nueva página del texto (beta)
Inglés (pdf)
Artículo en XML
Referencias del artículo
Traducción automática
Enviar artículo por email
Citado por SciELO 
Similares en
SciELO 
Permalink
INTRODUCTION
The Pacific white snook (Centropomus viridis) has a distribution that ranges from Baja California Sur and the Gulf of California to southern Ecuador and the Galapagos Islands (Fischer et al. 1995). Species in the Centropomidae family inhabit shallow coastal waters, estuaries, rivers, and brackish lagoons and migrate to fresh waters. Juveniles and pre-adults are commonly found in mangrove areas, where they show great tolerance to salinity fluctuations given their osmoregulatory capacity (Álvarez-Lajonchère et al. 2013). Snooks are considered species with high farming potential in Mexico because of their high economic value and overall demand in the national market given the flavor, color, and texture of their meat; their high growth rates; and their ability to adapt in captivity and consume balanced feeds (Álvarez-Lajonchère and Tsuzuki 2008, Labastida-Che et al. 2013). Today, in Mexico, natural populations of snook species are overexploited (Arreguín-Sánchez and Arcos-Huitrón 2011), so biotechnological developments have begun to be made at the marine fish production pilot plant at the Research Center for Food and Development (CIAD, for its acronym in Spanish), Mazatlán (Sinaloa, Mexico), for the production of high-quality C. viridis juveniles to promote the farming of this very important species in Mexico and to develop repopulation programs (Ibarra-Castro et al. 2017).
In the marine fish farming industry, the highest operational cost is feed, so proper design of the feeding strategy is critical for a successful mariculture enterprise (Baloi et al. 2016, D’Abramo 2019). Feeding frequency affects growth, food intake, feed conversion rate and efficiency, body chemical composition, fish survival, and the quality of farming water (Biswas et al. 2010, Shipton and Hasan 2013, Costa-Bomfim et al. 2014, Rahman and Lee 2017). Juvenile marine fish need high feeding frequencies on a daily basis to attain to good performance in the culture (Schnaittacher et al. 2005, da Cunha et al. 2013, Luo et al. 2015); however, overfeeding reduces feed efficiency, increases lipid accumulation, mainly in liver and viscera, deteriorates the quality of farming water, and increases production costs (Lee and Pham 2010, Mizanur and Bai 2014, Lee et al. 2016, Guo et al. 2018). On the other hand, low feeding frequency does not provide the required nutrients for normal growth and survival of fish and, therefore, promotes size dispersion and cannibalism; furthermore, it can cause oxidative damage and immunosuppression (Tucker et al. 2006, Oh and Venmathi-Maran 2015, Tian et al. 2015). Several authors have reported optimal feeding frequencies that improve food intake, digestion, nutrient absorption, growth, and survival in juveniles marine fishes such as Limanda ferrugínea (Dwyer et al. 2002), Pseudosiaena crocea (Xie et al. 2011), Trachinotus ovatus (Wu et al. 2015), Trachinotus blochii (Hamed et al. 2016), Megalobrama amblycephala (Xu et al. 2016), Sebastes inermis (Oh et al. 2018). Likewise, it has been reported that optimal feeding frequency depends on the species, age, size, environmental factors, feed quality, and cultivation system (Hamed et al. 2016, Xu et al. 2016, Oh et al. 2018). There are currently no published data on the frequency and optimal feeding time for C. viridis; therefore, the objective of the present study was to determine the appropriate feeding frequency and time for optimal performance of C. viridis juveniles and, in this way, contribute to the development of farming biotechnology for the species.
