Because populations of channel catfish Ictalurus punctatus have been difficult to sample in small impoundments, we determined whether tandem hoop net series (three nets tied in series and baited with cheese) fished for 3 d could effectively sample these populations. In 2000 we compared seasonal catch rates (catch per unit effort (CPUE), defined as the number of fish caught per series) and length frequencies (total length (TL); mm) of channel catfish captured with two types of tandem series (long versus short bridles) and monitored the mortality of channel catfish along with the CPUE and mortality of bycatch in five lakes. Mean CPUE ranged from 12 to 194 fish/series for channel catfish and from 5 to 101 fish/series for bycatch among series types, lakes, and sampling periods. Mean CPUE for channel catfish and bycatch was similar between series types and did not consistently vary with sampling period. Length frequencies of channel catfish were usually similar between series types. Mortality was low for both channel catfish (0.3% of total catch) and bycatch (8%). In 2001, using short‐bridled series, we assessed gear size bias in one lake; sampled 66 lakes that are stocked annually with channel catfish fingerlings at rates of 12, 37, or 74/ha; and estimated the sample sizes required for precise estimates of mean CPUE and accurate estimates of size structure. Hoop nets failed to capture channel catfish less than 250 mm TL in proportion to their abundance. Mean CPUE among 62 lakes ranged from 0.5 to 369.7 channel catfish/series and increased with stocking rate. Four lakes were excluded because of the high catch and mortality of turtles. Mean TL ranged from 276 to 463 mm among the lakes and decreased with stocking rate. For moderate precision (coefficient of variation (100·SE/mean) = 0.2) of mean CPUE, from 12 to nearly 50 series were required, and length measurements from 300 fish were necessary for accurate assessments of size structure. Tandem hoop netting provided adequate samples of channel catfish in most small impoundments, but obtaining precise estimates of CPUE may be difficult; moreover, turtle mortality can be problematic in some lakes.

Factors affecting abundance, growth, and survival of age-0 gizzard shad Dorosoma cepedianum were examined in Pomme de Terre and Stockton lakes. Missouri, during 1987Missouri, during -1991 Otolith ages were used to assign larvae (^25 mm total length, TL) and juveniles (>25 mm TL) to weekly cohorts so that cohort-specific estimates of abundance, growth, and survival could be made. The distribution of initial abundances of larvae among weekly cohorts was influenced by water temperature and reservoir water level but not by characteristics of adult females (abundance, size, condition, fecundity). Intense periods of spawning activity during rising water levels resulted in high peaks in larval abundance and relatively few large weekly cohorts (e.g., 1990). In the absence of water level rises, peaks in larval abundance were much lower and abundances of larvae were more evenly distributed among several cohorts (e.g., 1988). Spring warming also affected spawning; early cohorts in warm springs (1987 and 1991) were relatively more abundant than those during cool springs (1989 and 1990). The initial distribution of larvae among weekly cohorts influenced subsequent interactions between the larvae and their environment. Growth of larvae was positively affected by water temperature and, to a lesser degree, food abundance. Early cohorts grew slower because of lower water temperatures, and they suffered higher mortality than did late cohorts. Survival was also density dependent; weekly cohorts with high initial densities experienced lower survival than cohorts with small initial densities. High survival of larvae occurred in 1988 when densities of larvae were low and evenly distributed among weekly cohorts and food abundance was high. Growth and instantaneous mortality of juveniles were not closely tied to any single variable. Growth was positively related to food abundance and negatively related to gizzard shad density. Mortality was not significantly related to any measured variable. Overall, findings indicate that cohort dynamics of age-0 gizzard shad are initially driven by climatic factors and later by a combination of climatic and biotic factors.Gizzard shad Dorosoma cepedianum are impor-Despite their importance, factors affecting abuntant prey for many sport fish in reservoir ecosys-dance, growth, and survival of age-0 gizzard shad terns (Noble 1981;Storck 1986; Johnson et al. are not well documented. Initial abundances of lar-1988; Michaletz 1997, this issue), but they are also vae have been related to condition of adult females potentially detrimental competitors of young sport (Stock 1971;Kampa 1984;Willis 1987). Also, fish and other planktivores (Kirk and Davies 1987; i arger females may spawn earlier (Miranda and DeVries and Stein 1992;Welker et al. 1994). Muncy 1988) and over a shorter time (Willis 1987) Age-0 gizzard shad can be especially forceful comthan smaller counterparts, affecting temporal aspetitors because they are more zooplanktivorous pects of |arva , hatching and abun dance. Environthan their older cou...

