The role of sense organs in the feeding behaviour of
Chinese perch
X. F. LIANG*, J. K. LIU AND B. Y. HUANG*
*Department of Biology, Jinan University, Guangzhou 510632 and Institute of
Hydrobiology, Academia Sinica, Wuhan 430072, Peoples Republic of China
(Received 13 June 1997, Accepted 13 January 1998)
Experiments were conducted to identify the roles of the individual sense organs in the feeding
behaviour of Chinese perch Siniperca chuatsi by determining the consumption of natural food after
selective removal or blocking of eyes, lateral lines and olfactory organs, and also by observing the
behavioural response to visual, mechanical and chemical stimulation by artificial prey. Chinese
perch were able to feed properly on live prey fish when either eyes or lateral lines were intact or
functional, but could scarcely feed without these two senses. Chinese perch recognized its prey
by vision through the perception of motion and shape, and showed a greater dependence on
vision in predation when both visual and mechanical cues were available. Chemical stimulation
by natural food could not elicit any feeding response in Chinese perch, and gustation was only
important to the fish for the last stage of food discrimination in the oropharyngeal cavity. The
sensory basis of Chinese perch in feeding is well adapted to its nocturnal stalking hunting
strategy, and also explains its peculiar food habit of accepting live prey fish only and refusing
dead prey fish or artificial diets.
INTRODUCTION
The feeding behaviour of fishes results from the interaction of a variety of sense
organs that are receptive to light, mechanical, chemical, and electromagnetic
stimuli, depending on the habitat and the cues produced by potential food
(Keenleyside, 1979; Atema, 1980; Blaxter, 1988). The roles and function of
each of these individual senses have been reported extensively in the literature
(Ali, 1975; Schuijf & Hawkins, 1976; Lythgoe, 1979, 1980; Atema, 1980, 1988;
Tavolga et al., 1981; Fernald, 1988; Bleckmann et al., 1989; Montgomery, 1989;
Enger et al., 1989; Hara, 1992). In spite of existing knowledge of each of the
sensory organs and recognition of their co-participation in the feeding behaviour
of fishes, little information is available which identifies the individual involvement
of each sense in the feeding process (Tamura, 1952; Bardach et al., 1959;
Roberts & Winn, 1962; Daugherty et al., 1976; Batty & Hoyt, 1995).
The Chinese perch Siniperca chuatsi (Basilewsky) (Percichthyidae), or mandarin
fish, is a demersal piscivore, found only in the fresh waters of China and the
River Amur along the Russian borderlands. It has great commercial fishery
value due to its large size, fast growth, and delicious flesh (Liu & He, 1992). The
Chinese perch has very specialized feeding habits, for as soon as they start
feeding the fry of this fish feed solely on fry of other fish species (Chiang, 1959).
In rearing conditions, Chinese perch have been found to accept live prey fish
only, and refuse dead prey fish or artificial diets. Fry and fingerlings of mainly
Chinese-cultivated fish are used as the sole source of feed in the intensive culture
of Chinese perch. Study of the sensory mechanisms of Chinese perch in feeding
is of special significance, and thus has drawn much attention (Liang, 1996a). Wu
& Hardy (1988) observed Chinese perch to attack prey fish outside the aquarium,
and suggested that it recognized its prey using vision. Zhu & Fang (1991)
reported that Chinese perch stayed more frequently near a perforated opaque
cylinder containing live prey fish than near a control without prey fish, and
suggested that olfaction can be used in the searching behaviour. However, this
conclusion may not be justified since not only chemical stimuli, but also other
stimuli, e.g. a hydromechanical stimulus, may have been perceived by Chinese
perch through the holes in the cylinder of their experiment.
The purpose of this investigation was to identify the roles of the individual
sense organs in the feeding behaviour of Chinese perch by determining the
natural food consumption after selective removal or blocking of eyes, lateral
lines and nares, and by observing the behavioural response to visual, mechanical
and chemical stimulation by artificial prey.
