbiosis: Proximate causes of host behavior. Animal
Behaviour 27:669–685.
Poulin, R., and A.S. Grutter. 1996. Cleaning symbioses: Proximate and adaptive explanations.
BioScience 46:512–517.
R Core Team. 2013. R: A language and environment for statistical computing. R Foundation
for Statistical Computing, Vienna, Austria. Available online at http://www.R-project.
org. Accessed 16 September 2013.
Skelton, J., R.P. Creed, and B.L. Brown. 2014. Ontogenetic shift in host tolerance controls
initiation of a cleaning symbiosis. Oikos 123:677–686.
Tack, P.I. 1941. The life history and ecology of the crayfish Cambarus immunis Hagen.
American Midland Naturalist 25:420–446.
Thomas, M.J., R.P. Creed, and B.L. Brown. 2013. The effects of environmental context and
initial density on symbiont populations in a freshwater cleaning symbiosis. Freshwater
Science 32:1358–1366.
Venables, W.N., and B.D. Ripley. 2002. Modern Applied Statistics with S. 4th Edition.
Springer, New York, NY. 497 pp.
Young, W. 1966. Ecological studies of the branchiobdellidae (oligochaeta). Ecology
47:571–578.