North American Journal of Fisheries Management 25:1191–1201, 2005

q Copyright by the American Fisheries Society 2005

[Article]
DOI: 10.1577/M03-251.1

Population Abundance and Stock Characteristics of Flathead

Catfish in the Lower St. Joseph River, Michigan
DANIEL J. DAUGHERTY AND TRENT M. SUTTON*
Department of Forestry and Natural Resources, Purdue University,
195 Marsteller Street, West Lafayette, Indiana 47907-1159, USA

Abstract.—Little information exists regarding the biological attributes and stock dynamics of
flathead catfish Pylodictis olivaris in lotic systems throughout the northern United States. We
examined the population abundance, annual survival, growth, condition, size structure, and age
structure of flathead catfish in the lower St. Joseph River, Michigan, to direct future management
efforts in this system and increase our knowledge of northern flathead catfish stocks. Fish were
collected by means of electrofishing during June through September 2002 and 2003. Population
abundance was estimated at 5,452 individuals (range, 3,985–7,277 fish), and annual survival was
estimated at 67%. Analysis of pectoral spine cross sections revealed that growth was greatest for
fish younger than age 6 (range, 83–98 mm/year) and decreased among older age-classes (range,
18–70 mm/year). Relative weight was greatest for individuals less than 300 mm total length (TL;
115%) and declined with increasing fish length. The size structure and age structure of flathead
catfish were dominated by fish less than 400 mm TL and younger than age 4 (89% and 79%,
respectively), although individuals greater than 1,100 mm TL and up to age 17 were present in
the population. Relative stock density estimates indicated that 58% of fish greater than the minimum
stock size were of quality length, while flathead catfish of preferred, memorable, and trophy sizes
represented 28, 7, and 1% of fish collected, respectively. Despite their presence on the northern
fringe of the species’ geographic distribution, the flathead catfish in the lower St. Joseph River
exhibited biological characteristics and stock dynamics that were similar to those reported for
other lightly exploited stocks throughout the midwestern United States.

Flathead catfish Pylodictis olivaris are an im- catfish (Stauffer et al. 1996; Vokoun and Rabeni
portant component of the North American catfish 1999).
fishery (Jackson 1999). This species is native to Although a number of studies have examined
the Mississippi River, Mobile River, and Rio the population characteristics of flathead catfish in
Grande River drainages, the Laurentian Great lotic environments, most have focused on size
Lakes basin, and northeastern Mexico (Hubbs and structure, age distribution, and growth in systems
Lagler 1947; Lee and Terrell 1987; Jackson 1999) throughout the southern United States (Purkett
and supports recreational and commercial fisheries 1958; Minckley and Deacon 1959; Mayhew 1969;
throughout much of its geographic distribution Guier et al. 1984; Pisano et al. 1983; Young and
(Moss and Tucker 1989; Grussing et al. 2001). In Marsh 1990; Grussing et al. 2001). Purkett (1958)
many states, flathead catfish offer anglers an op- reported that flathead catfish grew an average of
portunity to catch fish by means other than hook 76 mm/year in the Salt River, Missouri, and
and line (e.g., handgrabbing, bankpoling, etc.), reached a mean total length (TL) of 300 mm by
while giving commercial fishers a high-quality age 4. Flathead catfish grew more slowly in the
product preferred by many consumers (Quinn Salt River system than in the Mississippi River
1993; Grussing et al. 2001; Jackson 1999). Despite (Illinois–Iowa and Missouri–Iowa), where fish
the popularity of this species, knowledge regarding reached 381 and 432 mm TL, respectively, at the
the population characteristics of flathead catfish age 4 (Barnickol and Starrett 1951). Similarly,
stocks is limited. This lack of information is most Minckley and Deacon (1959) reported that flathead
prevalent in moderate-size rivers, streams, and catfish reached 483 mm TL by age 4 in the Big
creeks, where fisheries managers have found as- Blue River, Kansas. However, Guier et al. (1984)
sessment of catfish populations difficult due to the reported faster growth in the Cape Fear River,
solitary nature and low population abundances of North Carolina, where individuals reached a mean
TL of 580 mm at the same age. The results of these
studies indicate that growth of flathead catfish is
* Corresponding author: [email protected] system specific and that the length of the growing
Received December 22, 2003; accepted January 12, 2005 season may not be a reliable predictor of growth
Published online August 2, 2005 rate.

