Spatio=temporal distribution of the fish Synodontis schall (Teleostei: Mochokidae) in Asa lake,Ilorin, Nigeria

Spatio,temporal distribution of Synodontis scha!l in Asa lake was studied for 24 months (March 1991 to February 1993). Distribution of individual was: 28.40 % (surface), 35.60 % (shore), and 36.0 % (bottom). Catches wilhin lhe habitat were nOl significantly differen!. Similarly catches within lhe habitats during the peri­ ods of wet (May lO October) and dry (November lO April) seasons were nOl significanlly differen!. There was seasonal occurrence of lhe fish al the bottom because the catches from lhis habitat in October (4.80 to 9.10%) and November (0.00 lO 8.70%), corresponding with the period of fiood and high water levels, were relatiyely low due lO feeding and reproductive phenomena. Although catch was inversely proportional to lhe water levels, this was not significant in this experimental gil! net catches. Synodontis schall was caught throughout lhe sam, pling period indicating its successful adaptation wílhin lhe environment due to low predatíon and its díverse feeding habits.

and Mochokus, of whieh the genus Synodontis is the most eommon and of great eommereial importanee (Reed et al. 1967).Synodontis speeies only oeeur in Afriea and, apart from those speeies present in the River Nile, they are restrieted to water systems within the trop ies (Willoughby 1974, Berra 1981).
Sorne early work on this genus includes the food and feeding habits by Pekkola [1919].Further work on this was reported by Sandon and EI-Tayed [1953] and studies on feeundity were made by Nawar (1958Nawar ( , 1959)).Reeent report on this genus in Nigeria includes that of Oní et al. (1983) on eompar ative physiology, Olatunde (1989) on the Biology of S. schall in Zaria and Araoye (1997) on Eeology of the Moehokid in Asa lake.Work done on distribution and abun dance was restrieted to very large water bod ¡es, including Lake Volta, the Upper Nile, Lake Chad and Kainji Lake of Nigeria.In the Nile, total eateh from the commereial fish eries was reported to ¡nerease with inerease in water levels (Bishai and Abu Gideiri 1967).
Twenty-one speeies have been reported to oeeur in Nigerian inland waters and most of them were found within the Sudanean and Guinean zones while very few speeies, like S. schall oeeur universally (MeConnell 1965, Reed et al. 1967).Ita (1978) observed in Kainji Lake that S. membranaceus and S. resupinatus exhibited distinet habitat preferenees.The for mer prefers unclear are as while the ¡atter prefers habitat with c10se proximity to the shoreline and fairly deep waters.
This works which is the first scientific report on S. schall Ín the lake, Ís aimed at pro viding informatÍon on the spatial and temporal distribution of the fish to enlighten artisanal fisheries on where and when to set their nets around the lake.

MATERIALS AND METHODS
Asa dam is located about 5 km South of lIorin, across River Asa in Ilorin, the capital of Kwara State (080 26' N, 040 29' E).The dam was constructed primarily for domestic water supply.The lake, with an area of 302 hectare, has four tributaries river Asa being the major one (Fig. 1).The sampling sites (stations 1 to 3) extended from the dam to Afon, near the river source, where the maximum mean depth at high water levels was 12 m.Close to the dam maximum mean depth was 14 m.Bimonthly collections of specimens of S. schall for 24 months (March 1991 to February 1993) were carr ied out at the sampling stations with the services of the local fishermen who were a1so in volved in the keeping of catch records.GiIl nets of different mesh sizes including 5.08, 6.35, 7.62, 8.89 and 10.16 cm in a fleet were used to sample randomly the surface, shore and bottom habitats.F1eet of nets for the shore and surface were in ply 2, while the bottom nets were in ply 4 in order to resist the effect of the bottom pressure which could foid up the neL Each net was 125 m long and it was set at sÍx randomly selected places (Le.two in each habitat) of the sampling sta tions making a total of 18 sets of net in all the three stations along the lake.Nets were set at about 6 pm and drawn the following day at about 7 am.
Specimens of S. schall were separated and numbers caught from the different habitats were recorded separately.Each specimen was allotted a serial number after which they were brought to the laboratory in three different ice containers for measuring and weighing.Standard length (cm) and body weight (g) of each specimen were determined using a gradu ated measuring board in centimetres and a top loading Metler balance in grams respectively.Length and weight measurement were taken to the nearest 0.1.The percentage catches from each habitat was computed and compared sta tistically using analysis of variance.Also catches from the habitats during the period of wet (May to October) and dry (November to April) seasons were compared statistically.The student t-test analysis was a1so used for statistical comparison of the catches between the habitats.
Mean monthly rainfaJl from March, 199] to February, 1993 was obtained from llorin intemational airport and mean monthly water levels were recorded from the water leve!board installed at the dam.The graph of the monthly variation of the mean weight of fish and mean water levels is also presented.

