Water wave discrimination in the surface-feeding fish Aplocheilus lineatus

The surface-feeding fish Aplocheilus lineatus uses its cephalic lateral line to detect water surface waves caused by prey insects. The ability of Aplocheilus to discriminate between surface waves with aid of the lateral line system was tested by go/no-go conditioning. Our results show that Aplocheilus can distinguish between single-frequency surface wave stimuli with equal velocity or equal acceleration amplitudes which differ only in frequency. Frequency difference limens were about 15%, i.e. fish distinguished a 20-Hz wave stimulus from a 23-Hz stimulus in 100% of the trials. Aplocheilus can also discriminate between pure sine-wave stimuli and sine waves which show abrupt frequency changes. In contrast, fish were unable to distinguish amplitude-modulated wave stimuli (carrier frequency 20, 40 and 60 Hz, modulation frequency 10 and 20 Hz) from pure sine waves of the same frequency, even if amplitude modulation depth was 80%. Accepted: 27 December 1996     

Peripheral lateral line responses to amplitude-modulated sinusoidal wave stimuli

This report describes the responses of single afferent fibers in the posterior lateral line nerve of the goldfish, Carassius auratus, to pure tone and to amplitude-modulated sinusoidal wave stimuli generated by a dipole source (stationary vibrating sphere). Responses were characterized in terms of output-input functions relating responses to vibration amplitude, peri-stimulus time histograms relating responses to stimulus duration, and the degree of phase-locking to both the carrier frequency and the modulation frequency of the amplitude-modulated stimulus. All posterior lateral line nerve fibers responded to a pure sine wave with sustained and strongly phase-locked discharges. When stimulated with amplitude-modulated sine waves, fibers responded with strong phase-locking to the carrier frequency and, in addition, discharge rates were modulated according to the amplitude modulation frequency. However, phase-locking to the amplitude modulation frequency was weaker than phase-locking to the carrier frequency. The data indicate that the discharges of primary lateral line afferents encode both the carrier frequency and the modulation frequency of an amplitude-modulated wave stimulus. Accepted: 2 June 1999     

CHANGES IN CLICK SOURCE LEVELS WITH DISTANCE TO TARGETS: STUDIES OF FREE-RANGING WHITE-BEAKED DOLPHINS LAGENORHYNCHUS ALBIROSTRIS AND CAPTIVE HARBOUR PORPOISES PHOCOENA PHOCOENA

ABSTRACT

Probably all odontocetes use echolocation for spatial orientation and detection of prey. We used a four hydrophone “Y” array to record the high frequency clicks from free-ranging White-beaked Dolphins Lagenorhynchus albirostris and captive Harbour Porpoises Phocoena phocoena. From the recordings we calculated distances to the animals and source levels of the clicks. The recordings from White-beaked Dolphins were made in Iceland and those from Harbour Porpoises at Fjord & Baelt, Kerteminde, Denmark during prey capture. We used stringent criteria to determine which clicks could be defined as being on the acoustic axis. Two dolphin and nine porpoise click series could be used to track individual animals, which presumably focused on the array hydrophones or a fish right in front of the array. The apparent source levels of clicks in the individual tracks increased with range. One individual White-beaked Dolphin and three Harbour Porpoises regulate their output signal level to nearly compensate for one-way transmission loss while approaching a target. The other dolphin regulated the output differently. For most of the recordings the sound level at the target remains nearly constant and the echo level at the animal increases as it closes on the target.     

Stimulus discrimination and wave source localization in fishing spiders (Dolomedes triton and D. okefinokensis)

We have studied the behavioral responses of fishing spiders (Dolomedes triton and Dolomedes okefinokensis) to water surface wave stimuli. D.okefinokensis responded to click-like wave stimuli (Fig. 3C) in less than 15% of the cases. Responsiveness did not increase if up to 20 clicks were elicited in quick succession from the same spot (Fig. 5). If longer lasting concentric stimuli were offered, the spiders determined the direction (Fig. 6) and the distance (Fig. 8) to the wave source. This was true for monofrequency stimuli and for narrow-band and broadband noise stimuli. If concentric multifrequency surface waves were offered, even a fivefold decrease in stimulus amplitude did not significantly change the mean running distance of D.triton. However, if multifrequency wave stimuli with a flat wave front were presented, the spiders (D.triton) no longer determined the source distance precisely (Figs. 11, 12). Our results indicate that fishing spiders of the genus Dolomedes mainly use the curvature of a concentric wave stimulus for distance determination.     