1191
1192 DAUGHERTY AND SUTTON

Whereas southern stocks of flathead catfish have head catfish in the St. Joseph River. However, the
received attention from fisheries managers, little current status of this fishery and the biological
information exists regarding the characteristics of attributes of the population remain unknown. The
populations in the upper Midwest and Great Lakes objective of this study was to provide information
regions. Historically, northern populations have regarding the population characteristics and stock
experienced lower exploitation rates and less an- dynamics of flathead catfish in the St. Joseph River
gling pressure than those in the south (Topp et al. system. Such information can be used to direct
1994; Stauffer et al. 1996). These fishery char- future management efforts and increase our knowl-
acteristics may result in differences in abundance, edge of flathead catfish populations in the northern
growth, condition, age structure, and size structure United States.
for flathead catfish stocks between northern and
southern systems. Stauffer et al. (1996) found that Study Site
growth rates of flathead catfish in the Minnesota The 338-km main-stem St. Joseph River is
River, Minnesota, were 13–52% lower than those joined by 2,641 km of tributaries, discharging an
of fish in the southern United States. The authors average of 130 m3/s from a watershed that drains
also noted that a high abundance of large fish was approximately 7,770 km2 in Michigan and 4,364
present in the population: 20% of flathead catfish km2 in Indiana. The free-flowing (hereafter re-
sampled were greater than 910 mm TL. These re- ferred to as lower) section of the St. Joseph River
sults suggest that differences in population char- is a 37.6-km reach between the Berrien Springs
acteristics exist between northern and southern Dam and the mouth of the river at Lake Michigan
flathead catfish stocks, supporting the need for ad- (Figure 1). Because flathead catfish are not found
ditional studies on fish in northern systems. above the Berrien Springs Dam (J. Dexter, Mich-
The popularity of recreational and commercial igan Department of Natural Resources, personal
fisheries for flathead catfish has increased sub- communication), the boundaries of this river reach
stantially throughout their geographic range in re- served as the upstream and downstream limits of
cent years (Summers 1986; Gilliland 1988; Quinn the sampling area. We divided the lower St. Joseph
1993; Cunningham 1995; Stauffer et al. 1996; River into four 9.4-km sampling sections to fa-
Jackson 1999). This trend is especially evident in cilitate collection of flathead catfish (Figure 1).
the northern United States, where flathead catfish Sampling sections 1 and 2 are dominated by gravel
have historically attracted little angler interest substrates and have a low to fair gradient (0–0.9
(Topp et al. 1994; Stauffer et al. 1996). Because m/km), whereas sections 3 and 4 are dominated
flathead catfish may be susceptible to overharvest by sand substrates and have a low gradient (0–0.5
due to such factors as their low abundance and m/km). Riparian land cover in the reach is dom-
aggressive behavior, knowledge of the life history inated by hardwood forests, although urban de-
characteristics and population status of this species velopment dominates section 4 as the river ap-
is important for the implementation of appropriate proaches Lake Michigan.
management strategies (Quinn 1993; Grussing et The lower St. Joseph River supports a resident
al. 2001; Jackson 1999). Without this information, warmwater fish community; the most abundant
management of flathead catfish populations would species are flathead catfish, channel catfish Ictal-
be difficult, particularly as angling pressure in- urus punctatus, freshwater drum Aplodinotus grun-
creases (Stauffer et al. 1996). niens, common carp Cyprinus carpio, and gizzard
The flathead catfish population in the St. Joseph shad Dorosoma cepedianum (Wesley and Duffy
River, a Lake Michigan tributary located in south- 1999). Mean channel width in the lower St. Joseph
western Michigan and northeastern Indiana, has River is 130 m (range, 50–250 m), and mean water
experienced an increase in angling pressure over depth is 1.6 m (range, 0.3–8.2 m), although lateral-
the past decade (J. Dexter, Michigan Department scour pools with water depths in excess of 7 m
of Natural Resources, personal communication). occur throughout the reach. Midsummer (June–
Prior to 1996, an average of 14 flathead catfish August) water temperature can be as high as 288C;
caught from the St. Joseph River each year was mean dissolved oxygen is 9.6 mg/L (range, 6.9–
entered into the annual state master angler award 14.1 mg/L) and turbidity is 16.5 nephelometric
program. Over the past 7 years, the average num- turbidity units (NTU; range, 11–27 NTU). In-
ber of entries submitted each year has decreased stream habitat is composed primarily of woody
by 50%. This decline in the catch of trophy fish debris and rip-rap; few aquatic macrophytes exist
suggests an increase in the exploitation rate of flat- in the system.
 ST. JOSEPH RIVER FLATHEAD CATFISH 1193

FIGURE 1.—Map of the lower, free-flowing reach of the St. Joseph River, Michigan. Lines bisecting the river
indicate the upper and lower boundaries of the four sections used for the sampling of flathead catfish.

Methods cm wide) were deployed approximately 4–5 m up-

Flathead catfish were collected during June stream and downstream of the boat. Electrical cur-
through September 2002 and 2003. We used the rent was applied continuously for 90 s at each
modified predator approach defined by Vokoun and sampling location. During August 2003, flathead
Rabeni (1999). Fish were sampled once per week catfish were also sampled from section 4 by means
from each section of the study area by use of 24– of low-frequency (20% pulse width), low-pulse
38-V AC produced by a three-bar magnetic motor, (7.5 pulse/s) DC boat electrofishing, as described
as described by Morris and Novak (1968). Al- by Stauffer and Koenen (1999). A chase boat was
though sampling efforts were conducted in all hab- used during all sampling periods in 2003 to capture
itat types (i.e., main-channel pools, riffles, and flathead catfish that surfaced downstream of the
runs with and without structure), efforts were con- boat carrying the electrofishing unit.
centrated in structural habitats typically selected Population characteristics.—All captured flat-
by flathead catfish (e.g., large woody debris jams, head catfish were measured for TL (nearest 1 mm)
timbered channels, undercut banks, rip-rap, etc.; and wet weight (nearest 1 g). The spine of the left
Cunningham 2000). At each sampling location, pectoral fin was removed and dried for subsequent
two 18-gauge insulated wires (each 6.1 m in age and growth analyses in the laboratory. Ab-
length) with the distal ends connected to solute population abundance of flathead catfish
aluminum-bar electrodes (30.5 cm long and 2.5 was calculated by use of the Schnabel estimator
1194 DAUGHERTY AND SUTTON