MATERIALS AND METHODS
Fish and experimental system
Centropomus viridis juveniles were obtained from the marine fish production pilot plant at CIAD in Mazatlán. The experiment was carried out in the CIAD area for bioassays, in 24 circular fiberglass tanks with black walls and white bottoms, each with 600-L capacity. The tanks had a central drain made with a 50-mm diameter PVC pipe, which was covered with a 5-mm mesh size net to prevent fish from escaping and to facilitate cleaning of the tanks. Each tank was provided with aeration and a flow-through system (approximately 6 L·min-1), with individual flow control valves to regulate flux. Seawater was pumped from Brujas Beach, Mazatlán, and passed through sand filters and cartridge filters with relative retention rating of 16 μm.
Experimental design
A completely randomized one-factor (feeding frequency) experimental design with 3 replicates was used. This design included 8 treatments, which is the number of treatments obtained after combining the number of feedings per day and feeding times (morning, noon, and evening). The assessed treatments were 1M, 1T, 2MD, 2DT, 2MT, 3MDT, 3DT, and 5MDT (Table 1). Each tank contained 20 juveniles weighing 0.36 ± 0.01 g, on average, which were fed a commercial feed for juvenile marine fish ad libitum for 6 weeks (Skretting, 0.8-1.5 mm), according to the dietary regimen (Table 1) that was randomly selected.
Table 1 Experimental design (modified from Van der Meer et al. 1997). Tested feeding frequencies and times for Centropomus viridis juveniles.
| Feeding frequency (Treatment code) |
Feeding time (h) | ||||
| 7:00 | 10:00 | 13:00 | 16:00 | 19:00 | |
| M | X | ||||
| 1T | X | ||||
| 2MD | X | X | |||
| 2DT | X | X | |||
| 2MT | X | X | |||
| 3MDT | X | X | X | ||
| 3DT | X | X | X | ||
| 5MDT | X | X | X | X | X |
M = morning, T = evening, D = noon.
Experimental conditions and development
Daily food intake by juveniles and the temperature, salinity, and dissolved oxygen of water in each tank were recorded. The bottom of the tanks was siphoned daily to remove feces and food remains. During the 6 weeks of the bioassay, water temperature was kept at 29 ± 0.05 ºC, salinity at 34 ± 1.0, and dissolved oxygen at 5.5 ± 0.3 mg·L-1. To assess growth, at the end of the experiment, all juveniles from each replica were individually sedated with clove essence (0.2 mL·L-1), and after removing excess moisture with blotting paper, each individual was weighed on a digital scale with ±0.05 g precision and measured for total length (TL) with a conventional vernier. Gained weight (GW), growth rate (GR), specific growth rate (SGR), and coefficient of variation (CV) were calculated using the following formulas:
Feeding efficiency was determined using the food conversión ratio (FCR), feeding efficiency rate (FER), hepatosomatic index (HI), and peritoneal fat index (PFI):
The survival percentage was calculated (S):
Statistical analysis of data
Percentage values were arcsine transformed to homogenize variances. All results were tested for normality (Bartlett’s test) and homoscedasticity (Levene’s test). Normal and homoscedastic data (FW, GW, GR, SGR, CV, FI, FCR, and FER) were analyzed using a one-way analysis of variance (ANOVA, P < 0.05), and significant differences between treatments were determined by Tukey’s multiple rank comparison tests (α = 0.05); data not showing a normal distribution (HI and PFI) were analyzed using a Kruskal-Wallis test (P < 0.05) and significant differences were determined using Levene’s test based on the median (Zar 1996). All statistical analyses were done using the Statgraphics Centurion XVI program v.16.204.
RESULTS
FW and GW were significantly higher (ANOVA, P =0.0000) in juveniles fed 5 times a day (5MDT). GR and SGR values for juveniles fed 3 times a day, regardless of feeding time (3MDT and 3DT), did not show significant differences (ANOVA, P > 0.05) with respect to the results obtained for organisms in the 5MDT treatment. Fish fed only once or twice a day, either in the morning (1M, 2MD) or in the evening (1T, 2DT), showed significantly less growth (ANOVA, P = 0.0000) than the rest of the juveniles in the other treatments. Growth in juveniles fed 2 times a day in the morning and the evening (2MT) was not significantly different (ANOVA, P > 0.05) from growth in juveniles fed 3 times a day (3MDT and 3DT). The CV showed no significant differences (ANOVA, P = 0.7548) between treatments (Table 2).
