Kenya Marine and Fisheries Research Institute,
P.O. Box 1881,
Prior to the introduction of Lates niloticus, a balance between predators and prey species had evolved in Lake Victoria. Adaptations by both predators and prey had ensured that extermination of a species would not occur as a result of predation.
Since its introduction into Lake Victoria, Lates niloticus has become established as the dominant species and now makes up 50% of the fish landings from the lake. L. niloticus demonstrates a natural ontogenic change in diet but also has the capability of adjusting its feeding habits to take advantage of the most abundant food source. In Lake Victoria, as in other lakes where Lates has been introduced, many endemic prey-species have been almost elimimated and the Lates population is now largely cannibalistic.
Lake Victoria was formed by tectonic movements during the mid-Pleistocene (Fryer and Iles, 1972). The lake has a surface area of about 68,800 km2, a maximum depth of 79m and a mean depth of 40m.
Of the estimated 200 species of fish occurring in Lake Victoria, about 30–40% are piscivorous. The principal predatory fishes are Lates niloticus, Bagrus docmac, Clarias mossambicus, Schilbe mystus and some species of Haplochromis (sensu lato). Other predators in Lake Victoria include fish-eating birds, fish-eating snakes, crocodiles and otters. The lake supports important commercial fisheries based mainly on Oreochromis esculentus, Oreochromis variabilis, Bagrus docmac, Clarias mossambicus, Protopterus aethiopicus, Labeo victorianus, Schilbe mystus, the introduced Lates niloticus and Oreochromis niloticus.
Predator-prey interactions between the fish species in Lake Victoria are poorly documented. Descriptions of the relationship between various fish species and their prey have been made by Graham (1929), Corbet (1961), Okedi (1970), Hamblyn (1966) and Gee (1969). The relationship between Bagrus docmac and its prey species in Lake Victoria has been described by Chilvers and Gee (1974) and Ochieng' (1982). Ogari (1984) described the relationship between Nile perch and its prey species in the Nyanza Gulf.
This paper covers a literature review on patterns of predation, including the major defences adopted by prey species, and includes a description of the relationship between Lates niloticus and its prey in the Nyanza Gulf of Lake Victoria.
Predators and prey have a wide spectrum of adaptive strategies to cope with their interactions. Predators are aided by morphological adaptations, which aid in pursuing, detecting and digesting prey. Most of the predatory fishes in Lake Victoria have streamlined bodies, large mouths and large eyes. Fryer (1965) described Lates niloticus as “swift” and Bagrus and Clarias as “lurkers”. Fryer and Iles (1972) reported that among members of the genus Haplochromis in Lake Victoria, the piscivorous species are larger in size (average maximum length 21.6cm). They described experiments to determine visual orientation for prey location using Haplochromis gowersi in Lake Victoria. Hamblyn (1966) performed similar experiments using Lates niloticus in Lake Albert. In both cases the predator seized and swallowed its prey whole, taking the prey in head first, however, H. gowersi chewed large prey before swallowing.
Prey fish also have morphological and behavioural adaptations for protecting themselves from predators. Lowe-McConnell (1975) listed various mechanisms fish use to avoid predators, such as schooling, modification of body shape and colour, possession of spines, oral incubation, guarding of eggs and young etc. These adaptations do not afford complete protection but can serve to increase the chance of survival (Popova, 1978).
Schooling of fish in Lake Victoria is common among such pelagic open-water species as Rastrionebola argentii, Alestes jacksoni and some Haplochromis species. Fryer and Iles (1972) explained that protection appears to be conferred by schooling because a predator presented with a single item of food receives only one set of stimuli which direct it to its prey, whereas confrontation with a school causes it to receive numerous conflicting stimuli which block the feeding response. They also mention that such an advantage will be greatest when the fishes are small and therefore vulnerable to predation. Hopson (1972 & 1975) observed leaping behaviour among small characins and Rastrionebola stellae in Lakes Chad and Turkana respectively. He concluded that such evasive movements were intended to assist the prey species escape predation. Comparable behaviour has been observed among the pelagic species of Lake Victoria (pers. obs.)
