- Email: [email protected]
Effects of intra- and interspecific competition on diet, growth and behaviour of Labeo calbasu (Hamilton) and Cirrhinus cirrhosus (Bloch)
Applied Animal Behaviour Science 128 (2010) 103–108
Contents lists available at ScienceDirect
Applied Animal Behaviour Science journal homepage: www.elsevier.com/locate/applanim
Effects of intra- and interspecific competition on diet, growth and behaviour of Labeo calbasu (Hamilton) and Cirrhinus cirrhosus (Bloch) Mohammad Mustafizur Rahman a,∗ , Marc Verdegem b a b
Aquatic Resource Science, Faculty of Fisheries, Kagoshima University, Shimoarata 4-50-20, Kagoshima 890-0056, Japan Aquaculture and Fisheries Group, Department of Animal Sciences, Wageningen University, P.O. Box 338, 6700 AH, Wageningen, The Netherlands
a r t i c l e
i n f o
Article history: Received 16 February 2010 Received in revised form 12 September 2010 Accepted 21 September 2010 Available online 16 October 2010 Keywords: Competition Labeo calbasu Cirrhinus cirrhosus Behaviour Niche shift
a b s t r a c t Effects of intra- and interspecific competition on diet, growth, grazing, swimming, resting and social behaviour of two carps calbasu (Labeo calbasu) and mrigal (Cirrhinus cirrhosus) were examined in single and mixed species treatments. Three treatments (tanks with 4 L. calbasu, 4 C. cirrhosus or 2 L. calbasu plus 2 C. cirrhosus) were randomly assigned to six 1 m2 glass-walled aquaria, in which pond conditions were simulated. Overall, both species preferred feeding on benthic macroinvertebrates, spending the majority of its grazing time near the tank bottom. Intraspecific food competition affected L. calbasu more than interspecfic food competition. The opposite was true for C. cirrhosus which was more affected by L. calbasu than by intraspecific competition. L. calbasu broadened its selection of food items and increased grazing time in response to intense (intraspecific) food competition. This behaviour allowed L. calbasu to maintain its food intake and hence growth. In presence of L. calbasu, C. cirrhosus continued to feed mainly on benthic macroinvertebrates, not changing its feeding behaviour. Therefore, C. cirrhosus’ total food consumption and growth diminished in the presence of L. calbasu. In addition to food competition, direct interaction (interference competition from L. calbasu) also played an important role in the behaviour, diet, and growth rate of C. cirrhosus. From an ecological, economic and fish welfare point of view, it can be suggested that C. cirrhosus is deprived when cultured together with L. calbasu in aquaculture ponds. © 2010 Elsevier B.V. All rights reserved.
1. Introduction The interaction between fishes is influenced by feeding habits and food availability (Schoener, 1982). For example, when two species forage on the same limited food resources, interspecific competition for food will be intense. In this situation some species will broaden its feeding niche or switch to less preferred food items to maximize its food intake, while other species will not (Werner and Hall, 1979; Balcombe et al., 2005). Understanding how fish species will exploit a limited food resource in single or
∗ Corresponding author. Tel.: +81 099 286 4161; fax: +81 099 286 4161. E-mail address: mustafi[email protected] (M.M. Rahman). 0168-1591/$ – see front matter © 2010 Elsevier B.V. All rights reserved. doi:10.1016/j.applanim.2010.09.015
mixed species conditions, will broaden our understanding on fish ecology and allow improving aquaculture practices. For example, fish with a broad feeding niche can be cultured at higher densities than the fish with a narrow feeding niche. Labeo calbasu (calbasu) and Cirrhinus cirrhosus (mrigal) are two important bottom feeder carps (cyprinids) stocked traditionally in south Asian polyculture ponds (Milstein et al., 2002; Rahman et al., 2006). In many cases, both species are stocked together. Many culturists believe that L. calbasu decreases the growth of C. cirrhosus while L. calbasu growth is not affected by the presence of C. cirrhosus, although there is no experimental evidence about it. Both L. calbasu and C. cirrhosus feed primarily on benthic macroinvertebrates (Islam et al., 1998; Kanak et al., 1999; Milstein et al., 2002)
104
M.M. Rahman, M. Verdegem / Applied Animal Behaviour Science 128 (2010) 103–108
therefore, food competition between them may affect the growth of C. cirrhosus. However, insight in how both species compete for limited food resources is still lacking. There are two common ways to quantify food competition. (1) Using growth and production data with the idea that winner will be more successful than the loser (Carpenter, 2005; Wuellner and Willis, 2008); or (2) observing changes in behavioural strategies and feeding patterns (Ward et al., 2006). However, each of these two methods cannot be used independently to explain the entire mechanisms involved in interactions among fishes. For example, growth data cannot distinguish between competition mediated by direct interaction (interference) or by resource depression. On the other hand, observations of behavioural strategies and feeding patterns cannot give a clear idea about the intensity of food competition. Therefore, combining fish behavioural observations with fish growth and production data will yield a more complete picture of the nature, intensity and mechanisms of interaction (Wuellner and Willis, 2008). In this experiment, growth, food selection and feeding behaviour of L. calbasu and C. cirrhosus were quantified in large aquaria in which pond conditions were simulated, stocking either single or mixed species groups. The objectives of this study were to investigate the effects of intra- and interspecific interaction on diet composition, growth and grazing, swimming and social behaviours of L. calbasu and C. cirrhosus.
2. Materials and methods 2.1. Experimental set-up The experiment was carried out in outdoor rectangular tanks (size: 2.5 m × 0.4 m × 0.9 m) at Bangladesh Agricultural University, Bangladesh. The short sides were constructed from concrete and the long sides from glass, allowing direct observation of fishes throughout the tank. To establish pond conditions, tank floors were covered with pond sediment, and tanks were filled with pond water (supplied from the same pond). The sediment and water column heights in each tank were 10 and 70 cm, respectively. Each tank was further treated with agricultural lime (CaCO3 ) at 25 g (250 kg ha−1 ), semi-decomposed cow manure at 125 g (1250 kg ha−1 ), urea at 3.1 g (31 kg ha−1 ) and triple super phosphate at 1.6 g (16 kg ha−1 ) one week before and again 2 weeks after fish stocking. Sunlight penetration through the glass walls was prevented by bamboo mat covers. The bamboo mats were only removed to record the fish behaviour. Treatments consisted of three different fish group combinations stocked in each tank: 4 L. calbasu (LC treatment), 4 C. cirrhosus (CC) or 2 L. calbasu plus 2 C. cirrhosus (mixed). Duplicate treatments were randomly assigned to 6 tanks. All fishes were within a similar weight range (L. calbasu: 103.6–112.5 g and C. cirrhosus: 103.7–116.7 g). The experiment ran for four weeks, relying solely on the available natural food. Every week, half of tank water was replaced with less-turbid pond water. In addition, the tank water was also diluted partially with less-turbid pond water if turbidity prevented observations during video recording.
