2009 SOUTHEASTERN NATURALIST 8(3):479–494
Life History and Ecology of Cambarus halli (Hobbs)
Susan Dennard1,3, James T. Peterson2,*, and Edwin S. Hawthorne1
Abstract - The life history of Cambarus halli, a crayfish endemic to the Tallapoosa
River Basin, GA, was studied at four sites within the Tallapoosa River. Two sites had
allopatric populations of C. halli, and two sites had populations of C. halli sympatric
with C. englishi. Three age classes existed across sites. For Cambarus halli, total
number of pleopodal eggs was positively related to carapace length, but egg size
was only weakly positively related to carapace length. Cambarus halli were smaller
across age classes at sympatric sites, but had greater growth rates than at allopatric
sites. Cambarus halli density estimates were lower at sympatric sites, while proportions
of reproductively active age-1 and age-2 individuals were higher at allopatric
sites (63% vs. 33%).
Introduction
Crayfish are important in aquatic ecosystems, making up a large proportion
of aquatic system biomass (Griffith et al. 1994, Rabeni 1992). They are
critical in food webs as processors of leaf litter (Griffith et al. 1994, Huryn
and Wallace 1987) and as important food for predatory fish (Probst et al.
1984, Roell and Orth 1993). Crayfish support recreational and commercial
bait fisheries, serve as a popular human food (Taylor et al. 1996), and are
used as bioindicators due to their sensitivity to organophosphates and carbamates,
two widely used classes of pesticides (Hyne and Maher 2003).
In North America, 122 of the 363 known species of crayfishes are either
imperiled or extinct (Taylor et al. 2007). Effective crayfish conservation
strategies need to be developed to protect and recover these important ecosystem
components. However, the development of effective conservation
measures for crayfish species requires information on reproductive biology
and population dynamics. This information is often lacking, as basic distribution
and life-history information is known for less than 40% of crayfish
species in North America (Taylor et al. 2007).
Taxa with small distributional ranges are considered particularly vulnerable
to extirpation, because of habitat degradation and destruction,
relative to more cosmopolitan species (Gilpin and Soule 1986). One such
stream-dwelling crayfish species, Cambarus halli Hobbs, is endemic to the
Tallapoosa River Basin, a small (12,121 km2) basin located in the northwest
Georgia Piedmont (Brouchard 1978, Ratcliffe and DeVries 2004). Cambarus
halli are similar in size (maximum carapace length = approximately 35–40
1Warnell School of Forestry and Natural Resources, University of Georgia, Athens,
GA 30602. 2US Geological Survey, Georgia Cooperative Fish and Wildlife Research
Unit, Warnell School of Forestry and Natural Resources, University of Georgia, Athens,
GA 30602.3Current address - Great Lakes Institute for Environmental Research,
University of Windsor, Windsor, ON, Canada, N9B 1G6. *Corresponding author -
peterson@warnell.uga.edu.
480 Southeastern Naturalist Vol. 8, No. 3
mm) and appearance (Freeman et al. 2003) to a congener, Cambarus englishi
Hobbs and Hall. It is believed that C. halli prefers fl owing-water habitats
and is often sympatric with C. englishi (Bouchard 1978, Freeman et al.
2003, Hobbs and Hall 1972). Allopatric C. halli reportedly use all habitat
types, but are found primarily in non-riffl e habitats when they occur with C.
englishi (Hobbs 1981). However, little is known about the reproductive and
life-history characteristics of C. halli.
Most life-history studies of crayfish in North America are based on studies
of a single population located at single location (Muck et al. 2002). Here
we studied the life-history characteristics of four populations of C. halli to
evaluate how life history can vary in the presence of a sympatric congener.
Thus, the goal of this study was to examine the life-history characteristics
and ecology of C. halli, by completing the following objectives: 1) to determine
the age-class structure, 2) to evaluate the chronology of reproductive
events and estimate fecundity, 3) to evaluate the seasonal growth, 4) to determine
the seasonal habitat use, and 5) to evaluate the potential infl uence of
an abundant congener, C. englishi.
Methods
Study sites
The Tallapoosa River fl ows south-southwest from its headwaters in
Paulding County, GA, eventually joining the Coosa River to form the Alabama
River (GA DNR 2002). Of the 12,121 km2 in the Tallapoosa River
drainage, 15% is in Georgia (GA DNR 2002). The river basin is characterized
by rolling hills and is almost entirely in the Upper Piedmont (GA DNR
2002). Four sites were chosen for this study based on an extensive survey of
the basin (Freeman et al. 2003): one on Blalock Creek, one on Kiser Creek,
and two areas on the mainstem of the Tallapoosa River (Fig. 1). Blalock and
Kiser Creeks represented allopatric populations of C. halli, whereas the Tallapoosa
River sites represented sympatric sites that contained both C. halli
and C. englishi.