Clicking in Shallow Rivers: Short-Range Echolocation of Irrawaddy and Ganges River Dolphins in a Shallow,Acoustically Complex Habitat

Toothed whales (Cetacea, odontoceti) use biosonar to navigate their environment and to find and catch prey. All studied toothed whale species have evolved highly directional, high-amplitude ultrasonic clicks suited for long-range echolocation of prey in open water. Little is known about the biosonar signals of toothed whale species inhabiting freshwater habitats such as endangered river dolphins. To address the evolutionary pressures shaping the echolocation signal parameters of non-marine toothed whales, we investigated the biosonar source parameters of Ganges river dolphins (Platanista gangetica gangetica) and Irrawaddy dolphins (Orcaella brevirostris) within the river systems of the Sundarban mangrove forest. Both Ganges and Irrawaddy dolphins produced echolocation clicks with a high repetition rate and low source level compared to marine species. Irrawaddy dolphins, inhabiting coastal and riverine habitats, produced a mean source level of 195 dB (max 203 dB) re 1 µPa_pp whereas Ganges river dolphins, living exclusively upriver, produced a mean source level of 184 dB (max 191) re 1 µPa_pp. These source levels are 1–2 orders of magnitude lower than those of similar sized marine delphinids and may reflect an adaptation to a shallow, acoustically complex freshwater habitat with high reverberation and acoustic clutter. The centroid frequency of Ganges river dolphin clicks are an octave lower than predicted from scaling, but with an estimated beamwidth comparable to that of porpoises. The unique bony maxillary crests found in the Platanista forehead may help achieve a higher directionality than expected using clicks nearly an octave lower than similar sized odontocetes.     

Electrosensory phase sensitivity in the weakly electric fish Eigenmannia in the detection of signals similar to its own

The electric organ discharge (EOD) of the South American knifefish Eigenmannia sp. is a permanently present wave signal of usually constant amplitude and frequency (similar to a sine wave). A fish perceives discharges of other fish as a modulation of its own. At frequency identity (F = 0 Hz) the phase difference between a fish's own electric discharge and that of another fish affects the superimposed waveform. It was unclear whether or not the electrosensory stimulus-intensity threshold as behaviourally determined depends on the phase difference between a fish's own EOD and a sine-wave stimulus (at F = 0 Hz). Also the strength of the jamming avoidance response (JAR), a discharge frequency shift away from a stimulus that is sufficiently close to the EOD frequency, as a function of phase difference was studied. Sine-wave stimuli were both frequency-clamped and phase-locked to a fish's discharge frequency (F = 0 Hz). In food-rewarded fish, the electrosensory stimulus-intensity threshold depended significantly on the phase difference between a fish's discharge and the stimulus. Stimulus-intensity thresholds were low (down to 3 V/cm, peak-to-peak) when the superimposed complex wave changed such that the shift in zero-crossings times relative to the original EOD was large but amplitude change minimal; stimulus-intensity thresholds were high (up to 16.9 V/cm, peak-to-peak) when the shift in zero-crossings times was small but amplitude change maximal. Similar results were obtained for the non-conditioned JAR: at constant supra-threshold stimulus intensities and F = 0 Hz, the phase difference significantly affected the strength of the JAR, although variability between individuals was higher than that observed in the conditioned experiments.Abbreviations ACP active phase coupling - EOD electric organ discharge - JAR jamming avoidance response - F frequency (fish) — frequency (stimulus) [Hz] - p-p peak-to-peak     

CLICK CHARACTERISTICS OF NORTHERN BOTTLENOSE WHALES (HYPEROODON AMPULLATUS)

Sounds produced by northern bottlenose whales ( Hyperoodon ampullatus ) recorded in the Gully, a submarine canyon off Nova Scotia, consisted predominately of clicks. In 428 min of recordings no whistles were heard which could unequivocally be attributed to bottlenose whales. There were two major types of click series, initially distinguished by large differences in received amplitude. Loud clicks (produced by nearby whales socializing at the surface) were rapid, with short and variable interclick intervals (mean 0.07 sec; CV 71%). The frequency spectra of these were variable and often multimodal, with peak frequencies ranging between 2 and 22 kHz (mean 11 kHz, CV 59%). Clicks received at low amplitude (produced by distant whales, presumably foraging at depth) had more consistent interclick intervals (mean 0.40 sec, CV 12.5%), generally unimodal frequency spectra with a mean peak frequency of 24 kHz (CV 7%) and 3 dB bandwidth of 4 kHz. Echolocation interclick intervals may reflect the approximate search distance of an animal, in this case 300 m, comparable to that found for sperm whales. The relationship between click frequency and the size of object being investigated, suggests that 24 kHz would be optimal for an object of approximately 6 cm or more, consistent with the size range of their squid prey.     

