Shelter-use and Interactions Between Banded Sculpin (Cottus carolinae) and Bigclaw Crayfish (Orconectes placidus) in Stream-pool habitats
Crystal Bishop, Brianna Begley, Christina Nicholas, Jessica Rader,
Elizabeth Reed, Kyle Sykes, Todd Williams, Elizabeth Young,
and Dennis Mullen
Southeastern Naturalist, Volume 7, Number 1 (2008): 81–90
Full-text pdf (Accessible only to subscribers.To subscribe click here.)
2008 SOUTHEASTERN NATURALIST 7(1):81–90
Shelter-use and Interactions Between Banded Sculpin
(Cottus carolinae) and Bigclaw Crayfish (Orconectes
placidus) in Stream-pool habitats
Crystal Bishop1, Brianna Begley1, Christina Nicholas1, Jessica Rader1,
Elizabeth Reed1, Kyle Sykes1, Todd Williams1, Elizabeth Young1,
and Dennis Mullen1,*
Abstract - The purpose of this study was to test for interference competition for shelter
between adult Cottus carolinae (Banded Sculpin) and adult Orconectes placidus
(bigclaw crayfish) in stream-pool habitats. Both species co-occur naturally in high
densities in Brawley’s Fork (Cumberland River Basin, TN), creating a potential for
strong interactions over shared resources. In-stream enclosures containing one rock
shelter were used to test for depth preference by adult crayfish (preference for pool
habitats has already been demonstrated for the sculpin), test for shelter preference by
both species and, determine if presence of one species affects shelter use of the other
species. Adult bigclaw crayfish displayed a strong preference for deep water over
shallow water in the enclosures, and both species used the shelter at a significantly
higher rate than expected from the null hypothesis of random habitat use. Neither
species, however, affected the shelter use of the other in sympatric trials, (in fact,
both species shared the shelter in about one third of the trials), indicating that these
species may not compete for shelter in this system. Although both species use rock
shelters in the pool habitat, the lack of predators in the pool habitats of this stream
may reduce the importance of shelter to the sculpin and crayfish, thereby reducing
the likelihood of strong interactions over shelter.
Introduction
Size-specific habitat use is a common phenomenon in stream fishes (Mahon
and Portt 1985; Mullen and Burton 1995, 1998; Power 1984) and crayfish
(Creed 1994, Englund and Krupa 2000). Most commonly, larger individuals
use deeper areas of streams to reduce predation risks, while smaller individuals
prefer, or are confined to, shallower areas. Because of these similar depth
preferences there is a potential for strong inter-specific interactions between
adult fish and adult crayfish for limited resources in stream-pool habitats.
Despite this potential, there are only a few studies of the interactions between
fishes and crayfish in stream ecosystems. In a laboratory study, Miller
et al. (1992) demonstrated that Orconectes virilis Hagen (northern crayfish)
suffered reduced feeding rates in the presence of Cottus cognatus Richardson
(Slimy Sculpin), but evicted sculpin from shelters when only one was provided.
Orconectes rusticus Girard (rusty crayfish) have been shown to evict
Etheostoma nigrum Rafinesque (Johnny Darters) from shelters when a predator
was present, increasing their risk of predation (Rahel and Stein 1988).
1Department of Biology, Middle Tennessee State University, Murfreesboro, TN
37132. *Corresponding author - dmullen@mtsu.edu.
82 Southeastern Naturalist Vol.7, No. 1
Interference competition for shelter space has also been demonstrated in
studies of the impacts of non-native Pacifastacus leniusculus Dana (signal
crayfish) on native benthic riffl e fishes (Guan and Wiles 1997), Salmo salar
Linnaeus (Atlantic Salmon) in a British lowland river (Griffiths et al. 2004),
and on native Cottus beldingi Eigenmann and Eigenmann (Paiute Sculpin) in
a California stream (Light 2005). In all three studies, the invasive crayfish displaced
the native fishes from shelters, potentially increasing the predation risk
to those fishes.
Recently, adult Cottus carolinae Gill (Banded Sculpin) were found to
prefer pool areas of streams, most likely to avoid predation by terrestrial
mammals and birds (Koczaja et al. 2005), while young-of-year (YOY) and
juvenile sculpin reside in the shallow waters. While examining size-specific
habitat use by Banded Sculpin in Brawley’s Fork, TN, Koczaja et al. (2005)
noticed, but did not quantify, a similar trend for Orconectes placidus Hagen
(bigclaw crayfish). Smaller individuals were encountered more frequently in
shallower areas, and adults occupied the same pool habitats as adult sculpin.
