Aphredoderus Research Articles

Ecological Aspects of the Pirate Perch (Aphredoderus Sayanus) in East Texas Streams

Aphredoderus sayanus (Pirate Perch) is a cryptic fish species inhabiting rivers throughout the eastern United States. We sampled Pirate Perch across 16 streams in 4 river basins (Trinity, Neches, Sabine, and Cypress) within east Texas to examine their relative abundance, habitat associations, and to describe their diet and reproductive biology. Ninety-five individuals were collected across all river basins, with streams within the Neches River containing the highest number of occurrences. Instream habitat conditions varied among basins, with major differences reflected by flow regimes, water depth, and substrate diversity. Pirate Perch were primarily associated with streams having lentic flows, shallow waters that contained woody structures, and substrates dominated by leaf litter and detritus. Insectivory was confirmed for the species. Microplastics were also documented for the first time in their stomach contents. Maximum fecundity was 2161 eggs, and ova development indicated multiple clutches of eggs during their reproductive period.

A wolf in sheep's clothing: Predatory fish have convergent consumptive effects but divergent predation‐risk effects

AbstractAlthough consumptive effects of predators have long been central to ecology, predation‐risk effects have emerged as major components of predator–prey interactions. Both consumptive and predation‐risk effects should vary with predator functional traits (e.g., hunting mode, gape size), where consumption rates and induction of morphological and behavioral defenses correlate with prey‐specific predator threat. Ambush predators, in contrast with active predators, may face selection pressure to be cryptic to avoid detection by prey. Thus, ambush predators may change prey density through consumptive effects but have reduced or absent predation‐risk effects. We performed two mesocosm experiments with free‐roaming and caged predators to explore post‐colonization interactions of the chemically camouflaged, large‐gaped, ambush predator, pirate perch (Aphredoderus sayanus), active, large‐gaped green sunfish (Lepomis cyanellus), and active, small‐gaped golden topminnow (Fundulus chrysotus), with larval gray treefrogs (Hyla chrysoscelis) and mole salamanders (Ambystoma talpoideum). We examined the consumptive and predation‐risk effects of each predator on amphibian mortality, growth rates, tail morphology, and polyphenisms. Large‐gaped pirate perch and green sunfish had strong, equivalent consumptive effects, but only the free‐roaming active predator, green sunfish, suppressed the growth of survivors through risk‐induced trait responses. Caged green sunfish induced much stronger non‐consumptive mortality than pirate perch in gray treefrogs, an effect that was absent for mole salamanders; golden topminnows had intermediate effects. Tail defenses were a function of prey mortality and only manifested in free‐roaming predator treatments, suggesting the necessity of predator attacks or conspecific alarm signals. Likewise, mole salamander polyphenism was unaffected by caged predators, whereas free‐roaming green sunfish prevented all metamorphosis. Free‐roaming golden topminnows increased the proportion of individuals remaining as larvae, and pirate perch increased the proportion of paedomorphs and metamorphs. Overall, the prey had positively correlated anti‐predator responses across life stages, suggesting multicomponent defenses. Predator effects varied with functional traits with large‐gaped predators having strong consumptive effects, but active predators causing stronger risk‐induced changes in growth compared to cryptic ambush predators, which provides more evidence for chemical camouflage. Our results emphasize the role of hunting mode and gape size in determining consumptive and predation‐risk effects, and that predation‐risk effects cannot be reliably predicted from consumptive effects.

Cue reduction or general cue masking do not underlie generalized chemical camouflage in pirate perch.

Avoiding detection is perhaps the ultimate weapon for both predators and prey. Chemosensory detection of predators via waterborne or airborne cues (predator-released kairomones) is a key prey adaptation in aquatic ecosystems. Pirate perch, Aphredoderus sayanus, a largely insectivorous mesopredatory fish, are considered to be chemically camouflaged because they are unavoided by all colonizing organisms tested, including treefrogs and aquatic insects, despite stronger predatory effects on target taxa than several avoided fish. To address the mechanism behind camouflage we used aquatic insect colonization as a bioassay to test (1) whether increasing pirate perch density/biomass leads to increased avoidance, and (2) whether pirate perch mask heterospecific fish kairomones. Insect abundances, species richness, and community structure showed no response to pirate perch density. Last, pirate perch did not mask the kairomones of heterospecific predatory fish. Results support the idea that fish kairomones are species-specific, and chemical camouflage is driven by a unique chemical signature that is either undetectable or has no negative associations for colonists.

