We studied the diet of Brycinus sadleri in Lake Victoria after extensive environmental changes during the 1980s. To check for diet expansion following these changes, as observed in some other fish species, we compared our results with data from the 1950s. Stomach contents were analysed in relation to fish size and time of the day to investigate ontogenetic and diel shifts in the diet.

Comparison of diets before and after the environmental changes showed a shift to more generalist feeding. The main food sources in the 1950s were plants during daytime and surface insects at night. Hardly any plant material was eaten during the late 1980s, but chironomid larvae, chaoborid larvae and pupae, shrimps, Odonata nymphs and fish contributed to a broader daytime diet. These prey species were also commonly taken at night, though, like in the 1950s, surface insects accounted for the major part of the diet. We found dietary shifts during ontogeny. Juveniles fed exclusively on zooplankton. Chironomid larvae, relatively large food items in comparison with zooplankton, were the key prey of sub-adult fish. Their importance decreased in the diet of adults, which included even larger prey like shrimps, Odonata nymphs and fish. At night, surface insects replaced chironomid larvae in the diet of adult fish.

Diet expansion may result from reduced competition after the decline in fish mass following the Nile perch boom. An alternative explanation, increasing the prey spectrum in response to deteriorated light conditions, is not likely to hold for B. sadleri. The ontogenetic shift to increasingly larger prey and the nocturnal shift to surface insects probably reflect optimal foraging behaviour and diel vertical migration patterns of the fish respectively. These dietary shifts emphasize the feeding flexibility of B. sadleri, which may have been essential in surviving the environmental changes in Lake Victoria.

Introduction

The species–rich fish community of Lake Victoria was characterized by a high number of specialist feeders, in particular among the dominant haplochromine cichlids (Greenwood, 1974). During the 1980s the upsurge of introduced predatory Nile perch (Lates niloticus), together with environmental perturbations, had dramatic consequences. Of more than 123 haplochromine species, belonging to 12 trophic groups, originally caught at a series of sampling stations across the Mwanza Gulf (Tanzania) ca. 80 had disappeared from the catches after 1986 (Witte et al., 1992b). Haplochromine catches in the sub–littoral part of this research transect decreased to virtually zero after the Nile perch boom. Even in the littoral area, where Nile perch were less common, both the number of haplochromine species and their abundance decreased considerably. Additionally, many non–cichlid species declined (Witte et al., 1992a; Goudswaard and Witte, 1997; Goudswaard et al., 2002a). Such a strong reduction in fish mass may have reduced competition for food among the survivors and, consequently, the need for feeding specialisation (Milinski and Parker, 1991).

By 1987, shortly after the Nile perch boom, the key taxa in the sub–littoral waters had reduced to Nile perch and the cyprinid Rastrineobola argentea (local name: dagaa). Dagaa, which showed a strong population increase (Wanink, 1999), was known as a specialist zooplanktivore (Corbet, 1961; Greenwood, 1974). However, dagaa expanded its diet after the collapse of the haplochromines, including macrobenthic invertebrates like chironomids, chaoborids and the shrimp Caridina nilotica whenever available (Wanink, 1998). Major survivors in the littoral waters were Nile tilapia (Oreochromis niloticus), an introduced tilapiine cichlid, and the characid Brycinus sadleri. The phytoplanktivorous Nile tilapia added zooplankton and macrobenthos to its diet (Gophen et al., 1993; Mwebaza–Ndawula, 1994). Brycinus sadleri has been categorized as a macrophyte feeder, with surface insects as important additional prey (Corbet, 1961). In this paper we show that also this species has expanded its diet after the environmental changes. Additionally, we present ontogenetic and diurnal shifts in the post-Nile perch diet of B. sadleri.

Methodology

Fish were collected between May 1987 and March 1988 at various locations (water depth: 1–9 m) in the Mwanza Gulf (Tanzania), in the southern part of Lake Victoria. In this area, the population boom of Nile perch took place between 1984 and 1986. During daytime, bottom trawls were made using a cod end mesh of 5 or 20 mm. In addition, gill nets of 1.5 m deep (stretched mesh: 12.5 or 25 mm) were set at bottom, mid water and surface, and emptied every 2–4 hours. At night, both bottom-and surface trawls were made using a cod end mesh of 5 mm, while the same gill nets as used during daytime were set and emptied every 2–4 hours.