Predatory fishes are able to consume cylindrical-shaped fishes which are of a larger size than the deep bodied fish which they may also prey upon. An effective combination of strong fin spines and a deep body probably protects Oreochromis niloticus from most of the predators in Lake Victoria. In Lake Kioga, fishermen have reported the appearance of Nile perch floating dead on the surface and in many cases a large Oreochromis niloticus has been found stuck inside the buccal cavity of the predator (Okedi, 1970). The shape and movements of long posterolateral spines in the rotifer Brachionus calyciflorus significantly decreases predation by Asplanchna (Wetzel, 1975). Wetzel (Op. cit.) found that adult Asplanchna sieboldi could capture nearly 100% of adult spineless B. calyciflorus contacted, but only about 78% of the long spined forms. Synodontis species which occur in Lake Victoria have a boney head region provided with stout spines on both pectoral and dorsal fins. When opened, these spines effectively increase the size of the fish thus making them more difficult to swallow and restricting their availability to larger predators. Hopson (1972) observed that Synodontis preyed upon by Lates niloticus in Lake Chad rarely exceeded 18% of the predators length. He also reported finding a dead Nile perch with a Synodontis jammed in its throat.
Another important protective measure is parental care for the eggs and young. Some cichlids are mouth-brooders, and in such cases the males, females or both carry the eggs to special brooding grounds. The parents eventually leave their young in nursery areas which are situated inshore. Welcomme (1964) has shown that juvenile Oreochromis variabilis migrate away from the sheltered nursery beaches when they are around 5.0cm total length. In both Lake Albert and Lake Turkana the young of prey-species use the cover of the sublittoral Ceratophyllum zone and the shallow lagoons in order to escape predation (Gee, 1969). Greenwood (1965) found that Protopterus aethiopicus spends much of its first year in the cover provided by the matted root systems of papyrus in Lake Victoria. Macan (1977) reported that Asellus on mud is at risk from predation by Sialis intaria (L) but amongst Elodea it is relatively safe from all predators.
Another important adaptation developed by prey-species is migration. Some species such as Barbus altianalis, Labeo victorianus and Schilbe mystus are potamodromous, moving up rivers to spawn. The young often remain upstream in shallow water where they are more likely to escape predation by large fish (Lowe McConnell, 1975). The majority of the fishes in Lake Victoria appear to be more demersal than pelagic (Kudhongania and Cordone, 1974), however, in order to partially segregate themselves from predators during periods of peak susceptibility, some Haplochromis species migrate vertically on a diel basis. This may also be a feeding strategy since the zooplankton upon which they feed also shows such diurnal migrations. Kudhongania and Cordone (op. cit.) observed higher catch rates for Haplochromis and Bagrus docmac in bottom-trawl hauls made during the day than during the night, whereas the mid-water trawl catches were higher during the night than during the day. This suggests that the prey fish may be gaining some protection by descending to dimmer lit zones during the day. In Lake Victoria there are probably two periods of peak feeding activity by Bagrus on Haplochromis, one in the morning the other in the evening (EAFFRO - Progress Report 1969). Ochieng (1982) confirmed two peak feeding periods in Bagrus docmac, one at sunset and the other at sunrise. Such crepuscular feeding activity by Bagrus may be due to low optimal light conditions required by the predator for feeding or may reflect diel behavoural changes in the haplochromine prey species. Elliot (1970) reported an increase in invertebrate drift at night and proposed that trout utilise some of this readily available drift food especially in the early hours of the night. The Lake Tanganyika clupeids (Limnothrissa and Stolothrissa) undergo vertical migration movements to escape large, diurnal predators such as Lates (Coulter, 1966).
Lates niloticus was introduced into Lake Victoria between 1959 and 1963 at Jinja in Uganda. The specimens introduced were from Lake Albert, Uganda. The species has established itself mainly in the shallow, inshore areas but is now colonising the deeper waters of the Lake. Goudswaard and Witte (1984) reported catches of 80 to 150 kg/hr of Nile perch at depths of 50–60m in Tanzanian waters. In Kenya, the present commercial fishery is based mainly on Lates which, in 1984, contributed 57% of the total landing. Similar figures have been noted in Uganda (Acere, 1984).
Lates niloticus is the largest carnivorous fish in Lake Victoria and is appreciably larger than the indiginous Bagrus docmac. The species is endowed with speed, large size and a massive mouth enabling it to capture large prey. The largest specimen caught in the gulf was 204cm TL (pers. obs.). Lates niloticus needs to make short bursts of high speed swimming to capture its prey (Holden, 1967).