Table 2 Growth performance of juvenile Centropomus viridis during the experiment. IW, initial average weight; FW, final average weight; GW, gained weight; GR, growth rate; SGR, specific growth rate; CV, coefficient of variation.
| Treatment | IW (g) | FW (g) | GW (g) | GR (%) | SGR (% d-1) | CV (%) |
| 1M | 0.35 ± 0.02 | 2.79 ± 0.20a | 2.43 ± 0.20a | 685.98 ± 13.60a | 4.59 ± 0.10a | 31.00 ± 6.00 |
| 1T | 0.34 ± 0.01 | 2.70 ± 0.10a | 2.35 ± 0.10a | 676.50 ± 66.80a | 4.56 ± 0.02a | 33.33 ± 4.60 |
| 2MD | 0.37 ± 0.01 | 4.47 ± 0.30b | 4.09 ± 0.30b | 1064.04 ± 93.80b | 5.50 ± 0.20b | 34.66 ± 2.00 |
| 2DT | 0.35 ± 0.01 | 4.49 ± 0.10b | 4.13 ± 0.10b | 1173.07 ± 103.40b | 5.66 ± 0.10b | 30.00 ± 6.00 |
| 2MT | 0.36 ± 0.02 | 4.98 ± 0.40bc | 4.61 ± 0.50bc | 1247.93 ± 185.60bc | 5.78 ± 0.30bc | 27.33 ± 5.10 |
| 3MDT | 0.36 ± 0.01 | 5.97 ± 0.70c | 5.60 ± 0.60c | 1519.45 ± 115.90cd | 6.23 ± 0.10cd | 32.66 ± 3.00 |
| 3DT | 0.34 ± 0.02 | 5.65 ± 0.60bc | 5.29 ± 0.60bc | 1506.00 ± 173.70cd | 6.19 ± 0.20cd | 34.33 ± 8.10 |
| 5MDT | 0.38 ± 0.01 | 7.32 ± 0.40d | 7.00 ± 0.30d | 1831.29 ± 75.47d | 6.57 ± 0.10d | 32.33 ± 5.50 |
Values (mean ± SD, n = 3) with different letters in the same column are significantly different according to Tukey’s multiple range test (( = 0.05).
Regarding feeding efficiency, juveniles in the 1M and 1T treatments showed significantly higher FCR (ANOVA, P = 0.0009) and significantly lower FER (ANOVA, P = 0.0009) in comparison with juveniles in the rest of the treatments. HI showed no significant differences (Kruskal-Wallis, P = 0.3928) between treatments. PFI was significantly different (Kruskal-Wallis, P = 0.00029) only between juveniles in the 5MDT treatment and juveniles in the 1M and 1T treatments. Regarding survival, no significant differences (P < 0.05) were observed between treatments (Table 3).
Table 3 Food utilization, somatic indexes, and survival of juvenile white snook Centropomus viridis during the experiment. FI, food intake; FCR, food conversion ratio; FER, feeding efficiency rate; HI, hepatosomatic index; PFI, peritoneal fat index; S, survival.
