The Effect of Dietary Protein Levels and Female Brood Stock Size On Size Heterogeneity Among Heterobranchus longifilis Fingerlings

Received 25 Dec, 2020

Accepted 30 Mar, 2021

Published 04 Jun, 2021

ABSTRACTBackground and Objectives: Heterobranchus longifilis and its sister species and corresponding hybrids are popular species cultured in the Sub-Saharan Africa. Studying the size variation of this species will help maximize product output and enhance aquaculture production, thus, the objective of this work was to assess how female brood stock size and dietary crude protein levels influence fingerlings size heterogeneity and at the same time seek for ways of obtaining better seed quality and survival in Heterobranchus longifilis culture. Materials and Methods: The experiment was conducted in a randomized factorial design with 4 brood stock size categories (size category 1-4) and 4 dietary protein levels (protein level 45, 40, 35 and PL 30%) in 3 replications. The fish in each replicate were fed twice daily at 3% bwt per day with the respective feed for the particular treatment combination for 6 months. Water exchange was maintained by daily intermittent flow through and the quality monitored weekly. After determination of the 3 best ripe fish for spawning, each fish was induced with Ovaprim at 0.5 ml kg^-1 bwt and 50 g egg aliquot fertilized with 1 ml of pooled milt from 16 homogenous males of 2,181.83 g mean bwt. 250 one day old larvae from each female were raised separately for 30 days. Each set of fingerlings was audited, sorted into sizes, survival rate, and percentage composition of each class of fingerlings determined. The growth performance and FCR of the brood stock in each replicate were also determined. Four size categories of fingerlings A, B, C and D in descending order of size were identified. The data generated were subjected to analysis of variance for significant difference and means separated with DMRT at p = 0.05. Results: The results revealed significant interaction for mean wt gain, SGR, and FCR of brood stocks, and composition of C and D fingerlings. No significant interactions were found in survival rate, and percentage composition of A and B fingerlings. Conclusion: Size heterogeneity of fingerlings appeared to be strongly maternal size dependent as a greater percentage of bigger larvae were obtained from larger maternal brood stock.

INTRODUCTION

Good quality seed is a critical resource which could prescribe the success of any aqua cultural project when its availability is guaranteed or failure if it is not attainable. Uncertainties in the availability and quality of fish seed are twine predicaments that could be significantly addressed via proper brood stock management. Key considerations in proper management of brood stock would include adequate nutrition and selection based on the desired trait such as body size. Brood stock nutrition and size are important factors when evaluating the reproductive performance of fish brood stocks. Proper nutrition and feeding has been reported to optimally enhance gamete quality as well as seed production in fish brood stocks^1-3. Protein is considered as a very important nutrient in brood stock diets^4-5 and its inclusion levels have been shown to have great implications in fish brood stock vitality, growth, reproductive performance and survival of the embryos^6-9. Based on the importance of dietary protein in routine body maintenance, growth as well as gonadal development, it is widely assumed that there would be disparity between the protein requirements of grow-out fish and the brood stocks^1,10,11,12.

The size at which individuals of even the same species attained sexual maturity tends to differ consequent on genetic and environmental factors operating at the individual levels. Some hatchery managers may consider the size of breeders for their operations based on such immediate factors as costs, availability, ease of management and sometimes, with little or no regards to seed quality based on their breeding goals. Nlewadim and Madu¹³ emphasized that the breeding goal of any aqua cultural project is expected to apply methods that would generate quality seeds capable of better survival, faster growth and resistance to endemic diseases and compromised environmental conditions. According to Bichi et al.¹⁴, the African catfish breeders in Nigeria prefer smaller sized breeders to larger ones due to the high cost of brood stocks, and the attendant cost of inducing hormones⁷. Where the breeding goal is largely on the premise of economic considerations, seed quality would be obviously compromised with. The brood stock quality undoubtedly influences the seed quality such that the effect of the brood stock size on the fish reproductive performance could not be over emphasized. Many literatures have associated improved seed quality with increased size of the female brood fish^15-19. Particularly in this wise has been the effect of the parent size on egg size which in many cases affect most other egg and hatchlings quality attributes^14,20. The seed quality in the African catfishes could be compromised by the presence of sundry sizes even in the same batch of seed from a particular hatchery with concomitant burden of stock management and losses to the grow-out fish farmers. Oyebola and Awodiran²¹ noted that poor brood stock productivity stems from possible faulty brood stock management and spawning techniques, poor brood stock nutrition and remarkable fry mortality. These tend to limit commercial fry production and subsequent scale enlargement in both seed and grow-out fish production.

The African Clariid catfishes notably, Heterobranchus longifilis, H. bidorsalis, Clarias gariepinus, and C. anguillaris and their inter generic hybrids remain the most dominant species cultured in the Sub-Saharan Africa. Their robust advantages have been highlighted by Baras²², Offem et al.²³, Owodeinde and Ndimele²⁴, Ndimele and Owodeinde²⁵, Baras et al.²⁶ identified H. longifilis to possess growth rate that parallels the fastest among the species in African aquaculture. However, the challenge of size variation and its consequential cannibalistic features still remain a serious concern notable in this species²⁶. According to Martins²⁷, understanding the principal cause of variation is essential to successful fish rearing as its decrease supports maximized production efficiency, food wastage minimization and improvement in water quality. It becomes imperative in this work to evaluate the effect of female brood stock size and dietary crude protein levels on fingerlings size heterogeneity with a view to improving seed quality and survival in Heterobranchus longifilis culture.

MATERIALS AND METHODS

The experiment was conducted at Safe Food MPCS fish farm, Uyo, Akwa Ibom State, Nigeria where the fish were obtained. About twelve months old brood stocks of Heterobranchus longifilis were used. Both sexes were stocked in separate tanks. Large concrete tanks were divided into 2.5 m × 3.0 m × 1.2 m partitions for the females feeding trial. The partitioning was done using fine meshed polyethylene monofilament netting anchored securely on tightly fitting mangrove purlin frames.

Experimental Feed: Four feeds were prepared at different crude protein levels: PL45% = 45%; PL40% = 40%; PL35% = 35% and PL30% = 30% according to the formula in Table 1. Samples of each feed were analyzed for proximate composition using the standard methods of AOAC²⁸ as presented in Table 2.