MATERIALS AND METHODS
SOURCE AND ACCLIMATIZATION OF FISH
Yearlings of Chinese perch were caught by seining in the Yangtze River at Jinkou
Town, Hubei Province, China. They were grouped according to size, and 1050 fish were
held in each of eight concrete indoor pools of 1 m3 volume in the Institute of
Hydrobiology, Academia Sinica. Each pool had several stones on the bottom to form
crevices for the fish to use as refuges. The water temperature ranged from 20·0 to 22·5) C.
The photoperiod was 12L : 12D. Lighting was provided by two 40-W fluorescent lights.
The fish were acclimatized to the laboratory conditions for a month before the start of the
experiment. During acclimatization, they were fed once every day to satiation on live
Pseudorasbora parva (Temminck & Schlegel).
EXPERIMENT ON FOOD CONSUMPTION AFTER SELECTIVE SENSE
BLOCKING
At the start of the experiment, uniform-size yearlings of the fish were selected and
distributed randomly into 24 tanks. Two fish were placed into each tank. The lengths of
the fish were 1012 cm. The tanks were divided into eight groups. Triplicate tanks were
assigned to each treatment of selective sense blocking (see Table I). The eyes were
enucleated after anaesthesia, the nares of the olfactory system were plugged with vaseline,
and the lateral lines were blocked by adding 0·1 mM CoCl2 to the rearing water (Karlsen
& Sand, 1987). Blinded Chinese perch could feed on live prey fish normally 2 or 3 days
after the surgical operation and were used in the experiment after a week. Experimental
fish with blocked lateral lines and nares were used immediately. The concentration of
the Co2+ was kept constant during the experiment, and the plugging of the nares was
inspected every day. Each tank (58#43#43 cm) contained 82·3 l of calcium-free
dechlorinated fresh water, and had six dark plastic holes on the bottom for the fish to use
as refuges. Eighty per cent of the water was replenished every day. Aeration was
provided and the oxygen content was kept above 5 mg l"1. The water temperature
ranged from 20·0 to 22·5) C. The photoperiod was 12L : 12D. Lighting was provided by
two 40-W fluorescent lights. Excess live P. parva (length, 3·54·0 cm) were put into each
tank on a daily basis each morning. The number of P. parva eaten was counted at the
same time (0830 hours) the following day and the number of P. parva eaten was then
added to each tank. The daily ration in prey fish number was calculated for each tank as:
daily ration=prey fish number of initial day"that of the following day/Chinese perch
number. The experiment lasted 40 days. The daily ration was the number of prey fish
SENSE ORGANS AND FEEDING IN CHINESE PERCH 1059
eaten per Chinese perch per day averaged for all 40 days in each tank. Data of three
tanks were used as replicates for statistical analysis (n=3).
EXPERIMENT ON FEEDING RESPONSE TO ARTIFICIAL STIMULI
The experiment was conducted in two aquaria (120#60#60 cm). Both aquaria were
divided into two parts (120#50#60 cm and 120#10#60 cm, respectively) by a glass
screen. Twelve Chinese perch (1215 cm total length) were held in the larger compartment,
which had 12 dark plastic refuges on the opposite side of the bottom to the glass
screen. The larger compartments were provided with aeration and flow-through
dechlorinated tap water at a rate of 800 ml min"1. Oxygen content was kept above 5 mg
l"1, and the water temperature ranged between 27·5 and 30·0) C. Purely visual stimulation
was obtained by holding artificial prey in the smaller compartment. Because of the
glass screen between the Chinese perch and the artificial prey, the fish could recognize
them only by vision. The smaller compartments were not provided with water. Fresh
P. parva killed by freezing and artificial prey of five shapes (all 2·5 cm in length or radius)
made of silvered stiff paper (see Table III) were used as visual stimuli. The distance
between the experimental fish and the artificial prey was about 35 cm. The motion of
artificial prey was driven by an experimental device, which had a motor and belt similar
to those used by Tamura (1952), and an electronic controller to make artificial prey move
in a predetermined fashion. The artificial prey hung from the belt by a length of thin
transparent thread. Since the belt was almost as long as the aquarium, the artificial prey
could be driven to move to and fro between the left and right sides of the aquarium in
the smaller compartment. The speeds of continuous and intermittent motion were
020 cm s"1, and the duration and interval of intermittent motion were both 0·9 s.