RESULTS
water levels ( h-tonthly mean range = 12.20 to Synodontis schall was caught throughout 12.80 cm) due to the rains that commenced in April (Table 1).Catches within the habitats durthe year, but highest catches were recorded in ing the wet and lliy seasons were not signifi-January and February, while total catch cantly', different (P > 0.05).Similarly catches decreased from September to November when between the habitats throughout the sampling the lake became flooded as a result of increased period were not significantIy different (P>O.05).Monthly variation of mean body weight and mean standard length is also shown in Table 1, while monthly variation of body weight and water levels is illustrated in Fig. 2 Lower values of mean standard length and mean body weight were also recorded in October/November.Monthly variation of the percentage catch within the tbree habitat is shown in the bottom habitat in October/November.AH the catches during this month were mainly from theshore and surface habitats hence, the monthsof October and November recorded highest catches along the shoreline (Ta�le 2).
They were caught along the surface through out the sampling period except in January when the catches were relatively low, and high catches were then recorded at the bottom.This The presence of S. schall within the three ecologícal ni ches was due to Íts di verse feed ing habíts.S. schall feeds on a variety of food items including vegetable materials, insects, molluscs, detritus, fish scales and plankton (Motwani;1970, Fagade;1983, Olatunde;1989, Araoye 1997)).The flooded liuoral zone of the lake was characterised with insect mul tiplication while the bottom layers were rich in invertebrates particularly molluscs (Araoye 1997).AIso a high percentage frequency of occurrence for plants and detritus were record ed in the gut of S. schall during the wet season and these items became more dispersed along the surface water column at this period due to flood and overtum (Araoye 1997).
The ability of S. schall to explore the sur face, shore and bottom habitats in search of food items accounted for the insignificant vari ation of distribution within these niches irre spective of season.Unlike other fish species, the presence of a large physostomous swim bladder, the bony shield of the head and high fat deposition avails this fish the opportunity to explore the bottom habitat despite higher pres sure (Willoughby 1976).The low abundance of this fish at the bottom in OctoberlNovember, when the lake became flooded is attributed to the spawning activities of the fish around this season (Araoye 1997).Over 50% of the adult population were in gonad stages IV (mature) to Vl 1 (spent) as from September to November (Araoye 1997).
The flooded littoral zone at this season provid ed an expanded ecological niche for shelter to the early life stages after fertilization (Araoye 1997).AIso flooding appeared to be a critical exogenous parameter because food (detritus) and other conditions including water current, pH and temperature changes essential for the survival of this species carne with flooding to sustain the new recruits into the environment (Ezenwaji 1992).
The paucity of relatively large speci mens in the catch at this season was also attributed to restricted movement due to the spawning activities, therefore these sizes becarne les s vulnerable to the gill nets.Berst (1961) reported that the fishing success of the passive nets depends on fish movement and their efficiency and selectivity may be affected by abrupt changes from shift in barometer pressure, wind-driven currents, water level fluctuation, turbiditYi and trans mitted light.The relative aburulance of S. schall irrespective of season can also be due to the success of the fish within the environ ment due to low predation.Araoye (1997) reported the absence of highly piscivorous fishes such as Lates niloticus, Hydrocynus forskahlii and Gymnarcus niloticus in the lalce due to its relatively small size compared with Chad and Kainji lakes.Seasonal abun dance of fish species was also reported to be influenced by a combination of physico chemical properties and the presence of food items (Fagade and Olaniyan 1974).
The inverse relationship between the water level and fish weight may be related with increase in fish concentration due to the draw down of water at dry season.Ita (1978) using experimental gill net data in a bigger lake also observed an inverse relationship between water level and catch rate, explaining the .prob ability of higher concentration of f ishat low water levels.Although this variation may not be significant as reported for S. schall in this report, however the relationship could be sig nificant with commercial• catches (Bazigos;1972, Biswas 1973)).The report of Bishai and Abu Gideiri (1967) which is at variance with this observation, although not explained by the authors was probably due to the different fish ing methods usually embarked upon by the commercial fisheries.The exclusion of fry and fingerlings from the catches in this experiment was due to their non economic sizes which was below 5.08 cm net meshes.The fry an d finger lings of S. schall were usually restricted to the flooded liuoral zone of the lake where they feed mainly on zooplankton and insect larvae (Araoye 1997).
This research has shown that a substan tial quantity of S. schall can be caught with gill nets set in any of the habitats at any sea son of the year.Feeding and reproductive phenomena were the main factors responsible for the spatio-temporal distribution of S. schall in Asa lake.
Fig. 2. Monthly variarían of mean weight of S. schall and mean water levels in Asa lake.

TABLE 1
Monthly variation of mean weight and mean standard length ofS.schl\lI and monthly mean values of rainfall and water levels of Asa lake(March 1991 to February 1993)

Table 2 .
Tbey were not caught from l i.6