Predator-induced changes in morphology of a prey fish: the effects of food level and temporal frequency of predation risk

In a series of experiments, we investigated the effects of food availability and risk frequency on the dynamics of predator-induced changes in growth and morphology of prey fish using goldfish (Carassius auratus) as our test species. In experiment 1, we fed goldfish high or low food rations and exposed them to either alarm cues from conspecifics, cues from swordtails or a water control. After 60 days, goldfish in the alarm cue treatment significantly increased their body depth and body weight but had smaller body length than goldfish exposed to swordtails cues or water, likely reducing their vulnerability to gape-limited predators. Importantly, food level had an impact on the amplitude of the morphological changes. In experiment 2, goldfish were exposed to two different frequencies of predation cues or a water control for 50 days. The cues were either continued or discontinued from day 51 to 100, and all cues were resumed from day 101 to 150. We found that goldfish exposed to predation cues increased their depth and weight at a faster rate than did the goldfish exposed to water, and of particular significance was the fact that frequency of risk had an effect on the amplitude of the change. When the cues were interrupted, the increase in growth rate parameters was reduced to the level of the goldfish exposed to water. However, when the cues were resumed, the rate increased to match the growth rate of the goldfish that were continuously exposed to the cues. Finally, we staged encounters between goldfish of differing morphologies and yellow perch (Perca flavescens) and found that deep-bodied goldfish had better survival than the shallow-bodied ones. These experiments illustrate the dynamic nature of inducible morphological defences.     

Dine or Dash?: Ontogenetic Shift in the Response of Yellow Perch to Conspecific Alarm Cues

During their first year of growth yellow perch, Perca flavescens, undergo an ontogenetic niche shift from invertebrate feeding to piscivory. They also undergo a similar shift in their response to heterospecific alarm cues, switching from anti-predator to foraging behaviour. We conducted laboratory trials to determine whether yellow perch experience a comparable ontogenetic shift in their response to conspecific alarm cues. When exposed to either young-of-year (YOY) or adult perch skin extract, YOY perch responded with decreased time in motion and number of feeding attempts as well as increased time spent with spines erect and latency to first feeding attempt, all of which are indicative of an anti-predator response. Adult perch, when exposed to the same cues, responded with increased time spent moving and number of feeding attempts as well as decreased time spent with spines erect and latency to first feeding attempt, indicative of a foraging response. These data suggest that yellow perch undergo an ontogenetic niche shift in response to conspecific alarm cues.     

Host activity and the risk of nest parasitism by brown-headed cowbirds

Proportions of nests parasitized by brown-headed cowbirds (Molothrus ater) vary greatly among host species, but factors underlyingthis variation remain poorly understood. Cowbirds are believedto find nests by watching host behavior. We tested the hypothesisthat the activity of hosts during nest building correlateswith the probability of parasitism among and within four sympatrichosts: dusky flycatchers (Empidonax oberholseri), warblingvireos (Vireo gilvus), yellow warblers (Dendroica petechia),and American redstarts (Setophaga ruticilla). Daily probabilityof parasitism varied substantially among these species, from3% for dusky flycatchers to more than four times that for warblingvireos. The four species did not differ in the proportion ofcowbirds fledged from their nests. Differences in nest placementdid not explain differences in probability of parasitism amongor within species. Parasitism frequencies increased among speciesthat made longer nest-building visits, had a greater propensityto perch during nest approach, spent more time near their nests,and had males that vocalized more often near nests. Within species, females that visited their nests less often, spent more timeon the nest per visit, and males that sang more and were activein a smaller area around their nests were more likely to beparasitized by cowbirds. These data support the hypothesisthat cowbirds use the activity of hosts during nest buildingto locate nests.     

Experiments on the Feeding Behavior of the Hematophagous Candiru, Vandellia cf. Plazaii

Hematophagous candirus (family Trichomycteridae Vandellia cf. plazaii) did not respond to potential chemical attractants (ammonia amino acids fresh fish slime and human urine) added to their aquarium water. They seemed to respond visually to live goldfish Carassius auratus and Amazonian cichlids Cichlasoma amazonarum. One hypothesis tested is that the goldfish (an ostariophysan) might be preferred to the cichlid (a perciform fish). Goldfish were attacked first significantly more often when both prey species were present (one-sided difference in proportions test, N = 15, p = 0.02). However neither the elapsed time to the first attack nor the duration of the attacks differed significantly (Mann-Whitney's U-test N = 32 p = 0.27 and p = 0.92). These two species seemed incapable (goldfish) or weakly capable (cichlid) of avoiding or repelling candiru attacks     