2.2. Measured variables Water quality (temperature, dissolved oxygen, pH, nitrate nitrogen, total ammonia nitrogen, total nitrogen, phosphate phosphorus and total phosphorus) and total phyto- and zooplankton within experimental tanks were quantified weekly. Total benthic macroinvertebrates were only quantified at the end of the experiment. Procedures for water quality analyses (APHA, 1998) and total phytoplankton, zooplankton and benthic macroinvertebrates enumerations were similar as reported by Rahman et al. (2008a). Fish behaviour was observed during the fourth week of the experiment during a full 24 h period starting at 08:00 h, video-recording the aquarium 15 min every 3 h (8:00, 11:00, 14:00, 17:00, 20:00, 23:00, 2:00 and 5:00 h). This system consisted of two analogue video cameras (model HEL30K1A000) connected with a Quard (model NB2010S), a video cassette recorder (SANYO, model TLS9924P) and TV (SONY, model KV-TG21M80). The combined camera images covered the entire tank. The video cameras were sensitive enough to yield clear images also during night without artificial light. The individual fish behaviour was quantified by analysing video data using THE OBSERVER software (version 4.1, Noldus Information Technology, Wageningen, Netherlands). Fish behaviours were categorized as ‘grazing’ (in the water column, on the tank wall or on the bottom), ‘swimming’ (in the water column or <10 cm from the bottom), ‘resting’ (motionless) and ‘schooling’ (at least two fish <10 cm apart) according to Rahman et al. (2008b). All behaviours were expressed as percentage of total time, pooled over the whole day. The sum of grazing, swimming and resting is 100%. Fish behaviours were also categorized as schooling (intraspecific schooling: at least two individual of same species less than 10 cm apart and moving in same direction; interspecific schooling: at least two individual of different species less than 10 cm apart and moving in same direction) and scattering (all fish were scattered more than 10 cm apart). Because the sum of schooling and scattering time is 100%, scattering behaviour is not presented in the study. At the end of the experiment all fishes were harvested and weighed individually up to the nearest 0.1 g. Specific growth rate (SGR, % body weight day−1 ) was calculated from the natural logarithm of mean final mass minus the natural logarithm of the mean initial mass and divided by the total number of experimental days expressed as a percentage (Hopkins, 1992). After weighing, the body cavity of each fish was carefully opened and 5 cm of the anterior gut was removed and preserved immediately in 10% buffered formalin. The gut contents were analysed using a Sedgewick–Rafter cell and a microscope according to Rahman et al. (2006). The volumes of phyto- and zooplankton and benthic macroinvertebrates were also calculated according to Rahman et al. (2006). 2.3. Statistical analysis All data were analysed using SPSS (version 14) after they were checked for normal distribution and homogeneity of variance. Only percent data had to be arcsine transformed
M.M. Rahman, M. Verdegem / Applied Animal Behaviour Science 128 (2010) 103–108
105
before analysis; however, non-transformed data are presented in tables and figures. Benthic macroinvertebrates availability in the bottom soil, gut content, specific growth rate and behavioural data were analysed through nested ANOVA to investigate treatment (fixed factor) effects and tank (random factor) effects. For each behavioural variable, 8 15-min observations were summed together before statistical analysis. Fish weight was used as a covariate in the nested ANOVA to analyse gut content data. Water quality and water column plankton abundance data were analysed through repeated measure one-way ANOVA. Where effects were significant, ANOVA was followed by post-hoc test (Tukey test) for unplanned multiple comparisons of means (P < 0.05). All data (except water quality and plankton and benthic macroinvertebrates abundance in the tanks) of L. calbasu and C. cirrhosus were analysed separately to compare between the single and the mixed species treatments. 3. Results 3.1. Water quality, plankton and benthic macroinvertebrates availabilty All water quality parameters except total nitrogen and total phosphorous were statistically similar between treatments (Table 1). Water quality within experimental tanks was more or less similar to pond water quality previously reported by Rahman et al. (2008c). The biovolume of total zooplankton in the water column and benthic macroinvertebrates in the bottom sediment were statistically greater in CC tanks than mixed tanks, followed by LC tanks. The biovolume of phytoplankton in the water column did not vary significantly between CC tanks and mixed tanks, but was lower in the LC tanks (P < 0.05). 3.2. Gut content and growth of fish The biovolume of macroinvertebrates in the foregut of L. calbasu was greater in the presence of C. cirrhosus, whereas the opposite was observed with regard to phytoand zooplankton biovolumes (Table 2). However, total food
Fig. 1. Effects of treatment on specific growth rate (SGR) of L. calbasu (A) (F(1,2) = 0.08, P > 0.05, nested ANOVA) and C. cirrhosus (B) (F(1,2) = 33.87, P < 0.05, nested ANOVA) based on nested ANOVA. Data are mean ±95% confidence intervals. Asterisk represents significant treatment difference and ns represents no significant treatment difference.
ingestion and specific growth rate of L. calbasu were similar between single species (LC) and mixed species tanks (Table 2, Fig. 1A). Plankton biovolume of C. cirrhosus was similar between single species (CC) and mixed species tanks, whereas total benthic macroinvertebrates and total food ingestions and SGR were higher in tanks containing only C. cirrhosus than compared with mixed population
Table 1 Water quality parameters and plankton and benthic macroinvertebrates availability in different treatments based on one-way repeated measure ANOVA. LC = tanks with 4 L. calbasu, CC = tanks with 4 C. cirrhosus and mixed = tanks with 2 L. calbasu and 2 C. cirrhosus. Variable
Temperature (◦ C) DO (mg l−1 ) pH range NO3 -N (mg l−1 ) TAN (mg l−1 ) TN (mg l−1 ) PO4 -P (mg l−1 ) TP (mg l−1 ) Total phytoplankton (mm3 l−1 ) Total zooplankton (mm3 l−1 ) Total benthic macroinvertebrates (mm3 cm−3 )
F-value and probability
Treatment means
Treatment (2, 12 d.f.)
LC
CC
Mixed
0.51ns 0.02ns – 0.52ns 0.21ns 30.85** 0.85ns 7.19* 10.22* 31.24** 65.33* (2, 3 d.f.)
26.03 5.83 7.20 0.21 0.09 0.72c 0.15 0.47c 0.24b 0.21c 0.11c
25.99 5.93 7.64 0.28 0.09 0.95a 0.17 0.60a 0.34a 0.68a 0.19a
26.11 5.90 7.71 0.23 0.11 0.81b 0.16 0.54b 0.33a 0.47b 0.16b
ns, not significant. Mean values in same the row with no superscript in common differ significantly (P < 0.05). If the effects are significant, ANOVA followed by Tukey test. * P < 0.01. ** P < 0.001.
106
M.M. Rahman, M. Verdegem / Applied Animal Behaviour Science 128 (2010) 103–108
Table 2 Mean biovolume (mm3 ) of plankton and benthic macroinvertebtates in foregut of L. calbasu and C. cirrhosus based on nested ANOVA using fish weight as a covariable. Alone = tanks with 4 L. calbasu or 4 C. cirrhosus and mixed = tanks with 2 L. calbasu and 2 C. cirrhosus. Variable
L. calbasu Total phytoplankton Total zooplankton Total benthic macroinvertebrates Total food C. cirrhosus Total phytoplankton Total zooplankton Total benthic macroinvertebrates Total food
F-value and probability
Treatment means
Treatment variation (1, 2 d.f.)
