1. INTRODUCTION
Clarias gariepinus (African catfish) is one of the most extensively cultured fish species in Africa [1,2]. However, the increased aquaculture production of this species is impeded by the high cost of fish feed [3,4]. Therefore, it is essential to investigate alternative methods that may help reduce the cost of fish feed. In recent years, periphyton-based aquaculture of tilapia has been promoted as a significant success [5-7]. Periphyton is a complex community of bacteria, algae, zooplankton, phytoplankton, fungi, and organic and inorganic detritus attached to submerged substrates [8,9]. It grows on submerged substrates in water and is nutritionally adequate to support the growth and reproduction of most aquatic organisms due to its nutrients and biologically active compounds [10-12]. Periphyton not only serves as food but also improves water quality [13-16]. Moreover, it enhances and maintains the health of cultured fish [9,15,17]. This is attributed to certain periphyton organisms within the community that can produce antimicrobial and bioactive compounds, which function as immunomodulators [18-20]. However, not all fish species can be cultured in periphyton-based aquaculture. To harvest periphyton efficiently, fish need a high degree of specialization in their feeding, filtering, and masticating apparatus [21].
The use of periphyton-based aquaculture as a source of natural food for fish has been widely explored in the production of tilapia [21] carps [22], freshwater prawns [23], and mullets [24,25]. African catfish, like tilapia and carp species, are opportunistic omnivorous feeders that readily feed on different natural food. Yet it is not commonly used in periphyton-based aquaculture as tilapia and carp species. Most studies [4,13,21] have reported optimal growth rates, feed efficiency, body composition, immune response, and overall health status of tilapia in ponds with periphyton substrates. A study [26] reported a net yield of 6700 kg/ha/year in a periphyton-based system compared to a net yield of 2340 kg/ha/year in the control system without periphyton substrates in polyculture systems. A study [27] reported a reduced commercial feed input by 30–40% without affecting tilapia growth rate in ponds with periphyton substrates as compared to ponds without periphyton substrates. Another study [28] reported a 42% reduction in commercial feed input in ponds with periphyton substrates covering 75% of the pond surface area. Generally, periphyton is nutritionally adequate for fish as it contains 27% crude protein, 18% lipid, and 52% carbohydrates, which are better than most commercial feeds used in aquaculture [9,29]. Thus, it has been successfully used as a cost-effective method to reduce commercial feed inputs in aquaculture. However, to date, the use of periphyton to reduce commercial feed input in African catfish farming remains less addressed in aquaculture compared to tilapia.
Tilapia species are commonly used in periphyton-based aquaculture due to their ability to efficiently ingest and digest periphyton of various sizes [6,17,30]. This ability is attributed to their morphological mouth features, which are effective for scraping, combing, and sucking various natural food items [21]. Cyprinid species, such as Labeo and Garra, also possess specialized mouth features that enable them to forage with their bodies parallel to the substrate. These species have inferior transverse mouths with a sharp cutting edge, characterized by thick papillose lips, a vomero-palatine organ, prominent rostral features, and labial folds [21]. The feeding apparatus of the African catfish is poorly documented. Studies [31-34] have primarily reported on the morphology of the cranium of the African catfish. This study will compare the feeding apparatus of the African catfish to that of tilapias.
The substrate structure plays a crucial role in determining consumer–resource interactions [35]. More complex substrates can hinder the movement of the fish, thereby reducing their grazing rate on periphyton [36]. As substrate complexity increases, the ability of fish to graze from the substrate decreases. Therefore, the mouth structure of the cultured fish species requires a certain level of premaxilla protrusibility and lower jaw extensibility to efficiently graze on periphyton. Previous studies on periphyton-based aquaculture have focused on the impact of substrate structure on periphyton colonization and abundance, demonstrating that higher substrate complexity leads to increased periphyton biomass [37-40]. This study aims to determine whether African catfish can utilize periphyton from net, plastic, and stone substrates in aqua dams.
In the wild, African catfish feed on a wide spectrum of food items, ranging from phytoplankton to other fish species [28,41]. Periphyton is a conglomeration of various food items. The omnivorous feeding habits of African catfish make them a good candidate for periphyton-based aquaculture. However, no studies have been undertaken to evaluate the utilization of periphyton by the African catfish in aqua dams. It is hypothesized that the type of substrate the periphyton is attached to affects the ability of African catfish to utilize it as food. The main objective of this study was to investigate the utilization of periphyton by African catfish from net, plastic, and stone substrates in aqua dams. The specific objectives of the study were:
To determine the effect of substrates on African catfish utilization of periphyton in aqua dams with net, plastic, and stone substrates
To determine the compatibility of the African catfish’s feeding structure to utilize periphyton from net, plastic and stone substrates.