The decline of the native fishes of lakes Victoria and Kyoga (East Africa) and the impact of introduced species,especially the Nile perch,Lates niloticus,and the Nile tilapia,Oreochromis niloticus

Synopsis There has been a decline, and in some cases an almost total disappearance, of many of the native fish species of lakes Victoria and Kyoga in East Africa since the development of the fisheries of these lakes was initiated at the beginning of this century. The Nile perch, Lates niloticus, a large, voracious predator which was introduced into these lakes about the middle of the century along with several tilapiine species, is thought to have caused the reduction in the stocks of several species. But overfishing and competition between different species also appear to have contributed to this decline. By the time the Nile perch had become well established, stocks of the native tilapiine species had already been reduced by overfishing. The Labeo victorianus fishery had also deteriorated following intensive gillnetting of gravid individuals on breeding migrations. L. niloticus is, however, capable of preying on the species which haven been overfished and could have prevented their stocks from recovering from overfishing. L. niloticus is also directly responsible for the decline in populations of haplochromine cichlids which were abundant in these lakes before the Nile perch became established. Even without predation by Nile perch, it has been shown that the haplochromine cichlids could not have withstood heavy commercial exploitation if a trawl fishery had been established throughout Lake Victoria. Their utilisation for human food has also posed some problems. The abundance of the native tilapiine species may also have been reduced through competition with introduced species which have similar ecological requirements. At present, the Nile perch and one of the introduced tilapiine species, Oreochromis niloticus, form the basis of the fisheries of lakes Victoria and Kyoga.Invited editorial     

Vigilance in Single- and Mixed-Species Groups of Tamarins (Saguinus labiatus and Saguinus fuscicollis)

By forming larger sizes of groups, individuals benefit from a decrease in vigilance, but the collective vigilance of the group as a whole is not compromised. We examined whether this group size effect is apparent in mixed-species groups of red-bellied tamarins (S. labiatus) and saddleback tamarins (S. fuscicollis) which form stable and permanent associations in the wild. We studied general vigilance and responses to hidden threatening stimuli in five captive groups of each species, while they were housed in single- and mixed-species groups. For vigilance, the individual rate was lower in the larger mixed-species groups than in the smaller single-species groups. In addition, the amount of time when at least one individual was vigilant was higher in mixed-species groups. This suggests that the tamarins alter their vigilance behavior in the presence of the other species. In response to hidden threats, both species performed brief vigilance checks and frequencies of checking did not differ in single- and mixed-species groups. However, both species had a significant reduction in the mean duration per check, and there was a reduced total amount of time spent vigilance checking in the mixed-species groups compared to the single-species groups, demonstrating the group size effect. Overall the mixed-species groups had a higher number and mean duration of checking than the smaller single-species groups. Given that the two species share a common set of predators, and respond to each other's alarm calls, these findings provide strong evidence that individuals of both tamarin species may be able to benefit from forming mixed-species groups via improved vigilance and monitoring of threats.     

Diel vertical migrations of bathypelagic perch fry

The behaviour of young‐of‐the‐year (YOY) perch Perca fluviatilis as a dominant species in the assemblage of fry in the pelagic of Slapy Reservoir (Czech Republic), was studied during late May and mid‐June 2002 using acoustic methods and complementary net catches. During the day, perch fry were present simultaneously in littoral, epipelagic and bathypelagic habitats. Bathypelagic perch fry, forming a scattering layer, migrated vertically each day between the epilimnion and hypolimnion, with an amplitude of 11·0 m in May and 12·5 m in June. At dusk, the migratory bathypelagic fry mixed in the epilimnion with non‐migrating epipelagic fry and spent the night close to the thermocline (abundance maximum at 3–4 m in May, 0–2 m in June). In June, shoaling behaviour by some of the bathypelagic perch fry was also observed: the shoaling fry remained higher in the water column than the non‐shoaling fry. Both depths of the scattering layer and the depths of the fry shoals were strongly controlled by the light intensity. The contribution of the bathypelagic part of the population to the total numbers of pelagic perch fry decreased from 28·1% in May to 4·7% in June, while the density of all pelagic perch fry increased (c. 96 000 individuals ha^?1 in May and 142 000 individuals ha^?1 in June). In May, the bathypelagic (average total length, L_T, 11·9 mm) and epipelagic (average L_T 14·6 mm) perch fry differed in size while, in June, the epipelagic fry were divided into two distinct size groups. The more abundant group, of small epipelagic perch fry (average L_T 14·6 mm), was similar in size to the bathypelagic fry (average L_T 14·6 mm) while the less abundant group, of larger epipelagic fry (average L_T 34·4 mm), was similar in size to littoral perch fry (average L_T 35·0 mm). The results suggest that in perch fry three different survival strategies with different risks can be used in the same locality, time and year.     