for closed populations (Schnabel 1938). Because when fish body length equals 0 mm (to) for the
assessments of flathead catfish movement in the von Bertalanffy growth model.
lower St. Joseph River indicated that individuals Fish condition was calculated as relative weight
remained within the study reach during the entire (Wr). Length-specific standard weights (Ws) for
study period (Daugherty and Sutton 2005), we as- flathead catfish were derived from the standard-
sumed the population to be geographically closed. weight equation, log10(Ws) 5 25.542 1 3.23log10(TL)
Ninety-five percent confidence intervals (CIs) (Bister et al. 2000). Relative weight was not cal-
were calculated with the Poisson distribution by culated for flathead catfish less than 130 mm in
substituting the confidence limits for the number length because the standard-weight equation ex-
of marked individuals at large in the population cluded smaller individuals due to variance-to-
(R) into the Schnabel estimator (Van Den Avyle mean errors greater than 0.02 (Bister et al. 2000).
1993). Flathead catfish density (i.e., the number of Relative stock density (RSD) indices (e.g., quality
individuals per river kilometer [rkm]) was calcu- [Q], preferred, memorable, and trophy) were cal-
lated by dividing the estimate of absolute abun- culated based on length categories developed by
dance by the length of the study reach. Bister et al. (2000). Minimum TLs for each cat-
In the laboratory, pectoral spines were sectioned egory were as follows: 350 mm for stock length,
by use of a rotary tool fitted with a diamond-coated 510 mm for quality length, 710 mm for preferred
cutting disk. To avoid potential fish age underes- length, 860 mm for memorable length, and 1,020
timation due to erosion of the central lumen (Turn- mm for trophy length.
er 1982; Nash and Irwin 1999), multiple cross sec- Due to variable capture rates of flathead catfish
tions were cut between the distal end of the basal ages 5–12, annual survival was estimated based
recess and the proximal end of the spine dentations on the Robson and Chapman (1961) method.
(Layher 1981; Crumpton et al. 1987). Spine sec- Ninety-five percent CIs for these estimates were
calculated by use of the methods described by
tions were then visually examined and discarded
Ricker (1975). The estimate of annual survival ob-
until enlargement of the central lumen was no lon-
tained from the Robson and Chapman (1961)
ger obvious (Munger et al. 1994). Cross sections
method was used to calculate the instantaneous
of each spine were mounted on microscope slides
total mortality rate Z. Confidence intervals for Z
and viewed with transmitted light at 36–1353
were calculated by substituting the confidence lim-
magnification under a zoom stereomicroscope fit-
its of the annual survival rate into the calculation
ted with a digital camera. Images of each spine
of Z. Age-classes younger than the first age-class
were analyzed with IPLab 3.6 image analysis soft-
fully recruited to the sampling gear, as well as age-
ware (Scanalytics, Fairfax, Virginia), which was
classes with limited sample sizes (N , 5), were
used to determine the number of annuli present,
omitted from the estimation of annual survival
the length of the spine radius, and the radial dis- (Ricker 1975; Van Den Avyle 1993).
tance to each annulus for back-calculation of Data analysis.—Piecewise linear regression was
length at age. All measurements were recorded to used to determine the relationship between mean
the nearest 0.01 mm. Each spine image was ana- annual growth increment and fish age. Simple lin-
lyzed independently by two readers, and disagree- ear regression was employed to determine the re-
ments between readers were reconciled with a sub- lationship between relative condition and fish TL.
sequent concert read. Mean TL of flathead catfish in the lower St. Joseph
Length at age for each flathead catfish was back- River was regressed as a function of fish age, and
calculated by means of a modification of the Fras- analysis of covariance (ANCOVA) was used to
er–Lee method (DeVries and Frie 1996). Back- compare this regression with growth relationships
calculated length-at-age estimates were used to de- for nine other flathead catfish populations. Signif-
termine the age at which flathead catfish were re- icant differences among regression coefficients
cruited to the harvestable fishery (minimum length were determined by the Tukey’s honestly signifi-
limit, 381 mm TL) and to calculate age-specific cant difference test. All statistical analyses em-
annual growth rates (mm/year). A von Bertalanffy ployed a significance level a of 0.05.
growth model was constructed for the population
in FASTq 2.0 (Auburn University, Auburn, Ala- Results
bama). Mean TL and age-at-capture data were used A total of 653 flathead catfish were collected
to estimate the theoretical maximum fish length from 759 sites on 32 sampling occasions during
(L`), Brody growth coefficient (K), and time period June through September 2002 and 2003. One-
 ST. JOSEPH RIVER FLATHEAD CATFISH 1195

FIGURE 2.—Length-frequency distribution of flathead FIGURE 3.—Age-frequency distribution of flathead

catfish collected from the lower St. Joseph River, Mich- catfish collected from the lower St. Joseph River, Mich-
igan, during 2002 and 2003. igan, during 2002 and 2003.

hundred forty-six flathead catfish were collected of flathead catfish ranged from 1 to 17 years (Fig-
by low-frequency electrofishing during 15 d of ure 3). Age-1 flathead catfish represented 22% of
sampling in 2002. In 2003, 447 individuals were the fish sampled, age-2 fish represented 36%, age-
collected during 17 sampling occasions. An ad- 3 fish represented 15%, and age-4 fish represented
ditional 60 fish were collected during August 2003 6%; these data indicate that the fish did not become
by means of DC boat electrofishing. fully vulnerable to the sampling gear until age 2.
Six percent (N 5 40) of the flathead catfish Capture rates of age-5–12 flathead catfish varied
tagged during the study period were recaptured among age-classes (Figure 3). Age-9 flathead cat-
during subsequent sampling occasions. The ab- fish composed the greatest proportion (4%) of fish
solute abundance of flathead catfish in the lower captured within this age range. Fish older than age
St. Joseph River was estimated at 5,452 individ- 12 made up 2% of the population (Figure 3).
uals (95% CI 5 3,985–7,277). The density of flat- Length-at-age estimates for flathead catfish in
head catfish was estimated at 145 fish/rkm. Annual the lower St. Joseph River indicated that flathead
survival of flathead catfish in the lower St. Joseph catfish reached a mean TL of 102 mm by age 1
River was 0.672 (95% CI 5 0.670–0.673). The Z and were recruited to the harvestable fishery by
derived from this approach was 0.397 (95% CI 5 age 4, when the mean TL was 401 mm (Figure 4).
0.396–0.400). The mean length at age of fish in the oldest age-
Flathead catfish captured during the study class (age 17) was 1,016 mm TL. Based on the
ranged from 87 to 1,132 mm TL (Figure 2). In- von Bertalanffy growth equation, the predicted age
dividuals less than 100 mm TL represented 1% of at which flathead catfish were recruited to the har-
fish sampled, whereas flathead catfish ranging vestable fishery was 4 years, while the maximum
from 100 to 400 mm TL made up the greatest attainable TL of flathead catfish in the St. Joseph
proportion of fish collected (89%). Capture rates River was estimated at 1,174 mm (Figure 4).
of flathead catfish greater than 400 mm TL varied Mean annual growth increments of flathead cat-
among length-classes (Figure 2). Fish ranging fish were greatest among fish younger than age 6
from 500 to 600 mm TL composed 3% of the catch, (range, 83–98 mm/year). The slope of the linear
whereas fish that were 700–800 mm TL repre- regression relationship over this age range indi-
sented 7% of fish sampled. Flathead catfish 900 cated that annual growth declined by less than 5
mm TL and larger made up 5% of the population. mm/year during this period (Figure 5). The great-
Pectoral spine cross sections were examined for est declines in the annual growth rate of flathead
94% (N 5 612) of the collected fish. Age and catfish in the lower St. Joseph River were observed
growth analyses were omitted for 41 individuals between ages 6 and 8 (range, 42–70 mm/year).
either because of missing or broken pectoral spines Piecewise regression of these age-classes indicated
or because of pectoral spine size. Pectoral spines that annual growth declined by approximately 14
collected from individuals less than 130 mm TL mm/year. Annual growth rates declined little
were too small to accurately cross-section. The age among fish older than age 8; however, fish growth
1196 DAUGHERTY AND SUTTON

