This article was downloaded by: [UQ Library]On: 13 July 2015, At: 07:25Publisher: Taylor & FrancisInforma Ltd Registered in England and Wales Registered Number: 1072954 Registered office: 5 Howick Place,London, SW1P 1WG

Transactions of the American Fisheries SocietyPublication details, including instructions for authors and subscription information:http://www.tandfonline.com/loi/utaf20

Habitat and Diet Partitioning between Shoal Bass andLargemouth Bass in the Chipola River, FloridaA. P. Wheeler a & Michael S. Allen aa Department of Fisheries and Aquatic Sciences, University of Florida , 7922 Northwest 71stStreet, Gainesville, Florida, 32653-3071, USAPublished online: 09 Jan 2011.

To cite this article: A. P. Wheeler & Michael S. Allen (2003) Habitat and Diet Partitioning between Shoal Bass andLargemouth Bass in the Chipola River, Florida, Transactions of the American Fisheries Society, 132:3, 438-449, DOI:10.1577/1548-8659(2003)132<0438:HADPBS>2.0.CO;2

To link to this article: http://dx.doi.org/10.1577/1548-8659(2003)132<0438:HADPBS>2.0.CO;2

PLEASE SCROLL DOWN FOR ARTICLE

Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) containedin the publications on our platform. However, Taylor & Francis, our agents, and our licensors make norepresentations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of theContent. Any opinions and views expressed in this publication are the opinions and views of the authors, andare not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon andshould be independently verified with primary sources of information. Taylor and Francis shall not be liable forany losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoeveror howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use ofthe Content.

This article may be used for research, teaching, and private study purposes. Any substantial or systematicreproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in anyform to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http://www.tandfonline.com/page/terms-and-conditions

438

Transactions of the American Fisheries Society 132:438–449, 2003q Copyright by the American Fisheries Society 2003

Habitat and Diet Partitioning between Shoal Bass andLargemouth Bass in the Chipola River, Florida

A. P. WHEELER*1 AND MICHAEL S. ALLEN

Department of Fisheries and Aquatic Sciences,University of Florida,

7922 Northwest 71st Street,Gainesville, Florida 32653-3071, USA

Abstract.—We investigated the macrohabitat use, microhabitat use, and food habits of shoal bassMicropterus cataractae and largemouth bass M. salmoides in the upper Chipola River, Florida. Weelectrofished two macrohabitats (pools and shoals) during the summer (May–August) and fall(September–December) of 1999 and 2000. The ratio of shoal bass to largemouth bass differedamong macrohabitats, being highest in the shoals and lowest in the pools. Age-0 and adult (age-1 and older) shoal bass were collected in areas of higher-than-average percentages of rocky substratein both shoals and pools. Age-0 and adult largemouth bass were associated with areas of reducedcurrent velocity and those with higher-than-average amounts of woody debris. Though the dietsof age-0 and adult shoal bass and largemouth bass were similar, a few differences were apparent.Age-0 largemouth bass diets contained grass shrimp Palaemonetes spp., whereas age-0 shoal bassdiets contained mostly mayflies (order Ephemeroptera: families Baetidae and Isonychidae). Cray-fish and fish were the primary food resources of adult shoal bass and adult largemouth bass, andcrayfish became more prevalent than fish in larger individuals of both species. Largemouth basstransitioned to a crayfish-dominated diet at a smaller size than did shoal bass. Considering thatwe found substantial differences in the habitat associations of these species but relatively few dietdifferences, habitat partitioning may be important for the coexistence of shoal bass and largemouthbass in rivers and streams. Future shoal bass conservation efforts should focus on maintaining adiversity of habitats where these species coexist and on protecting relatively rare shoals.

Shoal bass Micropterus cataractae, the most re-cently described species of black bass Micropterusspp., is believed to be threatened by habitat lossthroughout its limited range. Its endemic range isthe Apalachicola–Chattahoochee–Flint Riverdrainage of Florida, Alabama, and Georgia, andintroductions have been limited to the OcmulgeeRiver in Georgia (Williams and Burgess 1999).The range of shoal bass is diminishing due to thedestruction of natural riverine habitat by impound-ments (Williams and Burgess 1999). Althoughshoal bass can survive and reproduce in ponds(Smitherman and Ramsey 1972), they do not per-sist in impoundments (Ramsey 1975; Williams andBurgess 1999). Thus, shoal bass have disappearedfrom much of the Chattahoochee River and its trib-utaries in Georgia and Alabama (Williams andBurgess 1999). Within Florida, shoal bass are re-stricted to a segment of the upper Chipola Riverand an area below Jim Woodruff Dam on the Ap-alachicola River. These populations are threatened

* Corresponding author: wheelerap@brinet.com1 Present address: North Carolina Wildlife Resources

Commission, 20830 Great Smoky Mountain Express-way, Waynesville, North Carolina 28786, USA.

Received December 7, 2001; accepted October 3, 2002

by pollution and siltation in the Chipola River(Ogilvie 1980; Williams and Burgess 1999) andby dredging and irregular flows in the Apalachi-cola River (Williams and Burgess 1999). Shoalbass are considered a species of special concernin Alabama (Ramsey 1976) and threatened in Flor-ida (Gilbert 1992).

Unfortunately, a paucity of work has addressedthe biology and ecology of shoal bass. Williamsand Burgess (1999) concluded that future conser-vation efforts should include studies to determinethe microhabitat requirements of the species. Al-though shoal bass generally are thought to inhabitshoals in rivers and streams (Ramsey 1975; Gilbert1992; Williams and Burgess 1999), few previousstudies have attempted to verify this association.Wright (1967) collected shoal bass and largemouthbass M. salmoides at a 10:1 ratio from shoals anda 3:1 ratio from pools. Hurst (1969) collected shoalbass in both pool and riffle areas but collectedlargemouth bass mainly from large, deep pools.Shoal bass diets have been examined by few in-vestigators but have been found to consist of fishand crayfish (Wright 1967; Hurst 1969; Ogilvie1980).

Shoal bass share their range with endemic large-mouth bass. Largemouth bass are widely distrib-

Dow

nloa

ded

by [

UQ

Lib

rary

] at

07:

25 1

3 Ju

ly 2

015

439SHOAL AND LARGEMOUTH BASS HABITAT AND DIET PARTITIONING

uted, and many previous studies have examinedtheir biology and ecology. However, previous re-search has focused almost exclusively on lenticpopulations (Hamilton and Powles 1983), eventhough largemouth bass are common in a widerange of lotic habitats (Jenkins and Burkhead1993) and are often abundant in streams (Fajen1975).

