The Mafoubou River is located in the Niari Department, precisely in the Louvakou district near the town of Dolisie. This river is a tributary of the Loubomo and its source southwest of Dolisie where it flows in a SW-NW direction to the east of Dolisie. It receives two tributaries to the east and two other tributaries to the west before flowing in a northeast direction; South-west to the north of Dolisie a few meters before its confluence with the Loubomo ( Figure 1 ).

The intestinal coefficient (IC) was calculated for each individual according to the following formula described by [11] : IC = IL/SL, where IL is intestine length and SL represents the standard length.

The study of fish diet often includes two types of analyses: qualitative analysis and quantitative analysis [12] . Qualitative analysis which consists of drawing up a complete list of the different prey encountered in the stomachs, followed by a quantitative analysis which specifies the importance of the different prey and highlights the possible variations in the diet according to size and season. To characterize the diet, five indices were used.

The emptiness coefficient is the ratio expressed as a percentage between the number of empty stomachs Es and the total number of stomachs examined NT:

$V=\frac{Es}{NT}\times 100$

It is the dietary index used to analyze the dietary results of this species [13] , it is given by the following relationship:

$\%\text{OC}=\frac{Ni}{NT}\times 100$

Where, Ni is the number of stomachs containing a prey category i; NT represents the total number of full stomachs examined.

This percentage does not provide any indication of dietary preferences [14] .

$\%\text{P}=\frac{Ni}{PT}\times 100$

Where Pi = weight of item i and Pt = total weight of all items.

This index of [15] , modified by [16] makes it possible to quantify the proportion of each food item in the diet. Its calculation is based on the occurrence percentage [13] [17] [18] and the weight percentage [14] [19] . Its formula is as follows:

$\frac{\%\text{Oc}\times \%\text{P}}{{\displaystyle \sum \left(\%\text{Oc}\times \%\text{P}\right)}}\times 100$

This index which varies from 0 to 100, the classification of the different prey is as follows according to [12] : Ip < 10: accessory prey; 10 < Ip < 25: secondary prey; 25 < Ip < 50: important prey and Ip > 50: main prey.

The calculation of the Schoener index makes it possible to evaluate the degree of similarity in the composition of the diet between size class and seasons [20] Its formula is as follows:

$\alpha =1-0.5\left({\displaystyle {\sum }_{i=1}^n\left|Pxi-Pyi\right|}\right)$

Where, Pxi represents the proportion of prey consumed by individuals in a season x and Pyi is the proportion of prey consumed by individuals in a season y.

When the Schoener index is greater than or equal to 0.6, diets are significantly similar [21] .

The ascending hierarchical classification analysis based on Euclidean distance and by Ward’s method was carried out using the preponderance indices of prey consumed in each size class. The dendrogram thus obtained made it possible to distribute the size classes by groups.

The digestive tract of Bryconalestes tholloni has a thick-walled and muscular esophagus, followed by a developed U-shaped stomach. It has a well-individualized pyloric and cardiac branch. The pyloric caeca extend over the intestine which is short and folds on itself ( Figure 3 ).

The intestinal coefficient calculation made it possible to find values between 0.48 and 1.09, with an average of 0.74 ± 0.09. The fact that these values are between 0.3 and 2.18 indicates that Bryconalestes tholloni is an omnivorous, invertivorous species [22] and [11] . There is a significant linear relationship (r = 0.66) between gut length and standard fish length ( Figure 4 ).

Examination of the 75 stomachs showed that 7 were empty, which corresponds to an emptiness coefficient of 9.33%. During the dry season, 40 stomachs were examined, among which 6 were empty, the emptiness coefficient is 15%. During the rainy season, out of the 35 stomachs examined, only one stomach was empty, the emptiness coefficient is equal to 2.85%.

The general profile of the diet made it possible to identify 15 food items grouped into two fractions: the animal fraction (invertebrates) and the plant fraction (fruits). The animal fraction is made up of invertebrates forming 14 food items divided into 4 groups: insects (11 items), arachnids (1 item), nematodes (1 item), annelida (1 item). The plant fraction consists mainly of fruits ( Table 1 ). Bryconalestes tholloni mainly consume insects (Ip = 95.54%), including hymenoptera which represent important prey (Ip = 32.27%) and beetles which are secondary prey (Ip = 13.26%). Fruits (Ip = 3.98%) are accessory prey ( Figure 5 ).

Table 1. Diet composition of B. tholloni.

Specimens of Bryconalestes tholloni from the Mafoubou River have an omnivorous-insectivorous diet [11] [12] , because its intestinal coefficient is between 0.48 and 1.09. Similar results were obtained in Bryconalestes comptus on the right bank of Pool Malebo by [23] . Bryconalestes tholloni mainly consumes insects and secondarily fruits, nematodes, annelids and spiders. Hymenoptera and beetles are the most consumed prey. This essentially insectivorous diet could be explained as reported by [24] , by the presence of significant vegetation along the river. Similar diets of fish belonging to the same genus have been described by several authors in continental waterways. In the Niger River and the Chari, a diet essentially based on seeds, plants and insects was described by [25] . A diet based on macrophytes and terrestrial insects was demonstrated in Brycinus macrolepidotus by [26] as well as [23] . According to [27] , these species almost all feed on terrestrial insects. Bryconalestes comptus from the right bank of the Pool Malebo (Congo River) also consumes insects [22] .

Hymenoptera are important prey consumed by class 1 (38%), class 4 and class 5 (36%) of Bryconalestes tholloni. They constitute secondary prey (12%) for specimens of classes 3 and 6. Beetles are secondary prey (20.68%) consumed by class 5, dipterans are also secondary prey consumed by class 3. Mayfly larvae represent secondary prey for class 2 specimens ( Figure 6 ).

Overall, insects constitute the main prey for class 1 (90%), class 2 (98.34%), class 3 (84%), class 4 (91%), class 5 (82%), class 6 (78.51%); in class 7, insects are accessory prey with an Ip equal to 7%. On the other hand, the fruits which are the secondary prey (21.11%) consumed by the specimens of Bryconalestes tholloni of class 6, become the main prey (93.6%) consumed by the specimens of class 7, the insects are accessory prey (7%). Annelids and nematodes are accessory prey (0.04%) consumed by class 2 specimens, as well as arachnids (0.02%) for class 4 specimens ( Figure 7 ).

Examination of the food affinity dendrogram between the different size classes of Bryconalestes tholloni made it possible to discriminate, three groups of size classes at the aggregation threshold of 30%. Class 7 stands out by forming group I which stands out from the other two groups of size classes. Specimens in this group mainly consume fruits. Group II is made up of size classes 5, 1 and 4; group III is made up of classes 6, 2 and 3 ( Figure 8 ).

In the rainy season, insects remain the main prey of Bryconalestes tholloni, with a preponderance index of 98%, other prey (fruits, nematodes, annelids) are accessory prey. Among insects, Hymenoptera (Ip = 36.66%) and Coleoptera (Ip = 32.27%) are secondary prey ( Figure 9 ). In the dry season, insects represent the main prey of Bryconalestes tholloni, with a preponderance index of 87%, fruits are secondary prey with an Ip of 13%, Nematodes and Annelida are accessory prey (Ip = 0.04%). Among Insects, Hymenoptera and Beetles are secondary prey with an Ip of 20.25%. The diet does not present a significant difference between the two seasons, because the Schoener α index is greater than 0.6.