Implications of small scale variation in ecological conditions for the diet and density of red colobus monkeys

In this 3-year investigation we documented patterns of density, diet, and activity of red colobus monkeys (Procolobus tephrosceles) in six areas in or near Kibale National Park, Uganda and related these patterns to availability of food resources. There were large differences in the density and behavior of the red colobus among the sites. For example, the red colobus at one site with a diverse plant community of more than 61 tree species, had a diet that included at least 42 species. In contrast, at a second site red colobus spent 92% of their feeding time eating from one species that dominated the tree community. The density of important red colobus food trees varied among sites from 32 trees/ha to 204 trees/ha, and red colobus density ranged from 0.70 groups/km² to 7.41 groups/km². Among sites, red colobus density was related to the cumulative DBH of important food trees, when one apparently anomalous site was excluded, and populations with more plant species in their diets tended to be those that were found at higher densities. Activity budgets of the red colobus populations varied markedly among sites. For example, feeding time ranged among sites from 29 to 55%, and traveling varied from 5 to 20%. When faced with increased foraging demands, red colobus reduced the time spent resting, while the time spent socializing remained fairly constant. Comparative socioecological studies typically contrast species separated by large geographical distances to ensure there is sufficient variation in the environment to detect behavioral responses. The marked differences in ecological conditions and red colobus behavior we documented over short geographical distances, suggests that small-scale contrasts are a useful tool to examine ecological determinants of behavior and community structure.     

Temperature-dependent differences in foraging ability of two percids,Perca fluviatilis and Gymnocephalus cernuus

Synopsis The effects of temperature on capture rate, handling time, capture probability and routine swimming performance were studied in two fish species, perch Perca fluviatilis and ruffe Gymnocephalus cernuus. In addition, the reaction distance (RD) of both species was measured. Handling time decreased and capture rate increased with increasing temperature for both perch and ruffe. The response of handling time and capture rate to temperature was more pronounced for perch than for ruffe. Routine swimming performance increased with increasing temperature for perch but not significantly so for ruffe. The laboratory study showed that the foraging ability of ruffe was less sensitive to temperature than it was for perch. These results were consistent with the vertical distributions of both species in thermally stratified lakes; ruffe, which occurred at all depths, utilized a wider temperature range than perch.     

Activities and feeding behavior of free-ranging pregnant baboons

The activities and feeding behavior of pregnant baboons, Papio cynocephalus,were studied in two free-ranging groups in Amboseli National Park, Kenya. Both the environmental conditions and the age of their fetuses influenced females’ activity budgets and feeding behavior. Females fed more and rested less during the dry season than during the wet season, and most females spent progressively more time feeding and less time grooming than expected (based upon the amount of recent rainfall) as their fetuses matured. During the wet season, females also devoted significantly greater proportions of feeding time to consumption of grass blades, leaves, and flowers and significantly smaller proportions of their feeding time to feeding upon seeds, corms, and fruit. As their fetuses became older, the proportion of time spent feeding upon seeds increased, and the proportion of time spent feeding upon grass blades and fruit declined.     