New data on Henneguya postexilis Minchew, 1977, a parasite of channel catfish Ictalurus punctatus, with notes on resolution of molecular markers for myxozoan phylogeny.

Previous morphological and histological data are supplemented with molecular and ultrastructural data for a Henneguya sp. isolated from farm-raised channel catfish Ictalurus punctatus in Mississippi, USA. Myxospores were cryptic, encapsulated within a thin layer of epithelium in the gill lamellae with spore measurements consistent with the original description of Henneguya postexilis Minchew, 1977. Myxospores were 42.7-49.1 µm in total length with spore bodies 12.1-17.2 × 3.6-4.8 × 2.9-3 µm. Polar capsules were of unequal length, with the longer capsule being 4.4-6.7 × 1.1-1.6 µm and the shorter capsule being 4.4-6.4 × 1.1-1.6 µm. Polar tubules had 6-8 turns. Caudal processes were 25.7-38.1 µm in length. Spores were encapsulated in a thin layer of epithelium in the gill lamellae. Molecular data from the most commonly used markers for myxozoan identification and phylogeny, partial 18S small subunit ribosomal gene (SSU), partial 28S large subunit ribosomal gene (LSU), and elongation factor 2 (EF2) were generated for H. postexilis. Additionally, novel data for LSU and EF2 were generated for archived myxozoan specimens from farm-raised catfish (H. mississippiensis, H. ictaluri, H. exilis, H. adiposa, H. sutherlandi, H. bulbosus, Unicauda fimbrethilae), as well as archived specimens from wild fish (H. laseeae [from Pylodictis olivaris], Hennegoides flockae [from Aphredoderus sayanus], Myxobolus cloutmani [from Cycleptus elongatus]. These include the first EF2 sequence data for the genera Hennegoides and Unicauda. Phylogenetic analyses using these data placed H. postexilis in well supported clades with other ictalurid-infecting Henneguya species. Phylogenetic signal assessments on these analyses suggest that while SSU provided the greatest phylogenetic signal, LSU yielded comparable signal, supporting previous work implying this region may be underutilised in reconstructing myxobolid phylogenies.

Urocleidus sayani n. sp. (Monogenea: Dactylogyridae) from Gills of Pirate Perch (Aphredoderus sayanus).

Urocleidus sayani n. sp. is described from the gills of pirate perch (Aphredoderus sayanus) in the Wisconsin backwaters of the upper Mississippi River and was found in samples from the Southeastern United States. Urocleidus sayani n. sp. is the second monogenean described from the pirate perch and the first for this host within Dactylogyridae. The description includes a partial 18S rRNA gene sequence (623 bp), filling a void in sequence data from North American monogeneans.

Two Novel Myxozoans from Pirate Perch Aphredoderus sayanus (Gilliams, 1824) in the Upper Mississippi River, Including the First North American Species of Hennegoides Lom, Tonguthai, & Dyková, 1991

The pirate perch Aphredoderus sayanus is a relatively small fish species found in rivers throughout much of the eastern United States. Due to their cryptic nature, relatively little is known regarding their parasite fauna. A survey of pirate perch from the upper Mississippi River revealed 2 novel myxozoans. Hennegoides flockae n. sp. was observed in heavily infected gills where the lamellae featured irregular expansion by bulbous myxozoan polysporic plasmodia, typically affecting the middle to distal half of the filaments. When severe, infection of sequential filaments was such that the filaments were fused, forming what appeared as multicystic/lobular parasitic aggregates subdivided by fine epithelial cords. The total myxospore length of Hennegoides flockae was 35.4-46.4 (41.3 ± 3.3) and the spore body, asymmetrically ovoid in valvular view, was 15.4-18.7 (17.0 ± 0.7) × 7.1-8.7 (7.9 ± 0.4). Henneguya marcquenskiae n. sp. was observed in the liver with plasmodia present randomly and infrequently in the hepatocellular parenchyma. The total myxospore length for Henneguya marcquenskiae was 39.5-55.9 (48.4 ± 4.2), with the spore body being lanceolate, 13.9-16.5 (15.4 ± 0.7) × 7.1-9.0 (8.3 ± 0.5). Phylogenetic analysis of the SSU rRNA gene placed both Hennegoides flockae and Henneguya marcquenskiae as sisters to each other in a clade containing other Myxozoans known to infect the gills of esocids, percids, and centrarchids. These parasites represent the first reports of Henneguya and Hennegoides from pirate perch, with the latter being the first report of this genus outside of the Asian continent.