Effects of water conductivity on electrocommunication in the weak-electric fish Brienomyrus niger (Mormyriformes)

A characteristic electric organ discharge display in social encounters between mormyrid fish is a temporary discharge cessation. Using this response, we have investigated the useful range of electrocommunication under different water conductivity conditions in the mormyrid Brienomyrus niger. An individual fish was confined to a porous ceramic shelter tube and moved from a starting distance of 380 cm toward a similarly confined conspecific until discharge, cessation occurred. The moved fish was subsequently returned to its original, position. Water conductivity affects the peak-to-peak source voltage of the electric organ and the sensitivity of the fish's electroreceptors. Within a range of 10 to 36 000 μS/cm, the peak-to-peak amplitude of the electric organ discharge declined as a power function. At 120 μS/cm, the amplitude was 50%, and at 300μS/cm, 30% of the 10 μS/cm value. The interfish distance at which discharge cessation occurred and the associated electric field gradients were dependent on water conductivity and upon the spatial orientation of the two fish (end-to-end or parallel orientations of their shelter tubes). The respective ranges were from 135 cm and 0.02 mV/cm at 52 μS/cm (parallel orientation) to 22 cm and 0.36 mV/cm at 678 μS/cm (end-to-end orientation). When the data for both tube orientations were combined, the relationship between water conductivity (x) and the distance at which discharge cessation occurred (y) could be expressed by a power function, y=K·x^a (with K=10^2.97 and a=?0.56). When an electrically ‘silent’ fish was moved away from its conspecific, a discharge resumption in the form of a high-frequency rebound occasionally effected changes in the other fish's discharge activity at distances up to 157 cm (with an associated electric, field gradient of 0.01 mV/cm under the lowest conductivity condition).     

Transmission of Podisus maculiventris tremulatory signals through plants

Males of the predaceous stink bug Podisus maculiventris (Say) (Heteroptera: Pentatomidae: Asopinae) emit low frequency tremulatory signals. Laser vibrometry was used to record and analyze naturally emitted signals, focusing on variation in signal velocity and frequency during transmission through plants (Phaseolus vulgaris L. and Plumbago auriculata Lam.) as a function of distance from the vibrational source. Signal velocity varied individually between 2 and 15 mm/s recorded on a plant close to the calling male and decreased by 0.3 to 1.5 dB/cm on bean and 0.3 to 0.9 dB/cm on plumbago. The dominant frequency of signals was variable at frequencies below 50 Hz. On bean frequencies centered around 10 Hz or 20 Hz were dominant for signals recorded at the source. Transmission through bean resulted in an increase in the 20 Hz peak relative to other frequencies in the signal. Variation of the dominant frequencies of signals transmitted through plumbago stems were more predictable, showing typical changes in amplitude relative to the distance from the source. The regular variation of the dominant frequency along the stem with linear increase of signal velocity at decreasing distance from the source may provide plant-dwelling insects with information about the distance to the calling individual.     

Velocity and dominance affect prey capture and microhabitat selection in juvenile Chinook (<Emphasis Type="Italic">Oncorhynchus tshawytscha</Emphasis>)