Table 1: Percentage composition of the experimental feed fed to the brood stock of Heterobranchus longifilis for 6 months
Ingredient	PL45 %	PL40 %	PL35 %	PL30 %
Fish meal	29.40	26.60	24.00	20.25
Soybean cake	30.00	23.20	16.00	10.50
Groundnut cake	22.75	18.04	13.07	8.75
Wheat offal	9.10	23.41	38.18	51.75
Palm oil	0.50	0.50	0.50	0.50
Methionine	0.50	0.50	0.50	0.50
Lysine	1.00	1.00	1.00	1.00
Vitamin premix	2.00	2.00	2.00	2.00
Bone ash	2.10	2.10	2.10	2.10
Common salt	0.50	0.50	0.50	0.50
Composite wheat flour	2.00	2.00	2.00	2.00
Vitamin C	0.15	0.15	0.15	0.15
TOTAL	100	100	100	100

Table 2: Proximate composition of the experimental feed fed to the brood stock of Heterobranchus longifilis for 6 months
Proximate composition	PL45 %	PL40 %	PL35 %	PL30 %
Crude protein	45.50	40.77	35.64	30.22
Moisture	12.60	12.89	13.01	13.25
Lipid	12.56	12.04	11.98	11.88
Ash	11.55	11.45	11.45	11.38
Fibre	9.13	10.80	12.45	13.34
Nitrogen free extract	8.36	12.05	15.47	19.93

The ingredients were carefully weighed out according to the formulations and mixed thoroughly to homogenize. Diets PL45 and PL40%, and PL35 and PL30% were moistened with water at 60 and 70% respectively volume/weight to gelatinize. The dough was pelletized into 8 mm pellets and adequately sun dried. Each diet was prepared in batches of 10 kg throughout the experimental period. The dried diets were separately stored in well labelled dark polyethylene bags till each feeding. Parts of diet PL45% were ground into fine and coarse particles using a motorized classifier. These different grades of feed were packaged separately in dark polyethylene bags and stored in the freezer for feeding the hatchlings.

Stocking and feeding of brood stocks: Females of H. longifilis were grouped into four size categories: Sc1= 200 – 240 g; Sc2 = 500 – 540 g; Sc3 = 980 – 1095 g and Sc4 =1990 – 2080 g. Weight and length measurements were enhanced by sedating the fish with benzocaine at 100 ppm. A set of 4 fish with known weight and length from each size category was randomly picked and placed in separate tank partitions at random till all partitions were occupied and fully labelled. Each size category had 4 sets with a particular feed assigned randomly to a respective set within the category. The particular diet was clearly indicated on the respective label. All the fish in the given set were fed with the particular feed at 3 % body weight (bwt) given in 2 meals per day for 6 months. The experiment thus was laid out in a complete randomized 4 × 4 factorial design with 3 replications. The feeding experiment commenced at the onset of the dry season, when gonadal activities in H. longifilis arelikely to be at the minimum hence, the fish were more likely to be at a similar state of gonadal development. Fish were sampled for weight and length measurement bi-weekly and feeding subsequently adjusted to 3% of the new bwt. Water exchange was facilitated by daily intermittent continuous flow and the water level maintained at 0.90 m. Dissolved Oxygen (DO), temperature and pH of water in all the tanks were monitored weekly. Gonadal maturation was checked monthly effective from the fourth month of feeding. At the end of 6 months feeding trial, seed production started.

Seed propagation: The best 3 gravid females from each partition were selected, their individual weights and lengths noted and compared with the initials. The selected females were induced for artificial spawning with Ovaprim (Syndel Lab. Ltd, Canada) at 0.5 mL/kg bwt according to Umanah and Okure²⁹. The selection of both male and female ripe breeders, stripping of eggs from the females and fertilization followed standard procedures for catfish breeding. Three pools of milt were each obtained from a homogenous set of 16 males (mean wt = 2,181.88±7.21 g) by sacrifice and dissection of the testes. 1 ml of milt diluted with 0. 9% normal saline was used to fertilize the eggs from each female. 50 g aliquot of the fertilized eggs from a particular female was incubated in a separate labelled 45L plastic basin having a flow through device and containing 40L of clean bore – hole water.

After 26 hours post fertilization, the hatched larvae in each container were respectively transferred into another set of containers containing clean water. The incubation basins were thoroughly washed and disinfected with formalin. The sizes of the hatchlings were carefully observed. Two hundred and fifty (250) larvae from the respective containers were randomly transferred back into their incubation basins maintained at 40L of water by flow through system for rearing in near dark condition. Water in all the containers was aerated by a 1.5 HP air blower through networks of tubings with air stones.

Feeding of hatchlings began on the evening of the second day post hatching when the egg yolk was almost depleted. Artemia shell free (Inve Aquaculture Inc. Utah, USA) was used for 7 days in the first exogenous feeding. Feeding continued with diet PL45% fine powder for another 7 days and then, progressively with the coarse powdered feed as the fry grew older and bigger till the end of the 30 days rearing period. The hatchlings were fed to satiation 4 hourly per day in the first 2 weeks and then 6 hourly daily for the remaining days of the experiment. Water temperature, DO and pH were monitored respectively every morning (7:00 – 8:00 hrs) and late afternoon (15:30 – 16:30 hrs) with a nitrogen in glass thermometer (- 5to 105 , Brannan, England) and test metres (Jenway, England, model 3050 for DO; model 9070 for pH). When the experiment ended, fish from each replicate of a treatment combination were counted, their individual weight taken with a balance (Mettler P 165, Gallenhamp) and the standard length with a measuring board. The fish were sorted according to size and their respective numbers noted.

Data analysis: The following indices were determined.

Mean weight gain (MWG) by brood stock (g)

MWG = WT1 - WT2

equation(1)

Specific growth rate (SGR) of brood stock (%/day)

$$ \frac{100\left(\operatorname{Ln} W T_{2}-\operatorname{LnWT}_{1}\right)}{T_{2}-T_{1}} $$

equation(2)

Mean survival rate of hatchlings (SR) (%)

$$ S R=\frac{100 x N_{2}}{N_{1}} $$

equation(3)

Mean percentage size composition of the different categories of the hatchlings

$$ P C=\frac{100 \text { (no.of fry at a particular size })}{\text { total no.of fry }} $$

equation(4)

(%) Data were transformed into angular arcsine^30-31.

$$ X^{1}=\operatorname{ArcSin} \sqrt{x} $$

equation(5)

Food conversion Ratio of brood stock

$$ F R=\frac{\text { weight of feed consumed }(g)}{\text { weight gain by } f \text { ish }(g)} $$

equation(6)

Statistical analysis: The data obtained from the experiment were subjected to one - way analysis of variance and the treatment means compared using Duncan Multiple Range Test (p = 0.05). The Arcsine values were back transformed into percentages after the analyses for presentation. The SPSS version 25 was used to execute the analyses.