Figure 1. Diet of L. niloticus in 10 cm size groups assessed by percentage occurrence method. Figures in parenthesis are sample sizes.
|1. Lates niloticus||2. S. niloticus||3. E. argentii|
|4. Haplochromis||5. P. aethiopicus||6. Clarias sp.|
|7. Alestes sp.||8. M. frenetus||9. Synodontis sp.|
|10. Labeo sp.||11. Mormyrus sp.||12. Bivalves|
|13. Gastropods||14. Odonata nymphs||15. Chironomid larvae|
|16. Ephemeroptera nymphs||17. Corixid adults||18. Chaoborus larvae|
|19. Caridina nilotica||20. Crabs||21. Cladocerans|
|22. Ostracods||23. Copepods|
Fig. 2. Relationship between size of Lates niloticus and its prey (from stomach content analyses)
The results of the present study indicate a significant size related shift in the feeding behaviour of Lates. As it grows, its diet changes from mostly invertebrates to almost entirely fish. The findings made using percentage occurrence analysis of prey categories with respect to the size of the predator are presented in Fig.1. The results reveal that up to a predator size of 70cm (TL), crustacea were numerically the principle consituent of the diet, accounting for 44.2% of the food items eaten, followed by fish which made up 34.3% of the diet. In Lates larger than 70cm, fish was the dominant food item, contributing 69.2% whereas the number of crustacea in the diet declined to 20%. Hunter (1970) observed that juvenile Lates feed predominantly on invertebrates whereas the adults prey mainly on fish. Change in diet as a fish increases its size has been recorded in many species and Hellawell (1972) noted variation with age in the composition of the diet in Rutilus rutilus. In the present investigation, the number of large prey items tended to be larger than the the number of small items found in the stomachs of Lates. The maximum number of food items reported in a single stomach was 7 Oreochromis niloticus, 21 juvenile Lates and 105 Rastrionebola argentii (in a Lates in the 89–90 cm size group) and 628 Caridina nilotica in a specimen from the 40–50cm size group.
The relationship between predator size and prey size is presented in Fig.2. Analysis of the sizes of three fish prey-species (Lates niloticus, Oreochromis niloticus and Haplochromis spp.) recovered from Lates stomachs reveals an increase in prey size with increase in predator size. The figure further shows that most of the prey species selected by the predator were less than one third of its size. Another point Fig. 2 shows is that, whereas the maximum size of prey ingested increased with predator size, the minimum size of prey barely increased. Moore and Moore (1976), studying Anguilla anguilla, found that the maximum length of frequently ingested organisms increased from 3cm in fish of 15–25cm to 6.5cm in specimens measuring 45. 1–55cm.
Significant changes were observed in the composition of the diet of Lates from the five depth strata sampled. Fig. 3 shows that, with increase in depth, the occurrence of both insects and fish items (Lates of all lengths combined) declined from 17.6% to 5.2% and from 37.3% to 17.1% respectively. Crustacea, of which Caridina nilotica was the most important, constituted 23.1% of diet of Lates from the shallow waters increasing to 71.6% in specimens from the deeper zones. Molluscs made a small contribution to the diet but showed a slight increase within the intermediate depth stratum. Hunter (1970), when working on Lake Albert, indicated that Lates niloticus of less than 20cm in length preyed mainly on Odonata nymphs within the shallow water and in lagoons, whereas in the offshore areas, Caridina nilotica formed the bulk of the diet of juvenile Lates. Lock (1975) found that although Bagrus bayad preyed upon a variety of fish species, their relative importance in the diet varied with depth. Hopson (1972) made similar observations among Lates found in the inshore and offshore areas of Lake Chad.
Percentage occurrence of food items in the stomach of L. niloticus from five depth strata. Figures in parenthesis are sample sizes.
|1. Lates niloticus||2. O. niloticus||3. R. argentii|
|4. Haplochromis sp.||5. P. aethiopicus||6. Clarias sp.|
|7. Alestes sp.||8. M. frenetus||9. Synodontics sp.|
|10. Labeo sp.||11. Mormyrus sp.||12. Bivalves|
|13. Gastropods||14. Odonata nymphs||15. Chironomid larvae|
|16. Ephemeropteran nymphs||17. Corixid adults||18. Chaoborus larvae|
|19. Caridina nilotica||20. Crabs||21. Cladocerans|
|22. Ostracods||23. Copepods|
Results of the present investigations revealed that the dominant food items were, Lates niloticus juveniles and Rastrionebola argentii within the shallow inshore zones, whereas in the deeper zones the primary food item was Cardina nilotica.