Blalock Creek and Kiser Creek were second-order streams that emptied
into Walker Creek and Holcomb Creek, respectively. The sites were approximately
250 m long and characterized by runs, riffl es, and pools. The
substrate composition at each site was similar and was predominantly cobble
and gravel, with some boulders, bedrock, and sand. The two sites on the
mainstem of the Tallapoosa River were fourth-order, with the first located
upstream of the Highway 27 crossing and the second located downstream
of the Mount View Road crossing. Both mainstem sites were approximately
250 m long with varying combinations of runs, riffl es, and pools and also
containing sand, silt, cobble, gravel, and bedrock substrate.
Field sampling
Sampling occurred between December 2001 and June 2003. At approximately
monthly intervals, samples were used to determine phases of
the reproductive cycle of C. halli, including fecundity, time of oviposition,
hatching of young, and timing of the young’s departure from the female. A
2009 S. Dennard, J.T. Peterson, and E.S. Hawthorne 481
1-m2 quadrat sampler, described by Rabeni (1985), with 6-mm netting and
a 0.75-m bag was used to sample crayfish. Each study site was divided by
mesohabitat (i.e., pool, riffl e, and run) and two to four replicates of each
mesohabitat type were sampled. Within each mesohabitat, a minimum of 3
and maximum of 10 quadrat samples were taken to ensure the mesohabitat
was thoroughly sampled. Captured crayfish were placed into buckets designated
for the particular replicate mesohabitat in which they were captured,
thus providing for easy release back into the same location. Semi-monthly
sampling occurred during the months that oviposition, hatching of young,
and departure of young from the female were thought to occur (approximately
March to May). At least 50 individuals of the target species were
collected during each sampling period; if fifty were not collected via the
quadrat sampler, backpack electrofishing or simple hand capture was used to
supplement. Data collected for each crayfish at each mesohabitat included:
species, carapace length (CL; from tip of rostrum to posterior border of the
thoracic region), and sex. Additionally, reproductive capacity was recorded,
i.e., if males were Form I or II and if females had active glair. Glair was
identified as the white cement substance located around the base of each
pleopod that attaches the eggs to the pleopods (Stephens 1952a). Captured
crayfish then were released back into the same mesohabitat unit from which
they were originally sampled. However, approximately 10 randomly selected
female crayfish per sampling period were kept and preserved in 70%
Figure 1. The location of the Tallapoosa River Basin, GA and the four C. halli study
sites indicated by gray circles.
482 Southeastern Naturalist Vol. 8, No. 3
ethanol to facilitate fecundity estimation. Following sampling of crayfish,
water chemistry parameters (temperature, dissolved oxygen, conductivity,
and turbidity) were also measured with calibrated meters and recorded.
During the summer of 2001 and 2002, mesohabitat availability was
estimated at each study site at basefl ow discharge. The location of each mesohabitat
was recorded with a GPS unit. The boundaries of each mesohabitat
were then delineated using the depth-based criteria of McKenney (2001)
and the dimensions measured using a tape measure. The relative amount of
each mesohabitat was estimated for each site by summing the area of each
mesohabitat type and dividing by the total study site area (i.e., the total area
of all the habitats measured within each site; Table 1).
Analysis and laboratory procedures
Fecundity was determined by dissecting ovaries from the subsample
of females that were kept from each sampling period (following Stephens
1952b). Eggs on pleopods were also counted. Mean egg diameter for pleopod
eggs was determined by measuring ten randomly selected eggs with a
dissecting scope; if the total number of eggs was less than ten, all eggs were
measured. Linear regression analysis (Neter et al. 1996) was used to compare
the total number of ovarian and pleopod eggs and mean size of pleopod
eggs to carapace length, and the precision of model coefficients was assessed
by calculating 90% confidence intervals (CI).
Crayfish density in each sampled replicate mesohabitat was estimated
by dividing the total number of crayfish collected in the mesohabitat by the
total number of quadrat samples. Age-class structure was determined with
monthly length-frequency histograms, a method that was shown to be at least
80% accurate (Momot 1967). Because each size class was not distributed
equally throughout mesohabitats, mesohabitat-specific density estimates
were extrapolated by multiplying the estimated density of each 1-mm size
group by the mesohabitat availability proportion shown in Table 1.
Increases in the modal carapace length for an age group through time can
be used to estimate crayfish growth (Baker et al. 2008, Hamr and Berrill 1985,
Muck et al. 2002). We used a linear regression analysis (Neter et al. 1996) to
examine the relationship between age class-specific modal carapace length
and sample month, and interpreted the slope of the relationship as the monthly
growth rate. Because newly hatched C. halli (age 0) were not recruited to our
sampling gear until June (see Results), we coded months beginning with June
= 0 and ending with May = 11. We examined the relationship between C. halli
Table 1. Total surface area and mesohabitat composition of the four study sites expressed as a
percentage of the total surface area.
Pool Riffl e Run
Study site Area (m2) (% area) (% area) (% area)
Blalock Creek 1370.9 28.1 32.3 39.6
Kiser Creek 1175.1 36.3 32.2 31.5
Tallapoosa River at Highway 27 2073.8 15.7 17.3 67.0
Tallapoosa River at Mount View Road 2010.0 11.6 20.5 67.9
2009 S. Dennard, J.T. Peterson, and E.S. Hawthorne 483
monthly growth rate and crayfish age, and the presence of C. englishi by including
predictor variables for age and sympatry. The three C. halli age classes
(see Results) were categorized by assigning binary variables (0, 1) to age 0 and
age 2 crayfish, with age 1 crayfish retained as the baseline. Sympatric and allopatric
sites were similarly coded using a binary variable, with sympatric sites
coded as 1 and allopatric sites coded as 0.