Mosquito female response to the presence of larvivorous fish does not match threat to larvae

1. Several species with complex life‐history traits such as amphibians and insects with aquatic immature stages and terrestrial adults avoid ovipositing in pools containing larvivorous fish. This avoidance response was assumed to be a general one for most fish species.2. The generality of ovipositing Culex to the presence of three, widespread larvivorous fish species was tested in a set of field experiments with artificial oviposition pools using caged fish.3. Larval performance was further examined under actual predation by these three fish species.4. Results show that ovipositing females responded strongly to the presence of caged mosquitofish, Gambusia affinis, while showing no significant response to the presence of caged green sunfish, Lepomis cyanellus, or the pirate perch, Aphredoderus sayanus. All three fish species consumed similar amounts of larvae.5. This is the first example of species‐specific response differences to predators during mosquito oviposition habitat selection. These results point to the existence of predator‐released kairomones affecting mosquito behaviour. These kairomones may either be species‐specific or vary in concentration among fish, and probably have an important role in the understanding of mosquito spatial distribution.

Functional diversity of non-lethal effects, chemical camouflage, andvariation in fish avoidance in colonizing beetles.

Predators play an extremely important role in natural communities. In freshwater systems, fish can dominate sorting both at the colonization and post-colonization stage. Specifically, for many colonizing species, fish can have non-lethal, direct effects that exceed the lethal direct effects of predation. Functionally diverse fish species with a range of predatory capabilities have previously been observed to elicit functionally equivalent responses on oviposition in tree frogs. We tested this hypothesis of functional equivalence of non-lethal effects for four predatory fish species, using naturally colonizing populations of aquatic beetles. Among taxa other than mosquitoes, and with the exception of the chemically camouflaged pirate perch, Aphredoderus sayanus, we provide the first evidence of variation in colonization or oviposition responses to different fish species. Focusing on total abundance, Fundulus chrysotus, a gape-limited, surface-feeding fish, elicited unique responses among colonizing Hydrophilidae, with the exception of the smallest and most abundant taxa, Paracymus, while Dytiscidae responded similarly to all avoided fish. Neither family responded to A.sayanus. Analysis of species richness and multivariate characterization of the beetle assemblages for the four fish species and controls revealed additional variation among the three avoided species and confirmed that chemical camouflage in A.sayanus results in assemblages essentially identical to fishless controls. The origin of this variation in beetle responses to different fish is unknown, but may involve variation in cue sensitivity, different behavioral algorithms, or differential responses to species-specific fish cues. The identity of fish species occupying aquatic habitats is crucial to understanding community structure, as varying strengths of lethal and non-lethal effects, as well as their interaction, create complex landscapes of predator effects and challenge the notion of functional equivalence.

A New Species ofAzygia(Trematoda: Azygiidae) from Pirate Perch,Aphredoderus sayanus(Aphredoderidae), in the Big Thicket National Preserve, Texas, U.S.A

Azygia aphredoderi n. sp. is described from the stomach of the pirate perch Aphredoderus sayanus (Aphredoderidae) collected from the Menard Creek Corridor Unit of the Big Thicket National Preserve in Texas, U.S.A. The new species is similar to Azygia angusticauda in that the ventral sucker and gonads are placed posteriorly, characteristics that allow both species to be distinguished from all others in North America. Members of the new species are characterized by a ventral sucker that is larger than the oral sucker, small body size (<2 mm), small length-to-width ratio, and a forebody-to-hindbody ratio of nearly 1.0, all of which allow them to be distinguished from A. angusticauda.

Is the Pirate Really a Ghost? Evidence for Generalized Chemical Camouflage in an Aquatic Predator, Pirate PerchAphredoderus sayanus

Camouflage occupies a central role in arsenals of both predators and prey and invokes visions of organisms possessing specific characteristics or altering their shape, color, or behavior to blend into the visual background or confound identification. However, many organisms use modalities other than vision. Chemical communication is particularly important in aquatic systems, and chemicals cues are used by a broad array of colonizing organisms to recognize and avoid risky habitats. Here we describe a habitat selection experiment with aquatic beetles and summarize results of 11 experiments involving colonizing beetles and ovipositing tree frogs that provide evidence that pirate perch Aphredoderus sayanus are chemically camouflaged with respect to a diverse array of prey organisms. We believe this to be the first example of a predator possessing a generalized chemical camouflage effective against a broad array of prey organisms, and we suggest that it may constitute a novel weapon in the predator-prey arms race.