FIGURE 4.—Mean length at age and fitted von Ber- FIGURE 6.—Relationship between mean relative
talanffy growth equation (t 5 age in years; lt 5 length weight (Wr) and total length (TL) for flathead catfish
at time t) for flathead catfish collected from the lower collected from the lower St. Joseph River, Michigan,
St. Joseph River, Michigan, during 2002 and 2003. Error during 2002 and 2003. Individuals were pooled into 100-
bars represent 95% confidence intervals about the mean mm length groups. Error bars represent 95% confidence
length at age. Confidence intervals were not calculated intervals. Confidence intervals were not calculated for
for age 16 because only one fish was collected. the 1,100-mm size-class because only one fish was col-
lected.
was less than 30 mm/year for these older age-
classes.
greater than the minimum stock size (350 mm TL)
An inverse relationship was observed between
were quality length ($510 mm TL; Table 1). Flat-
Wr and TL of flathead catfish in the lower St. Jo-
head catfish in the preferred-size range ($710 mm
seph River (r2 5 0.75; P , 0.001; Figure 6). Fish
TL) composed 28% of the stock, while memorable-
condition was similar for individuals less than 300
size individuals ($860 mm TL) made up 7% of
mm TL (Wr 5 115%), whereas condition varied
fish collected. Trophy fish ($1,020 mm TL) rep-
among 300–700-mm fish. Mean Wr of 300–399-
resented 1% of fish captured during the study pe-
mm fish was 98%, while the mean Wr for 400–
riod.
700-mm fish was 113%. Relative weight was low-
est for flathead catfish greater than 700 mm TL Discussion
(mean, 88%).
Unexploited fish stocks are characterized by
The RSD estimates indicated that 58% of fish
high population abundance, a low rate of annual
mortality, a broad range of fish age-classes and
length-classes, and decreased annual growth (Cla-
dy et al. 1975; Goedde and Coble 1981). In ad-
dition, the presence of large fish with reduced con-
dition further suggests little or no exploitation due

TABLE 1.—Comparative relative stock density (RSD)

indices for flathead catfish in the lower St. Joseph River
(SJ), Michigan, the Flint River (FL), Georgia, and the
Cape Fear River (CF), North Carolina.

SJ FL CF
Index Minimum frequency frequency frequency
category TL (mm) (%)a (%)b (%)c
Quality 510 58 57 49
Preferred 710 28 28 21
FIGURE 5.—Relationship between age (X) and calcu- Memorable 860 7 14 9
lated mean annual growth increment (Y) for each age- Trophy 1,020 1 5 1
class (i) of flathead catfish collected from the lower St. a This study.
Joseph River, Michigan, during 2002 and 2003. Error b Quinn (1988).
bars represent 95% confidence intervals. c Ashley and Buff (1986).
 ST. JOSEPH RIVER FLATHEAD CATFISH 1197