An information-theoretic approach, described by Burnham and Anderson
(2002), was used to evaluate the relative plausibility of models relating sympatry
and age to C. halli growth rate. We constructed a global model (with all
predictors) and a subset of 14 candidate models representing hypotheses about
the relative effects of age and sympatry on growth (Table 2a). Our primary
hypothesis of interest was the infl uence of C. englishi presence on growth
and size at age. Candidate models also included a quadratic term for time of
year to account for likely nonlinearities in growth. To assess the plausibility
of each candidate model, we calculated AIC with the small-sample bias
adjustment (AICc; Hurvich and Tsai 1989) and Akaike weights, as described
in Burnham and Anderson (2002). We also computed model-averaged parameter
estimates of the individual coefficients and standard errors (Burnham
and Anderson 2002) for the predictor variables that occurred in one or more
candidate models with weights within 10% of the largest weight (i.e., the bestapproximating
model). The precision of model-averaged coefficients was
assessed by calculating 90% CI. Goodness-of-fit was assessed for the global
model by examining residual and normal probability plots (Neter et al. 1996).
Dependence among samples collected at a site was examined by conducting
an analysis of variance (ANOVA) on the residuals from the global model, with
significant differences indicating spatial dependence (i.e., autocorrelation).
The information-theoretic approach described above also was used to
evaluate the relative fit of various candidate models relating crayfish density
estimates to habitat, season, age class, and sympatry. Similar to above, we
constructed a global model (with all predictors) and a subset of 12 candidate
models representing hypotheses about the relative effects of age, season,
mesohabitat type, and the presence of C. englishi on the density of C.
halli (Table 2b). We defined seasons as: winter (January–February), spring
(March–May), summer (June–September), and fall (October–November).
The seasons were categorized using binary variables (0, 1) that were assigned
to spring, fall, and winter, with summer retained as the baseline.
Mesohabitat types were similarly coded for pool and riffl e habitats, with
runs as the baseline. The crayfish density candidate models then were fit
using linear regression, their plausibility evaluated by calculating AICc and
Akaike weights, and goodness-of-fit assessed by examining residuals from
the global model, as described above.
Results
Sampling
We collected 1434 crayfish (n = 150 samples) at Blalock Creek and 367
crayfish (n = 38 samples) at Kiser Creek, both allopatric sites. Cambarus
484 Southeastern Naturalist Vol. 8, No. 3
Table 2. Biological interpretation of predictors used in candidate models of the factors infl uencing (a) C. halli modal carapace length and (b) estimated C. halli
density. Asterisks indicate interaction between predictor variables. Two-way interactions and quadratic terms were only included in candidate models containing
associated main effects.
Predictor variables Biological interpretation (hypothesis)
(a) C. halli modal carapace length
Time Change in carapace length through time represents growth rate.
Time2 The increase in growth rate within age class is non-linear and decreases through time.
Sympatric sites Size of C. halli across age groups is different at sympatric sites due to interaction with C.
englishi
Age 0, age 2 Carapace length of C. halli differs among age classes.
Time*sympatric sites
Time2 *sympatric sites Growth rate of C. halli differs at sympatric sites due to interaction with C. englishi.
Time*age 0, time*age 2
Time2*age 0, time2*age 2 C. halli growth rate varies with age class.
Time*sympatric sites*age 0, time*sympatric sites*age 2
Time2*sympatric sites*age 0, time2*sympatric sites*age 2 The infl uence of sympatric C. englishi on crayfish growth rate differs with age class.
(b) Estimated C. halli density
Sympatric sites The density of C. halli across ages and habitat types differs at sympatric sites due to
interaction with C. englishi.
Fall, winter, spring The density of C. halli differs among seasons.
Age 0, age 2 The density of C. halli differs among age classes.
Pool, riffl e The density of C. halli differs among habitat types, due to differences in habitat use.
Sympatric sites*age 0, sympatric sites *age 2 The density of C. halli infl uence of sympatric C. englishi on varies with age.
Sympatric sites*pool, Sympatric sites *riffl e Habitat use of all age classes differs in sympatric sites due to interactions with C. englishi.
Fall*pool, fall*riffl e, winter*pool, winter*riffl e, Habitat use of C. halli differs among seasons.
spring*pool, spring*riffl e
Age 0*pool, age 0*riffl e, age 2*pool, age 2*riffl e Habitat use of C. halli differs among age classes.