Largemouth bass are generally reported to in-habit pools and backwater areas in rivers andstreams (Trautman 1957; Wydoski and Whitney1979) but may be habitat generalists in lotic sys-tems. For example, Schramm and Maceina (1986)found largemouth bass in a wide range of habitatsin the Santa Fe River, Florida. Sowa and Rabeni(1995) found largemouth bass biomass and densitywere positively correlated to maximum summertemperature, mean depth, pool area, and pool:riffleratio, whereas canopy cover and gradient werenegatively correlated to largemouth bass biomassand abundance in Missouri streams.

Largemouth bass are generally considered pis-civores (Heidinger 1975) but may prey on a widevariety of aquatic organisms (Jenkins and Burk-head 1993). Previous studies of largemouth bassdiets in lotic environments have determined thatfish were their primary prey (Scalet 1977; Davies1981; Hamilton and Powles 1983). However,Schramm and Maceina (1986) found that crayfishwere the primary food resource of largemouth bassin the Santa Fe River, Florida.

Shoal bass and largemouth bass are naturallysympatric throughout the range of shoal bass (Wil-liams and Burgess 1999). These species are po-tential competitors for prey and space because theyhave similar morphology and prey. Because thesefishes are naturally sympatric, they may exhibitresource partitioning to reduce interspecific inter-actions and facilitate coexistence. Previous studiesof sympatric black bass populations have variouslydocumented differences in spatial resource use(Sowa and Rabeni 1995; Sammons and Bettoli1999), spatial but not food resource use (Janssen1992; Scott and Angermeier 1998), food but notspatial resource use (Scalet 1977), neither food norspatial resource use (Hubert 1977), and both spa-tial and food resource use (Warden and Hubert1980; Schramm and Maceina 1986).

We investigated the habitat use and food habitsof shoal bass and largemouth bass in a lotic system.Our objectives were to test (1) whether the ratioof collected shoal bass to largemouth bass variedbetween macrohabitats (i.e., pools and shoals), (2)whether shoal bass or largemouth bass were as-

sociated with different microhabitat parameters(i.e., current velocity, substrate, woody debris, anddepth) than the average conditions within bothpools and shoals, and (3) whether diets were sim-ilar between shoal bass and largemouth bass.

Methods

Study site.—The Chipola River is a low-gradientstream (0.17 m/km; Bass and Cox 1985) that flows201 km south from Alabama to its confluence withthe Apalachicola River. Its watershed area is 3,124km2, and average discharge is 42.8 m3/s (Bass andCox 1985). Although the Chipola River receivessome surface runoff, it is primarily spring fed (Par-sons and Crittenden 1959; Bass and Cox 1985),resulting in low turbidity (Parsons and Crittenden1959) and relatively stable year-round water tem-peratures (range, 10–248C; Parsons and Crittenden1959; Bass and Cox 1985). The substrate is pri-marily sand and cobble, with limestone outcrops(Parsons and Crittenden 1959). The watershed isrelatively pristine, containing some agriculturaldevelopments and small towns (Winger et al.1987).

The 48-km section of river between the townsof Marianna and Clarksville, Florida, is inhabitedby shoal bass and largemouth bass. This sectionof river is characterized by two primary macro-habitats, pools and shoals, with pools being thedominant macrohabitat. Pools are characteristi-cally narrow (mean width, 31 m), long (.1 km),deep (mean, .2 m), and slow flowing. Shoals areshorter (,200 m), wider (mean, 66 m), shallower(mean depth, ,1 m), and have higher current ve-locities than the pools. Shoals also contain bedsof eelgrass Vallisneria americana, whereas poolsare devoid of aquatic macrophytes. We selectedthree pools and three shoals for sampling. Thethree shoals were locations where previous studiescollected shoal bass, and the pools were adjacentto the shoals. All study units were located withina 15-river-kilometer (rkm) section.

Fish collection.—Fish and microhabitat sam-pling were conducted in summer (May–August)and fall (September–December) of 1999 and 2000.Fish were collected by boat electrofishing withDC. Habitat differences between pools and shoalsrequired different electrofishing techniques for ef-fective fish collection. Shoals were electrofishedby moving upstream in such a way that all areasdeep enough to navigate were exposed to the elec-tric field. Due to the presence of shallow areas(,0.25 m), the entire area of the shoals could notbe navigated; however, many areas too shallow to

Dow

nloa

ded

by [

UQ

Lib

rary

] at

07:

25 1

3 Ju

ly 2

015

440 WHEELER AND ALLEN

navigate were within reach of the cathode probe.Pools were electrofished by drifting longitudinallydown the habitat unit while moving laterally be-tween the right bank, left bank, and the center ofthe channel. Units were sampled as many as threetimes in each season but always with at least 48h between samples. Because pools are the primarymacrohabitat in the upper Chipola River, weweighted our sampling effort in favor of pools.

Areas where shoal bass or largemouth bass werecollected were marked with a weighted buoy formicrohabitat measurements. The buoys were colorcoded to distinguish between age-0 fish and thoseolder than age 0 (hereafter referred to as adults)and between shoal bass versus largemouth bass.Ages of fish that appeared to be age 0 in the fieldwere verified by examining sagittal otoliths in thelaboratory. Ad hoc sampling in areas adjacent tothe three shoals and three pools was used to sup-plement fish collections for diet analysis.

Macrohabitat analysis.—Because habitat differ-ences between pools and shoals required differentelectrofishing techniques, we could not use elec-trofishing catch per effort to compare the relativeabundances of shoal bass and largemouth bass be-tween macrohabitats. We tested whether the pro-portion of shoal bass to largemouth bass differedamong macrohabitats with a weighted-least-squares analysis of proportions (WLSAP; Statis-tical Analysis System [SAS] Catmod procedure;SAS 1994; Stokes et al. 2000). The WLSAP testedwhether the proportion of shoal bass to largemouthbass differed across macrohabitat types, seasons,and years.

Microhabitat characterization and analysis.—We characterized the microhabitat associations ofage-0 and adult shoal bass and largemouth bass bymeasuring or estimating physical habitat at the lo-cation where individuals were collected. Micro-habitat parameters were measured at an area des-ignated longitudinally (i.e., upstream–downstreamaxis) as 10 m wide, centered at the point the in-dividual was captured, and latitudinally as the left,right or center third of the stream. The area con-tained in each interval was variable, dependent onstream width.