to increased competition for suitable prey resourc- Wyoming; high survival was attributed to a low
es and the increased occurrence of senescence rate of exploitation. The similarity in annual sur-
(Van Den Avyle 1993). Previous studies have char- vival rate between the lower St. Joseph River and
acterized the flathead catfish as a long-lived spe- these previously studied systems further suggests
cies ($15 years) that exhibits relatively fast relatively low exploitation of flathead catfish in
growth (approximately 100 mm/year; Barnickol the St. Joseph River.
and Starrett 1951; Layher and Boles 1979; Guier Because flathead catfish are long-lived and have
et al. 1984; Jackson 1999; Nash and Irwin 1999). relatively low fecundity (approximately 6,900–
In addition, flathead catfish are known to reach 11,300 eggs/female), the size structure and age
TLs in excess of 900 mm in systems with low rates structure of populations of this species may be
of exploitation (Hesse 1994; Stauffer et al. 1996). altered by excessive harvest (Jenkins and Burk-
Our study results suggest that the population char- head 1994; Stauffer et al. 1996; Jackson 1999).
acteristics and stock dynamics of flathead catfish Overexploited populations of flathead catfish have
in the lower St. Joseph River are similar to those been characterized by the presence of large num-
of other midwestern stocks that receive low ex- bers of small fish (#400 mm TL) and the absence
ploitation rates. of larger individuals. In contrast, populations that
Absolute population abundance estimates of receive little harvest pressure possess a more uni-
flathead catfish in lotic systems have not been re- form length-frequency distribution over a greater
ported; therefore, we could not make direct com- range of fish length-classes. Hesse (1994) reported
parisons with other studies. However, the esti- that the percentage of flathead catfish greater than
mated density (145 fish/rkm) of flathead catfish in the legal length limit (457 mm TL) in the highly
the lower St. Joseph River was comparable to den- exploited Missouri River was less than 5% be-
sities reported for other systems. Marsh et al. tween 1974 and 1993. In contrast, greater than
(1988) estimated that the density of flathead catfish 25% of flathead catfish collected in the lightly ex-
ranged from 155 to 259 fish/km in the lower Col- ploited Flint River, Georgia, were greater than this
orado River, Arizona. Similarly, flathead catfish length (Quinn 1989). Flathead catfish greater than
density estimates ranged from 159 to 249 fish/km 500 mm TL represented 47% of fish sampled in
in the Missouri River, Nebraska (Tondreau 1988). the Minnesota River (Stauffer et al. 1996), where
Therefore, our estimate of flathead catfish density exploitation of flathead catfish has been estimated
demonstrates that the lower St. Joseph River sup- at less than 1% (Leitch and Baltezore 1987). Al-
ports fish numbers similar to those seen in other though a large proportion (89%) of the flathead
riverine systems. catfish collected in the lower St. Joseph River were
Survival rates for flathead catfish have not been less than 400 mm, the relative frequency of fish
quantified in lotic systems because of the limited greater than this length and the presence of fish as
data availability and an overreliance on length- and old as age 17 suggest high survival and a low rate
age-frequency distributions to qualitatively de- of exploitation in this system. However, the sharp
scribe fish survival. For example, Stauffer et al. decline in abundance between ages 3 and 4, which
(1996) attempted to estimate survival and mortal- corresponds to the age at which most fish were
ity rates for flathead catfish and found that the recruited to the harvestable fishery ($381 mm
number of captured fish from each age-class did TL), suggests that flathead catfish do experience
not decline with increasing age. As a result, these some harvest pressure in the lower St. Joseph Riv-
authors concluded that flathead catfish survival er.
was high. Although estimates of flathead catfish The RSD estimates of flathead catfish in the low-
survival are not available, the annual survival of er St. Joseph River were within the range reported
fish in the lower St. Joseph River (67%) was in for other populations of this species (Table 1).
agreement with estimates for lightly exploited Quinn (1989) reported the RSD-Q of flathead cat-
populations of channel catfish. For example, Ku- fish in the Flint River fishery to be 57%, whereas
beny (1992) determined that the annual survival fish in the Cape Fear River had a RSD-Q of 49%
of channel catfish was 74% in the James River, (Ashley and Buff 1986). The proportions of pre-
South Dakota, whereas annual survival rates as ferred, memorable, and trophy sizes of fish in the
high as 87% have been reported for some Iowa lower St. Joseph River were also similar to pro-
streams (Paragamian 1990). Similarly, Gerhardt portions reported for these stocks. In other lotic
and Hubert (1991) estimated the annual survival systems receiving low rates of exploitation, trophy
of channel catfish to be 77% in the Powder River, fish have been found to make up less than 5% of
1198 DAUGHERTY AND SUTTON

TABLE 2.—Comparative regression relationships based on mean length-at-age data, mean annual regional surface air
temperature (MAST), and geographic location for flathead catfish populations in 10 U.S. rivers. The MAST data were
based on 1971–1999 averages published by the National Oceanic and Atmospheric Administration. Regression equations
with common letters were not significantly different (ANCOVA: F 5 3.59; df 5 9, 69; P 5 0.001). Fish older than
age 12 were not included in the analysis because of small sample sizes (N , 5 fish) or limited availability among
populations.

MAST
River Coordinates (8C) Regression r2 N P
Minnesota River (Minnesota)a 448N, 948W 7.5 TL 5 111.7 1 69.4 (Age) z 0.96 12 ,0.0001
St. Joseph River (Michigan)b 428N, 868W 7.5 TL 5 69.2 1 73 (Age) z 0.96 12 ,0.0001
Mississippi River (Iowa–Illinois)c 418N, 908W 10 TL 5 98.9 1 71.7 (Age) z 0.99 11 ,0.0001
Salt River (Missouri)d 408N, 928W 12.5 TL 5 34.6 1 61.7 (Age) z 0.99 9 ,0.0001
Missouri River (Nebraska)e 408N, 958W 10 TL 5 43.4 1 80.5 (Age) z 0.93 7 0.0004
Blue River (Kansas)f 398N, 968W 12.5 TL 5 46.0 1 108.4 (Age) y 0.99 7 ,0.0001
Flint River (Georgia)g 338N, 838W 17.5 TL 5 147.6 1 105.3 (Age) y 0.97 7 ,0.0001
Cape Fear River (North Carolina)h 358N, 798W 15 TL 5 117.7 1 101.5 (Age) y 0.97 8 ,0.0001
Colorado River (Arizona–California)i 328N, 1148W 22.5 TL 5 6.6 1 112.9 (Age) y 0.99 9 ,0.0001
Rio Grande River (Texas)j 308N, 1018W 20 TL 5 272.3 1 107.9 (Age) y 0.99 7 ,0.0001
a Stauffer et al. (1996).
b This study.
c Barnickol and Starrett (1951).
d Purkett (1958).
e Holz (1969).
f Minckley and Deacon (1959).
g Quinn (1988).
h Guier et al. (1981).
i Young and Marsh (1990).
j Pate (1980).