Gee (1969) and Okedi (1970) stated that the primary food of Lates in Lake Victoria was Haplochromis spp. The general absence of Haplochromis from the stomachs in the present investigations, reflects their scarcity in the area studied. This point is illustrated by the fact that the bottom trawl hauls usually yielded less than 1.0kg/hr of Haplochromis at all depth zones (Table 1). The preponderance of juvenile Lates and Caridina niloticus in the stomachs of Lates is also indicative of lack of other suitable prey. Ogutu-Ohwayo (1984), while working in Lake Kioga, found that his observations differed from those of Hamblyn (1961), Gee (1964 & 1969) and Okedi (1970) and indicated that, as a result of the change in type of prey available, the Nile perch in Lake Kioga has shifted from one prey to another over the years. Earlier studies had shown that the main prey was Haplochromis and mormyrids which were later replaced by Rastrionebola. However, by 1984, Oreochromis niloticus had become the major prey of Lates in Lake Kioga. The changes in diet with time is probably related to the availability and relative abundance of the prey items. Royce (1972) states that when a newly introduced predator population starts to consume a prey population that has been in equilibrium with its competitors and other predators, the first consequence is an increase in the mortality rate of the prey population.
Macan (1977) describes how the introduction of Salmo trutta into an artificial pond was followed by a greater reduction in the numbers of tadpoles, certain beetles and Notanecta species mainly found in open water. He concluded that only small invertebrates can survive predation in open water.
In the present investigations the dominant fish prey were Lates juveniles and this confirmed that cannibalism is prevalent. Hopson (1972) considers that cannibalism probably results partly from a dearth of alternative food sources. He noted scarcity of alternative food among cannibalistic Lates on the open shore at Malanfatori in Lake Chad. Both Okedi (1970) and Ogutu Ohwayo (1984) reported cannibalism in Lake Kioga. In the present study, a comparable scarcity of alternative prey is noted (Table 1.)
TABLE 1 Mean bottom-trawl catch rates (kg/hr) of the different organisms which form the food of Lates niloticus at different depth strata within the Nyanza Gulf.
|No. of Hauls||21||72||21||24||13|
+ Sample less than 1.0kg/hr
- No sample
Young tilapiine cichlids of various species are capable of tolerating water with an oxygen content as low as one part per million in Lake Victoria (Welcomme, 1964). Greenwood (1966) noted that aerial breathers such as Protopterus aethiopicus and Clarias are able to live in the swamps because of their ability to utilise atmospheric oxygen. Hunter (1970) quoted the experimental work of Fish (1956) which proved that L. niloticus has a relatively high oxygen requirement compared with many other freshwater fishes. The low oxygen demand by the above species apparently enables them to escape predation by Lates niloticus. Thorpe (1977) referred to a suggestion by Regier et al. (1969), that, together with the smelt (Osmerus mordax), perch fry are probably protected from walleye predation due to hypolimnial oxygen depletion which excludes walleye from their foraging base in Lake Erie.
Kudhongania and Cordone (1974) noted that the best catches of both Haplochromis and Bagrus were within a 10–60m depth range, and, from these results, concluded that Bagrus follows Haplochromis more closely than any other species, in terms of relative abundance, depth preference and diel vertical movements. Lock (1975) pointed out that vertical migration appeared to be an adaptation to maintain Bagrus bayad in contact with its prey in Lake Turkana. Echosounding in Lake Kivu has indicated that Oreochromis niloticus regani lives in shoals that disperse at dusk and reform at dawn (Fryer and Iles, 1972). Bagrus docmac seems to be most active during the transition between day and night when prey shoals are either forming or breaking up. Most of the prey items in Lake Victoria are probably either diurnal (pelagic species) or nocturnal (mainly demersal species) and are ill-equipped for rapidly changing conditions that prevail during the transition periods both in the evening and morning.