2. MATERIALS AND METHODS
2.1. Study Site
The study was conducted in experimental aqua dams (7000 L) at the Aquaculture Research Unit of the University of Limpopo, South Africa (23.8888° S, 29.7386° E). Experiments were carried out under real field conditions for 6 weeks. The study was approved, and a certificate (no. AREC/05/2023: PG) was issued by the University of Limpopo’s Animal Ethics Committee before the commencement of the experiments.
2.2. The Feeding Protocols of African Catfish in Aqua Dams with Net, Plastic, and Stone Substrates
Three substrates (net, plastic, and stone) were deployed separately in 7,000 L aqua dams using a completely randomized design. Each substrate was replicated 3 times, resulting in a total of 27 aqua dams. African catfish (200.2 ± 22.74 g; mean ± SD) were stocked in the aqua dams. Aqua dams are a static production system, and water lost through evaporation is replenished every 2 weeks. To determine the extent to which African catfish utilized periphyton, three feeding regimes were implemented. The first feeding regime served as the control treatment, where the African catfish were fed commercial pellets twice a day, every day. The fish were fed only commercial pellets daily, without any periphyton substrates in the aqua dams. This represents the standard feeding regime at fish farms, and it is equivalent to a 100% feeding regime. This feeding regime was designated as N100 for the net substrate, P100 for the plastic substrate, and S100 for the stone substrate. In the second feeding regime, the fish were fed commercial pellets every other day (twice a day), while periphyton was available for the fish from the deployed substrates. This was equivalent to a 50% feeding regime. This second feeding regime was designated as N50 for the net substrate, P50 for the plastic substrate, and S50 for the stone substrate. In the third feeding regime, the African catfish were fed commercial pellets every 3rd day, twice a day, with periphyton also available in the aqua dams. This was equivalent to a 33% feeding regime. This feeding regime was designated as N33 for the net substrate, P33 for the plastic substrate, and S33 for the stone substrate. The amount of commercial pellets fed to the fish in all aqua dams was recorded consistently. Temperature and dissolved oxygen levels were monitored in all the aqua dams to ensure that they did not become confounding variables. The temperature was 28.41 ± 3.59°C (mean ± SD) and dissolved oxygen was 9.76 ± 3.05 mg/L in all the treatments.
2.2.1. Determination of periphyton abundance in the aqua dams with net, plastic, and stone substrates
To determine the periphyton abundance and taxonomic composition on the net, plastic, and stone substrates in the aqua dams, periphyton samples were collected by carefully scraping the substrate (9 cm2 surface area). Samples from net and plastic substrates were collected from three depths: 20 cm, 50 cm, and 75 cm below the water surface. For the stone substrate, three random stones were sampled from different locations at the bottom of the aqua dam, which is 80 cm from the water surface. Three random stones were sampled since all stones were at the bottom, and none were on the surface of the water. The samples collected from the three depths (net and plastic) and locations (stone) were aggregated to form a single sample. The sample collection was standardized by scraping an equal surface area (9 × 9 cm2) on all substrates. The scraped periphyton was subsequently mixed with 20 mL of distilled water to enable enumeration under a light microscope using an improved double Neubauer chamber (W-Germany, 0.100 mm depth, 0.0025 mm2). The counting chamber and the coverslip were cleaned with 70% ethanol, and then 0.01 mL (10 μL) of the sample was loaded on the loading groove using a micropipette and counted. The concentration of cells in 1 μL was estimated by dividing the number of counted cells by the volume of the four main squares then the value was multiplied by 1000 to get the number of cells in 1 mL. The periphyton was also counted using a petri dish because some species were larger for the counting chamber. For this, a sample of 1 mL was loaded into the petri dish and examined with a light microscope (Zeiss, Axiolab, Germany). The periphyton was identified to genus and periphyton abundance was expressed as number per ml, and the graphs were plotted using the logarithm of the number per ml and the frequency of occurrence of genera.
The frequency of occurrence was calculated to determine the most occurring periphyton genera. The frequency of occurrence was calculated separately per substrate, each substrate’s frequency of occurrence is out of 100%. The frequency of occurrence and dominance index of the enumerated periphyton was calculated using the formula below. The dominance index was calculated to determine the most dominant periphyton genera. In the present study, only periphyton genera with a dominance index of >5 were considered.