RESULTS

The followings were the results obtained from the investigation. Table 3 below highlights the results of mean water quality recorded during the experiment. The values of water quality indices subsisted throughout the experimental period were adequate for the breeding and survival of the catfish.

Table 3: Mean values of water quality indices in the experimental tanks
Experimental tanks	Temperature (°C)	pH	DO (mg/L)
Brood stocks	27.88	7.71	10.88
Eggs incubation	27.84	7.44	11.56
Larval/Fry rearing	27.5	7.73	11.47

Results of the interaction between crude protein level in the brood stock feed and the size of the female brood stock; as well as the four size categories of fingerlings obtained were as recorded in Table 4. The results revealed that interaction between PL and Sc was significant in MWG, SGR and FCR of the female brood stock, as well as in fingerlings size categories C and D. The MWG and percentage of category C fingerlings increased with Sc peaking at Sc3 and minimum at Sc1 for every PL, and as PL approached 40% with a peak at 40% and least at 30% for every Sc. For an example in Table 4, it could be seen that for PL45%, maximum MWG and C fingerlings were 1191.6 g and 65.69% respectively at Sc3 while minimum MWG was 408 g, and C= 55.05% at Sc1. Similarly, for Sc3, maximum MWG = 1268.33 g and C = 66.26% at PL40% whereas the minimum MWG = 956.67 g and C = 64.93% at PL30%. On the other hand, SGR and category D fingerlings took some what a reverse trend. Both SGR and D fingerlings increased with decreasing Sc peaking at Sc1 for every PL but, while SGR rose correspondingly from PL30% till a zenith at PL40%, the percentage of D fingerlings was the opposite for every Sc (Table 4). Evidently in the Table, for PL 45%, SGR = 0.59 and D = 36% at Sc1 but 0.25 and 11.5% respectively at Sc4. Contrastingly though, for Sc1, SGR = 0.21 and D = 37.01% at PL 30% while SGR = 0.27 and DF = 35.76% at PL40%. The trend of FCR was exactly the reverse of the SGR. No significant interaction was observed in A and B fingerlings, and in the survival rate (SR) of the fingerlings.

From the results in Table 4, it was determined that Sc3/PL45, Sc4/PL45, Sc3/PL40, Sc4/PL40, Sc3/PL35 and Sc4/PL35% each generated mean combined A and D fingerlings not beyond 19.29% of the total number of fingerlings in the respective treatment combination, and also not exceeding 23.93% of mean combined A and D / B and C fingerlings in that treatment combination as shown in the following Table 5. The same treatment combinations also produced mean combined B and C fingerlings between 80.62% (in Sc3/PL35%) and 85.52% (Sc4/PL40%) of the total number of fingerlings in the respective treatment combinations indicated also in the table.

The main effects of the female brood stock size and the crude protein level in the female brood stock diets are respectively shown in Tables 6 and 7. The results in Table 6 indicate that 4 size categories of hatchlings ( A:70-89mm (2.1-3.0g); B: 50-69mm (1.21-2.0g); C:30-49mm (0.25-1.20g) and D: 10-29mm (0.03-0.249g) were obtained in Sc2 – 4 but 3 size categories (no category A) in Sc1, and significant differences existed across the different female brood stock sizes as regards MWG, SGR and FCR of brood stock, and the size composition of the fingerlings. However, the survival of fingerlings was significantly higher (70.2%) in Sc3 brood stock than in others which were similar (68.2 – 68.5%) without any significant difference. While SGR and category D fingerlings decreased progressively from Sc1 (SGR= 0.59, D = 36.43%) to Sc4 brood stocks (SGR = 0.24, D = 12.15%), reverse was the case for MWG, FCR and categories A, B and C fingerlings. C fingerlings however declined to a second highest position at Sc4 (60.19%) from the peak at Sc3 (65.6%).

In Table 7, the same four distinct categories of fingerlings were also observed as the results of feeding varying dietary protein levels (PL30, 35, 40 and 45%) to the female brood stocks. The female brood stock fed PL40% based diet showed the highest MWG (929.33 g) followed by those on dietary protein levels PL 45% (840.42 g), 35% (815.42 g) and 30% (691.67g) in descending order and all were significantly different. Fish on PL40% also gave highest significant values (SGR = 0.46, A = 1.65%, B = 15.24%, C = 60.97) as compared to PL45% (SGR = 0.42, A= 1.34%, B =14.67%, C = 60.19%) and PL35% (SGR = 0.42, A= 1.29%, B =14.67%, C = 59.99%) which showed no significant difference but differed from PL30% (SGR = 0.38, A= 1.04%, B =14.10%, C = 59.21%). The same significance pattern played out in FCR except that the range of values was reversed. The percentage of category D fingerlings rose from the lowest (20.19%) in fish fed PL40% diets to the highest (24.17%) in those on PL30%, all significantly different. The survival rate (SR) of the fingerlings was highest (70%) in PL40% though significantly similar to fingerlings from brood stocks fed diets PL45% (SR = 68.8%) and PL35% (SR = 68.9%) but significantly higher than the result from PL30% diet (67.2%).