We characterized microhabitat conditions in thepools and shoals using four permanent transectsin each habitat unit. Permanent transects were per-pendicular to the channel and approximately even-ly spaced through each unit. A tree on the bankadjacent to the transect was marked with paint,allowing the same fixed transects to be usedthroughout the study. During each sampling event

(i.e., summer and fall of 1999 and 2000), micro-habitat characteristics were recorded from threeintervals equally spaced across each transect (e.g.,two near the bank and one midchannel). The lon-gitudinal width (i.e., upstream–downstream axis)of the intervals was defined as 10 m on either sideof the transect. The area contained in each intervalwas variable, dependant on stream width. Thus,the intervals of the fixed transects were statisticallycomparable to the areas where fish were collected.

Microhabitat and cover variables were measuredor estimated for each interval of the fixed transectsand for the areas where individual shoal bass orlargemouth bass were collected. Mean depth wasestimated to the nearest 5 cm with a depth rod.Mean current velocity was measured at 60% ofdepth, at a point representative of the interval(judged visually), with a Marsh-McBirney model201M portable water current meter. Substrate wasrecorded as the visually estimated percentage ofsilt (,0.062 mm), sand (.0.062–2 mm), gravel(.2–16 mm), pebble (.16–64 mm), cobble (.64–256 mm), boulder (.256 mm), and flat bedrock.The total percentage of the substrate classified ascobble and boulder was summed and used to in-dicate rocky substrate particles. Rocky substratepercentage was used in the analysis because it pro-vided an easily interpretable index of substratecoarseness and because there is a reported asso-ciation of other black basses with these substrates(Leonard and Orth 1988; Todd and Rabeni 1989;Lobb and Orth 1991). The percent area coveredby eelgrass was also visually estimated. Individualpieces of woody debris were visually counted infour categories (I–IV) based on length and diam-eter, as done by Dolloff et al. (1993). The fourcategories we used were: (I) 1–5 m in length and5–10 cm in diameter, (II) 1–5 m in length and over10 cm in diameter, (III) over 5 m in length and 5–10 cm in diameter, and (IV) over 5 m in lengthand over 10 cm in diameter. In the case of fallentrees, branches connected to a larger trunk werecategorized and enumerated individually. Fromthese counts and categories, a woody debris index(WDI) was calculated as

WDI 5 (Wood I) 1 [2 3 (Wood II)]

1 [3 3 (Wood III)]

1 [4 3 (Wood IV)], (1)

where Wood I–IV represent numbers of woody de-bris of each size category in the interval. Thisindex was weighted so that larger size-classes of

Dow

nloa

ded

by [

UQ

Lib

rary

] at

07:

25 1

3 Ju

ly 2

015

441SHOAL AND LARGEMOUTH BASS HABITAT AND DIET PARTITIONING

debris received an arbitrarily higher value. Largervalues of WDI are used to indicate greater amountsof woody debris in the interval. Percent area cov-erage of eelgrass was estimated visually in eachinterval. The first author measured or estimated allmicrohabitat parameters except mean current ve-locity throughout the study.

We validated the visual estimation of substratecomposition. Substrate composition of 15 approx-imately 25-m2, shallow (depth, ,1 m) locations ina shoal was visually estimated as previously de-scribed. Locations were selected to reflect as muchvariation in substrate particle composition as pos-sible. After visually estimating the substrate at alocation, we randomly tossed a 0.5-m2 ring, di-vided by string into 16 equal sections, into thelocation 10 times. The dominant substrate particlecategory was recorded in each of the sections forevery toss. For each location, the percentage ofthe 160 sections (16 sections 3 10 tosses) domi-nated by rocky substrate was compared to the vi-sual estimation of percent rocky substrate. The ar-eas used to verify the visual estimation of substrateparticles contained substrates ranging from sandto boulder. There was a significant linear corre-lation (R 5 0.94, P , 0.001) between the visualestimation of rocky substrate and the observedproportion of sections dominated by rocky sub-strate. Thus, the visual estimation appeared to bean adequate index of substrate composition.

We used a multivariate analysis of variance(MANOVA) to test the null hypothesis that mi-crohabitat variables collectively were similar be-tween the permanent transects (average conditionsof the macrohabitat) and the areas where individ-uals were collected. The analysis was repeated forage-0 and adult individuals of both species. TheMANOVAs were conducted for individuals inpools and shoals (eight analyses total) based onthe microhabitat observations (i.e., intervals) fromthe fixed transects and the microhabitat data fromareas where individuals were collected. Currentvelocity, depth, percent rocky substrate, and WDIscore were used as response variables, and year(1999, 2000), season (summer, fall), species pres-ence (e.g., at least one individual was or was notcollected), and the interaction between speciespresence and season were used as predictor vari-ables. In addition, eelgrass abundance was used asa response variable for the shoals. Data were log10

transformed as necessary to help meet the as-sumption of a multivariate normal distribution(Zar 1988). We used Wilk’s lambda (SAS 1994)to test the null hypothesis that year, season, species

presence, and the species presence 3 season in-teraction had no effect on the mean microhabitatobservations. In cases of an overall significantMANOVA for the effect of species presence, one-way analyses of variance (ANOVA) were used todetermine microhabitat parameters that contrib-uted to the significant difference between the areaswhere individuals were collected and the perma-nent transects.

Diet collection and analysis.—Transparentacrylic tubes (Van Den Avyle and Roussel 1980)were used to remove the stomach contents of allshoal bass and largemouth bass larger than 200mm total length. A flexible claw retriever (Dimond1985) assisted in the removal of large items andcrayfish. The removed stomach contents wereplaced on ice and returned to the laboratory, wherethe items were enumerated, identified, andweighed. Shoal bass and largemouth bass 200 mmor smaller were placed on ice and returned to thelaboratory, where they were dissected and theirstomach contents removed, enumerated, identified,and weighed to the nearest 0.001 g. When possible,fish remains were identified to genus (based onOates et al. 1993) and insects to order.

Diet was quantified for each species in terms ofpercent by weight. Pianka’s (1973) index of dietoverlap (O) was used to index the similarity of thediets of the two species. Pianka’s Index is definedas:

0.5O 5 p p p 3 p , (2)@O 1O O 22i 1i 2i 1i

where O is the overlap in resources between spe-cies 1 and 2, p1i is the weight of resource categoryi found in species 1, and p2i is the weight of re-source category i found in species 2. The resourcecategories were fish, crayfish, insects, other in-vertebrates, and other prey items. Overlap (i.e.,similarity) values range from 0 (none) to 1 (com-plete). EcoSim software (Gotelli and Entsminger1997) was used to generate null models and tocalculate the probability of observed diet overlapoccurring by chance. Driscoll and Miranda (1999)used EcoSim software to evaluate diet overlapamong age-classes of yellow bass Morone missis-sippiensis in Mississippi River oxbow lakes. A ran-domization algorithm (RA2, Gotelli and Entsmin-ger 1997) was selected that generated null modelsfor each species based on only those prey itemsobserved in the diets.