A progression from planktivorous feeding habits in the smallest fry, through a transitional insectivorous phase to a predominantly piscivorous diet is well known in many predatory freshwater fish (Hopson, 1972). The transition in the food habits of Lates has been observed by Hamblyn (1961 & 1966), Okedi (1970), Holden (1967), Gee (1969), Hopson (1972 & 1975) and Coulter (1976). A similar pattern has been observed in Bagrus docmac (Chilvers and Gee, 1969 and Ochieng, 1982).
Fryer and Iles (1972) noted that the largest prey eaten by Haplochromis gowersi was 25% of the predator's length. Hamblyn (1966), Gee (1969) and Hopson (1972) observed a general increase of prey size with increase in predator size in Nile perch. Gee (1969) reported the proportion of predator body length to prey length to be between 25–30%, whereas Okedi (1970) and Coulter (1976) observed a proportion of about 35%. Hopson (1972) however, noted a maximum proportion of about 50%. Comparable increase in prey size to predator size has been reported in Bagrus bayad (Lock, 1975) and Bagrus docmac (Chilvers and Gee, 1974).
Corbet (1961) concluded that interspecific competition for food among the Lake Victoria non-cichlid fishes (Lates excluded) played a very minor part in their ecology. He observed that the few species that shared specialised feeding habits appeared to have a superabundance of food, and others coped with overlap by remaining mobile and facultative. Hamblyn (1966), when comparing the fish faunas of Lakes Albert and Rudolf (where Lates is endemic) with those of Lake Victoria and Kioga (where it is non-endemic), concluded that a large number of small species existed in the latter lakes compared with a small number of large species in the former lakes. Gee (1966) suggested that the behaviour of prey species in the lakes where Lates is endemic, has been modified so that they live in localities protected from predation until they reach a “safe-size”. He also states that in Lake Victoria, where Lates is not endemic, there are a large number of Haplochromis (and other species) which never grow large and are not confined to the sub-littoral weed cover.
The presence of large numbers of Haplochromis species in Lake Victoria was probably attributable to less severe predation by the endemic species.
The Lake Victoria ecosystem, prior to the introduction of Lates, appears to have been characterised by fishes schooling by day and dispersing to feed at night, a predominance of demersal species and a low natural mortality due to predation.
Lates niloticus has become highly successful in Lake Victoria due to the fact that it is a more efficient predator than any endemic species in the lake. The impact of a large population of Lates on the other species has led to a dramatic reduction in the catches of these prey species. With the establishment of Lates in the lake, the numbers of prey species have fallen to such a level that Lates has resorted to cannibalism. This trend might enable the populations of prey species to recover as the number of Lates declines, and the prey may also learn to adapt and co-exist with a population of large voracious carnivores. An oscillation in the size of the Lates and prey species populations continues but its amplitude is bound to decrease. Eventually the numbers of prey and predators should remain relatively constant.
Chilvers, R.K. and J.M. Gee, 1974 The food of Bagrus docmac and its relationship with Haplochromis in Lake Victoria. E. Africa J.Fish.Biol., 6:483–505
Corbet, P.S., 1961 The food of non-cichlid fishes in the Lake Victoria Basin with remarks on their evolution and adaptation to lacustrine conditions. Proc.Zool. Soc.Lond. 136:1–101
Coulter, G.W., 1976 The biology of Lates species in Lake Tanganyika and the status of the pelagic fishery for Lates species and Luciolates stappersii (Blgr.) J. Fish.Biol., 9:235–59
EAFFRO - Progress Report, 1969 April – June 1969.
Elliott, J.M., 1970 Diel changes in Invertebrate drift and the food of trout Salmo trutta. J.Fish.Biol., 2:161–65
Fish, G.R., 1956 Some aspects of the respiration of six species of fish in Uganda. J.Exp.Biol., 33:186–95
Fryer, G., 1965 Predation and its effects on migration and speciation in African fishes: a comment with further comments by P.H. Greenwood, a reply by P.B.N. Jackson and a footnote and postscript by G. Fryer. Proc.Zool. Soc.Lond., 144:301–22
Fryer, G. and T.D. Iles, 1972 The cichlid fishes of the Great Lakes of Africa. Edinburgh, Oliver and Boyd, 641p.