The mechanical stimulation without visual cues was obtained by placing artificial prey,
vibrating at about 2 Hz, 2 cm away from the mouth of a blinded Chinese perch in the
larger compartment. The vibration was driven by an experimental device similar to that
used by Tamura (1952). The device contained a motor and crank and produced a simple
harmonic oscillation, which could make the artificial prey vibrate simultaneously through
a steel needle. Fresh P. parva, rotten P. parva (made so by laying fresh fish in rearing
water at 30) C for 24 h), and custom-made plasticine fish, which were all 2·5 cm in length,
were used as artificial prey to determine the effect of the chemical nature of food on the
feeding response of Chinese perch.
Dual stimulation of eyes and lateral lines was obtained by placing a vibrating plasticine
ball (0·9 cm in radius) 2 cm away from the mouth of a normal fish in the larger
compartment, and a blinded fish was tested as an experimental control.
The experiments were conducted in the morning (08300900 hours) each day, and the
experimental fish were fed once every day in the evening to satiation on live P. parva. The
two aquaria were tested alternately, and all the 12 Chinese perch in either aquarium were
replaced totally after 1 month. At the experimental illumination background (2050 lx at
the water surface), all 12 Chinese perch in the larger compartment of either aquarium hid
inside the 12 dark refuges, normally one fish in one hole, with eyes gazing outward.
Whenever a prey fish appeared outside a refuge, the Chinese perch inside would stalk first
and then attack suddenly. The visual feeding responses of Chinese perch to artificial prey
behind the glass screen in the smaller compartment were classified as no response,
stalking, and attacking. Stalking rate and attacking rate were counted over 2 min, and
calculated as the following: stalking rate=number of stalking events/observation time
(min)#fish number; and attacking rate=number of attacking events/observation time
(min)#fish number. The stalking rate and attacking rate were both based on 12 fish in
each aquarium averaged for 15 days. Data of three aquaria were used as replicates for
statistical analysis (n=3). During the determination of the visual feeding response of
Chinese perch to artificial prey behind the glass screen occasionally they were fed small
live Oryzias latipes Temminck & Schlegel in order to prevent habituation and a cessation
of response.
The non-visual feeding responses of blinded Chinese perch to artificial prey in the
larger compartment were classified as no response, attacking, and swallowing. The
occurrence of attacking or swallowing behaviour was observed over a 2-min period, and
1060 X. F. LIANG ET AL.
its frequency was determined in 10 replicates (10 individuals in the process of searching
for food in each aquarium were chosen randomly as replicates) and calculated as follows:
frequency of occurrence of attacking behaviour=number of observations with occurrence
of attacking behaviour/number of replicates; and frequency of occurrence of swallowing
behaviour=number of observations with occurrence of swallowing behaviour/number of
replicates. The frequency of occurrence of attacking or swallowing behaviour was based
on 12 fish in each aquarium averaged for 15 days. Data of three aquaria were used as
replicates for statistical analysis (n=3).
STATISTICAL ANALYSIS
Data were subjected to one-way analysis of variance (ANOVA). NewmanKeuls
multiple range test was used to identify differences among means at the 0·05 level.
RESULTS
FOOD CONSUMPTION AFTER SELECTIVE SENSE BLOCKING
When one of the three sense organs (eyes, lateral lines, and nares) was blocked,
Chinese perch fed as much as the control fish (Table I). When two of the three
sense organs were blocked, Chinese perch could still feed normally except for the
case in which both eyes and lateral lines were blocked simultaneously. In the
latter case, Chinese perch could hardly feed, no matter whether olfaction was
working or not.
FEEDING RESPONSE TO VISUAL STIMULI
Chinese perch captured its prey by stalking first and attacking afterwards. The
typical sequence of events was: eyeing the prey, turning and slowly moving
towards it, leaping forward, snapping and finally swallowing it.