Habitat selection is an important phenomenon that may greatly affect individual fitness. Using an artificial stream, we examined the relationship between the percentage of prey captured, reactive distance, dominance, and water velocity for juvenile Chinook Salmon (Oncorhynchus tshawytscha) from the Chena River, Alaska, and tested the fitness-based microhabitat selection model of Grossman et al. (Ecol Freshw Fish 11:2–10, 2002). Recent declines in the abundance of Chinook accentuate our need for habitat selection studies on this species. We conducted three experiments: two with single fish (1st N?=?27, fish SL 58–84 mm, 2nd N?=?14, fish SL 49–56 mm) and one with pairs of dominant and subordinate fish (N?=?10 pairs, 64–96 mm, mean difference in SL?=?7 mm). We placed individual or pairs of fish in an artificial flume and recorded reactive distance and the percent prey capture with individual dead brine shrimp (Artemia spp.) as prey. Prey were presented at 10 cm/s velocity intervals ranging from 10 to 60 cm/s; velocities found in the natural habitat. Mean reactive distance in single fish experiments (henceforth SFE) averaged 33 and 29 cm respectively, and was not related to velocity. We detected a negative, curvilinear relationship between velocity and percent prey capture. Holding velocities for juvenile Chinook were significantly lower than prey capture velocities. The Grossman et al. (Ecol Freshw Fish 11:2–10, 2002) model yielded an optimal focal-point velocity prediction of 35 cm/s for juvenile Chinook, however focal-point velocities occupied by juveniles in the Chena River averaged 12 cm/s. Predicted optimal velocities were present in the Chena River; hence, this discrepancy suggests that other factors such as distraction from drifting debris or predation risk influenced habitat selection. There were no differences in reactive distances or holding velocity/capture velocity relationships for dominant and subordinate fish; however, dominants captured significantly more prey than subordinates. Being subordinate resulted in a decrease of 61% in mean percent prey capture (the difference between what was captured by the fish alone versus the difference with a dominant), whereas the mean cost to fish with dominant rank was a 21% decline between the percentage captured alone versus that with a subordinate.     

Independently evolved jamming avoidance responses employ identical computational algorithms: a behavioral study of the African electric fish,Gymnarchus niloticus

An African electric fish, Gymnarchus, and a South American electric fish, Eigenmannia, are believed to have evolved their electrosensory systems independently. Both fishes, nevertheless, gradually shift the frequency of electric organ discharge away when they encounter a neighbor of a similar discharge frequency. Computational algorithms employed by Gymnarchus for this jamming avoidance response have been identified in this study for comparison with those of extensively studied Eigenmannia.

1. Gymnarchus determines whether it should raise or lower its discharge frequency based solely upon the signal mixture of its own reafferent and the exafferent signal from a neighbor, and does not internally refer to the pacemaker command signal which drives its own discharge.
2. The signal mixture is analyzed in terms of the time courses of amplitude modulation and phase modulation at each area of the body surface.
3. Phase of the signal mixture at each area is compared with that of another area for the detection of phase modulation.
4. Unambiguous information necessary for the jamming avoidance response is extracted by integrating information from all body areas each of which yields ambiguous information.
5. These computational features are identical to those of Eigenmannia, suggesting that the neural circuit for jamming avoidance responses may have evolved from preexisting mechanisms for electrolocation in both fishes.

     

Spatial clumping of food increases its monopolization and defense by convict cichlids, Cichlasoma nigrofasciatum

The hypothesis that resource monopolization and defense increaseas the spatial clumping of resources increases was tested usinggroups of three convict cichlids competing for 120 Daphnia magnaprey. Spatial clumping was manipulated by varying the distance(3, 20, or 40 cm) between three tubes through which the preyappeared. As predicted, monopolization of prey (percentage eatenby the dominant fish) and frequency of aggression (chases perminute) by dominant fish increased significantly as the distancebetween the tubes decreased. However, there was no evidenceof individual flexibility in the aggressiveness (percentageof conspecifics chased) of dominant fish across treatments.Differences among dominant fish in aggressiveness were positivelycorrelated with their ability to monopolize prey, but the strengthof the correlation decreased as the distance between the tubesincreased. Aggression appears to be a more effective mechanismof interference competition when resources are clumped thanwhen resources are dispersed.     

Discrimination performance and echolocation signal integration requirements for target detection and distance determination in the CF/FM bat,Noctilio albiventris

Summary Bats of the speciesNoctilio albiventris were trained to detect the presence of a target or to discriminate differences in target distance by means of echolocation. During the discrimination trials, the bats emitted pairs of pulses at a rate of 7–10/s. The first was an 8 ms constant frequency (CF) signal at about 75 kHz. This was followed after about 28 ms by a short-constant frequency/ frequency modulated (short-CF/FM) signal composed of a 6 ms CF component at about 75 kHz terminating in a 2 ms FM component sweeping downward to about 57 kHz. There was no apparent difference in the pulse structure or emission pattern used for any of the tasks. The orientation sounds of bats flying in the laboratory and hunting prey under natural conditions follow the same general pattern but differ in interesting ways.The bats were able to discriminate a difference in target distance of 13 mm between two simultaneously presented targets and of 30 mm between single sequentially presented targets around an absolute distance of 35 cm, using a criterion of 75% correct responses.The bats were unable to detect the presence of the target or to discriminate distance in the presence of continuous white noise of 54 dB or higher SPL. Under conditions of continuous white noise, the bats increased their pulse repetition rate and the relative proportion of CF/FM pulses.The bats required a minimum of 1–2 successive CF/FM pulse-echo pairs for target detection and 2–3 to discriminate a 5 cm difference in distance. When the distance discrimination tasks were made more difficult by reducing the difference in distance between the two targets the bats needed to integrate information from a greater number of successive CF/FM pulse-echo pairs to make the discrimination.Abbreviations CF constant frequency - FM frequency modulation     