Notes on the Diet and Egg Clutches of the Pirate Perch (Aphredoderus Sayanus) from Central Arkansas

Abstract Aphredoderus sayanus (Pirate Perch) is a widespread species whose life history is the subject of only a few studies. I recorded life-history data from a small sample of Pirate Perch collected from an intermittent stream in midcontinental highlands of Arkansas (Batesville, Independence County) on 30 June 2004. Our small sample suggests variation in prey consumption rates between sexes during the breeding season and in allocation to reproduction among females. Our survey confirmed that Pirate Perch from Arkansas have similar insectivorous diets to those reported from other regions.

Proteocephalus pearsei (Cestoidea: Proteocephalidae) from the Pirate Perch, Aphredoderus sayanus (Percopsiformes: Aphredoderidae), in Northern Arkansas, U.S.A

One of 6 (17%) pirate perches, Aphredoderus sayanus, from Spring Creek, Independence County, Arkansas, was found to harbor in its small intestine a single proteocephalid tapeworm, Proteocephalus pearsei LaRue, 1919. This tapeworm has been reported previously from a variety of fish hosts, including those belonging to the families Centrarchidae, Cottidae, Gadidae, Moronidae, Percidae, Salmonidae, and Sciaenidae. The documented range of P. pearsei includes 6 northern states (Michigan, Minnesota, New York, Ohio, Pennsylvania, and Wisconsin), 2 Canadian provinces (Manitoba, Ontario), and the Great Lakes. We report a new host record and, for the first time, P. pearsei from Arkansas.

Patterns of Habitat Use among Vegetation‐Dwelling Littoral Fishes in the Atchafalaya River Basin, Louisiana

AbstractInformation regarding the relative value of different aquatic macrophytes as fish habitat is lacking for large floodplain river systems, such as the Atchafalaya River basin in south‐central Louisiana. We made seasonal comparisons of fish (total length, &lt;100 mm) density, diversity, and biomass and physicochemistry among submersed hydrilla Hydrilla verticillata, emergent bulltongue Sagittaria lancifolia, and floating water hyacinth Eichhornia crassipes. In addition, we examined the influence of habitat structure and physicochemistry on fish assemblage composition and density. Although there were no seasonal differences in fish biomass among plant types, hydrilla consistently supported the greatest fish density and diversity across seasons. While hydrilla and water hyacinth fish density changed little among seasons, density in bulltongue was higher during low water. Of the 12 most common fish species, the pirate perch Aphredoderus sayanus and yellow bullhead Ameiurus natalis showed no consistent pattern of habitat use. However, the densities of golden topminnow Fundulus chrysotus, bluegill Lepomis macrochirus, warmouth L. gulosus, redear sunfish L. microlophus, and pugnose minnow Opsopoeodus emiliae were highest in hydrilla, whereas the density of bantam sunfish L. symmetricus was highest in water hyacinth. A seasonally variable pattern was found for western mosquitofish Gambusia affinis, redspotted sunfish L. miniatus, sailfin molly Poecilia latipinna (higher in bulltongue), and least killifish Heterandria formosa (lower in water hyacinth). Overall, littoral fish distributions reflected responses to several spatially and temporally variable environmental factors (e.g., water quality, plant complexity, and seasonal depth changes). We suggest that the consistently high fish density and diversity in hydrilla across seasons were related to an optimization of favorable water quality (particularly dissolved oxygen [DO]), architectural complexity, and biovolume. However, during low water, as hydrilla and water hyacinth canopies became very dense and water hyacinth DO levels remained low, bulltongue was more heavily used by several species, presumably because of a more favorable combination of DO, complexity, and particularly depth.