Table 4: Effect of varying dietary crude protein levels and female broodstock size on size heterogeneity among the fingerlings of Heterobranchus longifilis - Interaction
Factors		Mean brood stock growth and feed utilization parameters						Mean percentage survival and composition of fingerlings
PL	Sc	WT1 (g)	WT2 (g)	MWG (g)	SGR (%)	FCR	SR	A	B	C	D
45%	1	213.33±14.91	621.67±16.87	408.33±8.45	0.59±0.01	1.25±0.01	68.80±1.00	0.00±0.005	8.34±0.001	55.05±0.001	36.63±0.004
	2	520.00±14.91	1110.00±16.87	590.00±8.45	0.42±0.01	1.27±0.01	69.20±1.00	0.58±0.005	12.74±0.001	59.31±0.001	27.32±0.004
	3	1016.67±14.91	2208.33±16.87	1191.67±8.45	0.43±0.01	1.26±0.01	70.00±1.00	3.24±0.005	15.60±0.001	65.69±0.001	15.45±0.004
	4	2020.00±14.91	3191.67±16.87	1171.67±8.45	0.25±0.01	1.41±0.01	67.20±1.00	4.18±0.005	23.83±0.001	60.48±0.001	11.50±0.004
40%	1	220.00±14.91	663.33±16.87	443.33±8.45	0.61±0.01	1.19±0.01	71.20±1.00	0.00±0.005	8.85±0.001	55.45±0.001	35.76±0.004
	2	524.33±14.91	1283.33±16.87	759.00±8.45	0.50±0.01	1.24±0.01	69.20±1.00	1.29±0.005	13.21±0.001	60.87±0.001	24.60±0.004
	3	1006.67±14.91	2275.00±16.87	1268.33±8.45	0.45±0.01	1.25±0.01	72.00±1.00	3.53±0.005	16.33±0.001	66.26±0.001	13.90±0.004
	4	2026.67±14.91	3273.33±16.87	1246.67±8.45	0.27±0.01	1.25±0.01	68.80±1.00	4.47±0.005	24.26±0.001	61.26±0.001	10.08±0.004
35%	1	215.00±14.91	625.00±16.87	410.00±8.45	0.59±0.01	1.27±0.01	68.00±1.00	0.00±0.005	8.62±0.001	54.85±0.001	36.43±0.004
	2	511.67±14.91	1110.00±16.87	598.33±8.45	0.43±0.01	1.26±0.01	68.40±1.00	0.59±0.005	12.67±0.001	59.50±0.001	27.32±0.004
	3	1041.67±14.91	2173.33±16.87	1131.67±8.45	0.41±0.01	1.34±0.01	70.40±1.00	3.03±0.005	15.31±0.001	65.31±0.001	16.26±0.004
	4	2023.33±14.91	3145.00±16.87	1121.67±8.45	0.24±0.01	1.38±0.01	68.80±1.00	4.06±0.005	23.66±0.001	60.09±0.001	12.21±0.004
30%	1	210.00±14.91	595.00±16.87	385.00±8.45	0.58±0.01	1.40±0.01	66.00±1.00	0.00±0.005	8.29±0.001	54.75±0.001	37.01±0.004
	2	511.67±14.91	1000.00±16.87	488.33±8.45	0.37±0.01	1.41±0.01	66.40±1.00	0.27±0.005	12.28±0.001	58.22±0.001	29.12±0.004
	3	1056.67±14.91	2013.33±16.87	956.67±8.45	0.36±0.01	1.45±0.01	68.40±1.00	2.73±0.005	14.81±0.001	64.93±0.001	17.53±0.004
	4	2063.33±14.91	3000.00±16.87	936.67±8.45	0.21±0.01	1.43±0.01	68.00±1.00	3.53±0.005	22.57±0.001	58.81±0.001	15.09±0.004
P value		0.396	0.000	0.000	0.001	0.000	0.531	0.056	0.572	0.036	0.003
P value 0.05 = Significant interaction along the column PL= Dietary crude protein level; Sc = Female Broodstock Size (Categories 1 – 4); WT1 = Initial Mean Weight; WT2 = Final Mean Weight; MWG = Mean Weight Gain; SGR = Specific Growth Rate; FCR = Feed Conversion Ratio; SR = Survival Rate; A, B, C and D = Fingerlings Size Categories A:70-89mm (2.1-3.0 g); B: 50-69 mm (1.21-2.0 g); C:30-49 mm (0.25-1.20 g) and D: 10-29 mm (0.03-0.249 g)

Table 5: Summary of fingerlings composition in Sc 3-4 and PL combination
Sc/PL Combination	A + D Hatchlings	B + C Hatchlings	Percentage ( A + D): (B + C)	SR (%)
3 / 45%	18.69	81.29	22.99	70
4 / 45%	15.68	84.31	18.60	67
3 / 40%	17.43	82.59	21.10	72
4/ 40%	14.55	85.52	17.01	68.8
3 / 35%	19.29	80.62	23.93	70.40
4 / 35%	16.27	83.75	19.43	68.8
PL= Dietary crude protein level; Sc = Female Broodstock Size (Categories 3 – 4); SR = Survival Rate; A, B, C and D = Fingerlings; Size Categories A:70-89mm (2.1-3.0g); B: 50-69mm (1.21-2.0g); C:30-49mm (0.25-1.20g) and D: 10-29mm (0.03-0.249g)

Table 6: Main Effect of female brood stock size on size heterogeneity among the fingerlings of Heterobranchus longifilis
Mean Brood stock Size, Growth and Feed utilization Parameters							Mean percentage survival and composition of fingerlings
Sc	WT1 (g)	WT2 (g)	MWG (g)	SGR (%)	FCR	SR	A	B	C	D
1	214.58±7.45d	626.25±8.44d	411.67±4.22d	0.59±0.00a	1.28±0.01d	68.50±0.50b	0.00±0.002d	8.51±0.000d	55.05±0.000d	36.43±0.001a
2	516.92±7.45c	1125.83±8.44c	608.92±4.22c	0.43±0.00b	1.29±0.01c	68.30±0.50b	0.64±0.002c	12.74±0.000c	59.50±0.000c	27.05±0.001b
3	1030.42±7.45b	2167.50±8.44b	1137.08±4.22a	0.41±0.00c	1.33±0.01b	70.20±0.50a	3.10±0.002b	15.53±0.000b	65.60±0.000a	15.74±0.001c
4	2033.33±7.45a	3152.50±8.44a	1119.17±4.22b	0.24±0.00d	1.37±0.01a	68.20±0.50b	4.06±0.002a	23.58±0.000a	60.19±0.000b	12.15±0.001d
*Means with different superscripts along the column are significantly different (p<0.05) Sc = Female Broodstock Size (Categories 1- 4); WT1 = Initial Mean Weight; WT2 = Final Mean Weight; MWG = Mean Weight Gain; SGR = Specific Growth Rate; FCR = Feed Conversion Ratio; SR = Survival Rate; A, B, C and D = Fingerlings Size Categories A:70-89 mm (2.1-3.0 g); B: 50-69 mm (1.21-2.0 g); C:30-49 mm (0.25-1.20 g) and D: 10-29 mm k (0.03-0.249 g)