flathead catfish populations (Hesse et al. 1978; ment of age-1–4 flathead catfish was 68 mm/year,
Pugibet and Jackson 1991; Insaurralde 1992). while the growth of fish older than age 4 decreased
These studies corroborate our results, indicating by 26%. Stauffer et al. (1996) estimated that the
that the size structure of flathead catfish in the growth rate of age-1–5 flathead catfish in the Min-
lower St. Joseph River provides further evidence nesota River averaged 120 mm/year. In contrast,
of low rates of exploitation. the mean annual growth increment for fish aged
Length-at-age data for flathead catfish in the 6–10 in this system decreased to 50 mm/year,
lower St. Joseph River were within the range of whereas the growth of fish older than age 10 av-
growth estimates reported for other midwestern eraged 34 mm/year. Mayhew (1969) reported that
populations of this species. However, fish growth annual growth increments were greater than 120
was slower than that reported for populations mm/year for fish younger than age 3, whereas the
throughout the southern United States (Table 2). growth of older age-classes declined to less than
The growth of fish in our system most closely re- 40 mm/year by age 7. Previous studies have de-
sembled the growth of flathead catfish in the Mis- termined that flathead catfish become reproduc-
sissippi River, Iowa–Illinois, and the Minnesota tively mature between ages 3 and 5 (Barnickol and
River, whereas the average growth rate of flathead Starrett 1951; Minckley and Deacon 1959; Turner
catfish in southern systems was 35% greater than and Summerfelt 1971). Although the onset of re-
that of St. Joseph River fish (Table 2). Similar productive maturity in flathead catfish may con-
latitudinal trends in growth have been reported for tribute to declines in the annual growth rate, the
other fish species (Power and McKinley 1997; variation among populations suggests that addi-
Braaten and Guy 2002). Our results suggest that tional factors (e.g., population density, prey avail-
the thermal gradient and length of the growing ability, etc.) may also affect annual growth on a
season associated with geographic location are re- system-specific basis.
liable indicators of flathead catfish growth rates. The relationship between Wr and TL of flathead
The declines in annual growth rates observed catfish in the lower St. Joseph River contrasted
for flathead catfish in the lower St. Joseph River with results reported for other populations of flat-
have been observed for other flathead catfish head catfish. Guier et al. (1984) and Lemmons
stocks throughout the species’ range. Purkett (1995) found that the condition of flathead catfish
(1958) found that the mean annual growth incre- increased with body length. Although food habits
 ST. JOSEPH RIVER FLATHEAD CATFISH 1199

and relative abundance of prey types for flathead Michigan Department of Natural Resources, Fish-
catfish were not examined in our study, the decline eries Division, provided assistance with project
in Wr with increasing length may indicate a lack development and sampling equipment. Construc-
of adequate prey sizes for larger fish. Adult flat- tive comments on earlier drafts of this manuscript
head catfish are known to be highly piscivorous by A. Benson, M. Hansen, P. Hrodey, R. Swihart,
(Minckley and Deacon 1959; Turner and Sum- H. Weeks, and two anonymous reviewers im-
merfelt 1971; Jackson 1999); forage fish make up proved this manuscript. Funding for this project
as much as 95% of the diet of fish greater than was provided by the Great Lakes Fishery Trust and
600 mm TL (Weller and Robbins 1999). Studies Purdue University Department of Forestry and
by Brown and Dendy (1961) and Haas et al. (2001) Natural Resources. This research was approved for
found a positive relationship between prey size and publication as manuscript number 17405 by the
the TL of flathead catfish, suggesting that larger Purdue University Agricultural Research Pro-
fish selected larger forage. The high relative abun- grams.
dance of large flathead catfish in the lower St. Jo-
seph River, which results from low exploitation References
and high survival, may limit the availability of Ashley, K. W., and B. Buff. 1986. Determination of
prey items selected by these fish. Future studies current food habits of flathead catfish in the Cape
should assess flathead catfish diet and the popu- Fear River. North Carolina Wildlife Resources
Commission, Raleigh.
lation structure and proportional stock densities of
Barnickol, P. G., and W. C. Starrett. 1951. Commercial
forage fishes to provide a better understanding of and sport fishes of the Mississippi River between
predator–prey dynamics in this system. Caruthersville, Missouri, and Dubuque, Iowa. Bul-
The results of our study indicate that flathead letin of the Illinois Natural History Survey 25:267–
catfish in the lower St. Joseph River are compa- 350.
rable in biological attributes and stock dynamics Bister, T. J., D. W. Willis, M. L. Brown, S. M. Jordan,
to populations throughout the midwestern United R. M. Neumann, M. C. Quist, and C. S. Guy. 2000.
Proposed standard weight (Ws) equations and stan-
States. The survival rate, size structure, and age dard length categories for 18 warmwater nongame
distribution of flathead catfish in the lower St. Jo- and riverine fish species. North American Journal
seph River were comparable to those of stocks of Fisheries Management 20:570–574.
known to receive low exploitation rates, suggest- Braaten, P. J., and C. S. Guy. 2002. Life history attri-
ing that harvest pressure is relatively low in this butes of fishes along the latitudinal gradient of the
system. However, little information exists regard- Missouri River. Transactions of the American Fish-
eries Society 131:931–945.
ing system-specific biotic and abiotic factors that Brown, B. E., and J. S. Dendy. 1961. Observations on
influence the stock characteristics and population the food habits of the flathead and blue catfish in
dynamics of flathead catfish. For example, future Alabama. Proceedings of the Annual Conference
studies should attempt to determine factors that Southeastern Association of Game and Fish Com-
influence the abundance, growth, and condition of missioners 15(1961):210–222.
flathead catfish. As angler attitudes and interests Clady, M. D., D. E. Campbell, and G. P. Cooper. 1975.
Effects of trophy angling on unexploited popula-
change regarding this species, efforts must be un-
tions of smallmouth bass. Pages 425–429 in R. H.
dertaken to quantify exploitation rates and sub- Stroud and H. Clepper, editors. Black bass ecology
sequent impacts on flathead catfish survival and and management. Sport Fishing Institute, Washing-
population structure. Studies that examine flathead ton, D.C.
catfish recruitment would increase our understand- Crumpton, J. E., M. M. Hale, and D. J. Renfro. 1987.
ing of stock structure and describe annual vari- Aging of three species of Florida catfish utilizing
ability in year-class strength and recruitment dy- pectoral spine sites and otoliths. Proceedings of the
Annual Conference Southeastern Association of
namics. These data would provide fisheries man- Fish and Wildlife Agencies 38(1984):335–341.
agers with the information necessary to develop Cunningham, K. K. 1995. Comparison of stationary and
appropriate management strategies for flathead mobile electrofishing for sampling flathead catfish.
catfish populations throughout the species’ geo- North American Journal of Fisheries Management
graphic range. 15:515–517.
Cunningham, K. K. 2000. Influence of environmental
Acknowledgments variables on flathead catfish electrofishing catch.
Proceedings of the Annual Conference Southeastern
We would like to thank S. Donabauer, L. Zurita, Association of Fish and Wildlife Agencies
R. Wyld, A. Gima, and S. Reed for their assistance 52(1998):125–135.
with field data collections. J. Wesley and J. Dexter, Daugherty, D. J., and T. M. Sutton. 2005. Seasonal
1200 DAUGHERTY AND SUTTON