All fish were stored in 5% formalin immediately after collection. The preserved fish were measured to the nearest mm standard length (SL) and split into 15–mm size classes (from 16–30 mm to 76–90 mm). Since B. sadleri mature at about 63 mm SL (Greenwood, 1966) our two largest size classes (61–75 and 76–90 mm) represent adults. We have defined fish up to 45 mm as juveniles and the size class 46–60 mm as sub-adults.

Nine categories of food were distinguished: plants (pieces bitten from macrophytes), zooplankton (cladocerans and copepods), surface insects (adult insects that could not have been taken from the water column), chironomid pupae, chironomid larvae, chaoborids (pupae and larvae lumped), shrimps (Caridina nilotica), Odonata nymphs and fish. Volume percentages were estimated for each food category over the total number of non-empty stomachs broken up per size category into day-and night catches.

Comparative data from before the Nile perch boom were taken from Corbet (1961), who collected fish during the 1950s in the northern part of the Lake Victoria basin. His data were split into day (bottom trawl and angling rod) and night (gill nets) catches. Fish sizes ranged from 5 till 9 cm total length (TL) in day catches and from 7 till 9 cm in night catches. These have been transformed to standard length using the formula: SL = 0.79* TL (Wanink and Joordens, 2001) but could not be broken up into 15–mm classes. Volume percentages per food type were calculated from their frequency of occurrence and the percentage “main contents”, assuming a contribution of 100% by a prey type in each stomach for which it was noted down as “main contents” and of 50% when no score for “main contents” could be given.

Results

Diet expansion

All data on diet composition per size class and time period are given in Figure 1. Before the environmental changes in Lake Victoria the daytime diet of adult and sub–adult B. sadleri consisted mainly of plant material and surface insects, with some chironomid pupae and less than 1% chironomid larvae. Thereafter, plants virtually disappeared from the diet. Within the comparable size classes (46–90 mm) the contribution of surface insects to the daytime diet strongly declined, while slightly more chironomid pupae were taken. Chironomid larvae became the most common food type. Chaoborids, shrimps, Odonata nymphs and fish formed the novel part of the diet, together comprising more than 25% of the diet of the adults and sub–adults. A substantial part of the fish taken consisted of juvenile Nile perch of 10–12 mm SL.

At night, plant material comprised only 1% of the stomach contents before the environmental changes and was not at all found thereafter. Before the changes surface insects made up 68% and chironomid pupae 31% of the nightly diet. While the contribution of surface insects to the diet remained at the same level after the environmental changes, the chironomid pupae were replaced by the same four novel food types as found in the daytime diet.

Ontogenetic and diel shifts

Figure 1 clearly shows the importance of zooplankton as prey for juvenile B. sadleri. Its contribution to the diet had declined to 5% in sub–adult fish and had practically disappeared in adults. Chironomid larvae became the major daytime food source of both subadults and adults, although their contribution to the diet dropped from more than 80% in the former to less than 40% in the latter group. The amount of chironomid larvae in the nightly diet of sub–adults could not be measured because these fishes were not caught. Adults, however, took much less chironomid larvae during the night than by day. At night, surface insects formed the main food source of the adults. A substantial part of the diet of adults consisted of relatively large prey. Shrimps, Odonata nymphs and fish made up about 20% of the daytime diet and about 30% of the nightly diet of this group.

Discussion

Diet expansion

After the environmental changes B. sadleri from the Mwanza Gulf took virtually no macrophytes, during the 1950s their main food in the northern part of the lake. However, most macrophytes disappeared from the lake after heavy rains between 1961 and 1964 only to recover during the 1990s (Witte et al., 1992a; Kudhongania and Chitamwebwa, 1995). Nevertheless, at least some water lilies (Nymphaea spp.) and fringing papyrus and reed swamps were present in the littoral waters of the Mwanza Gulf during our study period (Wanink and Joordens, University of Leiden, the Netherlands, pers. obs.). In spite of the loss of plant material, B. sadleri expanded its diet by adding macrobenthic invertebrates and fish. Diet expansions have also been observed in other surviving fish species after the environmental changes. In the 1960s Nile tilapia in Lake Victoria fed mainly on phytoplankton with some additional detritus (Welcomme, 1967). By the end of the 1980s, the species had added zooplankton, ostracods, chironomid larvae, shrimps, and molluscs to its diet (Balirwa, 1992; Gophen et al., 1993; Mwebaza-Ndawula, 1994). At the same time, zooplanktivorous dagaa started to feed on chironomid and chaoborid larvae and shrimps in the sub-littoral and littoral waters of the Mwanza Gulf (Wanink, 1998). Haplochromis (Yssichromis) pyrrhocephalus and H. (Y.) laparogramma showed a comparable diet expansion. These species belonged to a group of co-existing zooplanktivorous cichlids from the sub-littoral part of the Mwanza Gulf that virtually vanished during the 1980s. After a period in which they were rarely caught, these two species recovered during the 1990s (Witte et al., 2000). They have now added ostracods, chironomid larvae, shrimps, molluscs and fish to their diet (Wanink et al., University of Leiden, the Netherlands, unpublished data). Haplochromis tanaos, a surviving zooplanktivorous cichlid from the littoral part of the Mwanza Gulf, became much more insectivorous and also started to feed on ostracods, shrimps and fish (van Oijen and Witte, 1996; Wanink et al., University of Leiden, the Netherlands, unpublished data).