Table 7: Main effect of Varying levels of crude protein in female broodstock diets on size heterogeneity among the fingerlings of Heterobranchus longifilis
Dietary crude protein level, Mean growth and Feed utilization Parameters							Mean percentage survival and composition of fingerlings
PL (%)	WT1 (g)	WT2 (g)	MWG (g)	SGR (%)	FCR	SR	A	B	C	D
45	942.50±7.45a	1782.92±8.44b	840.42±4.22b	0.42±0.00b	1.30±0.01b	68.80±0.50a	1.34±0.002b	14.67±0.000b	60.19±0.000b	21.98±0.001b
40	944.42±7.45a	1873.75±8.44a	929.33±4.22a	0.46±0.00a	1.23±0.01c	70.30±0.50a	1.65±0.002a	15.24±0.000a	60.97±0.000a	20.19±0.001c
35	947.92±7.45a	1763.33±8.44b	815.42±4.22c	0.42±0.00b	1.31±0.01b	68.90±0.50a	1.29±0.002b	14.67±0.000b	59.99±0.000b	22.40±0.001b
30	960.42±7.45a	1652.08±8.44c	691.67±4.22d	0.38±0.00c	1.42±0.01a	67.20±0.50b	1.04±0.002c	14.10±0.000c	59.21±0.000c	24.17±0.001a
*Means with different superscripts along the column are significantly different (p<0.05) PL= Dietary crude protein level; WT1 = Initial Mean Weight; WT2 = Final Mean Weight; MWG = Mean Weight Gain; SGR = Specific Growth Rate; FCR = Feed Conversion Ratio; SR = Survival Rate; A, B, C and D = Fingerlings Size Categories A:70-89mm (2.1-3.0 g); B: 50-69 mm (1.21-2.0 g); C:30-49 mm (0.25-1.20 g) and D: 10-29 mm (0.03-0.249 g)

DISCUSSION

Four different size categories of fingerlings viz. A, B, C and D were identified in this study and this variation was even noticed from the first day of hatching. The experiment showed significant interaction between the size of female brood stocks and the level of crude protein in their diets as regards MWG, SGR and FCR of the brood fish, and fingerlings of categories C and D but no significant interaction in the survival rate (SR) of the hatchlings and the percentage of categories A and B fingerlings.

The significant interaction obtained in this work implies that the combination of female brood stock size and the level of dietary crude protein probably influenced MWG, SGR, FCR and categories C and D fingerlings in Heterobranchus longifilis. Different combinations of the tested factors yielded varying results with optimal levels of both factors producing the most significant values of MGR, SGR, and C and D fingerlings, and best value of FCR. The non – significant interaction indicates that in this fish the specific size of brood stocks and specific level of the dietary protein fed to the fish were possibly more important than their combinations in determining the fingerlings survival rate and the percentages of categories A and B fingerlings. These inform hatchery managers not to only be concerned with the size of their brood stocks or the level of dietary crude protein but both in order to optimize the growth performance and feed utilization of their brood stocks as well as control the percentages of the various categories of hatchlings.

The MWG was higher in the larger fish than in the smaller brood stocks though the growth rate of the smaller fish was faster than recorded in the larger brood stocks. The superior SGR and FCR recorded in this work in the smaller fish is in consonance with the works of Xie et al.³², Akbulut et al.³³ and Strand et al.³⁴ in which larger body size was implicated in poor SGR and FCR in some fish species. According to Xie et al.³², body size also affects feed consumption therefore these observations could probably arise from comparative higher feed intake by the smaller fish to the larger fish. Also, the smaller brood stocks were likely to be younger spawners³⁵ and younger fish are believed to have higher growth propensity than older ones³⁶. The heterogeneity in the size (size category) of the hatchlings here was higher in the larger brood stocks than in the smaller ones as similarly observed in older brood stocks of Clarias gariepinus by Umanah³⁷ and the percentages of A, B and C fingerlings also increased with increasing female body size as well as the crude protein level approached 40%. The effect of body size here may be related to larger fish (and presumably older fish) possibly producing larger eggs¹⁹ which also produce larger hatchlings^20,38-39 than the smaller fish. Subsequent rearing of these hatchlings under uniform conditions would eventually preserve this variation throughout their life trajectories. According to Martins²⁷, growth variation in the African catfishes might be inherent, while Grabowski et al.⁴⁰ as well as Tyor and Pahwa⁴¹ noted the presence of size heterogeneity in the same clutch of fish eggs. It is possible therefore that the variations observed in the eggs as well as hatchlings of these fishes based on brood stocks sizes might be a matter of proportions. The larger fish are believed to have higher body – energy reserves to be used during vitello genesis of the eggs than smaller ones¹⁶ thus could produce higher proportions of larger eggs than smaller ones. According to Strand et al.³⁴, the digestive energy need is dependent on body size, whereas relative feed intake decreases with size³². Reidel et al.⁴² noted that the demand of energy of catfish rises at the time of gonadal maturation particularly during vitello genesis. Reproduction might therefore be an additional burden to the routine energy budget of the fish. This could be very critical to the larger fish given the fecundity of the African catfishes being body size dependent^43-46, energy allocation to the individual eggs would therefore differ leading to possible variation in egg size and corresponding heterogeneity in the size of larvae and fingerlings even under homogeneous growing environments. This could take a bearing from the reports of other authors that resources related maternal effects could have pervading and sustainable implications for offspring growth, development and other characteristics throughout the life of the animals that would be reflexive of the specific maternal environmental conditions of birth^47-49. Thus, the higher degree of heterogeneity in the size of fingerlings from larger brood stocks compared to smaller ones could remain a practical reality. Similarly, though Sakai and Harada^50-51 opined that energy allocation to eggs in fish was egg specific and irrespective of fecundity or maternal size, it would be difficult to maintain this assertion generally in view of many literatures positively correlating egg size with maternal body size^{18,29,22,52-54} given that egg size in most cases is a function of maternal provisioning^49,55-56. Maternal effects are held to sometimes, modify the genetic effects^57-58.