Ontogenetic diet shifts were evaluated with lo-gistic regression. Logistic regression models theprobability of a binary response based on contin-

Dow

nloa

ded

by [

UQ

Lib

rary

] at

07:

25 1

3 Ju

ly 2

015

442 WHEELER AND ALLEN

FIGURE 1.—Length-frequency histogram showing size(TL; 20-mm groups) and number of shoal and large-mouth bass collected during 1999 and 2000 combined.

TABLE 1.—Mean 6 2 SEs observations for depth, current velocity, percentage rocky substrate, woody debris index(WDI), and eelgrass (% coverage) in the fixed transects of the pools (N 5 36) and shoals (N 5 36) during summer(May–August) and fall (September–December), 1999 and 2000, in the Chipola River.

Habitat Year Season Depth (cm)Velocity(cm/s)

Rockysubstrate WDI Eelgrass

Pool 1999

2000

SummerFallSummerFall

195 6 11.6173 6 9.6158 6 9.5154 6 9.8

39 6 2.126 6 2.015 6 1.613 6 1.4

23 6 4.228 6 3.829 6 3.733 6 4.7

3 6 1.32 6 0.82 6 0.5

,1 6 0.3

,1 6 0.3,1 6 0.1,1 6 0.110 6 4.1

Shoal 1999

2000

SummerFallSummerFall

105 6 9.569 6 6.163 6 5.767 6 6.3

51 6 5.331 6 2.627 6 3.219 6 2.1

29 6 3.343 6 2.642 6 3.042 6 3.2

2 6 0.62 6 0.72 6 0.61 6 0.3

9 6 3.126 6 4.81 6 0.8

22 6 5.2

uous and/or categorical predictor variables (SASprocedure Logistic; SAS 1994; Stokes et al. 2000).In this application, the binary response variablewas fish or crayfish remains dominating (byweight) the diet of shoal bass or largemouth bass.The linear model used was

logit(p) 5 a 1 b 3 TL 1 b 3 species1 2

1 b 3 (species 3 TL), (3)3

where logit(p) is the logistic probability of crayfish(versus fish) dominating the diet of either shoalbass or largemouth bass, a is the intercept value,TL is the total length (mm) of each species, speciesis the main effect of the categorical variable spe-cies type (shoal bass or largemouth bass), species3 TL is the interaction between species and TL,

and b1–b3 are the logistic regression coefficients.The estimate of the logit(p) was used to obtain thepredicted probability of a diet being crayfish dom-inated, that is,

logit(p) logit(p)p 5 e /(1 1 e ). (4)

Wald’s chi-square statistic (SAS 1994) was usedto test the significance of the individual modelterms.

Results

The total amount of electrofishing time was 88.6h, accumulated from pool (51.5 h), shoal (26.1 h),and ad hoc (10.9 h) collections. A total of 105adult shoal bass, 316 adult largemouth bass, 288age-0 shoal bass, and 125 age-0 largemouth basswere collected during this study. Shoal bass rangedfrom 70 to 480 mm TL, whereas largemouth bassranged from 60 to 560 mm TL (Figure 1).

The study area generally exhibited slightlylower-than-average monthly discharges for bothstudy years, and average monthly flows were high-er in 1999 than 2000. Mean depth and current ve-locity were greater in 1999 than 2000 in both poolsand shoals (Table 1) due to higher rainfall in 1999.The percent rocky substrate was relatively consis-tent throughout the study. Eelgrass was more abun-dant in 2000 (Table 1).

Macrohabitat Results

Age-0 shoal bass and largemouth bass were col-lected from both pools and shoals (Table 2). How-ever, the WLSAP detected differences in the ratioof age-0 shoal bass to age-0 largemouth bass inpools and shoals. We collected too few age-0 in-dividuals in 1999 to test for differences betweenthe two sample years (Table 2). Therefore, theyears were pooled for this analysis. Macrohabitattype was the sole significant predictor of the ratioof age-0 shoal bass to age-0 largemouth bass

Dow

nloa

ded

by [

UQ

Lib

rary

] at

07:

25 1

3 Ju

ly 2

015

443SHOAL AND LARGEMOUTH BASS HABITAT AND DIET PARTITIONING

TABLE 2.—Number of fish collected from pool andshoal sample sites in the Chipola River as well as from adhoc sampling of nearby areas that was used to supplementfish collections for diet contents.

Species and age-class

Sample

Pool Shoal Ad hoc

Age-0 shoal bassAdult shoal bassAge-0 largemouth bassAdult largemouth bass

1188382

252

137262034

339

2333

TABLE 3.—Results of MANOVA analyses testing fordifferences between the mean habitat observations in in-tervals where fish were collected (presence) versus thefixed intervals. In addition, effects of season (summer orfall), year (1999 or 2000), and the interaction betweenpresence and season were included in the MANOVA. Be-cause age-0 fish of both species did not fully recruit to thegear until fall sampling, fall habitat observations wereused in the MANOVA analyses for age-0 shoal and large-mouth bass.

EffectWilk’slambda F df P-value

Age-0 shoal bass in pools

PresenceYear

0.85310.8270

6.68.1

4, 1544, 154

,0.001,0.001

Age-0 shoal bass in shoals

PresenceYear

0.86080.8785

4.74.0

5, 1455, 145

,0.0010.002

Age-0 largemouth bass in pools

PresenceYear

0.80050.6588

7.114.8

4, 1144, 114

,0.001,0.001

Age-0 largemouth bass in shoals

PresenceYear

0.82030.8725

3.42.3

5, 775, 77

0.0080.058

Adult shoal bass in pools

PresenceSeasonPresence 3 seasonYear

0.93970.93400.98500.6229

3.33.10.7

26.5

4, 1754, 1754, 1754, 175

0.0120.0170.617

,0.001

Adult shoal bass in shoals

PresenceSeasonPresence 3 seasonYear

0.89680.94000.97030.7625

3.51.90.99.4

5, 1515, 1515, 1515, 151

0.0050.0930.468

,0.001

Adult largemouth bass in pools

PresenceSeasonPresence 3 seasonYear

0.74160.92990.97730.7030

20.04.31.3

24.2

4, 2294, 2294, 2294, 229

,0.0010.0020.260

,0.001

Adult largemouth bass in shoals

PresenceSeasonPresence 3 seasonYear

0.70680.95530.94030.7867

13.71.52.19.0

5, 1655, 1655, 1655, 165

,0.0010.1790.069

,0.001

(x2 5 41.77, df 5 1, P , 0.0001). The ratio ofage-0 shoal bass to age-0 largemouth bass wassignificantly higher in shoals (6.9:1) than in pools(1.4:1).