Sympatric sites*fall*age 0, sympatric sites*winter*age 0, Seasonal habitat use varies in the presence of C. englishi, through aggression, competition,
sympatric sites*spring*age 0, sympatric sites*fall*age 2, or predation throughout the stream.
sympatric sites*winter*age 2, sympatric sites*spring*age 2
2009 S. Dennard, J.T. Peterson, and E.S. Hawthorne 485
halli was collected with Cambarus latimanus LeConte, Cambarus striatus
Hay (Hay Crayfish), and Procambarus spiculifer LeConte in Blalock Creek
and with C. latimanus in Kiser Creek. Sampling at Kiser Creek was not
conducted for the entire sampling period and was stopped prematurely on
April 2, 2002 due to site-access issues. At the sympatric sites, Tallapoosa at
Highway 27 and Tallapoosa at Mount View Road, we collected 582 crayfish
(n = 69 samples) and 652 crayfish (n = 81 samples), respectively. At both
Tallapoosa sites, C. halli was collected with C. englishi, C. latimanus, and
P. spiculifer.
Age-class structure
Hatching began in March and April at all sites (see below); however, the
young (henceforth defined as age 0) did not recruit to the quadrat gear until
June, when the smallest individuals were approximately 5 mm in length
(Fig 2). Examination of carapace length-frequency distributions suggested
the presence of 3 age classes: age 0 (newly hatched that year), age 1, and
age 2 (Fig. 2). Graduation from age 0 to age 1 occurred in April/May at
approximately 19 mm, for both sympatric and allopatric sites; graduation
from age 1 to age 2 also occurred in April/May at approximately 30 mm, for
both sympatric and allopatric sites. Carapace length-frequency distributions
suggested that age-2 individuals were either dying in October/November or
indistinguishable from age 1.
Reproductive biology
Both male and female C. halli were in a nonreproductive state from June
to September (Fig. 3). By October, approximately 16.8% of C. halli were
Form I (reproductive) males. Female C. halli were becoming reproductively
active by January, as indicated by the presence of glair, and they began to
extrude eggs in late March and early April. Water temperatures during this
time varied between sites, ranging from 10 to 14 °C at Blalock Creek, where
C. halli was allopatric, and from 10 to 17 °C at the Tallapoosa sites, where
C. halli and C. englishi co-occured. The maximum number of ovigerous
females occurred in May, when water temperatures reached 19 °C. In June
2002, all males were in a nonreproductive (Form II) state. However, some
reproductive (Form I) males were collected in June 2003.
The smallest sexually mature female (i.e., females with glair or bearing
eggs) for C. halli was 13 mm, collected March 4, 2003 at the Tallapoosa
at Mount View Road site. The smallest sexually mature male (i.e., Form I)
for C. halli was 12 mm, collected October 17, 2002 at Blalock Creek. The
proportion of age-1 and age-2 reproductive individuals were greater at
allopatric sites (0.63; 90% CI: 0.38–0.86) compared to sympatric sites
(0.33, 90% CI: 0.14–0.51). At all sites, hatching began in March and April
(Fig. 3).
Eggs were only measured and counted for C. halli at Blalock Creek. The
total number of pleopod eggs carried by females ranged from 65 to 217. Linear
regression models indicated that the number of pleopod eggs was positively
related to female carapace length (Fig. 4a); carapace length accounted for
486 Southeastern Naturalist Vol. 8, No. 3
Figure 2. An example of C. halli carapace length-frequency histogram from April–
June 2002.
2009 S. Dennard, J.T. Peterson, and E.S. Hawthorne 487
83.7% of the variation in the number of pleopod eggs. Mean egg diameter
ranged from 2.280 to 2.476 mm. The egg diameter, however, was only weakly
positively related to female carapace length (Fig. 4b), which accounted for
only 0.8% of the variation in egg diameter. The carapace-length parameter estimate
for relating egg diameter was imprecise, as the confidence interval was
wide and included zero (Table 3).
Seasonal growth
The examination of residuals from the global linear models relating crayfish age and the presence of C. englishi to C. halli modal carapace length
indicated no departures from normality and no detectable spatial autocorrelation.
The best-approximating model of C. halli carapace length was the global
model that contained time, age, the sympatric indicator variable, and the interactions:
time by sympatric site, time by age, and time by sympatric site by age.
This model was 2 times more likely than the next-best approximating model
Figure 3. Reproductive state of mature male and female C. halli during February
2002 to June 2003. Percentage occurrence is the percentage of crayfish in a reproductive
state out of all mature individuals found in that sampling period.
Table 3. Parameter estimates, standard error (SE), and upper and lower 90% confidence limits
for models relating carapace length (mm) to total number of pleopod eggs and pleopod egg size
(mm) for C. halli.
Parameter Estimate SE Lower Upper
Total number of pleopod eggs
Intercept -137.859 24.617 -178.231 -97.488
Carapace length 8.667 0.797 7.360 9.975
Pleopod egg size
Intercept 2.361 0.243 1.962 2.760
Carapace length 0.002 0.008 -0.011 0.015
488 Southeastern Naturalist Vol. 8, No. 3
that did not contain the time by sympatric site by age interaction. There was no
support for a non-linear relationship between time and carapace length within
age class because models containing the quadratic term for time were among
the poorest approximating.