Gee, J.M., 1969 A comparison of certain aspects of the biology of Lates niloticus (L) in endemic and introduced environments in East Africa, in Man-Made lakes: the Accra Symposium, ed. L.E. Obeng, Ghana Univ. Press: 251–9
Graham, M., 1929 The Victoria Nyanza and its Fisheries. A report on the Fisheries Survey of Lake Victoria, 1927–1928. Crown Agents, London.
Greenwood, P.H., 1966 The Fishes of Uganda, 2nd rev. edn. Kampala, Uganda Society.
Hamblyn, E.L., 1961 Nile perch investigation. Ann.Rep.EAFFRO 1961 23–8
Hamblyn, E.L., 1966 The food and feeding habits of Nile perch :Lates niloticus (Linne) (Pisces: Centropomidae). Rev.Zool. Bot.Afr., 74:1–28
Hollawell, J.M., 1972 The growth, reproduction and food of the roach Rutilus rutilus (L) of the River Lugg, Herefordshire. J.Fish.Biol., 4:469–86
Holden, M.J., 1967 The systematics of the genus Lates niloticus (Teleostei: Centropomidae) in Lake Albert, East Africa. J.Zool.Lond., 151:329–42
Hopson, A.J., 1972 A study of the Nile perch Lates niloticus (L.) in Lake Chad. Foreign and Commonwealth Office: London HMSO Overseas Res. Pub., 19:93
Hopson, A.J., 1975 Preliminary observations on the biology of Lates niloticus in Lake Rudolf. Symposium on Hydrobiology and fisheries of Lake Rudolf, Molo 25 – 29 May, 1975.
Hunter, J.G., 1970 Observations on the taxonomy and biology of Lates (Cuvier, 1928) in Lake Albert. Uganda Occasional Papers No.3
Kudhongania, A.W. and A.J. Cordone, 1974 Bathospatial distribution patterns and biomass estimates of major demersal fishes in Lake Victoria. J.Trop.Hydrobiol.Fish., 3:15–31
Lock, J., 1975 Some preliminary results in the investigation of Bagrus bayad. Symposium in the Hydrobiology and Fisheries of Lake Rudolf. Molo 25 –29 May, 1975.
Lowe-McConnell, R.H., 1975 Fish communities in Tropical Freshwaters. London and New York, Longman 337p.
Macan, T.T., 1977 The influence of predation on the composition of freshwater animal communities. Biol.Rev., 52:45–70
Moore, J.W. and I.A. Moore, 1976 The basis of food selection in some estuarine fishes. Eels, Anguilla anguilla (L), whiting, Merlangius merlangus (L), sprat, Sprattus sprattus (L.) and stickleback, Gasterosteus esculeatus L. J.Fish.Biol., 9:375–90
Nikolsky, G.V., 1965 The Ecology of Fishes, London and New York, Academic Press 352p.
Ochieng, J., 1982 Reproductive biology and feeding ecology of a predatory siluroid catfish: Bagrus docmac Forskaal (Pisces: Bagridae) in Winam Gulf of Lake Victoria: E. Africa. M.Sc. Thesis, University of Nairobi.
Ogari, J., 1984 The biology of Lates niloticus (L) in the Nyanza Gulf of Lake Victoria (Kenya) with special reference to the food and feeding habits.
Okedi, J., 1970 Further observations on the ecology of the Nile perch (Lates niloticus Linne) in Lake Victoria and Kioga. Ann.Rep.E.Afr.Freshwat.Fish.Res.Org. p 42–55
Popova, O.A., 1978 The role of Predaceous Fish in Ecosystems. In S.D Gerking (ed.). Ecology of freshwater fish production, New York, Wiley, p 215–49
Royce, W.F., 1972 Introduction to the Fishery Sciences. New York, Academic Press, 351p.
Stroud, R.H. and H. Clepper, 1978 Predator-Prey Systems in Fisheries Management. International Symposium on Predator-Prey Systems in Fish Communities and their Role in Fisheries Management. Atlanta, Georgia, July 24–27, 1978.
Thorpe, J., 1977 Synopsis of biological data on the perch Perca fluvialitis (L) and Perca flavescens (Mitchell, 1814).
Welcomme, R.L., 1964 The Habits and Habitat Preferences of the young of the Lake Victoria Tilapia (Pisces Cichlidae). Rev.Zool.Bot.Afr., 70: 1–28.
Wetzel, R.G., 1975 Limnology. Philadelphia, W.B. Saunders Co., 743 p.