For the moving fresh prey fish (P. parva) driven by an experimental device, the
speeds of which were comparable to their natural swimming speeds, Chinese
perch stalked and attacked most frequently the intermittently fast-moving
(í¢10 cm s"1) prey fish, stalked and attacked less frequently the intermittently
TABLE I. Effect of selective blocking of sense organs on the food
consumption of Chinese perch (mean&S.E.)
Sense working (+) or blocked (") Number of
prey fish (P. parva)
consumed Vision Lateral lines Olfaction day"1
+ + + 1·5&0·25a
" + + 1·2&0·23a
+ " + 1·3&0·16a
+ + " 1·3&0·23a
+ " " 1·3&0·12a
" + " 1·2&0·25a
" " + 0·045&0·041b
" " " 0·005&0·005b
Values with the same superscript are not significantly different from each
other at the 0·05 level.
SENSE ORGANS AND FEEDING IN CHINESE PERCH 1061
or continuously slow-moving (í¡5 cm s"1) prey fish, and stalked most
frequently but scarcely or never attacked the continuously fast-moving prey fish
(Table II).
For the intermittently slow-moving (í=5 cm s"1) artificial prey, Chinese
perch stalked all the five shapes but attacked only two of them (no. 1 and no. 5)
(Table III).
FEEDING RESPONSE TO NON-VISUAL STIMULI
Blinded Chinese perch were able to attack various types of artificial prey
vibrating at low frequency (about 2 Hz), but did not respond to the still ones
(Table IV). The chemical nature of artificial prey did not affect the frequency
of occurrence of attacking behaviour, but greatly affected the frequency of
TABLE II. Visual feeding response of Chinese perch to devicedriven
fresh prey fish (P. parva) of different motion characters
behind a glass screen (mean&S.E.)
Speed of
artificial
prey (cm s"1)
Mode of
motion
Stalking rate
(events min"1)
Attacking rate
(events min"1)
0 Still 0&0a 0&0a
2 Continuous 0·44&0·034b 0·44&0·034b
2 Intermittent 0·49&0·036b 0·49&0·036b
5 Continuous 0·55&0·043c 0·12&0·017c
5 Intermittent 0·59&0·045c 0·59&0·045d
10 Continuous 0·89&0·030d 0·0032&0·013a
10 Intermittent 0·83&0·046d 0·83&0·046e
20 Continuous 0·96&0·031e 0&0a
20 Intermittent 0·95&0·049e 0·95&0·049e
Values in the same column with the same superscript are not significantly
different from each other at the 0·05 level.
TABLE III. Visual feeding response of Chinese perch to intermittently
slow-moving artificial prey of different shapes behind a
glass screen (mean&S.E.)
No. Shape of
artificial prey
Stalking rate
(events min"1)
Attacking rate
(events min"1)
1 . 0·50&0·043a 0·18&0·012a
2 . 0·24&0·022b 0&0b
3 . 0·25&0·026b 0&0b
4 , 0·25&0·023b 0&0b
5 , 0·70&0·069c 0·45&0·035c
Values in the same column with the same superscript are not significantly
different from each other at the 0·05 level.
1062 X. F. LIANG ET AL.
occurrence of swallowing behaviour. Chinese perch swallowed only fresh prey
fish, and rejected rotten prey fish or plasticine fish.
FEEDING RESPONSE TO DUAL STIMULATION OF EYES AND LATERAL
LINES
The feeding response of Chinese perch to mechanical stimuli with or without
visual cues is significantly different (P<0·05). Chinese perch attacked a small
vibrating plasticine ball at the frequency of 43&3·9% (mean&S.E.) when their
eyes had been enucleated, but they did not respond to it when their eyes were
functional.