2.3. Growth Performance Indices, Food Selection Indices, and Feeding Intensity of African Catfish in Aqua Dams with Net, Plastic, and Stone Substrates
African catfish were allowed to utilize periphyton for 6 weeks in aqua dams with net, plastic, and stone substrates before the growth performance indices were determined. The total weight (to the nearest 0.01 g) of all the African catfish in the aqua dams was measured using a bench scale. The standard length of the fish was determined using a tape measure to the nearest 0.5 mm. The weights and standard length of the fish were used to calculate growth performance indices. Three African catfish from each treatment group were then sacrificed and gutted to remove the stomachs. The stomachs were cut open to remove all contents into a petri dish and the periphyton found the stomach contents were identified to genus under a light microscope (Zeiss, Axiolab, Germany). The stomach contents were analyzed under a light microscope to identify the periphyton items ingested by the fish. The stomach fullness was ranked on a scale from 0 to 5, with 0 indicating an empty stomach and 5 indicating a full stomach. Additionally, the distal intestine was gutted from the sacrificed African catfish (mentioned above) and cut open to remove all contents into a petri dish. The periphyton found in the fecal matter was identified to genus level under a light microscope. The fecal matter was collected from the distal intestine of the fish and analyzed under a light microscope to identify which periphyton items were not digested by the fish. The analysis of stomach contents and faecal matter is reported for African catfish from aqua dams (N50, P50, S50, N33, P33, and S33) with periphyton substrates, while no data is provided from the control group (N100, P100, and S100). This omission occurred because the African catfish in the control group were only fed commercial pellets without access to periphyton on the substrates. Feeding intensity was determined by the percentage of full stomachs per treatment. Ranging from 0% to 100%. The growth performance indices calculated include:
Average Daily Weight Gain (ADWG) = Final Weight Gain (g)/Time (3)
Percentage Weight Gain = [(Final Weight – Initial Weight)/|Initial Weight|] × 100 (4)
Specific Growth Rate (SGR) = (ln Final Weight – ln Initial Weight)/Time (5)
Condition Factor = (Weight/Length3) × 100 (6)
Survival Rate (%) = (Final Number of Fish/Initial Number of Fish) × 100 (7)
A modified percentage Index of Relative Importance (%IRI), adopted from [42], was used to determine the importance of food items in the stomachs of African catfish across all treatments:
%IRI = %NO * %FO (8)
Where %NO is the percentage of the number of the food item, and %FO is the percentage frequency of occurrence of the food items.
Food selectivity was calculated using the following Chesson selectivity index (CSI), defined as:

Where n is the total number of periphyton species in the stomach contents; g and b are the proportional representation of periphyton species i or j in the stomach contents and the water (environment), respectively. The CSI value for a specific periphyton species i reflects its presence in the stomach contents relative to the environment, divided by the sum of the equivalent calculations for all periphyton species in the stomach contents. The CSI value ranges from −1 to +1, with −1 indicating complete avoidance of the food item, 0 indicating a random selection of the food, and +1 indicating exclusive feeding on that species.
2.3.1. The feeding structures of African catfish and tilapia species
A morphological analysis of the feeding structures of African catfish was conducted and compared to those of Oreochromis mossambicus (tilapia species). A total of 18 subadult individuals from each species were used. The width of the mouth, premaxilla protrusibility, and lower jaw extensibility were measured to the nearest 0.05 mm using the methods described by [43,44]. The width of the mouth aperture was expressed as a percentage of the standard length. The number of gill rakers was also counted. The protrusibility of the premaxilla is presented as the percentage difference in distance from the anterior margin of the lower eye socket to the outer edge of the premaxilla, measured from the tightly closed to the wide-open mouth [43]. The lower jaw extensibility is defined as the distance from the anterior margin of the lower eye socket to the tip of the lower jaw with the mouth closed. This distance is subtracted from the distance with the mouth open and expressed as a percentage of the former [43].
2.4. Statistical Analysis
Microsoft Excel and Power BI were used to plot graphs. The normality and homogeneity of growth performance indices were tested using the Shapiro-Wilk and Levene tests, respectively (Statistical Package for Social Science [SPSS] version 28). Analysis of similarities (ANOSIM), permutational multivariate analysis of variance (PERMANOVA), multivariate analysis of variance (MANOVA), and Shannon diversity index were performed in R (version 2024.4.2-764) using the vegan, MASS, effective, tidyverse, magrittr, and dplyr packages. ANOSIM was used to determine whether there were similarities in periphyton abundance on net, plastic, and stone substrates in aqua dams. PERMANOVA was employed to assess the statistical difference between the food items identified in the stomachs of African catfish utilizing periphyton. Additionally, ANOSIM was used to evaluate whether food items identified in the stomach were similar within each treatment or between the treatments. ANOSIM and PERMANOVA were used for data that was not normally distributed. Normality and homogeneity of variance data were tested using the Shapiro-Wilk and Levene tests, respectively. Analysis of Variance was used to test for significant differences in the growth performance of African catfish. Tukey’s post hoc analysis was used to determine which means were significantly different from each other (SPSS version 28). The data were tested at a level of significance of 0.05. A MANOVA was conducted to assess the significant differences in the growth performance indices among the substrates (net, plastic, and stone) using a permutation test.