Tank variation (2, 7 d.f)
Alone
Mixed
48.73* 330.84** 238.41** 1.25ns
10.21** 1.32ns 0.02ns 0.18ns
0.297 1.668 2.546 4.572
0.087 0.194 4.583 4.865
0.19ns 0.45ns 73.65* 66.14*
2.92ns 1.12ns 0.89ns 0.66ns
0.129 0.896 2.868 3.893
0.146 0.926 1.273 2.345
ns, not significant. * P < 0.05. ** P < 0.01.
tanks (Table 2, Fig. 1B). Between tanks variation was not significant (P > 0.05) on any gut content variable of L. calbasu and C. cirrhosus except total phytoplankton biovolume in the foregut of L. calbasu. 3.3. Grazing, swimming, resting and social behaviour of fish All grazing (except grazing on the tank wall), swimming (except total swimming) and resting variables of L. calbasu were influenced by the presence of C. cirrhosus (Table 3). L. calbasu fed and swam primarily in the water column in absence of C. cirrhosus, whereas they fed and swam pri-
marily close to the bottom in the presence of C. cirrhosus. L. calbasu’s resting time was nearly 4 times higher when C. cirrhosus was present instead of L. calbasu. There was no significant difference (P > 0.05) between tanks on any behavioural variable of L. calbasu. Apart from water column grazing and swimming, all behavioural variables of C. cirrhosus were significantly influenced by the presence of L. calbasu (Table 4). Presence of L. calbasu decreased the time spent for grazing close to the bottom, total grazing and resting of C. cirrhosus, while the time spent for swimming close to the bottom and total swimming were increased. Like L. calbasu, between tanks variation was not significant (P > 0.05) on any behavioural variables of C. cirrhosus.
Table 3 Grazing and swimming, resting and social behaviour of L. calbasu in two treatments based on nested ANOVA. Alone = tanks with 4 L. calbasu and mixed = tanks with 2 L. calbasu and 2 C. cirrhosus. Variable
Grazing in the water column Grazing on the wall Grazing on the bottom Total grazing Swimming in the water column Swimming near bottom Total swimming Resting (motionless)
F-value and probability
Treatment means
Treatment variation (1, 2 d.f.)
Tank variation (2, 8 d.f.)
Alone
Mixed
37.93* 2.73ns 25.97* 30.03* 52.97* 93.33* 2.90ns 40.69*
1.31ns 1.90ns 1.69ns 0.27ns 1.67ns 0.93ns 0.44ns 0.83ns
28.7 0.7 17.8 47.2 37.5 13.5 51.0 1.8
10.5 0.5 25.7 36.7 18.8 36.8 55.6 7.7
Results based on the percent duration of time for any given behaviour. ns, not significant. * P < 0.05 Table 4 Grazing and swimming, resting and social behaviour of C. cirrhosus in two treatments based on nested ANOVA. Alone = tanks with 4 C. cirrhosus and mixed = tanks with 2 L. calbasu and 2 C. cirrhosus. Variable
Grazing in the water column Grazing on the wall Grazing on the bottom Total grazing Swimming in the water column Swimming near bottom Total swimming Resting (motionless)
F-value and probability
Treatment means
Treatment variation (1, 2 d.f.)
Tank variation (2, 8 d.f.)
Alone
Mixed
1.58ns – 32.34* 26.26* 5.24ns 33.08* 52.44* 57.99*
0.41ns – 0.08ns 0.13ns 0.02ns 0.97ns 0.22ns 0.39ns
3.5 0.0 29.4 32.9 18.4 22.3 40.6 26.5
3.8 0.0 21.5 25.3 23.7 38.1 61.8 12.9
Results based on the percent duration of time for any given behaviour. ns, not significant. * P < 0.05.
M.M. Rahman, M. Verdegem / Applied Animal Behaviour Science 128 (2010) 103–108
Fig. 2. Effects of treatment on percent intraspecific schooling time of L. calbasu (A) (F(1,2) = 29.26, P < 0.05, nested ANOVA) and C. cirrhosus (B) (F(1,2) = 20.68, P < 0.05, nested ANOVA). Data are mean ±95% confidence intervals. Asterisks represent significant treatment differences.
There was no evidence of interspecific schooling or agonistic behaviour between L. calbasu and C. cirrhosus in mixed species treatment. Intraspecific schooling time of L. calbasu was higher in mixed species treatment than in single species treatment (Fig. 2A). In contrast, intraspecific schooling time of C. cirrhosus was lower in mixed species treatment than single species treatment (Fig. 2B). 4. Discussion Overall, both species fed primarily on benthic macroinvertebrates. These patterns concur with the commonly accepted view that L. calbasu and C. cirrhosus are bottom feeders, with macroinvertebrates its main food source (Islam et al., 1998; Kanak et al., 1999; Milstein et al., 2002). Since both species are dependent on similar food resources, competition for food between them should be intense (Ward et al., 2006). The present study describes outcomes of food competition within and between species. In mixed species treatment, L. calbasu did not change its feeding niche spending most of its time near tank bottom and feeding mainly on benthic macroinvertebrates. In single species (LC) treatment, mean total benthic macroinvertebrates availability in the tank bottom decreased. In this case, L. calbasu shifted feeding partially from benthic macroinvertebrates to plankton. This resulted in a shift by
107
L. calbasu from feeding primarily on the bottom towards the water column, spending less time grazing on the bottom and more time grazing in the water column. Moreover, L. calbasu increased its total grazing time, thereby maximizing food intake and maintaining growth rate. Such a pattern of shifting feeding niche and food habit are seen in many fishes (Gonzalez-Soles et al., 1997; Svanback and Bolnick, 2007; Rahman et al., 2010). For example, rohu (Labeo rohita) prefer zooplankton when available, but shift to phytoplankton when zooplankton is depleted (Rahman et al., 2006). The underlying factor driving these choices is the net energy value of total ingested food (MacArthur and Pianka, 1966). This could also explain why total grazing time was lower and resting time was higher when L. calbasu stayed with C. cirrhosus. This could have been achieved by a trade-off between feeding and resting time, hence, maximizing intake when feeding and reducing energy loss by resting. Under the experimental condition in this study, C. cirrhosus could not optimize its foraging behaviour. When food resources were lower in presence of L. calbasu, C. cirrhosus neither changed its feeding niche nor showed any dietary shift continuing to feed mainly on benthic macroinvertebrates. Hence, when L. calbasu and L. cirrhosus were together, and the availability of total benthic macroinvertebrates became limited, the growth rate of C. cirrhosus diminished. The inability of sympatric C. cirrhosus to optimize its feeding may have been a result of interference competition from L. calbasu resulting in the reduced consumption of food quantity and growth. This was supported by video observation: when L. calbasu came close to C. cirrhosus, C. cirrhosus stopped grazing or resting and started swimming. Therefore, C. cirrhosus increased time spent for swimming and decreased time spent for grazing and resting in the presence of L. calbasu. The interference competition could also explain why interspecific schooling was not seen in this study. Effects of L. calbasu on behaviour, food intake and growth of C. cirrhosus are similar to the effects of walleye (Sander vitreus) on the behaviour, food intake and growth of smallmouth bass (Micropterus dolomieu). In presence of walleye, smallmouth bass do not go after prey, therefore food intake and growth of smallmouth bass are diminished in presence of walleye (Inskip and Magnuson, 1983; Wuellner and Willis, 2008). It is widely established that schooling increases food competition (Pitcher and Parrish, 1993; Johnsson et al., 2005). Therefore, to reduce competition, many fish (e.g., Clupea harengus, Fundulus diaphanous, Mulloidichthys flavolineatu, etc.) prefer solitary grazing (Holland et al., 1993; Robinson, 1995; Ward et al., 2002). This concurs with the result of the present study where L. calbasu and C. cirrhosus decreased their schooling time when food competition was more intense (less food abundance). 5. Conclusion In conclusion, the preferred habitat of both L. calbasu and C. cirrhosus is primarily benthic where they feed mostly on benthic macroinvertebrates. L. calbasu were more efficient than C. cirrhosus at ingesting food in mixed species