Adults of both species appeared to be associated with rock shelters in this
habitat. Crayfish densities appeared to be high in the pool habitats, creating
a potential for strong interactions between adult Banded Sculpin and adult
bigclaw crayfish. Banded Sculpin forage nocturnally and are known to seek
refuge under rocks during the daytime (Greenberg and Holtzman 1987), and
many species of crayfish forage nocturnally and seek shelter during the day
(Gherardi and Barbaresi 2000, Griffiths et al. 2004, Guan and Wiles 1997,
Hazlett et al. 1974). However, specific information is lacking on activity
patterns and shelter use of the bigclaw crayfish.
The purpose of this study was to examine the interaction over shelter
space between Banded Sculpin and bigclaw crayfish in the pool habitat. We
used in-stream enclosures to: 1) determine if adult bigclaw crayfish exhibit
the same preference for pool habitats over shallow habitats exhibited by
adult sculpin (Koczaja et al. 2005), by testing the null hypothesis that adult
crayfish’s use of shallow and deep water areas is proportional to the availability
of these habitats in the environment; 2) determine if adult sculpin and
adult crayfish prefer to use the rock shelters in the pool area, by testing the
null hypothesis that both species use shelter in the same proportion that it
occurs in the environment; and 3) determine if crayfish and sculpin compete
for shelter in this habitat by testing the hypothesis that shelter use of crayfish
and sculpin is independent of the presence of the other species.
Methods
Field site description
The study was conducted in a pool in Brawley’s Fork (Cumberland River
Basin, Cannon County, TN) during September and October 2005. Brawley’s
Fork is a spring-fed, first-order stream with predominantly pebble/cobble
substrate and limestone outcrops and root masses usually associated with the
pool margins. Summer discharge is typically around 0.5 m3 per second.
2008 C. Bishop et al. 83
Crayfish depth preference and shelter use
Observations of crayfish depth and shelter use were conducted in enclosures
modeled after those used by Freeman and Stouder (1989) to study
size-specific depth segregation by Cottus bairdi Girard (Mottled Sculpin).
Four such enclosures were placed within a long pool in Brawley’s Fork. The
enclosures were 1-m x 1-m square and 0.5 m in depth, and consisted of a lumber
frame with untreated plywood sides and 6.35-mm2 hardware cloth covering
the front, back, and bottom to allow water movement through the enclosures.
A 0.67-m length of plywood extended 2/3 of the length down the middle of the
enclosure from the front to the back (see Koczaja et al. 2005 for an illustration
of the enclosures). This divided each enclosure into two equal sides with
an open area in the back so that crayfish could easily move between the two
sides. The enclosures were covered with a loose-fitting lid consisting of a 1-m
x 1-m lumber frame covered with 6.35-mm2 hardware cloth that could easily
be removed when sampling was conducted. The lid was necessary to reduce
crayfish escapes from the enclosures (although it did not completely eliminate
escapes). One side of each enclosure was filled with gravel and pebbles from
the stream bed to create a mean water depth of 10 cm, with depths ranging from
6 cm to 17 cm depending on stream water level (which fl uctuated some over the
course of the study). The other side was filled in the same way to create a mean
water depth of 29 cm, with depths ranging from 24 cm to 34 cm (depending on
water level and pool depth at that location). Throughout the sampling period,
the minimum difference in depth between the two sides of any enclosure was
13.8 cm, and the maximum difference was 23.3 cm.
Beginning from downstream, the enclosures were numbered 1–4. The sides
were assigned to shallow or deep in an alternating pattern from enclosure 1 to
enclosure 4. Each side of each enclosure contained one shelter (positioned in
the middle of the channel about 30 cm behind the front screen), which consisted
of a fl at stone, chosen from the stream, that was propped up on one side with
small rocks to create a crevice. Shelter area was estimated as follows: each shelter
stone used in the experiment was placed on aluminum foil, and the perimeter
was traced with a permanent marker. The area from the stone was cut out and
weighed and compared to the weight of a 100-cm2 piece of aluminum foil to
determine area of each shelter. Although efforts were made to choose natural
stones of similar size, the shelters ranged from 168 cm2 to 286 cm2, with a mean
area of 209 cm2. The shelters comprised from 3.4 to 5.7% of the area within the
enclosures, and the mean value of 4.2% was used as an estimate of available
shelter space in each enclosure. Since the rocks were propped on only one side,
this represents an overestimate of the actual shelter area.