Adult shoal bass and largemouth bass were col-lected from both pools and shoals (Table 2). How-ever, the WLSAP also detected differences in theratio of adult shoal bass to adult largemouth bassin these macrohabitats. The ratio of adult shoalbass to adult largemouth bass was predicted byyear (x2 5 10.88, df 5 1, P 5 0.0010), macro-habitat (x2 5 8.08, df 5 1, P 5 0.0945), and season(x2 5 21.41, df 5 1, P , 0.0001). The main effectof year was significant because the ratio of adultshoal bass to adult largemouth bass was lower in2000 (1:4.1) than in 1999 (1:3.51). The significantmain effect of season resulted from the collectionof a higher ratio of adult shoal bass to adult large-mouth bass in the summer sampling (1:2.0) thanin fall sampling (1:4.3). The ratio of adult shoalbass to adult largemouth bass was greater in theshoals (1:1.3) than in the pools (1:3.1).

Microhabitat Results

All MANOVA analyses detected significant dif-ferences between the microhabitats where age-0and adult shoal bass and largemouth bass werecollected and the average conditions of both poolsand shoals (Table 3). The presence 3 season in-teraction was never significant (all P . 0.06), in-dicating that generally these species did not changemicrohabitat associations seasonally. The main ef-fect of presence was always significant (all P ,0.008), indicating that both species were associ-ated with microhabitat parameters that differedfrom the mean microhabitat parameters in poolsand shoals (Table 3).

Age-0 and adult shoal bass showed similartrends in microhabitat associations. Age-0 andadult shoal bass were associated with deeper-than-average areas in shoals (one-way ANOVA: F1,149

5 18.1, P , 0.001; F1,155 5 6.1, P 5 0.015) and

shallower-than-average areas in pools (F1,157 512.4, P , 0.001; F1,178 5 5.5, P 5 0.020). Age-0and adult shoal bass were collected in areas ofhigher-than-average rocky substrate in shoals(F1,149 5 4.9, P 5 0.029; F1,155 5 9.9, P 5 0.002)and pools (F1,149 5 16.3, P , 0.001; F1,178 5 10.3,P 5 0.002). Adult shoal bass were found at lower-than-average eelgrass coverage in the shoals (F1,155

5 5.2, P 5 0.024).Age-0 and adult largemouth bass also showed

similar trends in microhabitat associations. Age-0and adult largemouth bass were associated with

Dow

nloa

ded

by [

UQ

Lib

rary

] at

07:

25 1

3 Ju

ly 2

015

444 WHEELER AND ALLEN

deeper-than-average areas in the shoals (one-wayANOVA: F1,81 5 8.5, P 5 0.005; F1,169 5 24.9,P , 0.001) and shallower-than-average areas inthe pools (F1,117 5 8.8, P 5 0.004; F1,232 5 5.8,P 5 0.017). Age-0 and adult largemouth bass werecollected from lower-than-average current veloc-ities in pools (F1,117 5 3.8, P 5 0.054; F1,232 55.3, P 5 0.022). In both pools and shoals, adultlargemouth bass were associated with areas ofhigher-than-average WDI scores (F1,232 5 67.5,P , 0.001; F1,169 5 22.0, P 5 0.001). Adult large-mouth bass were found at lower-than-average cov-erage of eelgrass in the shoals (F1,169 5 16.1, P ,0.001).

Diet Analyses

Stomach contents were examined for 288 (77empty) age-0 shoal bass and 125 (25 empty) age-0 largemouth bass. All diets collected in 1999 wereweighed, but because higher numbers of age-0shoal bass and age-0 largemouth bass were col-lected in 2000, the age-0 diets were subsampled.Individuals collected during 2000 were randomlysubsampled until 10 (nonempty) individuals perlength group were examined. There was a higheroverlap in the diets of age-0 shoal bass and age-0 largemouth bass than would be expected to occurby chance alone (O 5 0.88, P 5 0.054), indicatingthat their diets were generally similar. The dietsof most age-0 shoal bass and age-0 largemouthbass were dominated by unidentifiable fish, andboth species exhibited increasing prevalence offish in the diets with increasing TL (Figure 2).

However, the high diet overlap values may ex-aggerate diet similarity due to the prevalence ofrelatively heavy fish remains. Differences in dietswere apparent when prey items were compared bysize-classes, especially for individuals less than150 mm TL (Figure 2). Age-0 shoal bass diets weredominated by mayflies (order Ephemeroptera:families Baetidae and Isonychidae), whereas age-0 largemouth bass showed a higher prevalence ofgrass shrimp Palaemonetes spp. in their diets (Fig-ure 2). Diets consisting of over 200 individualmayflies were common in age-0 shoal bass, butthis phenomenon was not observed in age-0 large-mouth bass.

Stomach contents of 118 (35 empty) adult shoalbass and 319 (124 empty) adult largemouth basswere examined. There was a higher overlap in thediets of adults of both species than would be ex-pected to occur by chance alone (O 5 0.86, P 50.073). Crayfish and fish dominated the adult dietsof both species (Figure 2). Although the majority

of fish remains were unidentifiable, darters Percinaspp. and madtoms Noturus spp. were most com-mon in the diets of adult shoal bass, whereas dart-ers and sunfishes Lepomis spp. were most prevalentin the diets of adult largemouth bass.

The logistic regression analysis revealed the oc-currence of an ontogenetic shift in the diets of adultshoal bass and adult largemouth bass. Crayfish be-came increasingly more dominant than fish in thediets of large individuals of both species (maineffect of TL x2 5 46.9, P , 0.001, Figure 3). Thediet shift is modeled by:

logit(p) 5 23.2108 1 0.0113 3 TL

2 0.7767 3 (species)

1 0.00541 3 (species 3 TL)

where species is equal to 21 for shoal bass and11 for largemouth bass. Largemouth bass madethe transition to a crayfish-dominated diet fasterand at a smaller TL than did shoal bass (species3 TL interaction x2 5 10.8, P 5 0.001, Figure 3).For example, the logistic regression predicted thata 239-mm TL largemouth bass would have anequal probability of displaying a crayfish-domi-nated or fish-dominated diet, whereas for shoalbass, such diets did not reach equal probabilityuntil 413 mm TL.