movement patterns, habitat use, and home range of ternational ictalurid symposium. American Fisher-
flathead catfish in the lower St. Joseph River, Mich- ies Society, Symposium 24, Bethesda, Maryland.
igan. North American Journal of Fisheries Man- Jenkins, R. E., and N. M. Burkhead. 1994. Freshwater
agement 25:256–269. fishes of Virginia. American Fisheries Society, Be-
DeVries, D. R., and R. V. Frie. 1996. Determination of thesda, Maryland.
age and growth. Pages 483–512 in B. R. Murphy Kubeny, S. 1992. Population characteristics and habitat
and D. W. Willis, editors. Fisheries techniques, 2nd selection of channel catfish (Ictalurus punctatus) in
edition. American Fisheries Society, Bethesda, the lower James River, South Dakota. Master’s the-
Maryland. sis. South Dakota State University, Brookings.
Gerhardt, D. R., and W. A. Hubert. 1991. Population Layher, W. G. 1981. Comparison of annulus counts of
dynamics of a lightly exploited channel catfish stock pectoral and dorsal spines in flathead catfish. Pro-
in the Powder River system, Wyoming2Montana. gressive Fish-Culturist 43:218–219.
North American Journal of Fisheries Management Layher, W. G., and R. J. Boles. 1979. Growth of Py-
11:200–205. lodictis olivaris (Rafinesque) in a Kansas Reservoir.
Gilliland, E. 1988. Telephone, microelectronic, and Transactions of the Kansas Academy of Sciences
generator-powered electrofishing gear for collecting 82:36–48.
flathead catfish. Proceedings of the Annual Con- Lee, L. A., and J. W. Terrell. 1987. Habitat suitability
ference Southeastern Association of Fish and Wild- index models: flathead catfish. U.S. Fish and Wild-
life Agencies 41(1987):221–229. life Service, Biological Report 82, Washington,
Goedde, L. E., and D. W. Coble. 1981. Effects of fishing D.C.
on a previously fished and an unfished warmwater Leitch, J. A., and J. F. Baltezore. 1987. Attitudes of
fish community in two Wisconsin lakes. Transac- Minnesota anglers. Final report by the Center for
tions of the American Fisheries Society 110:594– Environmental Studies, North Dakota State Uni-
603. versity, to the Minnesota Department of Natural Re-
Grussing, M. D., D. R. DeVries, and R. A. Wright. 2001. sources, Division of Fish and Wildlife, St. Paul.
Stock characteristics and habitat use of catfishes in Lemmons, R. P. 1995. Habitat and populational char-
regulated section of four Alabama Rivers. Proceed- acteristics of flathead catfish in six Oklahoma
ings of the Annual Conference Southeastern As- stream systems. Master’s thesis. University of
sociation of Fish and Wildlife Agencies 53(1999): Oklahoma, Norman.
15–34. Marsh, P. C., W. L. Minckley, and K. L. Young. 1988.
Guier, C. R., L. E. Nichols, and R. T. Rachels. 1984.
Commercial potential of flathead catfish in the Col-
Biological investigation of flathead catfish in the
orado, Gila, Salt, and Verde rivers, Arizona. Ari-
Cape Fear River. Proceedings of the Annual Con-
zona Fish and Game Department, Final Report,
ference Southeastern Association of Fish and Wild-
Phoenix.
life Agencies 35(1981):607–621.
Mayhew, J. K. 1969. Age and growth of flathead catfish
Haas, J. J., L. A. Jahn, and R. Hilsabeck. 2001. Diets
in the Des Moines River, Iowa. Transactions of the
of introduced flathead catfish (Pylodictis olivaris) in
American Fisheries Society 98:118–121.
an Illinois reservoir. Journal of Freshwater Ecology
Minckley, W. L., and J. E. Deacon. 1959. Biology of
16:551–555.
the flathead catfish in Kansas. Transactions of the
Hesse, L. W. 1994. The status of Nebraska fishes in the
American Fisheries Society 88:344–355.
Missouri River, 4. Flathead catfish, Pylodictis oli-
varis, and blue catfish, Ictalurus furcatus (Ictaluri- Morris, L. A., and P. F. Novak. 1968. The telephone
dae). Transactions of the Nebraska Academy of Sci- generator as an electrofishing tool. Progressive
ences 21:89–98. Fish-Culturist 30:110–112.
Hesse, L. W., C. R. Wallace, and L. Lehman. 1978. Fish- Moss, J. L., and W. H. Tucker. 1989. Growth and harvest
es of the channelized Missouri River: age2growth, of flathead catfish in three Alabama public lakes.
length2frequency, length2weight, coefficient of Proceedings of the Annual Conference Southeastern
condition, catch curves and mortality of 25 species Association of Fish and Wildlife Agencies
of channelized Missouri River fishes. Nebraska 42(1988):149–156.
Game and Parks Commission, Technical Series Munger, C. R., G. R. Wilde, and B. J. Follis. 1994.
Number 4, Norfolk. Flathead catfish age and size at maturation in Texas.
Hubbs, C. L., and K. F. Lagler. 1947. Fishes of the Great North American Journal of Fisheries Management
Lakes region. Cranbrook Institute of Science, Bul- 14:403–408.
letin 26, Bloomfield Hills, Michigan. Nash, M. K., and E. R. Irwin. 1999. Use of otoliths
Insaurralde, M. S. 1992. Environmental characteristics versus pectoral spines for aging adult flathead cat-
associated with flathead catfish in four Mississippi fish. Pages 309–316 in E. R. Irwin, W. A. Hubert,
streams. Doctoral dissertation. Mississippi State C. F. Rabeni, H. L. Schramm, Jr., and T. Coon, ed-
University, Mississippi State. itors. Catfish 2000: proceedings of the international
Jackson, D. C. 1999. Flathead catfish: biology, fisheries, ictalurid symposium. American Fisheries Society,
and management. Pages 23–35 in E. R. Irwin, W. Symposium 24, Bethesda, Maryland.
A. Hubert, C. F. Rabeni, H. L. Schramm, Jr., and T. Paragamian, V. L. 1990. Characteristics of channel cat-
Coon, editors. Catfish 2000: proceedings of the in- fish populations in streams and rivers of Iowa with
 ST. JOSEPH RIVER FLATHEAD CATFISH 1201