Prey detection in antlions: propagation of vibrational signals deep into the sand

Trap‐building antlion larvae detect their prey according to the substrate vibrations produced during movement of the prey on the sand surface. Although most studies are devoted to surface vibrational waves, in the present study, we determine the role of vibrations travelling through deeper sand layers. A behavioural experiment confirms that vibrational stimuli from prey insects on the surface of the sand stimulate the antlions buried in deeper sand layers to move towards the surface. Sand depth and particle size both have a strong effect on signal transmission. The damping coefficient (α₁₀) varies from 0.49 dB to 3.30 dB cm^?1 and depends on frequency (in the range from 100 to 300 Hz), particle size (from finest to coarse sand) and distance from the source of the vibrations. The deeper the sand, the narrower the frequency range of the signal becomes. Sand is a filter for higher frequencies. The smaller the sand particles, the more intense the filtering becomes. Fine sand with a mean sand particle size of 360 μm is a more efficient filter than coarse sand; consequently, high frequencies (> 2.5 kHz) are eliminated at a depth of 3 cm. Mean frequency depends on both depth and particle size. However, low frequency signals still propagate at a certain distance, which is biologically important in prey detection. Although the most efficient signal propagation appears to occur in coarse sand, it contains overly large particles that are inconvenient for relatively small antlion larvae. Predators seek a compromise between fine and coarse sand choosing medium sand.     

Reduction of Reactive Distance and Foraging Success in Smallmouth Bass, Micropterus dolomieu, Exposed to Elevated Turbidity Levels

We determined how turbidity affected the reactive distance and foraging success of smallmouth bass, Micropterus dolomieu. Smallmouth bass reactive distance decreased exponentially with increasing turbidity, from 65cm in clear water to 10cm at the highest turbidity. Turbidity significantly decreased the probability of a fish reacting to a prey item, but did not influence foraging success following reaction to the prey. Elevated turbidity may reduce stream fish foraging efficiency and decrease prey consumption.     

Circumpapillary Course of Retinal Pigment Epithelium Can Be Fit to Sine Wave and Amplitude of Sine Wave Is Significantly Correlated with Ovality Ratio of Optic Disc

The purpose of this study was to develop a method of quantifying the degree of optic disc tilt in normal eyes. This was a prospective, observational cross sectional study of 126 right eyes of 126 healthy volunteers. The optic disc tilt was determined from the circular peripapillary optical coherence tomographic (OCT) scan images. The course of the retinal pigment epithelium (RPE) layer in the peripapillary cross sectional scan images was fit to a sine wave curve, and the amplitude of the sine curve was used to reflect the degree of the optic disc tilt in the optical axis. The repeatability of the amplitude determinations was calculated. The correlation between the amplitude and the ovality ratio of the optic disc was determined. The correlation between the amplitude and the body height was also calculated. The mean amplitudewas 36.6 ± 17.5 pixels, which was significantly and inversely correlated with the ovality ratio of the optic disc (R = -0.59, P<0.001). The intra-rater and inter-rater correlation coefficients of the amplitude were significant high (P<0.001, both). The amplitude was significantly and inversely correlated with the body height (R = -0.38, P<0.001), but not with the axial length. In conclusion, a sine wave function can be used to describe the course of the RPE in the circumpapillary OCT images. The results indicate that the amplitude of the sine wave can be used to represent the degree of optic disc tilt. Thus, the sine wave analyses can be used as a quantifiable and repeatable method to determine the optic disc tilt.     