Dietary proteins have similarly been identified to affect reproductive parameters which include gonadal maturation, fecundity, egg quality and larval viability in freshwater as well as marine fishes^8,59 and essentially its contribution through improved body size of the brood stocks¹¹. The female brood stock requires adequate protein for growth and development of the embryo⁶⁰. Influence of dietary protein in this work was evident in favourable mean weight gain, SGR and FCR of the brood stock, survival rate and percentage composition of the A, B, C and D categories of fingerlings. The best significant values of these parameters were obtained with 40 % dietary protein level except in survival rate of the hatchlings that it was significantly the same as 45 and 35%. The dietary protein levels 45% and 35% were congruent except in mean weight gain that 45% was significantly higher than 35% while 30% was significantly the least acceptable in all parameters. Protein and amino acids are considered the major nutrients influencing egg production given a fair balance of fatty acids⁶¹. The African catfish is a protracted spawner having a short vitello genic period and continues to grow even after sexual maturity hence its protein demand during gonadal activities would be higher. This dietary protein could form the major source of amino acids required for vitello genesis than muscle protein⁶¹ therefore, the growth and reproductive performance of the brood stock on higher dietary protein levels appeared better till the optimal level was attained at 40%. Aryani and Suharman³ reported better growth and reproductive performances with higher crude protein diets in Hemibagrus nemurus. Clarias gariepinus brood stocks fed 35% dietary protein were found to exhibit superior growth rate, FCR and apparent protein utilization than those on 40% but 40% crude protein recorded higher egg development and higher percentage protein for egg and carcass⁸. Muschlisin et al.¹¹ obtained higher growth rate and egg quality with 35 and 40% dietary protein while 40% also gave the highest egg protein content and egg diameter in the female brood stock of bagrid catfish (Mystus nemurus). The eggs with higher diameter are larger and where the African catfishes with heterogeneous egg sizes^27,41 are concerned then the proportions of larger eggs which are expected to produce larger hatchlings are most likely to be higher in those brood stocks fed higher levels of dietary protein till the optimal level.

The four size categories of fingerling obtained in this work were similar to the three noted by Oyebola et al.⁶². Fingerlings of category A in this work have been described as shooters⁶² or fast growers while the B and C fingerlings possibly grouped together as medium or average, and the D fingerlings as runt or slow growers⁶². The medium (aggregate of B and C) fingerlings are larger in number than the runts which are more than the shooters, or even their aggregate (A and D). The fast growers are generally known to be cannibalistic⁶² that Oyebola et al.⁶² suggested that they should be discarded in favour of the culture of the medium and runts in commercial fish production. Culturing the runts with the medium fingerlings as suggested is still plagued with danger of cannibalism considering the logistics of cannibalism (length or weight ratio of the cannibal to that of the prey) for the African catfishes. According to Baras²², the logistics for Clarias gariepinus is 2.1 or 10.4 (length or weight ratio respectively), and for Heterobranchus longifilis is1.75 or 6.00. Similarly, Umanah obtained the logistics for swallowed, and killed and mutilated hybrids of Heterobranchus longifilis × Clarias gariepinus respectively as 1.63 or 3.85 and 1.47 or 2.95. However, the logistics for the suggested stocking combination for example, in this work is between 2.03 and 3.05 or 5.18 and 11.46, therefore the runts would be very close to or much within the thresh hold of the medium fingerlings cannibalism. Rather, the B and C fingerlings should be grown together as the most viable and acceptable seeds while the shooters are separately groomed into future brood stocks due to their perpetual fast growth propensity. The runts could be discarded or possibly reared separately, though. This would prevent depreciation of the gene pool and subsequent decline in the vigor of future stocks as well as checkmate unnecessary extended culture period of small sized fish.

According to Izquierdo et al.¹ and Varghese et al.⁶¹ the gonadal development and fecundity in fishes with short vitello genic period could be manipulated by providing diets shortly before or during spawning. It is believed this could influence both the eggs and larval quality during this period⁵⁹. The directional advantage of larger brood stocks in generating B fingerlings and the significant interaction of brood stock size and protein levels in producing C fingerlings could be synergized in optimizing the aggregate percentage of these fingerlings. Gonadal maturation in larger brood stocks and fish fed higher level of protein than the minimum has been shown to be earlier and better^6,11,59. Brood stock Sc3 (WT1 = 1000 g; WT2 = 2000 g) and Sc (WT1 = 2000 g; WT2 = 3000 g) fed dietary protein levels from 35 – 4 % produced more acceptable fingerlings composition than brood stock sizes below 1300 g and dietary protein level of 30%. The dietary crude protein level of 40% was optimal and the brood stocks of 3000 g final mean weight category were the best in producing the highest aggregate levels of B and C fingerling and the least aggregate levels of A and D fingerlings.

CONCLUSION

Size heterogeneity of fingerlings was observed to be strongly maternal size dependent and remotely modified by dietary protein level towards the production of higher percentage of larger hatchlings. The crude protein level and brood stock size combination that could optimize the percentage aggregate of more compatible fingerlings and less ofhetero geneous or extreme individuals would be the better production strategy. 40% crude protein level and brood stock size of 3000 g category were considered optimal in this work so further researches should explore beyond the brood stock sizes here.

ACKNOWLEDGEMENT

The authors wish to deeply appreciate the enormous contributions of late Professor R. C. A. Orji, Department of Fisheries and Aquatic Resources Management, Michael Okpara University of Agriculture, Umudike, Nigeria to the success of this work. May his gentle soul rest in peace.