Crayfish used in this study were collected from the stream by hand capture
and baited minnow traps. The total length of each crayfish was measured (mm)
from the tip of the rostrum to the tip of the telson before use. One crayfish
was placed in the center of the downstream end of each enclosure between
the hours of 8:00 am and 10:00 am and allowed to acclimate for 24 hours.
The following morning, starting from downstream and working upstream, the
84 Southeastern Naturalist Vol.7, No. 1
enclosure covers were removed, and a wooden divider was placed in the space
between the 2 sides of the enclosure. A bottomless plastic 20-L bucket was
placed around each shelter stone simultaneously to prevent crayfish that were
under the shelter from moving to other parts of the enclosure. The position of
the crayfish (under or not under the shelter) and the depth preference of the
crayfish (shallow or deep) were recorded for each enclosure. The crayfish were
then released, and another replicate was initiated. To prevent using the same
individual more than once, each crayfish was marked after use by clipping the
corner of one uropod before releasing it back into the stream. Crayfish lengths
ranged from 51 mm to 75 mm. Sampling began on September 15 and ended on
September 22. Trials were conducted on 20 crayfish; however, 5 escaped the
enclosures (they were not found in the enclosures when the experiment was
dismantled on the last day), resulting in data for a total of 15 crayfish.
Sculpin shelter use
The in-stream enclosures were modified in order to observe shelter use of
adult sculpin. The plywood partition in each of the four enclosures was extended
back to completely separate the two sides creating eight 0.5-m x 1-m
enclosures. Each enclosure was situated in the same pool as above (Koczaja et
al. [2005] observed that the adult sculpin prefer the pool habitat) at an average
depth of 25 cm and contained one of the rock shelters used in the above study.
Sculpin were collected from the stream using a backpack electrofisher.
The total length of each sculpin (to the nearest mm) was measured before use.
One sculpin was placed in each of the eight enclosures between the hours of
8:00 am and 10:00 am and allowed to acclimate for 24 hr. Shelter use was determined
the next morning using the same procedures as in the crayfish study
described above. To prevent using the same sculpin more than once, sculpin
were released about 200 m downstream at the far side of a culvert that acted
as a barrier to prevent movement upstream to the study site. Sculpin lengths
ranged from 72 mm to 114 mm. Sampling was conducted from September
22 to September 26 on 32 sculpin (two of which escaped through tears in the
screens), and data on shelter use of 30 sculpin was obtained.
Competition for shelter between sculpin and crayfish
To test for competition for shelter access between sculpin and crayfish,
one sculpin and one crayfish were simultaneously added to each of the eight
mesocosms. The sculpin and crayfish were allowed to acclimate for 24 h.
The next day the position of the crayfish and sculpin relative to the shelter
were recorded (under or not under the shelter). Sculpin lengths ranged from
62 mm to 113 mm with a mean of 87.8 mm, and the crayfish lengths ranged
from 53 mm to 81 mm with a mean of 66.0 mm. Individuals of each species
were assigned haphazardly into each enclosure, and there was no attempt to
control for size differences between the sculpin and crayfish in each enclosure.
The size differences between the crayfish and sculpin (sculpin length
- crayfish length) ranged from -10 mm to 53 mm, with the sculpin being
larger than the crayfish an all but 2 of the trials. Sampling took place from
2008 C. Bishop et al. 85
September 26 to October 6, and 28 replicates were obtained (either the sculpin
[one] or the crayfish [three] escaped in four of the 32 trials).
Data analysis
A chi-square goodness-of-fit test was used to test the null hypothesis that
crayfish used the shallow and deep sides of the enclosures equally. Fisher
exact tests were used to test the null hypotheses that crayfish and sculpin
use shelter in the same proportion that it occurs in the environment (in other
words, they lack a shelter preference). Expected values for shelter use were
generated by multiplying the mean proportion of the available habitat that was
shelter (0.042) by the total number of observations. Because the generated
expected values were less than five, a chi-square goodness-of-fit test could not
be used. Two-by-two contingency table analysis (using the chi-square statistic)
was used to test the null hypothesis that shelter use of crayfish and sculpin
is independent of the presence of the other species. Goodness-of-fit tests and
two-by-two contingency table analyses were conducted with the Yates’ correction
for continuity (Zar 1984).
Because the range of size differences between the two species was so large
in the sympatric trials and the strength and outcome of interactions may be
size-dependent, a three-dimensional contingency table was used to examine
the effect of the magnitude of the size difference on the outcome of these trials.