We surveyed resource agencies in the United States and Canada to assess the status of the management of catfish species \[primarily channel catfish (Ictalurus punctatus), blue catfish (I. furcatus), and flathead catfish (Pylodictis olivaris)\]. Thirty‐two of the fifty‐three agencies we surveyed considered catfish to be either moderately or highly important to anglers. Twenty‐eight states allowed commercial fishing in selected waters. Catfish populations were managed primarily by creel limits and gear restrictions, and less frequently by size limits. Put‐grow‐take fisheries for channel catfish were popular in small impoundments, and most catfish stockings were associated with these fisheries. Managers most commonly used gill nets and electrofishing to sample catfishes and used catch‐per‐unit effort, size structure, and creel information to evaluate catfish populations. Major constraints to the management of catfishes included (1) low priority or angler interest, (2) inadequate habitat, (3) inadequate data, and (4) inadequate sampling. A shortage of catfish for stocking also was a major constraint for small impoundments. Although catfish provide important fisheries throughout a wide geographical area, most resource agencies reported they do not intensively manage catfish populations, although put‐grow‐take and put‐take fisheries in small impoundments were notable exceptions.

Population characteristics of gizzard shad Dorosoma cepedianum were examined in 14 Missouri reservoirs to determine if they were related to reservoir productivity (chlorophyll‐a concentrations, mg/m3, CHLA), mean depth, and sport fish growth. Gizzard shad catch per unit effort (CPUE, >age 1), mean length at age 1, mortality, and a recruitment variability index (RVI, larger numbers indicate more consistent recruitment) increased with increasing reservoir productivity and decreasing mean depth, while proportional stock density (PSD), relative weight (Wr), and mean length at age 3 decreased. In stepwise regression models, gizzard shad mean length at age 1 was negatively related to mean depth; mean length at age 3 was negatively related to CHLA; mortality was negatively related to mean depth and positively related to CPUE; and RVI was negatively related to PSD and positively related to mean length at age 1. Mean lengths at age 3 for largemouth bass Micropterus salmoides and white crappie Pomoxis annularis (used as indicator of growth) were negatively related to gizzard shad mean length at age 1. Sport fish grew faster in deep, less‐productive reservoirs, in which gizzard shad grew more slowly during their first year, than they did in shallow, more‐productive reservoirs, in which gizzard shad grew faster.

Relationships among age-0 gizzard shad Dorosoma cepedianum and their predators were examined over 5 years (1987)(1988)(1989)(1990)(1991) in Pomme de Terre and Stocklon lakes. Missouri, to determine the influence of gizzard shad abundance and size on diets, diet overlap, and growth of predators. Within years, temporal changes in age-0 gizzard shad size affected predator diets and diet overlap of predators more than abundance. Age-0 gizzard shad were not important as prey until they grew larger than 25 mm total length, even though larval densities were much higher. They were the most common prey for most predators during June through October and diet overlap among predators was usually highest during that period. Lengths of age-0 gizzard shad ingested by predators resembled those available in the reservoirs throughout much of the season except in early summer, when predators ingested larger-than-average individuals, and in the fall, when smaller predators ingested smaller-than-average individuals. Among years, variations in summer production (kg/ha) and mean weight (g) of age-0 gizzard shad were significantly correlated with their importance in most predator diets and with diet overlap among predator species in Stockton Lake but not in Pomme dc Terre Lake, possibly because age-0 gizzard shad were always abundant in Pomme de Terre Lake and annual variation in gizzard shad variables was less in Pomme de Terre than in Stockton Lake. One or both of these gizzard shad variables were usually significant in regression models explaining variation in growth increments of predators. When significant, production of age-0 gizzard shad was positively related to the species' importance in predator diets, to diet overlap among predators, and to predator growth, but mean weight of age-0 gizzard shad was negatively related to those variables. Because mean weight and mean density of age-0 gizzard shad were strongly negatively correlated, predators may have been responding more strongly to prey density than to prey size. Size of age-0 gizzard shad appeared to most strongly influence growth of smaller predators, whereas age-0 gizzard shad production most strongly influenced growth of larger predators. I conclude that changes in age-0 gizzard shad abundance and size can significantly influence predator diets, diet overlap among predators, and predator growth.