The model-averaged parameter estimates indicated that carapace length
was positively related to time and that age-1 crayfish at allopatric sites grew
2.39 mm per month, whereas age-0 and age-2 growth rates were lower than the
age-1 growth rate (Table 4). The sympatric site parameter estimate indicated
that C. halli were on average 2.07 mm smaller at sympatric sites compared to
allopatric sites. However, the growth rates of age-0 and age-2 crayfish at sympatric
sites were on average, 0.58 and 0.97 mm greater than allopatric sites,
respectively. The parameter estimates for the time by sympatric site interaction
Figure 4. Relation of female C. halli size to (a) total number of pleopod eggs, and (b)
pleopod egg size in May and June of 2003 at Blalock Creek.
2009 S. Dennard, J.T. Peterson, and E.S. Hawthorne 489
suggested that the growth rate for age-1 C. halli at sympatric sites was slightly
greater than at allopatric sites; however, this estimate was unreliable, because
the confidence interval was wide and included zero (Table 4).
Habitat use and density estimates
An examination of the normal probability plot of residuals from the global
linear models relating age, habitat, season, and the presence of C. englishi to
C. halli density estimates indicated that the residuals departed from expected
(i.e., the plots were curvilinear rather than linear). To normalize these data, we
natural log transformed the data and re-fit the candidate models. The ANOVA
of residuals from the global C. halli density model fit to the transformed data
indicated no detectable dependence among samples within sites (P = 0.27). The
most plausible model of C. halli density contained season, age, mesohabitat
type, the sympatric indicator variable, and the interactions age by sympatric
site and mesohabitat type by sympatric site, and was 1.4 times more likely to
predict density than the next-best approximating model, the global model.
Density estimates of C. halli were greatest during the summer months and
lowest during the spring (Table 5). Density estimates were also lower at sites
where they were sympatric with C. englishi, with the greatest differences between
density estimates at allopatric and sympatric sites occurring for age-1
C. halli. The parameter estimates for mesohabitat type and mesohabitat type
by sympatric site interaction indicated that density estimates of all age classes
of C. halli were generally greatest in runs and pools at allopatric sites, but were
lower in pools and runs and greatest in riffl es at sympatric sites (Table 5). The
remaining parameter estimates were unreliable for interpretation, as the confi-
dence intervals were wide and included zero.
Discussion
Reproductive characteristics of C. halli in the Tallapoosa River Basin
were similar to other stream-dwelling crayfish, such as Orconectes luteus
Creaser (Golden Crayfish) in the Missouri Ozarks (Muck et al. 2002).
Table 4. Model-averaged parameter estimates, standard errors (SE), and upper and lower 90%
confidence limits for composite linear regression model of C. halli carapace length. Time
(month = zero) begins at June and ends in May; age 1 was used as the baseline age class in the
regression.
Upper Lower
Parameter Estimate SE (90% CI) (90% CI)
Intercept 15.705 1.114 17.531 13.879
Time 2.391 0.270 2.834 1.947
Sympatric sites -2.067 1.210 -0.083 -4.051
Age 0 -7.957 1.279 -5.859 -10.054
Age 2 14.252 1.809 17.219 11.286
Time*sympatric sites 0.223 0.322 0.751 -0.304
Time*age 0 -0.663 0.365 -0.063 -1.262
Time*age 2 -2.447 0.777 -1.172 -3.721
Time*sympatric sites*age 0 0.580 0.265 1.014 0.146
Time*sympatric sites*age 2 0.972 0.557 1.885 0.059
490 Southeastern Naturalist Vol. 8, No. 3
Reproductive timing for O. luteus was comparable to C. halli, with O. luteus
females extruding eggs in late March and early April, eggs hatching in
April, and independence of young in the summer (Muck et al. 2002). Growth
rates of the three O. luteus age classes examined by Muck et al. (2002) were
comparable to C. halli; for example, age-0 C. halli reached 19 mm after one
year of growth, while young-of-year O. luteus grew to approximately 20 mm
before graduating to age 1 (Muck et al. 2002).
Fecundity for C. halli differed slightly from O. luteus. Muck et al. (2002)
found that both number of pleopod eggs and mean egg diameter increased
with increasing carapace length of the female. We observed a relationship
between number of eggs and carapace length, but not between egg diameter
and carapace length. Fecundity for C. halli somewhat refl ected the life-history
strategy of an r-selected species: maturity at a young age (3% of age-0
crayfish were reproductively active at the end of their first summer), rapid
growth rates (19 mm in the first year, and 11 mm more the second year),
and large numbers of offspring (more than 200 for a female with CL of 35
mm). However, it is not as fecund as invasive species, such as Procambarus
clarkii Girard (Red Swamp Crayfish), whose average-sized females produce
Table 5. Model-averaged parameter estimates, standard errors (SE), and upper and lower 90%
confidence limits (CI) for composite linear regression model of estimated C. halli density.