Discussion

Age-0 and adult shoal bass and largemouth basswere collected from both pools and shoals, indi-cating that both species were somewhat general-istic in their macrohabitat associations. The upperChipola River contains primarily pool habitat, andshoal habitat makes up a relatively minor propor-tion of the area. Thus, we invested greater effortin sampling pools, and we collected more shoalbass (except age-0 fish) and largemouth bass fromthe pools. Physical differences in pools and shoalsprevented us from using electrofishing catch pereffort to assess differences in the abundance ofthese species across macrohabitats. However, wedetected differences in the ratio of one species toanother in the different macrohabitats, which in-dicated habitat differences between these two spe-cies. The ratio of shoal bass to largemouth basswas highest in the shoals and lowest in the pools,indicating that shoal bass represented a signifi-cantly greater percentage of the black bass assem-blage in shoals. This trend was consistent for bothage-0 and adult fish and consistent with previousstudies (Wright 1967; Hurst 1969).

Dow

nloa

ded

by [

UQ

Lib

rary

] at

07:

25 1

3 Ju

ly 2

015

445SHOAL AND LARGEMOUTH BASS HABITAT AND DIET PARTITIONING

FIGURE 2.—Percent by weight of prey items in the diets of shoal and largemouth bass collected during 1999 and2000 combined. The number of stomachs containing prey items is listed over the respective length category column.Diets of individuals smaller than 200 mm TL were subsampled in 2000.

Macrohabitat associations of age-0 individualsmay have been influenced by differing nestingpreferences of shoal bass and largemouth bass.Previous studies have observed shoal bass nestsin a shoal (Wright 1967) and just upstream of ariffle (Hurst 1969), whereas largemouth bass inlotic systems are believed to nest in pools (Jenkinsand Burkhead 1993). Given the reported nestingpreferences of shoal bass and the high proportion

of age-0 shoal bass in shoals, these areas may rep-resent spawning habitat for adults, nursery habitatfor shoal bass, or both. However, we did not doc-ument spawning habitat or movement of eitherspecies, which warrants further investigation.

Age-0 and adult individuals of both speciesshowed differences in habitat use from the meanmicrohabitat observations in the pools and shoals.Age-0 and adult individuals of both species were

Dow

nloa

ded

by [

UQ

Lib

rary

] at

07:

25 1

3 Ju

ly 2

015

446 WHEELER AND ALLEN

FIGURE 3.—The relationship between TL (mm) andthe probability of an individual largemouth bass or shoalbass exhibiting a crayfish-dominated diet (by weight).Results of the logistic regression analysis are shown (seetext).

associated with deeper-than-average areas in theshoals and shallower-than-average areas in thepools. These observations likely result from a gearbias, because our electrofishing equipment was notable to reach all the shallow areas in the shoalsand was likely ineffective in sampling the deepestareas in the pools (.2 m). Therefore, the habitatassociations we identified probably exclude shal-low areas in shoals and deep sections of the pools.

Shoal bass were associated with areas of higher-than-average percentages of rocky substrate,whereas largemouth bass were not. Age-0 shoalbass were associated with higher-than-average per-centages of rocky substrate in pools, and adultshoal bass were associated with higher-than-average percentages of rocky substrate in bothpools and shoals. Cobble and boulder substrate iscommonly considered an important habitat typefor smallmouth bass M. dolomieu. Age-0 small-mouth bass commonly associate with cobble sub-strate (Leonard and Orth 1988; Livingstone andRabeni 1991), and adults are known to use boul-ders as cover in areas that are surrounded withother substrates, such as cobble (Leonard and Orth1988; Todd and Rabeni 1989; Lobb and Orth1991). Our results suggest similar habitat associ-ations for shoal bass.

Largemouth bass are believed to inhabit poolsand backwater areas in lotic situations (Trautman1957; Wydoski and Whitney 1979). We found ev-idence to support this presumption. The ratio ofshoal bass to largemouth bass was lowest in the

pools, and age-0 and adult largemouth bass werecollected in lower-than-average current velocitiesin pools. Shoal bass were not associated with areasof lower-than-average current velocity in eitherpools or shoals. Miller (1975) reported that large-mouth bass inhabited pools and backwater areaswhen they occurred in streams with smallmouthbass, spotted bass M. punctulatus, and Suwanneebass M. notius. Sowa and Rabeni (1995) foundlargemouth bass were most abundant in streamswith high pool:riffle ratios. In addition, Schrammand Maceina (1986) found intermediate-sized(149–299 mm TL) largemouth bass were mostabundant in areas of relatively low current velocity(3–25 cm/s) in the Santa Fe River, Florida. Thus,pools may be the preferred habitat of largemouthbass in lotic systems. However, Schramm and Ma-ceina (1986) collected large (.300 mm TL) large-mouth bass primarily from a turbulent area withhigh current velocity (30–92 cm/s) and bedrocksubstrates. Similarly, we collected some large-mouth bass in shoals, suggesting that largemouthbass may be more general in their stream habitatassociations than previously believed.

Adult largemouth bass were associated withhigher-than-average WDI scores in both the poolsand shoals. Age-0 largemouth bass were also as-sociated with higher-than-average WDI scores inthe pools. The importance of woody debris inwarmwater streams is well documented. Woodydebris provides increased invertebrate production(Benke et al. 1985), protection from strong currentvelocity (Todd and Rabeni 1989), and camouflagefrom predators or prey (Angermeier and Karr1984). However, Lehtinen et al. (1997) found norelationship between largemouth bass habitat as-sociations and woody debris in the upper Missis-sippi River, and Sowa and Rabeni (1995) foundno relationships between woody debris and large-mouth bass density or abundance in Missouristreams. We found the mean WDI for areas wherelargemouth bass were collected was higher thanthe average values in both pools and shoals. Large-mouth bass may use woody debris as structurefrom which to ambush crayfish, as shelter fromcurrent, or they may simply be responding to aninnate association with this type of structure.Woody debris index scores for microhabitatswhere shoal bass were collected did not differ fromaverage WDI scores in the pools or shoals.

Generally, we found similar food habits betweenshoal bass and largemouth bass, with high (O .0.8) diet overlap. The estimate of diet overlap maybe inflated since only five prey categories (fish,

Dow

nloa

ded

by [

UQ

Lib

rary

] at

07:

25 1

3 Ju

ly 2

015

447SHOAL AND LARGEMOUTH BASS HABITAT AND DIET PARTITIONING

crayfish, insects, other invertebrates, and other)were used in the analysis. Some differences in foodhabits were apparent when food taxa were sepa-rated into smaller categories. For example, fishwere the primary prey (by weight) of age-0 large-mouth bass over 80 mm TL and age-0 shoal bassover 120 mm TL. Prior to the onset of piscivory,age-0 shoal bass primarily consumed baetid may-fly larvae. Age-0 largemouth bass rarely consumedmayflies but frequently consumed grass shrimp,which were almost absent in the diets of age-0shoal bass. The consumption of grass shrimp andthe overall diet of age-0 largemouth bass were sim-ilar to the previous observations of Davies (1981)and Schramm and Maceina (1986). Minor differ-ences in the diets of age-0 shoal bass and age-0largemouth bass may have resulted from differ-ences in prey availabilities in the microhabitatsand macrohabitats occupied by the two species.Because of low sample size (Figure 2) for somesize categories of shoal bass and largemouth bass,we were unable to analyze diet differences be-tween the pools and shoals. Nevertheless, the preyconsumed by both species were generally similarin this study.