varying habitats. Journal of the Iowa Academy of sampling methods of flathead catfish Pylodictis oli-
Science 97:37–45. varis in the Minnesota River. Minnesota Department
Pisano, M. S., M. J. Inansci, and W. L. Minckley. 1983. of Natural Resources, Division of Fish and Wildlife,
Age and growth and length2weight relationship for Section of Fisheries, Study IV, Job 389, St. Paul.
flathead catfish, Pylodictis olivaris, from Coachella Summers, G. L. 1986. Oklahoma anglers opinion survey
Canal, southeastern California. California Fish and 1985. Oklahoma Department of Wildlife Conser-
Game 69:124–128. vation, Federal Aid in Sport Fisheries Restoration,
Power, M., and R. S. McKinley. 1997. Latitudinal var- Project F-37-R, Bulletin 18, Oklahoma City.
iation in lake sturgeon size as related to the thermal Tondreau, R. 1988. A population study of flathead cat-
opportunity for growth. Transactions of the Amer- fish in the channelized Missouri River near Sioux
ican Fisheries Society 126:549–558. City, Iowa. Iowa Department of Natural Resources,
Pugibet, E. E., and D. C. Jackson. 1991. Sampling flat- Final Report, Sioux City.
head catfish in small streams. Proceedings of the Topp, D. H., H. Drewes, M. Henry, G. Huberty, and P.
Annual Conference Southeastern Association of Jacobson. 1994. Assessment of the Red River fish-
Fish and Wildlife Agencies 43(1989):133–137. ery, with special emphasis on channel catfish. Min-
Purkett, C. A., Jr. 1958. Growth of the fishes in the Salt nesota Department of Natural Resources, Study IV,
River, Missouri. Transactions of the American Fish- Job 242 Completion Report, St. Paul.
eries Society 87:116–131. Turner, P. R. 1982. Procedures for age determination
Quinn, S. P. 1988. Investigations into the biology of the and growth rate calculations of flathead catfish. Pro-
flathead catfish (Pylodictis olivaris) in the lower ceedings of the Annual Conference Southeastern
Flint River and the promotion of this species as a Association of Fish and Wildlife Agencies
fisheries resource for the Flint River fishermen. 34(1980):253–262.
Georgia Department of Natural Resources, Game Turner, P. R., and R. C. Summerfelt. 1971. Food habits
of adult flathead catfish, Pylodictis olivaris (Rafin-
and Fish Division, Federal Aid in Sport Fish Res-
esque), in Oklahoma reservoirs. Proceedings of the
toration, Project F-28- 15, Atlanta.
Annual Conference Southeastern Association of
Quinn, S. P. 1989. Flathead catfish abundance and
Game and Fish Commissioners 24(1970):387–401.
growth in the Flint River, Georgia. Proceedings of
Van Den Avyle, M. J. 1993. Dynamics of exploited fish
the Annual Conference Southeastern Association of
populations. Pages 105–135 in C. C. Kohler and W.
Fish and Wildlife Agencies 42(1988):141–148.
A. Hubert, editors. Inland fisheries management in
Quinn, S. P. 1993. Development of a multiuse fishery
North America. American Fisheries Society, Be-
for flathead catfish. North American Journal of Fish- thesda, Maryland.
eries Management 13:594–599. Vokoun, J. C., and C. F. Rabeni. 1999. Catfish sampling
Ricker, W. E. 1975. Computation and interpretation of in rivers and streams: a review of strategies, gears,
biological statistics of fish populations. Fisheries and methods. Pages 271–286 in E. R. Irwin, W. A.
Research Board of Canada Bulletin 191. Hubert, C. F. Rabeni, H. L. Schramm, Jr., and T.
Robson, D. S., and D. G. Chapman. 1961. Catch curves Coon, editors. Catfish 2000: proceedings of the in-
and mortality rates. Transactions of the American ternational ictalurid symposium. American Fisher-
Fisheries Society 90:181–189. ies Society, Symposium 24, Bethesda, Maryland.
Schnabel, Z. E. 1938. The estimation of the total fish Weller, R. R., and C. Robbins. 1999. Food habits of
population of a lake. American Mathematical flathead catfish in the Altamaha River System,
Monographs 45:348–368. Georgia. Proceedings of the Annual Conference
Stauffer, K. W., and B. D. Koenen. 1999. Comparison Southeastern Association of Fish and Wildlife
of methods for sampling flathead catfish in the Min- Agencies 53(1999):34–41.
nesota River. Pages 329–339 in E. R. Irwin, W. A. Wesley, J. K., and J. E. Duffy. 1999. St. Joseph River
Hubert, C. F. Rabeni, H. L. Schramm, Jr., and T. assessment. Michigan Department of Natural Re-
Coon, editors. Catfish 2000: proceedings of the in- sources, Fisheries Division, Special Report 24, Ann
ternational ictalurid symposium. American Fisher- Arbor.
ies Society, Symposium 24, Bethesda, Maryland. Young, K. L., and P. C. Marsh. 1990. Age and growth
Stauffer, K. W., R. C. Binder, B. C. Chapman, and B. of flathead catfish in four southwestern rivers. Cal-
D. Koenen. 1996. Population characteristics and ifornia Fish and Game 76:224–233.