Apparently, diet expansion has been a common response of surviving fish species in the Mwanza Gulf to the environmental changes during the 1980s. Key features of these changes were: 1) increased eutrophication, marked by deteriorated light-and oxygen conditions; 2) substantial loss of fish species and fish mass; 3) increased abundance of macrobenthic invertebrates, in particular chironomid larvae, shrimps and molluscs (Witte et al., 1992a, 2000; Wanink et al., 2001). At least two hypotheses that may explain the observed cases of diet expansion arise from these key features: 1) reduced visibility causes a decrease in effective encounter rate with preferred prey and, consequently, a broader diet (Seehausen et al., 2002); 2) decreased fish mass results in reduced competition for food, thus causing a shift from specialist to generalist feeding (Milinski and Parker, 1991). The increased abundance of macrobenthos may have reinforced a reduction in competition caused by the decline in fish mass.

So far, a general explanation for diet expansion has not been found. A study on dagaa provided support for the “competitive release” hypothesis (Wanink, 1998). However, reduced competition is an unlikely explanation for the diet expansion of Nile tilapia. This introduced species is regarded as a dominant competitor, probably having caused the decline in the two indigenous tilapiine cichlids in Lake Victoria and that of several fish species in other lakes where it was introduced (Ogutu-Ohwayo, 1990; Lowe-McConnell, 2000; Goudswaard et al., 2002b). Neither of the two hypotheses has as yet been tested on the zooplanktivorous haplochromines, though foraging experiments with captive fish under various light conditions are in progress (P.C. Chifamba et al., University Lake Kariba Research Station, Kariba, Zimbabwe, pers. comm.). The “dimmed light” hypothesis is unlikely to explain the diet expansion of B. sadleri, because Secchi disc values in the inshore habitat of this species showed little change between 1928 (Worthington, 1930) and 1987 (de Beer, 1989). In shallow areas the effect of daily breezes stirring up the water and bringing into suspension the fine silt from the bottom makes water turbidity less dependent on the abundance of phytoplankton (Worthington, 1930). Moreover, B. sadleri added larger food items to its diet, whereas the dimmed light model predicts a shift to smaller prey (Seehausen et al., 2002). Although there is no evidence for a formerly high competition pressure on B. sadleri, we regard competitive release as a plausible explanation for its diet expansion, based on the decline in potential competitors, i.e. the many insectivorous haplochromines in shallow waters (Witte et al., 1992b), after the Nile perch boom and the concomitant increase in macrobenthos.

Ontogenetic and diel shifts

The observed shift during ontogeny from zooplankton to increasingly larger prey is a common feature in fish. It is thought to reflect optimal prey choice, which allows for rapid growth. In turn, this would shorten the time spent in vulnerable stages and lower the age of reproductive maturity. As the fish grow, they can process larger food items in progressively less time, which make large prey more profitable than small ones (Galis and de Jong, 1986). The diel switch in sub-adult and adult B. sadleri between benthic chironomid larvae and surface insects can be related to diel vertical migration patterns of the fish, which dwell near the bottom during daytime and near the surface at night (Wanink and Joordens, University of Leiden, the Netherlands, unpublished data). Feeding on surface insects may be regarded as an opportunistic response to mass emergence of insects, since feeding usually is done during daylight mainly (Wanink and Joordens, University of Leiden, the Netherlands, unpublished data).

Feeding flexibility and survival

Corbet (1961) already called B. sadleri a potentially opportunistic feeder. The dietary shifts presented in the current paper confirm the species' feeding flexibility. This flexibility may have contributed significantly to its survival after the Nile perch boom. There are indications that B. sadleri has advanced its reproduction (Wanink and Joordens, University of Leiden, the Netherlands, unpublished data). Such a reproductive strategy is expected under increased predation pressure and could be made possible by maximization of the rate of food intake.