Fish and crayfish were the primary prey of adultshoal bass and adult largemouth bass, and the prev-alence of crayfish in their diets increased with fishsize. However, this transition to a crayfish-dominated diet occurred at a smaller size for large-mouth bass than for shoal bass. Adult largemouthbass larger than 239 mm TL preyed almost entirelyon crayfish, whereas shoal bass diets were not asexclusive. Roell and Orth (1993) observed a sim-ilar trend in the diets of smallmouth bass, andSchramm and Maceina (1986) documented thisphenomenon in the diets of largemouth bass andSuwannee bass in a Florida River. This studyagrees with previous studies, showing that fish(Wright 1967; Ogilvie 1980) and crayfish (Hurst1969) are the primary food resources of shoal bass.

Similar species in sympatric situations frequent-ly exhibit resource partitioning, presumably to fa-cilitate coexistence (Hardin 1960). We found sub-stantial differences in the habitat associations ofshoal bass and largemouth bass in this study, butcomparatively few differences in food habits. Pre-vious studies have shown spatial resource parti-tioning is more common than food resource par-titioning among similar species (Schoener 1974),congenerics (reviewed in Ross 1986), and cen-trarchids (Werner and Hall 1977; George and Had-ley 1979; Janssen 1992; Scott and Angermeier1998). Our study also suggests that habitat parti-

tioning may be more important than diet differ-ences for the coexistence of shoal bass and large-mouth bass in the Chipola River, Florida.

Maintaining shoal and pool habitats, as well asa diversity of microhabitats such as rocky sub-stratum and woody debris, may be important forfacilitating the coexistence of shoal bass and large-mouth bass in streams and rivers. Shoal bass areconsistently extirpated from impounded sectionsof rivers (Ramsey 1975; Williams and Burgess1999), whereas largemouth bass persist. Given thatshoal bass can reproduce in ponds, Ramsey (1975)suggested that competition with lake-dwellingfishes eliminates shoal bass from impoundments.Our results suggest that shoal bass and largemouthbass partitioned resources spatially in the ChipolaRiver. Thus, future shoal bass conservation effortsshould focus on maintaining or enhancing streamhabitat diversity and on protecting the relativelyuncommon shoals.

Acknowledgments

We thank T. Bonvechio, M. Bledsoe, K. Dock-endorf, C. Hanson, K. Henry, J. Howard, B. Pine,and K. Tugend for assistance with field work.Comments by P. Angermeier, P. Bettoli, T. Frazer,D. Murie, W. Neal, and T. Newcomb improved thismanuscript. This manuscript represents publica-tion number R-09272 of the Florida AgriculturalExperiment Station.

References

Angermeier, P. L., and J. R. Karr. 1984. Relationshipsbetween woody debris and fish habitat in a smallwarmwater stream. Transactions of the AmericanFisheries Society 113:716–726.

Bass, D. G., and D. T. Cox. 1985. River habitat andfishery resources of Florida. Pages 122–188 in W.Seaman Jr., editor. Florida aquatic habitat and fish-ery resources. American Fisheries Society, FloridaChapter, Eustis.

Benke, A. C., R. L. Henry III, D. M. Gillespie, and R.J. Hunter. 1985. Importance of snag habitat for an-imal production in southeastern streams. Fisheries10(5):8–13.

Davies, J. H. 1981. Food habits of largemouth bass intwo coastal streams of North Carolina. Pages 346–350 in L. A. Krumholz, editor. The warmwaterstreams symposium. American Fisheries Society,Southern Division, Bethesda, Maryland.

Dimond, W. F. 1985. Device to increase efficiency ofacrylic tubes for removing stomach contents of fish.North American Journal of Fisheries Management5:214.

Dolloff, C. A., D. G. Hankin, and G. H. Reeves. 1993.Basinwide estimation of habitat and fish populations

Dow

nloa

ded

by [

UQ

Lib

rary

] at

07:

25 1

3 Ju

ly 2

015

448 WHEELER AND ALLEN

in streams. U.S. Forest Service General TechnicalReport SE-83.

Driscoll, M. P., and L. E. Miranda. 1999. Diet ecologyand yellow bass, Morone mississippiensis, in an ox-bow of the Mississippi River. Journal of FreshwaterEcology 14:477–486.

Fajen, O. 1975. Population dynamics of bass in riversand streams. Pages 195–203 in H. Clepper, editor.Black bass biology and management. Sport FishingInstitute, Washington, D.C.

George, E. L., and W. F. Hadley. 1979. Food and habitatpartitioning between rock bass (Ambloplites rupes-tris) and smallmouth bass (Micropterus dolomieu)young of year. Transactions of the American Fish-eries Society 108:253–261.

Gilbert, C. R. 1992. Rare and endangered biota of Flor-ida, volume 2. University of Press of Florida,Gainesville.

Gotelli, N. J., and G. L. Entsminger. 1997. EcoSim: nullmodels software for ecology. Quarterly Review ofBiology 50:237–266.

Hamilton, J. G., and P. M. Powles. 1983. Fish predationand other distinctive features in the diet of NogiesCreek, Ontario, largemouth bass, Micropterus sal-moides. Canadian Field-Naturalist 97:47–56.

Hardin, G. 1960. The competitive exclusion principle.Science 131:1292–1297.

Heidinger, R. C. 1975. Life history of the largemouthbass. Pages 11–20 in H. Clepper, editor. Black bassbiology and management. Sport Fishing Institute,Washington, D.C.

Hubert, W. A. 1977. Comparative food habits of small-mouth and largemouth basses in Pickwick Reser-voir. Journal of the Alabama Academy of Science48:167–178.

Hurst, H. N. 1969. Comparative life history of the red-eye bass, Micropterus coosae Hubbs and Bailey, andthe spotted bass, Micropterus p. punctulatus (Raf-inesque), in Halawakee Creek, Alabama. Master’sthesis. Auburn University, Auburn, Alabama.

Janssen, F. W. 1992. Ecology of three species of blackbass in the shoals reach of the Tennessee River andPickwick Reservoir, Alabama. Master’s thesis. Au-burn University, Auburn, Alabama.

Jenkins, R. E., and N. M. Burkhead. 1993. Freshwaterfishes of Virginia. American Fisheries Society, Be-thesda, Maryland.