108
M.M. Rahman, M. Verdegem / Applied Animal Behaviour Science 128 (2010) 103–108
treatment, indicating L. calbasu is a superior competitor to C. cirrhosus. In contrast, C. cirrhosus was more affected by interspecific food competition from L. calbasu than by intraspecfic food competition. When food competition was more intense (less food abundance) L. calbasu broadened its feeding niche to maximize food intake, but C. cirrhosus did not broaden its feeding niche resulting in reduced food consumption and growth. These results demonstrate that L. calbasu has the ecomorphology that enables them to change its food selectivity and feeding niche in response to food resource variability (Rahman et al., 2009). This trait may help L. calbasu to maximize fitness by reducing food competition and by maximizing growth through dietary shifts. Thus, L. calbasu is better than C. cirrhosus for food resource utilization and fish production when they are cultured together in non-fed pond. However, culturing them (alone or both) with other species might change its behaviour, food selectivity and food intake. Therefore, any prediction about culturing with additional species should be based on direct observation of the intended species combination. This study was conducted in simulated pond condition and therefore, more research is recommended before extrapolating results to an industrial level. Acknowledgments We gratefully acknowledge the European Commission for financial support through Pond–Live project. We also thank to Prof. Dr. M.A. Wahab, Bangladesh Agricultural University, Mymensingh, Bangladesh, for his cooperation during conducting this experiment. References APHA, 1998. Standard Methods for the Examination of Water and Waste Water. American Public Health Association, Washington, DC. Balcombe, S.R., Bunn, S.E., Davies, P.M., McKenzie Smit, F.J., 2005. Variability of fish diets between dry and flood periods in an arid zone floodplain river. J. Fish. Biol. 67, 1552–1567. Carpenter, J., 2005. Competition for food between an introduced crayfish and two fishes endemic to the Colorado River basin. Environ. Biol. Fish. 72, 335–342. Gonzalez-Soles, J., Oro, D., Jover, L., Ruiz, X., Pedrocchi, V., 1997. Trophic niche width and overlap of two sympatric gulls in the southwestern Mediterranean. Oecologia 112, 75–80. Holland, K.N., Peterson, J.D., Lowe, C.G., Wetherbee, B.M., 1993. Movements, distribution and growth rates of the white goatfish Mulloides flavolineatus in a fisheries conservation zone. Bull. Mar. Sci. 52, 982–992. Hopkins, K.D., 1992. Reporting fish growth: a review of the basics. J. World Aquacult. Soc. 23, 173–179.
Inskip, P.D., Magnuson, J.J., 1983. Changes in fish populations over an 80year period: Big Pine Lake. Wis. Trans. Am. Fish. Soc. 112, 378–389. Islam, M.R., Miah, M.S., Alam, M.S., Khair, M.A., Ahmed, S.S., Rokeya, J.A., 1998. Effects of various inputs on fish production under carp polyculture system. Bangladesh J. Fish. 21, 47–51. Johnsson, J.I., Winberg, S., Sloman, K.A., 2005. Social interactions. In: Sloman, K.A., Wilson, R.W., Balshine, S. (Eds.), Behaviour and Physiology of Fish. Elsevier Academic Press, Amsterdam, pp. 151–196. Kanak, M.K., Dewan, S., Salimullah, M., 1999. Performance of exotic fishes with Indian major carps in polyculture under three different species combinations. Bangladesh J. Fish. 22, 1–6. MacArthur, R.H., Pianka, E.R., 1966. On optimal use of a patchy environment. Am. Nat. 100, 603–609. Milstein, A., Wahab, M.A., Rahman, M.M., 2002. Environmental effects of common carp Cyprinus carpio (L.) and mrigal Cirrhinus mrigala (Hamilton) as bottom feeders in major Indian carp polycultures. Aquacult. Res. 33, 1103–1117. Pitcher, T.J., Parrish, J.K., 1993. Functions of shoaling behaviour in teleosts. In: Pitcher, T.J. (Ed.), Behaviour of Teleost Fishes. Chapman and Hall, pp. 363–439. Rahman, M.M., Kadowaki, S., Balcombe, S.R., Wahab, M.A., 2010. Common carp (Cyprinus carpio L.) alters its feeding niche in response to changing food resources: direct observations in simulated ponds. Ecol. Res. 25, 303–309. Rahman, M.M., Hossain, M.Y., Jo, Q., Kim, S.-K., Ohtomi, J., Meyer, C., 2009. Ontogenetic shift in dietary preference and low dietary overlap in rohu (Labeo rohita) and common carp (Cyprinus carpio) in semi-intensive polyculture ponds. Ichthyol. Res. 56, 28–36. Rahman, M.M., Verdegem, M.C.J., Nagelkerke, L.A.J., Wahab, M.A., Milstein, A., Verreth, J.A.J., 2006. Growth, production and food preference of rohu Labeo rohita (H.) in monoculture and in polyculture with common carp Cyprinus carpio (L.) under fed and non-fed ponds. Aquaculture 257, 359–372. Rahman, M.M., Verdegem, M.C.J., Nagelkerke, L.A.J., Wahab, M.A., Milstein, A., Verreth, J.A.J., 2008c. Effects of common carp Cyprinus carpio (L.) and feed addition in rohu Labeo rohita (Hamilton) ponds on nutrient partitioning among fish, plankton and benthos. Aquacult. Res. 39, 85–95. Rahman, M.M., Verdegem, M.C.J., Nagelkerke, L.A.J., Wahab, M.A., Verreth, J.A.J., 2008b. Swimming, grazing and social behaviour of rohu Labeo rohita (Hamilton) and common carp Cyprinus carpio (L.) in tanks under fed and non-fed conditions. Appl. Anim. Behav. Sci. 113, 255–264. Rahman, M.M., Verdegem, M.C.J., Wahab, M.A., Hossain, M.Y., Jo, Q., 2008a. Effects of day and night on swimming, grazing and social behaviours of rohu Labeo rohita (Hamilton) and common carp Cyprinus carpio (L.) in simulated ponds. Aquacult. Res. 39, 1383–1392. Robinson, C.J., 1995. Food competition in a shoal of herring: the role of hunger. Mar. Freshw. Behav. Physiol. 24, 237–242. Schoener, T.W., 1982. The controversy over interspecific competition. Am. Sci. 70, 586–595. Svanback, R., Bolnick, D.I., 2007. Intraspecific competition drives increased resource use diversity within a natural population. Proc. R. Soc. B 274, 839–844. Ward, A.J.W., Webster, M.M., Hart, P.J.B., 2006. Intraspecific food competition in fishes. Fish Fish. 7, 231–261. Ward, A.J.W., Botham, M.S., Hoare, D.J., James, R., Broom, M., Godin, J.-G.J., Krause, J., 2002. Association patterns and shoal fidelity in the threespined stickleback. Proc. R. Soc. Lond. [Biol.] 269, 2451–2455. Werner, E.E., Hall, D.J., 1979. Foraging efficiency and habitat switching in competing sunfish. Ecology 60, 256–264. Wuellner, M.R., Willis, D.W., 2008. On northern pond: fish competition. Pond Boss Mag. 17, 20–22.