Parameter Estimate SE Upper CI Lower CI
Intercept 1.366 0.098 1.526 1.206
Sympatric sites -0.901 0.100 -0.736 -1.065
Fall -0.202 0.106 -0.029 -0.375
Winter -0.241 0.132 -0.024 -0.457
Spring -0.176 0.090 -0.029 -0.324
Age 0 -0.378 0.081 -0.245 -0.511
Age 2 -0.349 0.081 -0.216 -0.481
Pool 0.099 0.109 0.277 -0.080
Riffl e -0.841 0.156 -0.585 -1.096
Sympatric sites*age 0 0.401 0.122 0.601 0.202
Sympatric sites*age 2 0.340 0.119 0.536 0.145
Sympatric sites*pool -0.080 0.048 -0.002 -0.159
Sympatric sites*riffl e 0.535 0.112 0.719 0.351
Fall*pool 0.056 0.164 0.326 -0.213
Fall*riffl e 0.164 0.164 0.434 -0.105
Winter*pool 0.158 0.179 0.452 -0.137
Winter*riffl e 0.256 0.179 0.550 -0.039
Spring*pool 0.080 0.127 0.290 -0.129
Spring*riffl e 0.202 0.127 0.411 -0.007
Age 0*pool -0.218 0.127 -0.010 -0.426
Age 0*riffl e 0.124 0.127 0.332 -0.084
Age 2*pool -0.079 0.127 0.129 -0.287
Age 2*riffl e 0.172 0.127 0.380 -0.037
Sympatric sites*fall*age 0 0.106 0.180 0.400 -0.189
Sympatric sites*winter*age 0 0.503 0.292 0.983 0.024
Sympatric sites*spring*age 0 -0.285 0.143 -0.051 -0.519
Sympatric sites*fall*age 2 -0.197 0.179 0.097 -0.491
Sympatric sites*winter*age 2 -0.182 0.292 0.297 -0.661
Sympatric sites*spring*age 2 0.225 0.142 0.458 -0.009
2009 S. Dennard, J.T. Peterson, and E.S. Hawthorne 491
400 pleopodal eggs (Gherardi 2006) compared to C. halli that produced on
average 127 eggs.
The density estimates of C. halli were significantly lower at sites where
they were sympatric with C. englishi compared to those without C. englishi.
Previous studies have suggested that the density of stream-dwelling macroinvertebrates
is related to local stream habitat characteristics (Allan 1995),
the types and amounts of nutrients (Allan 1995), and species interactions
(Bovbjerg 1970). Of these factors, we believe that habitat characteristics
were probably not responsible for the observed differences in C. halli density
estimates for several reasons. First, the streams were physically similar
with riffl e-pool-run sequences and large amounts of cobble and gravel substrate.
Second, C. halli density estimates were lower across all mesohabitat
types, except riffl e habitats at the sympatric sites. Third, all of the study sites
were located relatively close together, had similar climatic conditions (e.g.,
temperature, precipitation), and fl owed through similar geologies, which
suggests similar groundwater nutrient inputs. We believe that the differences
in C. halli density could be due to differences in terrestrial nutrient inputs,
species interactions, or a combination of these two factors. Identifying the
exact mechanism, however, requires an understanding of how these factors
infl uence the production of crayfish and other macroinvertebrates.
Low-order (small) streams rely on allocthonous nutrient inputs for consumer
communities; however, as the river broadens, energy inputs change,
allowing for autochthonous production (Vannote et al. 1980). Because
crayfish are processors of leaf litter and detritus (Griffith et al. 1994, Huryn
and Wallace 1987), C. halli require some type of coarse particulate organic
matter (CPOM) input—usually provided by streamside vegetation. Blalock
Creek and Kiser Creek are lower order, smaller systems than the Tallapoosa
River sites. The fact that low-order streams are typically associated with
a higher CPOM resource base could, in part, be responsible for the higher
density estimates at the allopatric Blalock and Kiser creeks.
In addition to nutrient differences between sites, we believe interactions
with C. englishi could be responsible for the lower density estimates at the
sympatric sites. Crayfishes that use the same habitats are often limited to one
or two species, and closely related species usually exhibit disjunct distributions
(Rabeni 1985). Both C. halli and C. englishi reportedly prefer stream
habitats, suggesting that these two species often interact and may compete
for resources (Bouchard 1978, Freeman et al. 2003, Hobbs and Hall 1972).
We found evidence that C. halli did alter their habitat use in the presence
of C. englishi. Both species also use leaf litter and detritus as their food
base (Hobbs and Hall 1972), which suggest that there is strong potential for
competition of resources at the sympatric sites. Other crayfish species were
also collected at the allopatric and sympatric study sites, but we believe that
it was unlikely that they competed with C. halli. The differences in habitat
preferences between C. halli and C. latimanus, C. striatus, and P. spiculifer
account for the coexistence of these species at the study sites. Cambarus
latimanus prefers small burrows in the stream bottom or between rocks
alongside the stream (Yarbrough 1973), C. striatus are primary burrowers
492 Southeastern Naturalist Vol. 8, No. 3
that tend to remain in their burrows continuously (Hobbs 1981), and P.
spiculifer are reported to be habitat generalists (Ratcliffe and Devries 2004),
though we primarily found them in slow-fl owing, edge areas with vegetation.
Interaction between C. halli, C. latimanus, C. striatus, and P. spiculifer
was unlikely due to resource partitioning.