DISCUSSION
The experiments reported here indicate that Chinese perch use mainly vision
and mechanoreception for the detection and capture of prey, and gustation in the
oropharyngeal cavity is essential for the fish to swallow prey. These results
support the earlier work of Wu & Hardy (1988) and Zhu & Fang (1991), who
found that Chinese perch relied on vision in feeding. In addition, this study
demonstrated the importance of mechanoreception for Chinese perch to capture
prey, and found that pure chemical cues of prey could not elicit any searching or
attacking behaviour by the fish. Zhu & Fang (1991) observed that Chinese perch
stayed more frequently near a perforated opaque cylinder containing live prey
fish than near a control without prey fish. The Chinese perch appears to detect
live prey fish by perceiving hydromechanical stimulation instead of chemical
stimulation through the holes in such cylinders. It is also possible that the
Chinese perch could see some motion of prey through the holes. As prey fish
would pass there would be changes in the light intensity coming through the
holes that might cause a predator to investigate. In the present experiment,
Chinese perch could hardly feed after both eyes and lateral lines were blocked
simultaneously, indicating that the other sense organs of Chinese perch have
little function in seeking and obtaining prey. Enger et al. (1989) found that
bluegill sunfish Lepomis macrochirus Rafinesque with simultaneously blocked
eyes and lateral lines, could still attack live goldfish Carassius auratus L.
successfully in a very small tank using the tactile sense (the frequency of
TABLE IV. Non-visual feeding response of blinded Chinese perch to various still and
vibrating artificial prey (mean&S.E.)
Artificial prey
Frequency of
occurrence of
attacking behaviour
(%)
Frequency of
occurrence of
swallowing behaviour
(%)
Vibrating fresh prey fish (P. parva) 55&5·1a 55&5·1a
Vibrating rotten prey fish (P. parva) 54&4·9a 3·5&0·42b
Vibrating plasticine fish 51&5·5a 0&0c
Still artificial prey 0&0b 0&0c
Values in the same column with the same superscript are not significantly different from each other at
the 0·05 level.
SENSE ORGANS AND FEEDING IN CHINESE PERCH 1063
occurrence of attacking behaviour after touch was 58%). This difference may be
caused by a species-specific sensory basis of fish feeding as well as different
experimental conditions in the laboratory. Few data are available on the
function of tactility in the feeding behaviour of piscivorous fish fed live natural
prey fish. Experiments using easily captured prey types, high prey density, small
experimental tanks, and a simply-structured environment, will enable the
enhanced involvement of some sense organs that may not play an important role
in fish feeding in the field. The effect of experimental conditions on the
utilization of senses in the feeding behaviour of Chinese perch needs further
study.
The Chinese perch is a typical stalking predator, and has a split-head colour
pattern in the form of a stripe running from the snout back along the mid-dorsal
line to the dorsal fin, which may well allow the fish to approach its prey
undetected. As in other stalking predators, Chinese perch rely on vision for
following the movement of prey before leaping forward and snapping at them
(Keenleyside, 1979). Liang et al. (1995) found that Chinese perch hid in shelter
in the deep water of Fuqiaohe Reservoir, China, during daylight, and attacked
suddenly and rapidly prey fish swimming accidentally near the shelter. At night
the Chinese perch left its shelter, and searched for prey fish in shallow water. The
water depth of the feeding area at night was related significantly to the intensity
of ambient light (moonlight or starlight). Chinese perch preferred to feed in
deeper water when the ambient light intensity at the water surface was higher.
The ability of Chinese perch to capture prey fish in low light and even in total
darkness supports the findings of the present behavioural experiments, indicating
that mechanoreception is especially important in addition to vision in the feeding
behaviour of Chinese perch.
The prey fish of Chinese perch are mostly diurnal fishes (Chiang, 1959), whose
eyes have colour vision and high acuity, but cannot function at night (Pulotasov,
1968). Consequently, these species have a great ability to avoid predation during
the day time but are seized relatively easily at night. As shown by electrophysiological
and histological studies, the retina of Chinese perch which possesses rare
cones is well suited to dim light vision and it improves its photo-sensitivity by the
abandonment of colour vision and a decrease in acuity (Liang, 1994; Liang et al.,
1994). The experiment reported here showed that the eyes of Chinese perch can
perceive the motion of its prey fish within a certain distance (at least 35120 cm),
which determines the far-range stalking behaviour of the fish. In addition, its
eyes can perceive the approximate shape of its prey fish within closer range,
which determines the near-range stalking and attacking behaviour of the fish.