REFERENCES

1. Izquierdo, M.S., H. Fernandez-Palacios and A.G.J. Tacon, 2001. Effect of brood stock nutrition on reproductive performance of fish. Aquaculture, 197: 25-42.
2. Adewumi, A.A., V.F. Olaleye and E.A. Adesulu, 2005. Egg and sperm quality of the African catfish, Clarias gariepinus (Burchell) brood stock fed differently heated soybean-based diets. Res. J. Agric. Biol. Sci., 1: 17-22.
3. Aryani, N. and I. Suharman, 2015. Effect of dietary protein level on the reproductive performance of female of green catfish (Hemibagrus nemurus Bagridae). J. Aquacult. Res. Dev.
4. Coward, K. and N.R. Bromage, 2000. Reproductive physiology of female tilapia brood stock. Rev. Fish Biol. Fish., 10: 1-25.
5. Zehra, S. and M.A. Khan, 2011. Dietary protein requirement for fingerling Channa punctatus (bloch), based on growth, feed conversion, protein retention and biochemical composition. Aquacult. Int., 20: 383-395.
6. Chong, A.S.C., S.D. Ishak, Z. Osman and R. Hashim, 2004. Effect of dietary protein level on the reproductive performance of female swordtails Xiphophorus helleri (Poeciliidae). Aquaculture, 234: 381-392.
7. Sule, O. and I. Adikwu, 2004. Effect of broodstock size on egg and larval size and survival of larvae of the African catfish, Clarias gariepinus Under laboratory conditions. J. Aquat. Sci.
8. Sotolu, A.O. and A.A. Kigbu, 2011. Growth and gonad quality of Clarias gariepinus (burchell, 1822) broodstock fed varying dietary protein levels. Prod. Agric. Technol., 7: 61-67.
9. Ochokwu, I.J., T.G. Apollos and J.O. Oshoke, 2015. Effect of egg and sperm quality in successful fish breeding. IOSR J. Agric. Vet. Sci., 8: 48-57.
10. National Research Council, 2015. Nutrient requirements of fish and shrimp. The National Academies Press, United States.
11. Abidin, M.Z., R. Hashim and A.C.S. Chien, 2006. Influence of dietary protein levels on growth and egg quality in brood stock female bagrid catfish (Mystus nemurus Cuv. & Val.). Aquacult. Res., 37: 416-418.
12. Quintero, H.E., A. Hutson, A. Chaimongkol, A. Davis, R. Dunham and A. Abebe, 2009. Effects of varying dietary protein levels and feeding frequencies on condition and reproductive performance of channel catfish to produce hybrid catfish. J. World Aquacult. Soc., 40: 601-615.
13. Nlewadim, A.A. and O.C. Madu, 2004. Growth response and survival of F1 hybrid fry of heterobranchus longifilis and clarias gariepinus reared in glass aquaria and plastic basins. Proceedings of the 19th Fisheries Society of Nigeria Annual Conference, November 29-December 3, 2004, Ilorin, pp: 334-340.
14. Bichi, A., S. Isyaku, E. Danba, I. Kurawa and A. Ayawo, 2015. Effect of brood stock size on egg fertilization, hatchability and fry survival rate of African catfish (Clarias gariepinus). Bayero J. Pure Appl. Sci., 7: 150-154.
15. Marteinsdottir, G. and G. Begg, 2002. Essential relationships incorporating the influence of age, size and condition on variables required for estimation of reproductive potential in atlantic cod Gadus morhua. Mar. Ecol. Prog. Ser., 235: 235-256.
16. Heinimaa, S. and P. Heinima, 2004. Effect of the female size on egg quality and fecundity of the wild Atlantic salmon in the sub-arctic River Teno. Boreal Environ. Res., 9: 55-62.
17. Rideout, R., E. Trippel and M. Litvak, 2005. Effects of egg size, food supply and spawning time on early life history success of haddock Melanogrammus aeglefinus. Mar. Ecol. Prog. Ser., 285: 169-180.
18. Kennedy, J., A.J. Geffen and R.D.M. Nash, 2007. Maternal influences on egg and larval characteristics of plaice (Pleuronectes platessa L.). J. Sea Res., 58: 65-77.
19. Goncalves, I.B., I. Ahnesjö and C. Kvarnemo, 2011. The relationship between female body size and egg size in pipefishes. J. Fish Biol., 78: 1847-1854.
20. Segers, F.H.I.D. and B. Taborsky, 2010. Egg size and food abundance interactively affect juvenile growth and behaviour. Fun. Ecol., 25: 166-176.
21. Oyebola, O. O. and M.O. Awodiran, 2015. Effect of spawning methods on fertilization, hatchability and fry size variation in Clarias gariepinus. Ife J. Sci., 17: 305-311.
22. Baras, E., 1999. Sibling cannibalism among juvenile vundu under controlled conditions. i. cannibalistic behaviour, prey selection and prey size selectivity. J. Fish Biol., 54: 82-105.
23. Offem, B.O., E.O. Ayotunde and G.U. Ikpi, 2008. Dynamics in the reproductive biology of Heterobranchus longifilis val, (Pisces: 1840) in the Inland Wetlands of Cross river, Nigeria. Res. J. Fish. Hydrobiol., 3: 22-31.
24. Owodeinde, F.G. and P.E. Ndimele, 2011. Survival, growth and feed utilization of two clariid catfish (Clarias gariepinus, Burchell 1822 and Heterobranchus bidorsalis, Geoffroy, 1809) and their reciprocal hybrids. J. Applied Ichthyol., 27: 1249-1253.
25. Ndimele, P.E. and F.G. Owodeinde, 2012. Comparative reproductive and growth performance of Clarias gariepinus (Burchell, 1822) and its hybrid induced with synthetic hormone and pituitary gland of Clarias gariepinus. Turk. J. Fish. Aquat. Sci., 12: 619-626.
26. Baras, E., R. Dugué and M. Legendre, 2014. Do cannibalistic fish forage optimally? an experimental study of prey size preference, bioenergetics of cannibalism and their ontogenetic variations in the African catfish Heterobranchus longifilis. Aquat. Living Resour., 27: 51-62.
27. Martins, C.I.M., J.W. Schrama and J.A. Verreth, 2005. Inherent variation in growth efficiency of African catfish Clarias gariepinus (Burchell, 1822) juveniles. Aquacult. Res., 36: 868-875.
28. AOAC., 2000. Official Methods of Analysis of the Association of Official Analytical Chemists. 17th Edn., AOAC International, Washington, DC USA.
29. Umanah, S.I. and G.P. Okure, 2017. Comparative studies of cannibalism on diploid and triploid African catfish-Heterbranchus longifilis. Am. J. Inno. Res. Appl. Sci.,, 5: 53-61.
30. Riley, J., 2002. Presenation of statistical analysis. Exp. Agric., 37: 115-123.
31. McDonald, J.H., 2014. Handbook of Biological Statistics. 3rd Edn., Sparky House Publishing, Baltimore, Maryland, USA., Pages: 305.