Contents lists available at ScienceDirect
Applied Animal Behaviour Science journal homepage: www.elsevier.com/locate/applanim
Effects of intra- and interspecific competition on diet, growth and behaviour of Labeo calbasu (Hamilton) and Cirrhinus cirrhosus (Bloch) Mohammad Mustafizur Rahman a,∗ , Marc Verdegem b a b
Aquatic Resource Science, Faculty of Fisheries, Kagoshima University, Shimoarata 4-50-20, Kagoshima 890-0056, Japan Aquaculture and Fisheries Group, Department of Animal Sciences, Wageningen University, P.O. Box 338, 6700 AH, Wageningen, The Netherlands
a r t i c l e
i n f o
Article history: Received 16 February 2010 Received in revised form 12 September 2010 Accepted 21 September 2010 Available online 16 October 2010 Keywords: Competition Labeo calbasu Cirrhinus cirrhosus Behaviour Niche shift
a b s t r a c t Effects of intra- and interspecific competition on diet, growth, grazing, swimming, resting and social behaviour of two carps calbasu (Labeo calbasu) and mrigal (Cirrhinus cirrhosus) were examined in single and mixed species treatments. Three treatments (tanks with 4 L. calbasu, 4 C. cirrhosus or 2 L. calbasu plus 2 C. cirrhosus) were randomly assigned to six 1 m2 glass-walled aquaria, in which pond conditions were simulated. Overall, both species preferred feeding on benthic macroinvertebrates, spending the majority of its grazing time near the tank bottom. Intraspecific food competition affected L. calbasu more than interspecfic food competition. The opposite was true for C. cirrhosus which was more affected by L. calbasu than by intraspecific competition. L. calbasu broadened its selection of food items and increased grazing time in response to intense (intraspecific) food competition. This behaviour allowed L. calbasu to maintain its food intake and hence growth. In presence of L. calbasu, C. cirrhosus continued to feed mainly on benthic macroinvertebrates, not changing its feeding behaviour. Therefore, C. cirrhosus’ total food consumption and growth diminished in the presence of L. calbasu. In addition to food competition, direct interaction (interference competition from L. calbasu) also played an important role in the behaviour, diet, and growth rate of C. cirrhosus. From an ecological, economic and fish welfare point of view, it can be suggested that C. cirrhosus is deprived when cultured together with L. calbasu in aquaculture ponds. © 2010 Elsevier B.V. All rights reserved.
1. Introduction The interaction between fishes is influenced by feeding habits and food availability (Schoener, 1982). For example, when two species forage on the same limited food resources, interspecific competition for food will be intense. In this situation some species will broaden its feeding niche or switch to less preferred food items to maximize its food intake, while other species will not (Werner and Hall, 1979; Balcombe et al., 2005). Understanding how fish species will exploit a limited food resource in single or
∗ Corresponding author. Tel.: +81 099 286 4161; fax: +81 099 286 4161. E-mail address: mustafi[email protected] (M.M. Rahman). 0168-1591/$ – see front matter © 2010 Elsevier B.V. All rights reserved. doi:10.1016/j.applanim.2010.09.015
mixed species conditions, will broaden our understanding on fish ecology and allow improving aquaculture practices. For example, fish with a broad feeding niche can be cultured at higher densities than the fish with a narrow feeding niche. Labeo calbasu (calbasu) and Cirrhinus cirrhosus (mrigal) are two important bottom feeder carps (cyprinids) stocked traditionally in south Asian polyculture ponds (Milstein et al., 2002; Rahman et al., 2006). In many cases, both species are stocked together. Many culturists believe that L. calbasu decreases the growth of C. cirrhosus while L. calbasu growth is not affected by the presence of C. cirrhosus, although there is no experimental evidence about it. Both L. calbasu and C. cirrhosus feed primarily on benthic macroinvertebrates (Islam et al., 1998; Kanak et al., 1999; Milstein et al., 2002)
104
M.M. Rahman, M. Verdegem / Applied Animal Behaviour Science 128 (2010) 103–108
therefore, food competition between them may affect the growth of C. cirrhosus. However, insight in how both species compete for limited food resources is still lacking. There are two common ways to quantify food competition. (1) Using growth and production data with the idea that winner will be more successful than the loser (Carpenter, 2005; Wuellner and Willis, 2008); or (2) observing changes in behavioural strategies and feeding patterns (Ward et al., 2006). However, each of these two methods cannot be used independently to explain the entire mechanisms involved in interactions among fishes. For example, growth data cannot distinguish between competition mediated by direct interaction (interference) or by resource depression. On the other hand, observations of behavioural strategies and feeding patterns cannot give a clear idea about the intensity of food competition. Therefore, combining fish behavioural observations with fish growth and production data will yield a more complete picture of the nature, intensity and mechanisms of interaction (Wuellner and Willis, 2008). In this experiment, growth, food selection and feeding behaviour of L. calbasu and C. cirrhosus were quantified in large aquaria in which pond conditions were simulated, stocking either single or mixed species groups. The objectives of this study were to investigate the effects of intra- and interspecific interaction on diet composition, growth and grazing, swimming and social behaviours of L. calbasu and C. cirrhosus.
2. Materials and methods 2.1. Experimental set-up The experiment was carried out in outdoor rectangular tanks (size: 2.5 m × 0.4 m × 0.9 m) at Bangladesh Agricultural University, Bangladesh. The short sides were constructed from concrete and the long sides from glass, allowing direct observation of fishes throughout the tank. To establish pond conditions, tank floors were covered with pond sediment, and tanks were filled with pond water (supplied from the same pond). The sediment and water column heights in each tank were 10 and 70 cm, respectively. Each tank was further treated with agricultural lime (CaCO3 ) at 25 g (250 kg ha−1 ), semi-decomposed cow manure at 125 g (1250 kg ha−1 ), urea at 3.1 g (31 kg ha−1 ) and triple super phosphate at 1.6 g (16 kg ha−1 ) one week before and again 2 weeks after fish stocking. Sunlight penetration through the glass walls was prevented by bamboo mat covers. The bamboo mats were only removed to record the fish behaviour. Treatments consisted of three different fish group combinations stocked in each tank: 4 L. calbasu (LC treatment), 4 C. cirrhosus (CC) or 2 L. calbasu plus 2 C. cirrhosus (mixed). Duplicate treatments were randomly assigned to 6 tanks. All fishes were within a similar weight range (L. calbasu: 103.6–112.5 g and C. cirrhosus: 103.7–116.7 g). The experiment ran for four weeks, relying solely on the available natural food. Every week, half of tank water was replaced with less-turbid pond water. In addition, the tank water was also diluted partially with less-turbid pond water if turbidity prevented observations during video recording.