Conclusions

We studied the diet of Brycinus sadleri in Lake Victoria after extensive environmental changes during the 1980s. To check for diet expansion following these changes, as observed in some other fish species, we compared our results with data from the 1950s. Stomach contents were analysed in relation to fish size and time of the day to investigate ontogenetic and diel shifts in the diet.

Comparison of diets before and after the environmental changes showed a shift to more generalist feeding. The main food sources in the 1950s were plants during daytime and surface insects at night. Hardly any plant material was eaten during the late 1980s, but chironomid larvae, chaoborid larvae and pupae, shrimps, Odonata nymphs and fish contributed to a broader daytime diet. These prey species were also commonly taken at night, though, like in the 1950s, surface insects accounted for the major part of the diet. We found dietary shifts during ontogeny. Juveniles fed exclusively on zooplankton. Chironomid larvae, relatively large food items in comparison with zooplankton, were the key prey of sub–adult fish. Their importance decreased in the diet of adults, which included even larger prey like shrimps, Odonata nymphs and fish. At night, surface insects replaced chironomid larvae in the diet of adult fish.

Diet expansion may result from reduced competition after the decline in fish mass following the Nile perch boom. An alternative explanation, increasing the prey spectrum in response to deteriorated light conditions, is not likely to hold for B. sadleri. The ontogenetic shift to increasingly larger prey and the nocturnal shift to surface insects probably reflect optimal foraging behaviour and diel vertical migration patterns of the fish respectively. These dietary shifts emphasize the feeding flexibility of B. sadleri, which may have been essential in surviving the environmental changes in Lake Victoria.

Acknowledgements

We thank Prof. P.O.J. Bwathondi (director-general TAFIRI), Frans Witte (IBL, Leiden), Wim van Densen (FCFG, Wageningen), Michiel Berger (CEES, Groningen) and our colleagues from TAFIRI (Mwanza Centre) and HEST for their help. The Tanzania Commission for Science and Technology (COSTECH) is thanked for providing the necessary research permit. Frans Witte commented on an earlier draft. Financial support was obtained from WOTRO (grants W 87–189; W 84-488), the section DPO of the Netherlands Ministry of Foreign Affairs, Stichting Leids Universiteits-Fonds, Schure-Beijerinck-Popping Fonds, the van Tienhoven Stichting and Yellow Springs Instruments.