Lehtinen, R. M., N. D. Mundahl, and J. C. Madejczyk.1997. Autumn use of woody snags by fishes inbackwater and channel border habitats of a largeriver. Environmental Biology of Fishes 49:7–19.

Leonard, P. M., and D. J. Orth. 1988. Use of habitatguilds of fishes to determine instream flow require-ments. North American Journal of Fisheries Man-agement 8:399–409.

Livingstone, A. C., and C. F. Rabeni. 1991. Food-habitatrelations of underyearling smallmouth bass in anOzark stream. Pages 76–83 in D. C. Jackson, editor.The first international smallmouth bass symposium.Mississippi State University, Mississippi Agricul-tural and Forestry Experiment Station, Starkville.

Lobb, D. M., III, and D. J. Orth. 1991. Habitat use by

an assemblage of fish in a large warmwater stream.Transactions of the American Fisheries Society 120:65–78.

Miller, R. J. 1975. Comparative behavior of centrarchidbasses. Pages 85–94 in H. Clepper, editor. Blackbass biology and management. Sport Fishing Insti-tute, Washington, D.C.

Oates, D. W., L. M. Krings, and K. L. Ditz. 1993. Fieldmanual for the identification of selected NorthAmerican freshwater fish by fillets and scales. Ne-braska Game and Parks Commission, NebraskaTechnical Series 19, Lincoln.

Ogilvie, V. E. 1980. Shoal bass investigation. FloridaGame and Freshwater Fish Commission, Endan-gered Wildlife Project E-1, Study I-L, CompletionReport, Tallahassee.

Parsons, J. W., and E. Crittenden. 1959. Growth of theredeye bass in the Chipola River, Florida. Trans-actions of the American Fisheries Society 88:191–192.

Pianka, E. R. 1973. The structure of lizard communities.Annual Review of Ecology and Systematics 4:53–74.

Ramsey, J. S. 1975. Taxonomic history and systematicrelationships among species of Micropterus. Pages67–75 in H. Clepper, editor. Black bass biology andmanagement. Sport Fishing Institute, Washington,D.C.

Ramsey, J. S. 1976. Freshwater fishes. Pages 53–65 inH. T. Boschung, Jr., editor. Endangered and threat-ened plants and animals of Alabama. Bulletin of theAlabama Museum of Natural History 2.

Roell, M. J., and D. J. Orth. 1993. Trophic basis ofproduction of stream-dwelling smallmouth bass,rock bass and flathead catfish in relation to inver-tebrate bait harvest. Transactions of the AmericanFisheries Society 122:46–62.

Ross, S. T. 1986. Resource partitioning in fish assem-blages: a review of field studies. Copeia 1986:352–388.

Sammons, S. M., and P. W. Bettoli. 1999. Spatial andtemporal variation in electrofishing catch rates ofthree species of black bass (Micropterus spp.) fromNormandy Reservoir, Tennessee. North AmericanJournal of Fisheries Management 19:454–461.

SAS Institute. 1994. SAS/STAT user’s guide, version4, 4th edition. SAS Institute, Cary, North Carolina.

Scalet, C. G. 1977. Summer food habits of sympatricstream populations of spotted bass, Micropteruspunctulatus, and largemouth bass, M. salmoides,(Osteichthyes: Centrarchidae). Southwestern Nat-uralist 21:493–501.

Schoener, T. W. 1974. Resource partitioning in ecolog-ical communities. Science 185:27–39.

Schramm, H. J., and M. J. Maceina. 1986. Distributionand diet of Suwannee bass and largemouth bass inthe lower Santa Fe River, Florida. EnvironmentalBiology of Fishes 15:221–228.

Scott, M. C., and P. L. Angermeier. 1998. Resource useby two black bass in impounded and riverine sec-tions of the New River, Virginia. North AmericanJournal of Fisheries Management 18:221–235.

Dow

nloa

ded

by [

UQ

Lib

rary

] at

07:

25 1

3 Ju

ly 2

015

449SHOAL AND LARGEMOUTH BASS HABITAT AND DIET PARTITIONING

Smitherman, R. O., and J. S. Ramsey. 1972. Observa-tions on spawning and growth of four species ofbasses (Micropterus) in ponds. Proceedings of theAnnual Conference Southeastern Association ofFish and Wildlife Agencies 25(1971):357–365.

Sowa, S. P., and C. F. Rabeni. 1995. Regional evaluationof the relation of habitat to distribution and abun-dance of smallmouth bass and largemouth bass inMissouri streams. Transactions of the AmericanFisheries Society 124:240–251.

Stokes, M. E., S. C. Davis, and G. G. Koch. 2000. Cat-egorical data analysis using the SAS system, 2ndedition. SAS Institute, Cary, North Carolina.

Todd, B. L., and C. F. Rabeni. 1989. Movement andhabitat use by stream-dwelling smallmouth bass.Transactions of the American Fisheries Society 118:229–242.

Trautman, M. B. 1957. The fishes of Ohio with illus-trated keys. Ohio State University Press, Columbus.

Van Den Avyle, M. J., and J. E. Roussel. 1980. Eval-uation of a simple method for removing food itemsfrom live black bass. Progressive Fish-Culturist 42:222–223.

Warden, R. L., and W. A. Hubert. 1980. Comparative

life history of young-of-year smallmouth and large-mouth bass in Pickwick Reservoir. Tennessee Acad-emy of Science 55(2):58–106.

Werner, E. E., and D. J. Hall. 1977. Competition andhabitat shift in two sunfishes. Ecology 58:869–876.

Williams, J. D., and G. H. Burgess. 1999. A new speciesof bass, Micropterus cataractae (Teleostei: Cen-trarchidae), from the Apalachicola River basin inAlabama, Florida, and Georgia. Bulletin of the Flor-ida Museum of Natural History 42:80–114.

Winger, P. V., D. P. Shultz, and W. W. Johnson. 1987.Contamination from battery salvage operations onthe Chipola River, Florida. Proceedings of the An-nual Conference Southeastern Association of Fishand Wildlife Agencies 39(1985):139–145.

Wright, S. E., IV. 1967. Life history and taxonomy ofthe Flint River redeye bass (Micropterus coosae,Hubbs and Bailey). Master’s thesis. University ofGeorgia, Athens.

Wydoski, R. S., and R. R. Whitney. 1979. Inland fishesof Washington. University of Washington Press,Seattle.

Zar, J. H. 1988. Biostatistical analysis, 2nd edition.Prentice-Hall, Englewood Cliffs, New Jersey.

Dow

nloa

ded

by [

UQ

Lib

rary

] at

07:

25 1

3 Ju

ly 2

015