The silvery body of the prey fish of Chinese perch is obviously distinguishable
against a dark background, and is also large enough to be recognized easily. It
is thus advantageous for Chinese perch to catch prey fish at night through the
perception of motion and shape with the help of its well-developed scotopic
vision.
The habitat of Chinese perch is often rich in macrophytes and sometimes very
turbid during the rainy season. Furthermore, the intensity of illumination on the
bottom at night is extremely low. The vision of Chinese perch may be greatly
restricted in such circumstances, and it is necessary that a non-visual sense of the
fish should aid in feeding. The study of the cephalic lateral line of Chinese perch
1064 X. F. LIANG ET AL.
has revealed rather elaborate supra- and infra-orbital canals, which are highly
sensitive to hydromechanical stimulation (Liang, 1996b). Results reported here
confirmed that the lateral line of Chinese perch can serve as an effective
substitute sensory organ for eyes to detect moving prey. Tricas & Highstein
(1990) showed that visual stimuli can activate the octavolateralis efferent system
and inhibit the lateral line primary afferent activity during predation in the
free-swimming toadfish Opsanus tau L. Janssen & Corcoran (1993) found that
lateral line stimuli can override vision to determine green sunfish Lepomis
cyanellus Rafinesque strike trajectory, and suggested that the class of neurons in
green sunfish inhibited by visual stimuli may not be involved in feeding. Such
fish may be more sensitive to prey because unaffected neurons responsive to prey
signals become relatively more important. The present results also demonstrated
that the hydrodynamically elicited feeding response of Chinese perch can be
inhibited by the presence of visual cues of non-prey. These results support the
work of Tricas & Highstein (1990), and also indicate that different fish species
may have different relationships between vision and the lateral line in predation.
Because prey discrimination in low light and especially in total darkness may
not be very accurate, it is reasonable for Chinese perch to recognize the captured
object again before ingestion. The study of taste buds in the oropharyngeal
cavity of Chinese perch showed mostly two kinds of elevated taste buds in close
association with teeth, which are sensitive to both chemical and mechanical
stimuli (Liang, 1996c). The present results confirmed that taste buds in the
oropharyngeal cavity of Chinese perch are used in swallowing prey. As tactility
is of the same importance as gustation for the discrimination of large food in the
oropharyngeal cavity, it can make the final stage of food discrimination more
reliable for Chinese perch to have mainly elevated taste buds and determine
whether to ingest the captured object or not through both chemical and
mechanical information.
The peculiar feeding habit of Chinese perch of accepting live prey fish only and
refusing dead prey fish or artificial diets may be explained by these findings.
Unlike Ictalurus or Cyprinus spp. (the main warm-water cultivated fishes),
chemical stimuli of food cannot elicit any feeding response in Chinese perch and
gustation is used by the fish only in swallowing prey (Hara, 1992). Although
Salmo spp. (the main cold-water cultivated fishes) also show the selection of
food motion and shape to some extent, the offered food pellet can be captured
immediately before it falls down to the bottom of the tank because they have
high visual acuity and can thus feed swiftly by darting (Irvine & Northcote, 1983;
Stradmeyer et al., 1988; Stradmeyer, 1989). This contrasts with the case of
Chinese perch, for it has low visual acuity and can feed only slowly by stalking.
Chinese perch may not accomplish its relatively long process of prey recognition
before the offered food pellet has fallen down to the bottom and can no longer
be perceived by its sensory organs. The more strict selection of prey motion and
shape also makes it more difficult to feed Chinese perch with artificial diets.
This work was supported by the Chinese Academy of Sciences and the State Key
Laboratory for Freshwater Ecology and Biotechnology of China. We are grateful
to J. Janssen, Y. Cui and Daren He for their critical comments on the manuscript;
Shouqi Xie for his help in data analysis; and I. K. Chiang for his encouragement during
the study.
SENSE ORGANS AND FEEDING IN CHINESE PERCH 1065