32. Xie, S., K. Zheng, J. Chen, Z. Zhang, X. Zhu and Y. Yang, 2010. Effect of water temperature on energy budget of Nile Tilapia, Oreochromis niloticus. Aquacult. Nutr., 17: e683-e690.
33. Akbulut, B., T. Ahin, N. Aksungur and M. Aksungur, 2002. Effect of initial size on growth rate of rainbow trout, Oncorhynchus mykiss, reared in cages on the Turkish black sea coast. Turk. J. Fish. Aquat. Sci., 2: 133-136.
34. Strand, Å., C. Magnhagen and A. Alanärä, 2011. Growth and energy expenditures of eurasian perch Perca fluviatilis (linnaeus) in different temperatures and of different body sizes. J. Aquacult. Res. Dev., Vol. 02.
35. Bavand S. E., and H. Khara 2017. Effect of age on reproductive performance of kutum, Rutilusfrisii (nordmann, 1840) in Shirood river, the southern coast of the Caspian sea. Caspian J. Environ. Sci., 15: 205-212.
36. Hseu, J.R., P.P. Hwang and Y.Y. Ting, 2004. Morphometric model and laboratory analysis of intracohort cannibalism in giant grouper Epinephelus lanceolatus fry. Fish. Sci., 70: 482-486.
37. Umanah, S.I., 2020. Maternal age influence on fry survival, growth and size variation in Clarias gariepinus. Asian J. Anim. Sci., 14: 145-152.
38. Pereira, L.S., A.A. Agostinho and K.O. Winemiller, 2017. Revisiting cannibalism in fishes. Rev. Fish Biol. Fish., 27: 499-513.
39. Negi, D., G. Singh, B. Pant and R.N. Ram, 2018. Broodstock age as a determinant of embryonic development and growth of rohu (Labeo rohita) under captive conditions. Int. J. Curr. Microbiol. App. Sci., 7: 2802-2812.
40. Grabowski, T.B., S.P. Young, J.J. Isely and P.C. Ely, 2012. Age, growth, and reproductive biology of three catostomids from the Apalachicola river, Florida. J. Fish Wildl. Manage., 3: 223-237.
41. Tyor, A.K. and K. Pahwa, 2017. Ovarian development of African sharptooth catfish (Clarias gariepinus, Burchell 1822) from Delhi segment of River Yamuna. J. Fish. Aqua. Sci., 12: 117-126.
42. Reidel, A., W.R. Boscolo, A. Feiden and E. Romagosa, 2010. The effect of diets with different levels of protein and energy on the process of final maturation of the gametes of Rhamdia quelen stocked in cages. Aquacult, 298: 354-359.
43. Ataguba, G.A., V.T. Okomoda and M.C. Onwuka, 2013. Relationship between broodstock weight combination and spawning success in African catfish (Clarias gariepinus). Croatian J. Fish., 71: 176-181.
44. Yusuf, K., S.A. Dada and M.A. Abari, 2013. Length-weight relationship, fecundity and gonadal development of the African catfish (Clarias gariepinus) from doma dam, Nasarawa state, Nigeria. Prod. Agric. Technol., 9: 47-58.
45. Yousif, O.M., A. A-Fatah, A. Rahman and M. Krishnakumar, 2016.. Fecundity of three different size groups of African catfish, Clarias gariepinus (burchell, 1822), females and growth performance of larvae fed different starter food types. Aquaculture.
46. Absalom, K.V., J.A. Anpe, L.E. Igoche and S.A. Okunsebor, 2017. Fecundity and egg size of Clarias gariepinus in Pandam lake, Quan-Pan L.G.A. Plateau state, Nigeria. IOSR J. Agric. Vet. Sci., 10: 19-22.
47. Forchhammer, M.C., T.H. Clutton-Brock, J. Lindström and S.D. Albon, 2001. Climate and population density induce long-term cohort variation in a Northern Ungulate. J. Anim. Ecol., 70: 721-729.
48. Beckerman, A.P., T.G. Benton, C.T. Lapsley and N. Koesters, 2005. How effective are maternal effects at having effects? Proc. Royal Soc. B: Bio. Sci., 273: 485-493.
49. Benton, T.G., J.J.H.S. Clair and S.J. Plaistow, 2008. Maternal effects mediated by maternal age: from life histories to population dynamics. J. Anim. Ecol., 77: 1038-1046.
50. Sakai, S. and Y. Harada, 2001. Why do large mothers produce large offspring? theory and a test. Am. Nat., 157: 348-359.
51. Sakai, S. and Y. Harada, 2001. Sink-limitation and the size-number trade-off of organs: production of organs using a fixed amount of reserves. Evolution, 55: 467-476.
52. Einum, S. and I.A. Fleming, 2000. Highly fecund mothers sacrifice offspring survival to maximize fitness. Nature, 405: 565-567.
53. Badyaev, A.V., T.L. Hamstra, K.P. Oh and D.A.A. Seaman, 2006. Sex-biased maternal effects reduce ectoparasite-induced mortality in a passerine bird. Proc. Nat. Acad. Sci., 103: 14406-14411.
54. Bestion, E., A. Teyssier, F. Aubret, J. Clobert and J. Cote, 2014. Maternal exposure to predator scents: offspring phenotypic adjustment and dispersal. Proc. Royal Soc. B: Bio. Sci., Vol. 281.
55. Berkeley, S.A., C. Chapman and S.M. Sogard, 2004. Maternal age as a determinant of larval growth and survival in a marine fish, Sebastes melanops. Ecol., 85: 1258-1264.
56. Einum, S. and I.A. Fleming, 2004. Environmental unpredictability and offspring size: conservative vs. diversified bet- hedging. Evol. Ecol. Res., 6: 443-455.
57. Arnold, L.M., W.D. Smith, P.D. Spencer, A.N. Evans, S.A. Heppell and S.S. Heppell, 2018. The role of maternal age and context-dependent maternal effects in the offspring provisioning of a long-lived marine teleost. R. Soc. Open Sci.
58. Reed, W.L., M.E. Clark and C.M. Vleck, 2009. Maternal effects increase within‐family variation in offspring survival. Am. Nat., 174: 685-695.
59. El-Sayed, A.F.M., C.R. Mansour and A.A. Ezzat, 2003. Effects of dietary protein level on spawning performance of Nile tilapia (Oreochromis niloticus) broodstock reared at different water salinities. Aquaculture, 220: 619-632.
60. James, R. and K. Sampath, 2003. Effect of animal and plant protein diets on growth and fecundity in ornamental fish, Beta splendens (Regan). J. Aquacult. Bamid., 55: 39-52.
61. Varghese, B., R. Paulraj, G. Gopakumar and K. Chakraborty, 2009. Dietary influence on the egg production and larval viability in true sebae clownfish Amphiprion sebae bleeker 1853. Asian Fish. Sci.,, 22: 7-20.
62. Oyebola, O.O., A.K. Kajopaye and M.O. Awodiran, 2015. Genetic and biochemical basis of growth variation in bred Clarias gariepinus (burchell, 1822). Acad. Res. J. Agric. Sci.,, 3: 111-120.