The primary infl uence of C. englishi on C. halli appears to occur during
early life-history stages. As expected, allopatric C. halli populations
exhibited demographic patterns consistent with the life-history strategy of
an r-selected species: many young-of-year were present shortly after hatching,
followed by sharp decreases in estimated density (i.e., low survival)
till age-1. Sympatric C. halli, however, exhibited a different pattern than
expected with low estimated densities of hatchlings, followed by a slight
decrease in estimated density till age-1. We hypothesize that the different
patterns can be explained by two potential mechanisms: (1) reproduction
of C. halli was lower at sympatric sites or (2) age-0 C. halli had a higher
mortality rate before gear recruitment at the sympatric sites. The observed
lower proportion of reproductively active adults at the sympatric sites is consistent
with the first mechanism. Support for the second mechanism could
not be evaluated with the existing data. However, the smaller decreases in
estimated density from age-0 to age-2 at sympatric sites compared to allopatric
sites suggests that sympatric age 0 individuals that were large enough
to be collected generally had higher survival than their counterparts at the
allopatric sites. Further, the greater growth rates of age-0 and age-1 C. halli
observed at sympatric sites suggests that competition for resources (food)
may have been lower than at allopatric sites. Both of these lines of evidence
suggest that the mortality of age-0 crayfish that occurred before they were
recruited to the sampling gear may not have been higher at sympatric sites.
Clearly further study is needed to determine the mechanisms, and we recommend
that these studies focus on reproductive dynamics and very early life
stages when age-0 crayfish are smaller than 5 mm carapace length.
The results of our study suggest that C. halli life history was similar to
other stream-dwelling crayfish, with larger females focusing energy into
the production of more eggs, rather than larger, higher-quality eggs. Hobbs
(1981) reported that allopatric C. halli use all habitat types, but are primarily
found in non-riffl e habitats when they occur with C. englishi. However, we
found that C. halli preferred pools and runs over riffl e habitats at allopatric
sites for all age classes. Differences in estimated density among sites were
likely caused by differences in terrestrial nutrient inputs and by species interactions,
though identifying the exact mechanism requires further research.
For instance, the life history of C. englishi must be known, specifically its
reproductive biology, growth, and habitat use, to determine the specific interactions
occurring between C. halli and C. englishi.
Because the counties containing the Tallapoosa River Basin will likely
become part of Metro-Atlanta in the near future, understanding the ecological
roles of both C. halli and C. englishi will be crucial for conservation of
these species. Both species are listed as being “of special concern” by the
Georgia Natural Heritage Program, and C. englishi is also on Georgia’s state
2009 S. Dennard, J.T. Peterson, and E.S. Hawthorne 493
list of protected species. Effective conservation strategies require knowledge
of reproductive biology, reproductive timing, growth, and seasonal habitat
use. This study adds to our understanding of C. halli life history and provides
a starting point for the formation of conservation strategies.
Acknowledgments
We thank crew members N. Banish, S. Craven, A. Meadows, D. McPherson, J.
McCargo, and J. Ruiz for data collection. The manuscript was improved with suggestions
from J. Shelton, N. Nibbelink, C. Rabeni, B. Albanese, and anonymous
reviewers. The US Fish and Wildlife Service provided funding for the field portion
of the project. The Georgia Cooperative Fish and Wildlife Research Unit is jointly
sponsored by the US Geological Survey, the Georgia Department of Natural Resources,
the US Fish and Wildlife Service, the University of Georgia, and the Wildlife
Management Institute.
Literature Cited
Allan, J.D. 1995. Stream Ecology: Structure and Function of Running Waters. Kluwer
Academic Publishers, Boston, MA. 400 pp.
Baker, A.M., P.M. Stewart, and T.P Simon. 2008. Life-history study of Procambarus
suttkusi in southeastern Alabama. Journal of Crustacean Biology 28:451–460.
Bouchard, R.W. 1978. Taxonomy, ecology, and phylogeny of the subgenus Depressicambarus,
with the description of a new species from Florida and redescriptions
of Cambarus graysoni, Cambarus latimanus, and Cambarus striatus (Decapoda:
Cambaridae). Bulletin of the Alabama Museum of Natural History 3:2760.
Burnham, K.P., and D.R. Anderson. 2002. Model Selection and Inference: A Practical
Information Theoretic Approach. Second Edition. Springer-Verlag, New
York, NY. 353 pp.
Bovbjerg, R.V. 1970. Ecological isolation and competitive exclusion in two crayfish
(Orconectes virilis and Orconectes immunis). Ecology 51:225–236.
Freeman, M.C., J.T. Peterson, E.R. Irwin, and B.J. Freeman. 2003. Distribution and
status of at-risk aquatic taxa in the Upper Tallapoosa River System, Georgia and
Alabama. Final Report to US Fish and Wildlife Service, Ecologic Services, Athens,
GA. 65 pp.
Georgia Department of Natural Resources (GA DNR). 2002. Section 2: River basin
characteristics. Pp. 2-1–2-28, In Tallapoosa River Basin Management Plan 1998.
Available online at http://www.gaepd.org/Files_PDF/plans/tallapoosa/chapt-2.
pdf. Acessed 19 March 2007.
Gherardi, F. 2006. Crayfish invading Europe: The case study of Procambarus clarkii.