References

Balirwa, J. S.
1992
.
The evolution of the fishery of Oreochromis niloticus (Pisces, Cichlidae) in Lake Victoria
.
Hydrobiologia
,
232
:
85
90
.
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
.
De Beer, M.
1989
.
Light measurements in Lake Victoria, Tanzania
.
Ann. Mus. Roy. Afr. Centr., Sc. Zool.
,
257
:
57
60
.
Galis, F. and de Jong, P.
1986
.
Changes in foraging behaviour during the ontogeny of a zooplanktivorous cichlid from Lake Victoria
.
Ann. Mus. Roy. Afr. Centr., Sc. Zool.
,
251
:
53
56
.
Gophen, M., Ochumba, P. B.O., Pollingher, U. and Kaufman, L. S.
1993
.
Nile perch (Lates niloticus) invasion in Lake Victoria (East Africa)
.
Verh. Internat. Verein. Limnol.
,
25
:
856
859
.
Goudswaard, P. C. and Witte, F.
1997
.
The catfish fauna of Lake Victoria after the Nile perch upsurge
.
Env. Biol. Fish.
,
49
:
21
43
.
Goudswaard, P. C., Witte, F. and Chapman, L. J.
2002a
.
Decline of the African lungfish (Protopterus aethiopicus) in Lake Victoria (East Africa)
.
J. Afr. Ecol.
,
40
:
42
52
.
Goudswaard, P. C., Witte, F. and Katunzi, E. F.B.
2002b
.
The tilapiine fish stock of Lake Victoria (East Africa) before and after the Nile perch upsurge
.
J. Fish Biol.
,
60
:
838
856
.
Greenwood, P. H.
1966
.
The Fishes of Uganda,
, 2nd edition,
Kampala
:
The Uganda Society
.
Greenwood, P. H.
1974
.
The cichlid fishes of Lake Victoria, East Africa: the biology and evolution of a species flock
.
Bull. Br. Mus. (Nat. Hist.), Zool., Suppl.
,
6
:
1
134
.
Kudhongania, A. W. and Chitamwebwa, D. B.R.
1995
. “
Impact of environmental change, species introductions and ecological interactions on the fish stocks of Lake Victoria
”. In
The Impact of Species Changes in African Lakes
, Edited by: Pitcher, T. J. and Hart, P. J.B.
19
32
.
London
:
Chapman and Hall
.
Lowe-McConnell, R. H.
2000
. “
The roles of tilapias in ecosystems
”. In
Tilapias: Biology and Exploitation
, Edited by: Beveridge, M. C.M. and McAndrew, B. J.
129
162
.
Dordrecht
:
Kluwer Academic Publishers
.
Milinski, M. and Parker, G. A.
1991
. “
Competition for resources
”. In
Behavioural Ecology: an Evolutionary Approach,
, 3rd edition, Edited by: Krebs, J. R. and Davies, N. B.
137
168
.
Oxford
:
Blackwell Scientific Publications
.
Mwebaza-Ndawula, L.
1994
.
Changes in relative abundance of zooplankton in northern Lake Victoria, East Africa
.
Hydrobiologia
,
272
:
259
264
.
Ogutu-Ohwayo, R.
1990
.
The decline of the native fishes of lakes Victoria and Kyoga (East Africa) and the impact of introduced species, especially the Nile perch, Lates niloticus, and the Nile tilapia, Oreochromis niloticus
.
Env. Biol. Fish.
,
27
:
81
96
.
Seehausen, O., van Alphen, J. J.M. and Witte, F.
2002
. “
Implications of eutrophication for fish vision, behavioural ecology and species coexistence: A theoretical framework
”. In
Aquatic Conservation and Management in Africa
, Edited by: Crisman, T. L., Chapman, L., Chapman, C. A. and Kaufman, L. A.
268
287
.
Gainesville
:
University Press of Florida
.
Van Oijen, M. J.P. and Witte, F.
1996
.
Taxonomical and ecological description of a species complex of zooplanktivorous and insectivorous cichlids from Lake Victoria
.
Zool. Verh. Leiden.
,
302
:
1
56
.
Wanink, J. H.
1998
.
The Pelagic Cyprinid Rastrineobola argentea as a Crucial Link in the Disrupted Ecosystem of Lake Victoria: Dwarfs And Giants – African Adventures
,
PhD Thesis
Leiden
:
Leiden University
.
Wanink, J. H.
1999
.
Prospects for the fishery on the small pelagic Rastrineobola argentea in Lake Victoria
.
Hydrobiologia
,
407
:
183
189
.
Wanink, J. H. and Joordens, J. C.A.
2001
. “
Is Brycinus sadleri from Lake Victoria an anadromous fish?
”. In
Academician Leo Berg – 125 Years: Collection of Scientific Articles
, Edited by: Trombitsky, I.
58
62
.
Bendery, Moldova
:
BIOTICA
.
Wanink, J. H., Kashindye, J. J., Goudswaard, P. C. and Witte, F.
2001
.
Dwelling at the oxycline: does increased stratification provide a predation refugium for the Lake Victoria sardine Rastrineobola argentea?
.
Freshw. Biol.
,
46
:
75
85
.
Welcomme, R. L.
1967
.
Observations on the biology of the introduced species of Tilapia in Lake Victoria
.
Rev. Zool. Bot. Afr.
,
76
:
249
279
.
Witte, F., Goldschmidt, T., Goudswaard, P. C., Ligtvoet, W., van Oijen, M. J.P. and Wanink, J. H.
1992a
.
Species extinction and concomitant ecological changes in Lake Victoria
.
Neth. J. Zool.
,
42
:
214
232
.
Witte, F., Goldschmidt, T., Wanink, J., van Oijen, M., Goudswaard, K., Witte-Maas, E. and Bouton, N.
1992b
.
The destruction of an endemic species flock: quantitative data on the decline of the haplochromine species from the Mwanza Gulf (Lake Victoria)
.
Env. Biol. Fish.
,
34
:
1
28
.
Witte, F., Msuku, B. S., Wanink, J. H., Seehausen, O., Katunzi, E. F.B., Goudswaard, P. C. and Goldschmidt, T.
2000
.
Recovery of cichlid species in Lake Victoria: an examination of factors leading to differential extinction
.
Rev. Fish Biol. Fish.
,
10
:
233
241
.
Worthington, E. B.
1930
.
Observations on the temperature, hydrogen-ion concentration, and other physical conditions of the Victoria and Albert Nyanzas
.
Int. Rev. Gesamt. Hydrobiol.
,
24
:
328
357
.