Aggression and nest spacing in single and mixed species groups of seabirds

When heterospecific seabirds are part of a nesting colony, there may be less opportunity for conspecifics to come in direct contact with each other, resulting in lower intraspecific aggressiveness. To determine if individuals spend less time in aggressive behavior when nesting in conspecific rather than heterospecific groups, we compared the behavior of black skimmers (Rhynchops niger) nesting with gull-billed terns (Sterna nilotica) in three mixed species subcolonies to those of black skimmers in three single species subcolonies. In contrast to our predictions, black skimmers spent significantly less time in aggressive behaviors when nesting in single species subcolonies than when nesting with heterospecifics. Although skimmers in mixed species subcolonies tended to have more aggressive interactions with skimmers than terns, this may be a function of subcolony composition; the proportions of aggressive interactions with conspecifics were similar to the proportions of conspecifics in each subcolony. However, within the mixed species subcolonies, skimmers that nested nearer to terns were involved in aggressive interactions significantly less than skimmers that nested closer to conspecifics. Also, skimmers nested closer to their nearest neighbor when it was a gull-billed tern than when it was another skimmer. Regardless of which species they nested closest to, skimmers were more aggressive towards other skimmers than to terns within the mixed species subcolonies. Distance to nearest neighbor's nest did not differ significantly between the colony types, and did not seem to influence the duration of aggressive activity in the single species subcolonies. In the mixed species subcolonies, however, the time spent in aggressive behavior increased as the distance to nearest neighbor increased. It appears that of the several benefits that have been proposed of mixed species colonies, reduced time spent in conspecific aggression is not among them. However, within a mixed species colony, an individual can reduce time spent in aggressive interactions by nesting near heterospecifics. Received: 16 September 1996 / Accepted: 24 February 1997     

Behavioral responses to prey density by three acarine predator species with different degrees of polyphagy

Behavioral responses by three acarine predators, Phytoseiulus persimilis, Typhlodromus occidentalis, and Amblyseius andersoni (Acari: Phytoseiidae), to different egg and webbing densities of the spider mite Tetranychus urticae (Acari: Tetranychidae) on rose leaflets were studied in the laboratory. Prey patches were delineated by T. urticae webbing and associated kairomones, which elicit turning back responses in predators near the patch edge. Only the presence of webbing affected predator behavior; increased webbing density did not increase patch time. Patch time increased with increased T. urticae egg density in the oligophagous P. persimilis, but was density independent in the polyphagous species T. occidentalis and A. andersoni. Patch time in all three species was more strongly correlated with the number of prey encounters and attacks than with the actual prey number present in the patch. Patch time was determined by (a) the turning back response near the patch edge; this response decayed through time and eventually led to the abandonment of the patch, and (b) encounters with, and attacks upon, prey eggs; these prolonged patch time by both an increment of time spent in handling or rejecting prey and an increment of time spent searching between two successive prey encounters or attacks. Although searching efficiency was independent of prey density in all three species, the predation rate by P. persimilis decreased with prey density because its searching activity (i.e. proportion of total patch time spent in searching) decreased with prey density. Predation rates by T. occidentalis and A. andersoni decreased with prey density because their searching activity and success ratio both decreased with prey density. The data were tested against models of predator foraging responses to prey density. The effects of the degree of polyphagy on predator foraging behavior were also discussed.     

Habitat use and feeding behavior of thirteen species of benthic stream fishes

Synopsis The densities, habitat use, and feeding behaviors of 13 fish species belonging to the benthic invertebrate-feeding guild were studied by snorkeling at three localities in the Little River of eastern Tennessee, U.S.A. Resource partitioning occurred by habitat, feeding behavior and time of activity. Differences were also found at the generic level.Cottus was a nocturnal feeder, whereasPercina andEtheostoma were, for the most part, diurnally active.Percina moved about rapidly and spent most of its time above the bottom. In contrast,Etheostoma varied considerably in the amount of time spent under cover, spent little time above the bottom, and exhibited low levels of swimming activity. Nearly all species sought cover at night, suggesting they may be particularly sensitive to predation at night. Species with small adult sizes (Etheostoma, Cottus andP. evides) were concentrated in shallow water habitats, whereas species with large adult sizes (Percina) were more abundant in deep water habitats. The habitat use data are consistent with the hypothesis that size-selective predation by centrarchid bass may cause smaller fish to avoid deep water areas. Large species should have a lower risk of predation due to their size and behavior.     