Marine and Freshwater Behaviour and Physiology 39:175–191.
Gilpin, M.E., and M.E. Soule. 1986. Minimum viable populations: Processes of
species extinction. Pp. 19–34, In M.E. Soule (Ed.). Conservation Biology: The
Science of Scarcity and Diversity. Sinauer Associates, Inc., Sunderland, MA.
584 pp.
Griffith, M.B., S.A. Perry, and W.B. Perry. 1994. Secondary production of
macroinvertebrate shredders in headwater streams with different basefl ow alkalinity.
Journal of the North American Benthological Society 13:345–356.
Hamr, P., and M. Berrill. 1985. The life histories of north-temperate populations
of the crayfish Cambarus robustus and Cambarus bartonii. Canadian Journal
of Zoology 62:2313–2322.
494 Southeastern Naturalist Vol. 8, No. 3
Hobbs, H.H., Jr. 1981. The Crayfishes of Georgia. Smithsonian Institution Press,
Washington, DC. 549 pp.
Hobbs, H.H., Jr., and E.T. Hall, Jr. 1972. A new crayfish from the Tallapoosa River
in Georgia (Decapoda: Astacidae). Proceedings of the Biological Society of
Washington 85:151–162.
Hurvich, C.M., and C.L. Tsai. 1989. Regression and time-series model selection in
small samples. Biometrika 76:297–307.
Huryn, A.D., and J.B. Wallace. 1987. Production and litter processing by crayfish in an
Appalachian mountain stream. Freshwater Biology 18:277–286.
Hyne, R.V., and W.A. Maher. 2003. Invertebrate biomarkers: Links to toxicosis that
predict population decline. Ecotoxicology and Environmental Safety 54:366–374.
McKenney, R. 2001. Channel changes and habitat diversity in a warm-water, gravelbed
stream. Pp. 57–71, In J. Dorava, D. Montgomery, B. Palcsak, and F. Fitzpatrick
(Eds.). Geomorphic Processes and Riverine Habitat. Water Science and Application
Series, Volume 4. American Geophysical Union, Washington, DC. 253 pp.
Momot, W.T. 1967. Population dynamics and productivity of the crayfish Orconectes
virilis in a marl lake. American Midland Naturalist 78:55–81.
Muck, J.A., C.F. Rabeni, and R.J. Distefano. 2002. Reproductive biology of the crayfish Orconectes luteus (Creaser) in a Missouri stream. American Midland Naturalist
147:338–351.
Neter, J., M.H. Kutner, C.J. Nachtsheim, and W. Wasserman. 1996. Applied Linear
Statistical Models. 4th Edition. Irwin, Chicago, IL. 1408 pp.
Probst, W.E., C.F. Rabeni, W.G. Covington, and R.E. Marteney. 1984. Resource use by
stream-dwelling Rock Bass and Smallmouth Bass. Transactions of the American
Fisheries Society 113:283–294.
Rabeni, C.F. 1985. Resource partitioning by stream dwelling crayfish: The infl uence of
body size. American Midland Naturalist 113:20–29.
Rabeni, C.E. 1992. Trophic linkage between stream centrarchids and their crayfish
prey. Canadian Journal of Fisheries and Aquatic Sciences 49:1714–1721.
Ratcliffe, J.A., and D.R. DeVries. 2004. The crayfishes (Crustacea: Decapoda) of the
Tallapoosa River drainage, Alabama. Southeastern Naturalist 3:417–430.
Roell, M.J., and D.J. Orth. 1993. Trophic basis of production of stream-dwelling
Smallmouth Bass, Rock Bass, and Flathead Catfish in relation to invertebrate bait
harvest. Transactions of the American Fisheries Society 122:46–62.
Stephens, G.C. 1952a. The control of cement gland development in the crayfish Cambarus.
Biological Bulletin 103:242–258.
Stephens, G. J. 1952b. Mechanisms regulating the reproductive cycle in the crayfish,
Cambarus. I. The female cycle. Physiological Zoology 75:70–84.
Taylor, C.A., M.L. Warren, Jr., J.F. Fitzpatrick, Jr., H.H. Hobbs III, R.F. Jezerinac,
W.L. Pfl ieger, and H.W. Robison. 1996. Conservation status of crayfishes of the
United States and Canada. Fisheries 21:25–38.
Taylor, C.A., G.A. Schuster, J.E. Cooper, R.J. DiStefano, A.G. Eversole, P. Hamr, H.H.
Hobbs III, H.W. Robison, C.E. Skelton, and R.F. Thoma. 2007. A reassessment
of the conservation status of crayfishes of the United States and Canada after 10+
years of increased awareness. Fisheries 32:372–389
Vannote, R.L., G.W. Minshall, K.W. Cummins, J.R. Seddell, and C.E. Cushing. 1980.
The river continuum concept. Canadian Journal of Fisheries and Aquatic Sciences
37:130–137.
Yarbrough, J.E. 1973. An ecological and taxonomic survey of the crayfish of the
Tallapoosa and Chattahoochee river drainages in Lee County. Alabama. M.Sc.
Thesis. Auburn University, Auburn, AL. 96 pp.