The data were arbitrarily divided into two groups based on the magnitude
of the size difference (small difference [sculpin length - crayfish length < 20
mm, n = 15] and large difference [>20 mm, n = 13]) and log linear analysis (using
the G statistic) of the three-dimensional contingency table was conducted.
The three factors were small size difference/large size difference, crayfish
used shelter/ crayfish did not use shelter, and sculpin used shelter/sculpin did
not use shelter.
Results
Crayfish depth preference
Crayfish used the deep side of the enclosures more frequently than the
shallow side (Yates’ corrected χ2 = 6.66; P = 0.01; Fig. 1) despite the fact
that in one enclosure the depth difference between the sides was only about
14 cm (because it was in a slightly shallower area of the pool). Four of the 5
crayfish from that enclosure were found in the deep side.
Shelter use
In the absence of the other species, both sculpin and crayfish were found
more frequently under the shelter (which composed about 4% of the available
habitat) (Figs. 2A and 2B). This trend was significant for sculpin (Fisher
exact test: P = 0.0001) and crayfish (Fisher exact test: P = 0.0012). Ten of the
15 crayfish were found under the shelter, and the remaining five were found
burrowed next to the enclosure walls.
Sixteen of the 30 sculpin were found under the shelter, and the remaining
14 were found throughout the enclosures (the actual location of non-sheltering
86 Southeastern Naturalist Vol.7, No. 1
sculpin was not determined because they were well camoufl aged and we could
not reliably determine their position if they were not under the shelter).
Shelter competition
Crayfish and sculpin shelter use was independent of presence of the other
species (Yates’ corrected χ2 = 0.26 and 0.34; P = 0.61 and 0.56 respectively).
With crayfish present, sculpin were found under the shelter 18 times out of
28 replicates (Fig. 2A). With sculpin present, crayfish were found under the
shelter 15 times out of 28 replicates (Fig. 2B). Crayfish and sculpin shared
the shelter in 10 of the 28 replicates. Log linear analysis of the affect of the
relative size difference between the crayfish and sculpin on the outcome of
the interactions indicates that the size difference was not an important factor
in determining the outcome (overall G2 = 3.04, P = 0.22; size difference/
shelter use interaction G2 = 1.34, P = 0.25).
Discussion
Even though both adult Banded Sculpin and adult bigclaw crayfish prefer
deeper areas of the stream and use rock shelters during they day, they do not
appear to compete for those shelters, even when only one is provided. In this
study, neither crayfish nor sculpin altered their shelter use in the presence
of the other species, and both species shared the shelter in 36% (10 of 28)
of the trials. This result was independent of the magnitude of the size difference
between the sculpin and crayfish in the enclosure. These results differ
from the results of other studies of crayfish/fish interactions.
Invasive signal crayfish have been shown to have a strong impact on
shelter use, and potential survival, of native stream fishes (Griffiths et al.
2004, Guan and Wiles 1996, Light 2005). Several studies (e.g., Gherardi
and Daniels 2004, Klocker and Strayer 2004, Usio et al. 2001) have demonstrated
that invasive crayfish species tend to be more aggressive than their
Figure 1. Frequency of crayfish observations in deep and shallow sides of enclosures.
2008 C. Bishop et al. 87
native counterparts. This aggressive behavior is likely responsible for the
success of these species as invaders. Native crayfish species that have coevolved
with the native fish fauna may interact less aggressively with those
Figure 2. Frequency of shelter use (A) by sculpin with (gray) and without (white)
crayfish, and (B) by crayfish with (gray) and without (white) sculpin.
88 Southeastern Naturalist Vol.7, No. 1
fishes (especially over a resource such as shelter that, as in this case, is used
during periods of inactivity) than do invasive crayfish.
A few studies of interactions between native crayfish species and native
fish species have demonstrated an affect of one species on the other.
Northern crayfish evicted Slimy Sculpin from shelters in a laboratory study
investigating interactions between the two species (Miller et al. 1992). Cottus
bairdi Girard (Mottled Sculpin) actually increased their use of shelters
in the presence of Orconectes putnami Faxon (phallic crayfish) and predatory
Micropterus dolomieu Lacepède (Smallmouth Bass) (McNeely et al.