2.2. Measured variables Water quality (temperature, dissolved oxygen, pH, nitrate nitrogen, total ammonia nitrogen, total nitrogen, phosphate phosphorus and total phosphorus) and total phyto- and zooplankton within experimental tanks were quantified weekly. Total benthic macroinvertebrates were only quantified at the end of the experiment. Procedures for water quality analyses (APHA, 1998) and total phytoplankton, zooplankton and benthic macroinvertebrates enumerations were similar as reported by Rahman et al. (2008a). Fish behaviour was observed during the fourth week of the experiment during a full 24 h period starting at 08:00 h, video-recording the aquarium 15 min every 3 h (8:00, 11:00, 14:00, 17:00, 20:00, 23:00, 2:00 and 5:00 h). This system consisted of two analogue video cameras (model HEL30K1A000) connected with a Quard (model NB2010S), a video cassette recorder (SANYO, model TLS9924P) and TV (SONY, model KV-TG21M80). The combined camera images covered the entire tank. The video cameras were sensitive enough to yield clear images also during night without artificial light. The individual fish behaviour was quantified by analysing video data using THE OBSERVER software (version 4.1, Noldus Information Technology, Wageningen, Netherlands). Fish behaviours were categorized as ‘grazing’ (in the water column, on the tank wall or on the bottom), ‘swimming’ (in the water column or <10 cm from the bottom), ‘resting’ (motionless) and ‘schooling’ (at least two fish <10 cm apart) according to Rahman et al. (2008b). All behaviours were expressed as percentage of total time, pooled over the whole day. The sum of grazing, swimming and resting is 100%. Fish behaviours were also categorized as schooling (intraspecific schooling: at least two individual of same species less than 10 cm apart and moving in same direction; interspecific schooling: at least two individual of different species less than 10 cm apart and moving in same direction) and scattering (all fish were scattered more than 10 cm apart). Because the sum of schooling and scattering time is 100%, scattering behaviour is not presented in the study. At the end of the experiment all fishes were harvested and weighed individually up to the nearest 0.1 g. Specific growth rate (SGR, % body weight day−1 ) was calculated from the natural logarithm of mean final mass minus the natural logarithm of the mean initial mass and divided by the total number of experimental days expressed as a percentage (Hopkins, 1992). After weighing, the body cavity of each fish was carefully opened and 5 cm of the anterior gut was removed and preserved immediately in 10% buffered formalin. The gut contents were analysed using a Sedgewick–Rafter cell and a microscope according to Rahman et al. (2006). The volumes of phyto- and zooplankton and benthic macroinvertebrates were also calculated according to Rahman et al. (2006). 2.3. Statistical analysis All data were analysed using SPSS (version 14) after they were checked for normal distribution and homogeneity of variance. Only percent data had to be arcsine transformed
M.M. Rahman, M. Verdegem / Applied Animal Behaviour Science 128 (2010) 103–108
105
before analysis; however, non-transformed data are presented in tables and figures. Benthic macroinvertebrates availability in the bottom soil, gut content, specific growth rate and behavioural data were analysed through nested ANOVA to investigate treatment (fixed factor) effects and tank (random factor) effects. For each behavioural variable, 8 15-min observations were summed together before statistical analysis. Fish weight was used as a covariate in the nested ANOVA to analyse gut content data. Water quality and water column plankton abundance data were analysed through repeated measure one-way ANOVA. Where effects were significant, ANOVA was followed by post-hoc test (Tukey test) for unplanned multiple comparisons of means (P < 0.05). All data (except water quality and plankton and benthic macroinvertebrates abundance in the tanks) of L. calbasu and C. cirrhosus were analysed separately to compare between the single and the mixed species treatments. 3. Results 3.1. Water quality, plankton and benthic macroinvertebrates availabilty All water quality parameters except total nitrogen and total phosphorous were statistically similar between treatments (Table 1). Water quality within experimental tanks was more or less similar to pond water quality previously reported by Rahman et al. (2008c). The biovolume of total zooplankton in the water column and benthic macroinvertebrates in the bottom sediment were statistically greater in CC tanks than mixed tanks, followed by LC tanks. The biovolume of phytoplankton in the water column did not vary significantly between CC tanks and mixed tanks, but was lower in the LC tanks (P < 0.05). 3.2. Gut content and growth of fish The biovolume of macroinvertebrates in the foregut of L. calbasu was greater in the presence of C. cirrhosus, whereas the opposite was observed with regard to phytoand zooplankton biovolumes (Table 2). However, total food
Fig. 1. Effects of treatment on specific growth rate (SGR) of L. calbasu (A) (F(1,2) = 0.08, P > 0.05, nested ANOVA) and C. cirrhosus (B) (F(1,2) = 33.87, P < 0.05, nested ANOVA) based on nested ANOVA. Data are mean ±95% confidence intervals. Asterisk represents significant treatment difference and ns represents no significant treatment difference.
ingestion and specific growth rate of L. calbasu were similar between single species (LC) and mixed species tanks (Table 2, Fig. 1A). Plankton biovolume of C. cirrhosus was similar between single species (CC) and mixed species tanks, whereas total benthic macroinvertebrates and total food ingestions and SGR were higher in tanks containing only C. cirrhosus than compared with mixed population
Table 1 Water quality parameters and plankton and benthic macroinvertebrates availability in different treatments based on one-way repeated measure ANOVA. LC = tanks with 4 L. calbasu, CC = tanks with 4 C. cirrhosus and mixed = tanks with 2 L. calbasu and 2 C. cirrhosus. Variable
Temperature (◦ C) DO (mg l−1 ) pH range NO3 -N (mg l−1 ) TAN (mg l−1 ) TN (mg l−1 ) PO4 -P (mg l−1 ) TP (mg l−1 ) Total phytoplankton (mm3 l−1 ) Total zooplankton (mm3 l−1 ) Total benthic macroinvertebrates (mm3 cm−3 )
F-value and probability
Treatment means
Treatment (2, 12 d.f.)
LC
CC
Mixed
0.51ns 0.02ns – 0.52ns 0.21ns 30.85** 0.85ns 7.19* 10.22* 31.24** 65.33* (2, 3 d.f.)
26.03 5.83 7.20 0.21 0.09 0.72c 0.15 0.47c 0.24b 0.21c 0.11c
25.99 5.93 7.64 0.28 0.09 0.95a 0.17 0.60a 0.34a 0.68a 0.19a
26.11 5.90 7.71 0.23 0.11 0.81b 0.16 0.54b 0.33a 0.47b 0.16b
ns, not significant. Mean values in same the row with no superscript in common differ significantly (P < 0.05). If the effects are significant, ANOVA followed by Tukey test. * P < 0.01. ** P < 0.001.
106
M.M. Rahman, M. Verdegem / Applied Animal Behaviour Science 128 (2010) 103–108
Table 2 Mean biovolume (mm3 ) of plankton and benthic macroinvertebtates in foregut of L. calbasu and C. cirrhosus based on nested ANOVA using fish weight as a covariable. Alone = tanks with 4 L. calbasu or 4 C. cirrhosus and mixed = tanks with 2 L. calbasu and 2 C. cirrhosus. Variable
L. calbasu Total phytoplankton Total zooplankton Total benthic macroinvertebrates Total food C. cirrhosus Total phytoplankton Total zooplankton Total benthic macroinvertebrates Total food
F-value and probability
Treatment means
Treatment variation (1, 2 d.f.)