Composition and diversity patterns in metazoan parasite communities and anthropogenic disturbance in stream ecosystems

The composition and diversity of metazoan parasite communities in naturally depauperate ecosystems are rarely studied. This study describes the composition of helminth endoparasite communities infecting fish that are part of naturally acidic stream ecosystems in the coastal-plains region of the State of New Jersey (USA) known as the Pinelands, and compares the diversity of parasites between six streams that differ in anthropogenic disturbance. A total of 514 fish were examined representing 6 species native but restricted to the Pinelands region, 5 species native and widespread throughout the region and State, and 6 species introduced to the Pinelands and State. Fish (prevalence: 78%) were infected with 18 helminth endoparasite species. In most streams, prevalence of infection, mean abundance, and total number of individuals for the 5 most common parasites were higher in pirate perch, a native fish species. The diversity of helminth endoparasite communities measured as species richness and Shannon index was higher in degraded streams, and especially in native or introduced fish at these sites. Parasite diversity was positively correlated with anthropogenic disturbance, which was measured by water pH, water conductance, and the proportion of agricultural and developed land surrounding streams. Helminth community composition included parasites intimately tied to trophic interactions in food webs, and disturbance to these ecosystems results in changes to these communities. Understanding structure and function of animal communities from these naturally depauperate ecosystems is important before continued anthropogenic changes result in the extirpation or extinction of their unique fauna.

Observations of the Pirate Perch, Aphredoderus sayanus (Gilliams), with Comments on Sexual Dimorphism, Reproduction, and Unique Defecation Behavior

ABSTRACT Analysis of variance results showed significant differences in measurements on 71 specimens of Aphredoderus sayanus comparing standard length (F = 21.91, P < 0.0001), maximum depth (F = 8.44, P = 0.006), and standard length to body depth ratio (F = 9.11, P = 0.004), indicating that this fish is sexually dimorphic with females being larger and more full-bodied than males. In laboratory observations, a pair of A. sayanus created a nest and successfully spawned, which strengthens the argument that this fish might not be a branchial brooder but instead might be a nest builder with the female releasing adhesive eggs in the proximity of the nest. On three separate daytime occasions, A. sayanus ventured outside its refuge approximately to the point that only its head and anus were exposed, defecated, and then backed into its refuge. The jugular position of the anus might allow adult A. sayanus to partially leave its refuges to defecate to avoid altering water quality and without fully exposing itself to predators. This unique defecation behavior might explain the evolutionary reason for the position of the anus in adult A. sayanus.

Spawning Behavior and Genetic Parentage in the Pirate Perch (Aphredoderus sayanus), a Fish with an Enigmatic Reproductive Morphology

We describe for the first time reproductive behaviors in the Pirate Perch (Aphredoderus sayanus), a secretive nocturnal fish whose urogenital opening is positioned far anteriorally, under its throat. Some naturalists had speculated that this peculiar morphological condition might serve to promote egg transfer to the fish's branchial chamber for gill-brooding; others hypothesized that Pirate Perch spawn in the substrate of streams but offered no adaptive rationale for the odd placement of the fish's urogenital pore. Here we solve the conundrum through a combination of intensive field investigations, underwater filming, and molecular parentage analyses. We show that Pirate Perch spawn in underwater root masses, the first documentation of such nesting behavior in any species of North American fish. Female Pirate Perch thrust their heads and release their eggs into sheltered canals of these masses. Males congregate at these sites and likewise enter the narrow canals headfirst, to release sperm. Thus, the forward-shifted urogenital pore may facilitate spawning under this special nesting circumstance. We found no evidence of extended parental care. Fish formed their own canals or used burrows made by aquatic macro-invertebrates and salamanders. Genetic analyses based on three polymorphic microsatellite loci demonstrate that a total of at least five to 11 sires and dams were the parents of embryos within each of three assayed root-mass nests (of a total of 23 nests found). Males defended the oviposition sites by body-plugging canal entrances after spawning. This and more direct aggressive behaviors by males probably relate to selection pressures imposed by intense competition for fertilization success under these group-spawning conditions.

Functional equivalence of non‐lethal effects: generalized fish avoidance determines distribution of gray treefrog, Hyla chrysoscelis, larvae

How patterns of prey abundance are generated determines how predation influences population and community level dynamics. We investigated how a natural population of ovipositing treefrogs, Hyla chrysoscelis, partitioned their eggs amongst experimental ponds differing in the non‐lethal presence of five fish predators, each representing a distinct family. Four fish predators were functionally equivalent in terms of the behavioral response they induced in H. chrysoscelis females and hence the distribution of larval H. chrysoscelis. Mean number of H. chrysoscelis eggs deposited was significantly decreased by the mere presence of these predators relative to controls, with three species eliciting a complete avoidance (e.g. no eggs deposited). One fish predator, Aphredoderus sayanus, was statistically indistinguishable from the control treatment. These data mimic patterns of species distribution observed in nature in response to the presence of fish, but suggest an alternative mechanism for generating patterns of prey abundance amongst habitats differing in their predator composition. Furthermore, our results expand the importance of predator induced non‐lethal effects as a process that could dramatically affect population and community level dynamics.