1990). In the absence of crayfish, the sculpin responded to bass by reducing
movements instead of seeking shelter. The presence of crayfish apparently
distracted the bass and allowed the sculpin to move into shelters, even
though the crayfish frequently evicted the sculpin from the shelters. Similarily,
in the presence of predatory bass, rusty crayfish were shown to evict
Johnny Darters from shelters (Rahel and Stein 1988).
The lack of a strong interaction over shelter might refl ect the fact that
the sculpin that were used in this study were usually larger than the crayfish
that were used, and in other studies, crayfish are consistently the aggressor
species. In most of the studies (discussed above) where crayfish evicted fish
from shelters, the crayfish were larger than the fish (Guan and Wiles 1996,
Miller et al. 1992, Rahel and Stein 1988), with the exception of McNeely
et al. (1990) in which the fish were as large, if not larger, than the crayfish.
Since the methods used to capture sculpin and crayfish for this study were
not size selective (the openings in the minnow traps were sufficiently large
enough to allow entry of the largest crayfish observed in the stream) and all
sculpin and crayfish captured were used in the study, we feel that the sizes
used accurately refl ect the sizes of the sculpin and crayfish in the stream.
It is possible that, at a mean size of 209 cm2, the rocks were large enough
that the shelter they provided was not actually a limiting resource. However, the
shelter sizes are within the range of sizes used in other studies of fish/crayfish
interactions (144 cm2 for McNeely et al. [1990] and 270 cm2 for Guan and Wiles
[1996]) in which crayfish evicted fish from shelters. Additionally, we did not
manipulate densities of either species, instead we provided one shelter per
crayfish/sculpin pair. A more thorough test of interference competition would
include manipulations of the competitor/shelter ratio. However, Guan and
Wiles (1996) demonstrated competition with a ratio of one shelter per crayfish/
fish pair, and Miller et al. (1992) demonstrated competition with a ratio of two
shelters per crayfish/sculpin pair (the crayfish evicted the sculpin in all 17 cases
where one species entered a shelter occupied by the other species).
Both species used the shelters in the allopatric trials in a significantly
higher proportion than expected from the hypothesis of random distribution
within the enclosures. However, since the location of the shelter within each
enclosure was consistent across all trials (in the middle of the enclosure) and
not randomized, it is possible that avoidance of the enclosure walls serves
as a partial explanation for shelter use by the animals. However, crayfish not
using the shelter were consistently found burrowed adjacent to the enclosure
2008 C. Bishop et al. 89
walls, indicating that wall avoidance was not occurring for these animals.
We cannot rule this possibility out for the sculpin. However, adult sculpin
were consistently found adjacent to structures (tree roots and large rocks) in
the stream (D. Mullen, pers. observ.), and it seems unlikely that they would
actively avoid the untreated wooden walls of the enclosures.
A large proportion of individuals of both species were found outside the
shelter (33% for crayfish and 47% for sculpin in the allopatric trials). This
suggests that rock shelters may not be a very important resource for adult
sculpin and crayfish in Brawley’s Fork pools or that the enclosure walls were
also perceived as effective shelter by the crayfish and possibly the sculpin. An
alternative approach to testing for a shelter preference would be to generate
expected values of 50% under shelter and 50% not under shelter. This analysis
would have indicated that neither the sculpin nor the crayfish exhibited a preference
for the shelter (even though 53% of the sculpin and 67% of the crayfish
were found in an area that occupied just 4.2% of the enclosure and Greenberg
and Holtzman (1987) found that Banded Sculpin spend the daylight hours under
rock shelters). Although we feel that this approach is not appropriate, the
results ultimately lead to the same biological conclusion —that the crayfish and
sculpin do not compete for, and will in fact share, rock shelters in this system.
Two of the three studies of interactions between native crayfish and native
fishes mentioned above (McNeely et al. 1990, Rahel and Stein 1988) found
strong interactions over shelter only in the presence of a potential predator
(Smallmouth Bass in both cases). Since there are no potential in-stream (fish)
predators in Brawley’s Fork (D. Mullen, pers. observ.), deep water may serve as
sufficient refuge from predation (by terrestrial and avian predators), reducing
the importance of physical shelter, and therefore the likelihood of strong interactions
over shelter. Both species have other means of reducing predation risk
(burrowing for crayfish and crypsis for sculpin). The crayfish observed outside
the shelter in this study had burrowed into small crevices between the rocks and
the walls of the shelter. Sculpin frequently rely on immobility and crypsis to
avoid detection by predators (McNeely et al. 1990) and were difficult to locate
in this study when they were not using the shelter that was provided.