Searching and oviposition behavior of a mymarid egg parasitoid, Anagrus nigriventris, on five host plant species of its leafhopper host, Circulifer tenellus

Searching and oviposition behavior and parasitization ability of Anagrus nigriventris Girault (Hymenoptera: Mymaridae), an egg parasitoid of beet leafhopper, Circulifer tenellus (Baker) (Homoptera: Cicadellidae), were examined on five host plant species of beet leafhopper: sugar beet (Beta vulgaris L.), red stem filaree (Erodium cicutarium[L.]), peppergrass (Lepidium nitidum Nuttall), desert plantain (Plantago ovata Forsskal), and London rocket (Sisymbrium irio L.). Beet leafhopper embeds its eggs in the tissues of these plant species. For each plant species, A. nigriventris behavior was examined on plants with and without beet leafhopper eggs. Experimental design was a 5 (plant species) by 2 (host eggs present/absent) factorial. Additionally within each treatment, parasitoid behavior was observed over a 22-h period at five different observation periods: t=0, 3, 6, 9, and 22 h where t=0 h represents initial exposure of the insect with the plant. The behavioral events observed were: `fast walking' (general searching), `slow walking' (intensive searching), ovipositor probing, grooming, feeding, and resting. Significant differences (=0.05) among plant species in time spent on the plant, percentage of host eggs parasitized, and behavioral variables associated with intensive searching and oviposition all indicated that the plant species fell into two groups: `preferred' plants (sugar beet, London rocket, and peppergrass), and `unpreferred' plants (filaree and plantago). These variables also indicated that the parasitoids spent more time on, searched more, probed more, and oviposited more in plants with host eggs than plants without host eggs. Consistent effects of time (over the observation periods from t=0 to t=22 h) generally were detected only in the preferred plant species that had host eggs present. In these cases, intensive searching and probing decreased as time advanced, while variables related to general searching (`fast walking') and abandoning host egg patches (leaving the plant) tended to increase over time.     

Species differences in the interest shown in infants by juvenile female macaques (Macaca radiata andM. mulatta)

Using a focal animal technique, 16 juvenile female rhesus and bonnet macaques (2–3 years old) were observed with respect to the frequency and duration of their contacts with infants (0–1 year old). Each of these subjects was a member of one of four seminatural groups of macaques housed in 0.5-acre field cages at the California Primate Research Center. Rhesus juvenile females became very interested in infants when the latter were quite young, and maintained that interest throughout the study. When the infants were approximately 3–4 months old, the rhesus juveniles spent, on the average, 3.9 min/hr with those infants. By the time the infants were yearlings, the time spent with infants was approximately 6 min/hr. Relative to the rhesus, the bonnet juveniles showed less interest (about 2.2 min/hr) in young (3- to 4- month-old) infants, although by the time the infants were yearlings, the bonnets spent about 5.6 min/hr with them. Rhesus juveniles preferred sibling over nonsibling infants, but the presence of a sibling was not a necessary requirement for infant-directed behavior. The two bonnet juveniles who had infant siblings showed no preference for the latter. These data were interpreted within the framework of Quiatt’s [(1979) Am. Anthropol. 81: 310–319 conclusions about allomaternal behavior. Specifically, it was argued that, to the extent that there are species differences in maternal behavior, there should also be species differences in allomaternal behavior. The species differences in the maternal styles of rhesus and bonnet macaques provide a basis for an examination of this hypothesis.     

Sequential foraging of dusky dolphins with an inspection of their prey distribution

The aim of this work was to analyze the sequential foraging behavior of dusky dolphins (Lagenorhynchus obscurus). Foraging sequences were defined when more than two feeding bouts occur with a traveling bout between them. We hypothesized that traveling costs of searching for prey patches were related to the time spent feeding on a patch. In addition, the distribution and seasonal variation of anchovy schools were studied in the area to better understand dolphins' behavior. We observed dolphins from a research vessel from 2001 to 2007, and recorded their location and behavior. Anchovy data were collected during two hydro‐acoustic surveys. Dusky dolphin behaviors varied seasonally; they spent a greater proportion of time traveling and feeding in the warm season (Kruskal‐Wallis: H = 172.07, P < 0.01). During the cold season dolphin groups were more likely to exhibit diving behavior and less surface feeding. We found a positive correlation between searching and foraging time (r = 0.88, P = 0.019), suggesting that the costs associated with searching were compensated by an increase in the energy intake during the foraging bout. There was an association between dusky dolphin and anchovy distribution, in that they co‐varied spatially and seasonally.     