Tank variation (2, 7 d.f)
Alone
Mixed
48.73* 330.84** 238.41** 1.25ns
10.21** 1.32ns 0.02ns 0.18ns
0.297 1.668 2.546 4.572
0.087 0.194 4.583 4.865
0.19ns 0.45ns 73.65* 66.14*
2.92ns 1.12ns 0.89ns 0.66ns
0.129 0.896 2.868 3.893
0.146 0.926 1.273 2.345
ns, not significant. * P < 0.05. ** P < 0.01.
tanks (Table 2, Fig. 1B). Between tanks variation was not significant (P > 0.05) on any gut content variable of L. calbasu and C. cirrhosus except total phytoplankton biovolume in the foregut of L. calbasu. 3.3. Grazing, swimming, resting and social behaviour of fish All grazing (except grazing on the tank wall), swimming (except total swimming) and resting variables of L. calbasu were influenced by the presence of C. cirrhosus (Table 3). L. calbasu fed and swam primarily in the water column in absence of C. cirrhosus, whereas they fed and swam pri-
marily close to the bottom in the presence of C. cirrhosus. L. calbasu’s resting time was nearly 4 times higher when C. cirrhosus was present instead of L. calbasu. There was no significant difference (P > 0.05) between tanks on any behavioural variable of L. calbasu. Apart from water column grazing and swimming, all behavioural variables of C. cirrhosus were significantly influenced by the presence of L. calbasu (Table 4). Presence of L. calbasu decreased the time spent for grazing close to the bottom, total grazing and resting of C. cirrhosus, while the time spent for swimming close to the bottom and total swimming were increased. Like L. calbasu, between tanks variation was not significant (P > 0.05) on any behavioural variables of C. cirrhosus.
Table 3 Grazing and swimming, resting and social behaviour of L. calbasu in two treatments based on nested ANOVA. Alone = tanks with 4 L. calbasu and mixed = tanks with 2 L. calbasu and 2 C. cirrhosus. Variable
Grazing in the water column Grazing on the wall Grazing on the bottom Total grazing Swimming in the water column Swimming near bottom Total swimming Resting (motionless)
F-value and probability
Treatment means
Treatment variation (1, 2 d.f.)
Tank variation (2, 8 d.f.)
Alone
Mixed
37.93* 2.73ns 25.97* 30.03* 52.97* 93.33* 2.90ns 40.69*
1.31ns 1.90ns 1.69ns 0.27ns 1.67ns 0.93ns 0.44ns 0.83ns
28.7 0.7 17.8 47.2 37.5 13.5 51.0 1.8
10.5 0.5 25.7 36.7 18.8 36.8 55.6 7.7
Results based on the percent duration of time for any given behaviour. ns, not significant. * P < 0.05 Table 4 Grazing and swimming, resting and social behaviour of C. cirrhosus in two treatments based on nested ANOVA. Alone = tanks with 4 C. cirrhosus and mixed = tanks with 2 L. calbasu and 2 C. cirrhosus. Variable
Grazing in the water column Grazing on the wall Grazing on the bottom Total grazing Swimming in the water column Swimming near bottom Total swimming Resting (motionless)
F-value and probability
Treatment means
Treatment variation (1, 2 d.f.)
Tank variation (2, 8 d.f.)
Alone
Mixed
1.58ns – 32.34* 26.26* 5.24ns 33.08* 52.44* 57.99*
0.41ns – 0.08ns 0.13ns 0.02ns 0.97ns 0.22ns 0.39ns
3.5 0.0 29.4 32.9 18.4 22.3 40.6 26.5
3.8 0.0 21.5 25.3 23.7 38.1 61.8 12.9
Results based on the percent duration of time for any given behaviour. ns, not significant. * P < 0.05.
M.M. Rahman, M. Verdegem / Applied Animal Behaviour Science 128 (2010) 103–108
Fig. 2. Effects of treatment on percent intraspecific schooling time of L. calbasu (A) (F(1,2) = 29.26, P < 0.05, nested ANOVA) and C. cirrhosus (B) (F(1,2) = 20.68, P < 0.05, nested ANOVA). Data are mean ±95% confidence intervals. Asterisks represent significant treatment differences.
There was no evidence of interspecific schooling or agonistic behaviour between L. calbasu and C. cirrhosus in mixed species treatment. Intraspecific schooling time of L. calbasu was higher in mixed species treatment than in single species treatment (Fig. 2A). In contrast, intraspecific schooling time of C. cirrhosus was lower in mixed species treatment than single species treatment (Fig. 2B). 4. Discussion Overall, both species fed primarily on benthic macroinvertebrates. These patterns concur with the commonly accepted view that L. calbasu and C. cirrhosus are bottom feeders, with macroinvertebrates its main food source (Islam et al., 1998; Kanak et al., 1999; Milstein et al., 2002). Since both species are dependent on similar food resources, competition for food between them should be intense (Ward et al., 2006). The present study describes outcomes of food competition within and between species. In mixed species treatment, L. calbasu did not change its feeding niche spending most of its time near tank bottom and feeding mainly on benthic macroinvertebrates. In single species (LC) treatment, mean total benthic macroinvertebrates availability in the tank bottom decreased. In this case, L. calbasu shifted feeding partially from benthic macroinvertebrates to plankton. This resulted in a shift by
107
L. calbasu from feeding primarily on the bottom towards the water column, spending less time grazing on the bottom and more time grazing in the water column. Moreover, L. calbasu increased its total grazing time, thereby maximizing food intake and maintaining growth rate. Such a pattern of shifting feeding niche and food habit are seen in many fishes (Gonzalez-Soles et al., 1997; Svanback and Bolnick, 2007; Rahman et al., 2010). For example, rohu (Labeo rohita) prefer zooplankton when available, but shift to phytoplankton when zooplankton is depleted (Rahman et al., 2006). The underlying factor driving these choices is the net energy value of total ingested food (MacArthur and Pianka, 1966). This could also explain why total grazing time was lower and resting time was higher when L. calbasu stayed with C. cirrhosus. This could have been achieved by a trade-off between feeding and resting time, hence, maximizing intake when feeding and reducing energy loss by resting. Under the experimental condition in this study, C. cirrhosus could not optimize its foraging behaviour. When food resources were lower in presence of L. calbasu, C. cirrhosus neither changed its feeding niche nor showed any dietary shift continuing to feed mainly on benthic macroinvertebrates. Hence, when L. calbasu and L. cirrhosus were together, and the availability of total benthic macroinvertebrates became limited, the growth rate of C. cirrhosus diminished. The inability of sympatric C. cirrhosus to optimize its feeding may have been a result of interference competition from L. calbasu resulting in the reduced consumption of food quantity and growth. This was supported by video observation: when L. calbasu came close to C. cirrhosus, C. cirrhosus stopped grazing or resting and started swimming. Therefore, C. cirrhosus increased time spent for swimming and decreased time spent for grazing and resting in the presence of L. calbasu. The interference competition could also explain why interspecific schooling was not seen in this study. Effects of L. calbasu on behaviour, food intake and growth of C. cirrhosus are similar to the effects of walleye (Sander vitreus) on the behaviour, food intake and growth of smallmouth bass (Micropterus dolomieu). In presence of walleye, smallmouth bass do not go after prey, therefore food intake and growth of smallmouth bass are diminished in presence of walleye (Inskip and Magnuson, 1983; Wuellner and Willis, 2008). It is widely established that schooling increases food competition (Pitcher and Parrish, 1993; Johnsson et al., 2005). Therefore, to reduce competition, many fish (e.g., Clupea harengus, Fundulus diaphanous, Mulloidichthys flavolineatu, etc.) prefer solitary grazing (Holland et al., 1993; Robinson, 1995; Ward et al., 2002). This concurs with the result of the present study where L. calbasu and C. cirrhosus decreased their schooling time when food competition was more intense (less food abundance). 5. Conclusion In conclusion, the preferred habitat of both L. calbasu and C. cirrhosus is primarily benthic where they feed mostly on benthic macroinvertebrates. L. calbasu were more efficient than C. cirrhosus at ingesting food in mixed species