Acknowledgments
This research was funded by the Biology Department of Middle Tennessee State
University. We thank 2 anonymous reviewers for helpful comments on the manuscript.
We also thank landowner James Ervin for providing access to Brawley’s Fork.
Literature Cited
Creed, R.P., Jr. 1994. Direct and indirect effects of crayfish grazing in a stream community.
Ecology 75(7):2091–2103.
Englund, G., and J.J. Krupa. 2000. Habitat use by crayfish in stream pools: Infl uence
of predators, depth, and body size. Freshwater Biology 43(1):75–83.
Freeman, M.C., and D.J. Stouder 1989. Intraspecific interactions infl uence size-specific depth distribution in Cottus bairdi. Environmental Biology of Fishes 24(3):
231–236.
90 Southeastern Naturalist Vol.7, No. 1
Gherardi, F., and S. Barbaresi. 2000. Invasive crayfish: Activity patterns of Procambarus
clarkii in the rice fields of the Lower Guadalquivir (Spain). Archiv fur
hydrobiologia 150(1):153–168.
Gherardi, F., and W.H. Daniels. 2004. Agonism and shelter competition between
invasive and indigenous crayfish species. Canadian Journal of Zoology 82(12):
1923–1932.
Greenberg, L.A., and D.A. Holtzman. 1987. Microhabitat utilization, feeding periodicity,
home range, and population size of the Banded Sculpin, Cottus carolinae.
Copeia 1987(1):19–25.
Griffiths, S.W., P. Collen, and J.D. Armstrong. 2004. Competition for shelter among
over-wintering signal crayfish and juvenile Atlantic Salmon. Journal of Fish Biology
65(2):436–447.
Guan, R.Z., and P.R. Wiles.1997. Ecological impact of introduced crayfish on benthic
fishes in a British lowland river. Conservation Biology 11(3):641–647.
Hazlett, B., D. Rittschoff, and D. Rubenstein. 1974. Behavioral biology of the
crayfish Orconectes virilis I. Home range. American Midland Naturalist 92:
301–319.
Klocker, C.A., and D.L. Strayer. 2004. Interactions among an invasive crayfish (Orconectes
rusticus), a native crayfish (Orconectes limosus), and native bivalves
(Sphaeriidae and Unionidae). Northeastern Naturalist 11(2):167–178.
Koczaja, C., L. McCall, E. Fitch, B. Glorioso, C. Hanna, J. Kyzar, M. Niemiller,
J. Spiess, A. Tolley, R. Wyckoff, and D. Mullen. 2005. Size-specific habitat
segregation and intraspecific interactions in Banded Sculpin (Cottus carolinae).
Southeastern Naturalist 4(2):207–218.
Light, T. 2005. Behavioral effects of invaders: Alien crayfish and native sculpin in a
California stream. Biological Invasions 7(3):353–367.
Mahon, R., and C.B. Portt. 1985. Local size-related segregation of fishes in streams.
Archiv fur Hydrobiologia 103:267–271.
McNeely, D.L., B.N. Futrell, and A. Sih. 1990. An experimental study on the effects
of crayfish on the predator-prey interaction between bass and sculpin. Oecologia
85(1):69–73.
Miller, J.E., J.F. Savino, and R.K. Neely. 1992. Competition for food between crayfish (Orconectes virilis) and the Slimy Sculpin (Cottus cognatus). Journal of
Freshwater Ecology 7(2):127–136.
Mullen, D.M., and T.M. Burton. 1995. Size-related habitat use by longnose dace
(Rhinichthys cataractae). American Midland Naturalist 133:177–183.
Mullen, D.M., and T.M. Burton. 1998. Experimental tests of competition in stream
riffl es between juvenile and adult Longnose Dace (Rhinichthys cataractae). Canadian
Journal of Zoology 76:855–862
Power, M.E. 1984. Depth distribution of armored catfish: Predator-induced resource
avoidance? Ecology 65:523–528.
Rahel, F.J., and R.A. Stein. 1988. Complex predator-prey interaction and predator
intimidation among crayfish, piscivorous fish, and small benthic fish. Oecologia
75:94–98.
Usio, N., M. Konishi, and S. Nakano. 2001. Species displacement between an introduced
and a “vulnerable” crayfish: The role of aggressive interactions and shelter
competition. Biological Invasions 3(2):179–185.
Zar, J.H. 1984. Biostatistical Analysis, 2nd Edition. Prentice Hall, Englewood Cliffs,
NJ. 718 pp.