Ecological half-life of 137Cs in fish from a stream contaminated by nuclear reactor effluents

Radiocesium ( 137Cs) concentrations were determined during 1974, 1981 and 1998 for seven species of fish inhabiting a stream (Steel Creek) contaminated by effluents from a nuclear reactor to examine the decline of this radionuclide in a natural ecosystem. Median 137Cs concentrations were highest in Micropterus salmoides (largemouth bass) during each year of the investigation (1974=6.67 Bq g −1 dry wt. of whole body; 1981=3.72 Bq g −1; 1998=0.35 Bq g −1), but no patterns of differences were observed among Aphredoderus sayanus (pirate perch), Esox americanus (redfin pickerel), Lepomis auritus (redbreast sunfish), L. gulosus (warmouth), L. punctatus (spotted sunfish), and Notropis cummingsae (dusky shiner). Results demonstrated a rapid decline in 137Cs within fish from Steel Creek during the 24-year period. For example, 137Cs concentrations in all fish species declined significantly among years, even after accounting for radioactive decay. The observed percent declines in 137Cs concentrations of individual species were 3–4 times greater between 1974 and 1981 compared to that expected by physical decay alone, and 2–3 times greater during 1981–1998. Ecological half-lives (EHLs) of 137Cs in fish ranged from 4.43 years in A. sayanus to 6.53 years in L. gulosus. The EHL for 137Cs in all fish species combined was 5.54 years. Current levels of 137Cs in fish from Steel Creek (1.16 Bq g −1 dry wt. of whole body to below detection limits) indicate that the consumption of fish from this ecosystem poses little risk to humans and sensitive wildlife species. These results demonstrate the importance of incorporating the concept of ecological half-life into determinations concerning the length and severity of potential risks associated with radiocontaminants.

Observations on the Reproductive Ecology of Pirate Perch Aphredoderus sayanus

ABSTRACT The pirate perch (Aphredoderus sayanus) is common throughout its range; however, little is known about its reproductive ecology. Reports of spawning periods range from January to April, but it is unclear whether the pirate perch is a nest spawner or a branchial brooder. Results of this study indicated the major spawning period for pirate perch in the Atchafalaya River Basin, Louisiana, is February through March (12.6—22.8 °C). Larval pirate perch were collected in limnetic areas at 3.6-mm total length and were absent at 14-mm total length. It appears that adult pirate perch are not branchial brooders but release their adhesive eggs over leaf litter and woody debris.

Characteristics of Woody Cover Used by Brown Madtoms and Pirate Perch in Coastal Plain Streams

We measured or described microhabitat characteristics of woody debris accumulations (i.e., structural complexity, cavity space, stein diameter, suspended and benthic leaves, depth, inside and outside flow, undercut bank, and lateral position) found in headwater coastal plain streams of central Louisiana. Debris characteristics at sites used as diurnal cover by adult brown madtoms Noturus phaeus and pirate perch Aphredoderus sayanus were compared with characteristics at unoccupied debris sites. Woody debris used by brown madtoms was characterized as having greater cavity space, structural complexity, and suspended leaves. Additionally, brown madtoms more frequently occupied debris areas located beneath undercut banks or next to areas of high flow. Of all the habitat variables, undercut bank and cavity space showed the strongest relationship with the presence of brown madtoms in woody debris. Brown madtoms demonstrated no significant difference in use of debris characteristics between small (34–83 mm TL), medium (84–101 mm TL), and large (102–140 mm TL) size-groups. Relative to unoccupied sites, pirate perch more frequently occupied sites with benthic leaves and leaves suspended within the debris. Debris in deep water with undercut banks and slow inside flows were also used more by this species regardless of its size-group. The structural complexity of debris at sites occupied by large and medium size pirate perch (59–95 mm TL) was significantly greater than that at unoccupied sites and at sites occupied by small pirate perch (27–58 mm TL). Overall, submerged debris tended to have specific physical characteristics that acted as important determinants in diurnal habitat use of these two structure-associated fishes.