108
M.M. Rahman, M. Verdegem / Applied Animal Behaviour Science 128 (2010) 103–108
treatment, indicating L. calbasu is a superior competitor to C. cirrhosus. In contrast, C. cirrhosus was more affected by interspecific food competition from L. calbasu than by intraspecfic food competition. When food competition was more intense (less food abundance) L. calbasu broadened its feeding niche to maximize food intake, but C. cirrhosus did not broaden its feeding niche resulting in reduced food consumption and growth. These results demonstrate that L. calbasu has the ecomorphology that enables them to change its food selectivity and feeding niche in response to food resource variability (Rahman et al., 2009). This trait may help L. calbasu to maximize fitness by reducing food competition and by maximizing growth through dietary shifts. Thus, L. calbasu is better than C. cirrhosus for food resource utilization and fish production when they are cultured together in non-fed pond. However, culturing them (alone or both) with other species might change its behaviour, food selectivity and food intake. Therefore, any prediction about culturing with additional species should be based on direct observation of the intended species combination. This study was conducted in simulated pond condition and therefore, more research is recommended before extrapolating results to an industrial level. Acknowledgments We gratefully acknowledge the European Commission for financial support through Pond–Live project. We also thank to Prof. Dr. M.A. Wahab, Bangladesh Agricultural University, Mymensingh, Bangladesh, for his cooperation during conducting this experiment. References APHA, 1998. Standard Methods for the Examination of Water and Waste Water. American Public Health Association, Washington, DC. Balcombe, S.R., Bunn, S.E., Davies, P.M., McKenzie Smit, F.J., 2005. Variability of fish diets between dry and flood periods in an arid zone floodplain river. J. Fish. Biol. 67, 1552–1567. Carpenter, J., 2005. Competition for food between an introduced crayfish and two fishes endemic to the Colorado River basin. Environ. Biol. Fish. 72, 335–342. Gonzalez-Soles, J., Oro, D., Jover, L., Ruiz, X., Pedrocchi, V., 1997. Trophic niche width and overlap of two sympatric gulls in the southwestern Mediterranean. Oecologia 112, 75–80. Holland, K.N., Peterson, J.D., Lowe, C.G., Wetherbee, B.M., 1993. Movements, distribution and growth rates of the white goatfish Mulloides flavolineatus in a fisheries conservation zone. Bull. Mar. Sci. 52, 982–992. Hopkins, K.D., 1992. Reporting fish growth: a review of the basics. J. World Aquacult. Soc. 23, 173–179.
Inskip, P.D., Magnuson, J.J., 1983. Changes in fish populations over an 80year period: Big Pine Lake. Wis. Trans. Am. Fish. Soc. 112, 378–389. Islam, M.R., Miah, M.S., Alam, M.S., Khair, M.A., Ahmed, S.S., Rokeya, J.A., 1998. Effects of various inputs on fish production under carp polyculture system. Bangladesh J. Fish. 21, 47–51. Johnsson, J.I., Winberg, S., Sloman, K.A., 2005. Social interactions. In: Sloman, K.A., Wilson, R.W., Balshine, S. (Eds.), Behaviour and Physiology of Fish. Elsevier Academic Press, Amsterdam, pp. 151–196. Kanak, M.K., Dewan, S., Salimullah, M., 1999. Performance of exotic fishes with Indian major carps in polyculture under three different species combinations. Bangladesh J. Fish. 22, 1–6. MacArthur, R.H., Pianka, E.R., 1966. On optimal use of a patchy environment. Am. Nat. 100, 603–609. Milstein, A., Wahab, M.A., Rahman, M.M., 2002. Environmental effects of common carp Cyprinus carpio (L.) and mrigal Cirrhinus mrigala (Hamilton) as bottom feeders in major Indian carp polycultures. Aquacult. Res. 33, 1103–1117. Pitcher, T.J., Parrish, J.K., 1993. Functions of shoaling behaviour in teleosts. In: Pitcher, T.J. (Ed.), Behaviour of Teleost Fishes. Chapman and Hall, pp. 363–439. Rahman, M.M., Kadowaki, S., Balcombe, S.R., Wahab, M.A., 2010. Common carp (Cyprinus carpio L.) alters its feeding niche in response to changing food resources: direct observations in simulated ponds. Ecol. Res. 25, 303–309. Rahman, M.M., Hossain, M.Y., Jo, Q., Kim, S.-K., Ohtomi, J., Meyer, C., 2009. Ontogenetic shift in dietary preference and low dietary overlap in rohu (Labeo rohita) and common carp (Cyprinus carpio) in semi-intensive polyculture ponds. Ichthyol. Res. 56, 28–36. Rahman, M.M., Verdegem, M.C.J., Nagelkerke, L.A.J., Wahab, M.A., Milstein, A., Verreth, J.A.J., 2006. Growth, production and food preference of rohu Labeo rohita (H.) in monoculture and in polyculture with common carp Cyprinus carpio (L.) under fed and non-fed ponds. Aquaculture 257, 359–372. Rahman, M.M., Verdegem, M.C.J., Nagelkerke, L.A.J., Wahab, M.A., Milstein, A., Verreth, J.A.J., 2008c. Effects of common carp Cyprinus carpio (L.) and feed addition in rohu Labeo rohita (Hamilton) ponds on nutrient partitioning among fish, plankton and benthos. Aquacult. Res. 39, 85–95. Rahman, M.M., Verdegem, M.C.J., Nagelkerke, L.A.J., Wahab, M.A., Verreth, J.A.J., 2008b. Swimming, grazing and social behaviour of rohu Labeo rohita (Hamilton) and common carp Cyprinus carpio (L.) in tanks under fed and non-fed conditions. Appl. Anim. Behav. Sci. 113, 255–264. Rahman, M.M., Verdegem, M.C.J., Wahab, M.A., Hossain, M.Y., Jo, Q., 2008a. Effects of day and night on swimming, grazing and social behaviours of rohu Labeo rohita (Hamilton) and common carp Cyprinus carpio (L.) in simulated ponds. Aquacult. Res. 39, 1383–1392. Robinson, C.J., 1995. Food competition in a shoal of herring: the role of hunger. Mar. Freshw. Behav. Physiol. 24, 237–242. Schoener, T.W., 1982. The controversy over interspecific competition. Am. Sci. 70, 586–595. Svanback, R., Bolnick, D.I., 2007. Intraspecific competition drives increased resource use diversity within a natural population. Proc. R. Soc. B 274, 839–844. Ward, A.J.W., Webster, M.M., Hart, P.J.B., 2006. Intraspecific food competition in fishes. Fish Fish. 7, 231–261. Ward, A.J.W., Botham, M.S., Hoare, D.J., James, R., Broom, M., Godin, J.-G.J., Krause, J., 2002. Association patterns and shoal fidelity in the threespined stickleback. Proc. R. Soc. Lond. [Biol.] 269, 2451–2455. Werner, E.E., Hall, D.J., 1979. Foraging efficiency and habitat switching in competing sunfish. Ecology 60, 256–264. Wuellner, M.R., Willis, D.W., 2008. On northern pond: fish competition. Pond Boss Mag. 17, 20–22.