Meiofauna as food source for small-sized demersal fish in the southern North Sea

Meiofauna play an essential role in the diet of small and juvenile fish. However, it is less well documented which meiofaunal prey groups in the sediment are eaten by fish. Trophic relationships between five demersal fish species (solenette, goby, scaldfish, dab <20 cm and plaice <20 cm) and meiofaunal prey were investigated by means of comparing sediment samples and fish stomach contents collected seasonally between January 2009 and January 2010 in the German Bight. In all seasons, meiofauna in the sediment was numerically dominated by nematodes, whereas harpacticoids dominated in terms of occurrence and biomass. Between autumn and spring, the harpacticoid community was characterized by Pseudobradya minor and Halectinosoma canaliculatum, and in summer by Longipedia coronata. Meiofaunal prey dominated the diets of solenette and gobies in all seasons, occurred only seasonally in the diet of scaldfish and dab, and was completely absent in the diet of plaice. For all fish species (excluding plaice) and in each season, harpacticoids were the most important meiofauna prey group in terms of occurrence, abundance and biomass. High values of Ivlev’s index of selectivity for Pseudobradya spp. in winter and Longipedia spp. in summer provided evidence that predation on harpacticoids was species-selective, even though both harpacticoids co-occurred in high densities in the sediments. Most surficial feeding strategies of the studied fish species and emergent behaviours of Pseudobradya spp. and Longipedia spp. might have caused this prey selection. With increasing fish sizes, harpacticoid prey densities decreased in the fish stomachs, indicating a diet change towards larger benthic prey during the ontogeny of all fish species investigated.     

Behavioural aspects of predation by juvenile Atlantic salmon (Salmo salar L.) on particulate,drifting prey

A recirculatory flume tank simulating a simplified stream environment was used to study the feeding behaviour of juvenile Atlantic salmon (Salmo salar L.), 5.1 to 9.4 cm in fork length (from tip of snout to fork of tail), on artificial particulate prey passively drifting in the water current. Changes in feeding behaviour at two different times of the year and when fish were presented with prey of different sizes are described and quantified. Responsiveness to food was greatly reduced in autumn as compared to summer. The maximum distances at which prey elicited a response decreased in autumn to 40% of the summer value, and the maximum distances which fish traversed in order to capture prey decreased by 80% over the same period. During the peak growing season, the response to a range of prey sizes from 0.013 to 0.102 × fish fork length was directly related to prey size and could be accounted for on the basis of visual theory alone. Capture distances were closely related to fixation distances. Maximum capture distance increased to a peak value for prey of between 0.025 and 0.069 × fork length, while larger prey were never captured and the smallest prey rarely evoked a response. Prey size selectivity also operated after capture, through rejection versus retention of the prey.     

Piscivorous cyprinid fish modulates suction feeding kinematics to capture elusive prey

Previous studies have shown that evasive prey generally elicit a different kinematical pattern of prey capture from suction feeding fish compared to non-evasive types of prey. However, no evidence exists that predatory fish can modulate their prey capture kinematics in response to whether or not an elusive prey performs an escape response. Here, we analyse prey capture kinematics of a specialist piscivore (asp, Aspius aspius) during feeding on untethered, live goldfish, which regularly displayed escape attempts when attacked by the asp. Significant modulation occurred in function of the escape attempts of prey: mouth opening was prolonged and increased in magnitude, and one individual also showed an increased hyoid depression when feeding on prey trying to escape. As the orientation of the prey with respect to the predator prior to the start of mouth opening was related to the probability of observing an escape attempt, asp could theoretically perform this type of modulation by a priori choosing a pre-programmed motor pattern. However, since contact between the prey and the asp's mouth appeared to be a factor improving the timing of mouth closing, this fine-tuning of prey capture kinematics is more likely to be caused by reflexive neural feedback control.     

A model for switching between particulate-feeding and filter-feeding in the common bream,Abramis brama

Synopsis A model has been developed to describe the process of switching between particulate- and filter-feeding in common bream, Abramis brama, in relation to fish size and zooplankton density. The model assumes that the encounter rate of fish and zooplankton is determined by the density of zooplankton and the swimming speed of fish. However, if zooplankton density is so high as to allow at least one prey to be engulfed per random snap, the encounter rate is determined by the volume of the buccal cavity and by zooplankton density, but is independent of swimming speed. The snapping frequency will be maximal at the time of switching, decreasing with increasing zooplankton density because of the extra time needed for intra-oral prey handling. The model predicted switching from particulate- to filter-feeding only for bream> 15 cm standard length at zooplankton densities < 500 l^-1. The snap frequency of six size classes of bream (7.5, 10.4, 12.5, 15, 24 and 29.5 cm) was measured at varying densities of Daphnia. The model predictions for snap frequencies of all size classes corresponded to the highest values observed. The average of the observed snap frequencies was only 50% of the predicted values, probably because the calculated average distance between prey animals assumed an ideal swimming route of the fish and error-free vision for particulate-feeding, and the handling time was ignored.