Feeding behavior ofLemur catta andLemur fulvus

The feeding behavior of two sympatric species of lemurs, Lemur cattaand Lemur fulvus,was studied in an enclosure simulating a natural habitat at the Duke University Primate Center. L. fulvusspent less time feeding during the day than L. catta.But the former species ate more fruit and had longer feeding bouts on preferred food items than L. catta.They also had a shorter food passage time than L. cattaand their choice of resting places was more influenced by food distribution. Furthermore, the two lemur species ate parts of different plant species and showed different reactions to chemical plant components. According to these results, L. fulvusis a more conservative feeder than L. catta.These interspecific differences in feeding behavior may be one of a number of differences that allow the two species to coexist. In allopatry, however, L. fulvusmay also adopt feeding patterns similar to those of L. catta.But L. cattawas never found to change its feeding strategies in different areas. It may be this option of L. fulvusto adopt different feeding strategies in different situations that allows this species to have the widest range of all Malagasy lemurs. Duke University Primate Center Publication No. 259.     

The foraging behavior of the central mudminnow and yellow perch: the influence of foraging site,intraspecific and interspecific competition

Summary The central mudminnow (Umbra limi) and yellow perch (Perca flavescens) are two species of fishes that commonly co-occur in small bog lakes in the Great Lakes region of North America. Both species are dietary generalists with a high degree of dietary overlap, and perch populations have been shown to have a strong negative effect on mudminnow populations. I compared the foraging rates of the two species in a series of laboratory experiments in which four foraging sites were simulated (open bottom substrate, water column, submerged aquatic macrophytes, and water surface). Fish were observed as they foraged singly, in monospecific pairs, and in mixed species pairs to evaluate the effects of intra- and interspecific competition on foraging success and the potential for resource partitioning based on foraging site. Single species trials showed that each species had similar foraging rates at each of the four sites. Across all sites combined, interspecific competition had a greater negative effect on the foraging success of mudminnows than did intraspecific competition. Conversely, intraspecific competition had a greater negative effect on the foraging success of perch than did interspecific competition. In mixed species trials perch took more food items from each of the four foraging sites than did mudminnows and caused shifts in site-use patterns of mudminnows. Differences in the foraging success of the two species reflected the superior ability of perch to discover the presence of food and to search rapidly for additional items, but were unrelated to handling times. These differences were linked to the species' foraging modes. The competitive superiority of perch, demonstrated in these experiments, could contribute to the negative impact of perch on mudminnow populations under natural conditions.     

Ontogenetic niche shifts and resource partitioning in a subarctic piscivore fish guild

The feeding ecology of three piscivorous fish species (perch (Perca fluviatilis), pike (Esox lucius) and burbot (Lota lota)), was studied in the subarctic Pasvik watercourse (69 °N), northern Norway and Russia. All three species primarily occupied the benthic habitats in the watercourse. Perch and burbot exhibited distinct ontogenetic niche shifts in food resource use, perch changing from a dominance of zooplankton to zoobenthos to fish, and burbot from zoobenthos to fish. Fish prey dominated the diet of all the investigated size-classes of pike, but small-sized pike (<20 cm) were not represented in the sample. Fish prey size was positively related to predator size in all three species. Whitefish (Coregonus lavaretus) was the dominant prey of pike and large-sized burbot and perch. Nine-spined sticklebacks (Pungitius pungitius) was also an important prey and appeared to be a dietary stepping-stone enhancing the transition from invertebrate feeding to consumption of large-sized whitefish prey for all three predators. A cluster analysis separated the different size groups of the three predator species into five functional feeding groups, most of them containing two or all three species. Within these feeding groups, and especially among the piscivorous size-classes, there was a strong and significant interspecific overlap in prey selection, and the dietary similarities between the species were in general much larger than the intraspecific similarities between ontogenetic stages. All three piscivorous species are important top predators in the aquatic food web of the watercourse, and their ontogenetic diet shifts and resource partitioning patterns generate a substantial food web complexity in this subarctic ecosystem.     

Growth responses of age 0 white perch and yellow perch from field-enclosure experiments

In this paper, we investigate the potential for competition between age 0 white perch Morone americana and yellow perch Perca flavescens in field-enclosure experiments. In 1986 and 1987, different densities of white perch and yellow perch were stocked into cages in the western basin of Lake Erie and their growth responses determined. Fish were weighed and measured pre- and post-stocking. In 1986 when fish were food-limited, individual growth of both species were negatively affected by increased total fish density; yellow perch to a greater extent. In 1987, experiments were carried out later in the year at lower temperatures. Yellow perch were shown to feed and continue to grow at lower water temperatures than white perch. We conclude that competition in summer may be mitigated in the fall because of differences in thermal optima between these two species.