| Treatment | FI (g) | FCR | FER (%) | HI (%) | PFI (%) | S (%) |
| 1M | 3.74 ± 0.30ab | 1.55 ± 0.20b | 65.90 ± 12.30a | 1.01 ± 0.70 | 2.28 ± 0.80a | 100 |
| 1T | 3.14 ± 0.06a | 1.33 ± 0.06b | 74.90 ± 3.30a | 1.45 ± 0.80 | 2.04 ± 0.40a | 100 |
| 2MD | 4.49 ± 0.40bc | 1.09 ± 0.07a | 91.23 ± 6.90b | 1.55 ± 0.10 | 3.25 ± 1.00ab | 96.66 ± 2.80 |
| 2DT | 4.19 ± 0.10abc | 1.01 ± 0.05a | 97.24 ± 2.40b | 2.52 ± 1.90 | 2.96 ± 0.70ab | 100 |
| 2MT | 4.70 ± 0.40bc | 1.02 ± 0.10a | 94.98 ± 8.60b | 1.53 ± 0.40 | 3.45 ± 0.70ab | 96.66 ± 2.80 |
| 3MDT | 5.82 ± 0.20d | 1.04 ± 0.16a | 96.70 ± 16.00b | 1.50 ± 0.70 | 2.99 ± 1.90ab | 100 |
| 3DT | 5.13 ± 0.60cd | 0.96 ± 0.07a | 103.6 ± 8.10b | 1.56 ± 0.60 | 3.22 ± 0.40ab | 96.66 ± 2.80 |
| 5MDT | 6.96 ± 0.10e | 1.00 ± 0.04a | 98.03 ± 1.80b | 1.38 ± 0.40 | 4.18 ± 0.90b | 100 |
Values (mean ± SD, n = 3) with different letters in the same column are significantly different according to Tukey’s multiple range test (( = 0.05), except for IGP, which follows Levene’s median contrast.
DISCUSSION
Several studies have shown that growth and feeding efficiency increase when feeding frequency is increased up to a certain number of times (Ribeiro et al. 2015, Baloi et al. 2016, Guo et al. 2018, Oh et al. 2019). The present study showed that growth (FCR and FER) in C. viridis juveniles increased significantly when they were fed 3 times a day and that growth was not significantly different when they were fed 5 times a day. These results are similar to those reported by Mendes-de-Oliveira et al. (2019), who tested 2 to 6 feeding frequencies per day in Centropomus undecimalis and concluded that juveniles should be fed 4 times a day for adequate growth. Moreover, Tian et al. (2015) assessed the effect of 1 to 6 feeding frequencies per day on the growth of Megalobrama amblycephala juveniles and determined that optimal feeding frequency was 3 times a day. On the contrary, if feeding frequency is reduced to 1 or 2 times a day, growth will decrease because fish cannot obtain the necessary nutrients and meet the energy requirements for their performance and somatic development; likewise, FER will decrease and FCR will increase (Salama 2008, Biswas et al. 2010, Lee and Pham 2010, Hamed et al. 2016). This was observed in this study when C. viridis juveniles fed 1 or 2 times a day showed lower GW, GR, and SGR compared with fish fed 3 and 5 times a day. FCR was significantly higher and FER was significantly lower in juveniles fed once a day. Increasing feeding frequency, on the other hand, has been reported to produce uniform fish sizes, probably because small fish have a better chance of consuming food and competition between fish in the same tank decreases (Mihelakakis et al. 2002, Tucker et al. 2006, Biswas et al. 2010, Ribeiro et al. 2015). However, in the present study, no significant differences were observed in CV values for weight between the different treatments, which has also been observed for C. undecimalis (Mendes-de-Oliveria et al. 2019) and other marine fish species such as L. ferruginea (Dwyer et al. 2002),
Pagrus auratus (Booth et al. 2008), and Platichthys flesus luscus (Küçük et al. 2014).
Regarding feeding time, the present study showed that there is no significant effect on juvenile C. viridis growth and feeding efficiency if feeding is done in the morning or the evening. This finding was also reported for Sigamus rivulatus (Barakat et al. 2011) and Colossoma macropomum (van der Meer et al. 1997). However, in some species, such as P. flesus luscus (Küçük et al. 2014) and Oplegnathus fasciatus (Oh and Venmathi-Maran 2015), reports show that food intake is higher in the first hour of the morning and in other species, such as Sardinella brasiliensis (Baloi et al. 2016) and Hapalogenys nigripinnis (Oh et al. 2019), intake is higher in the evening. In the present study, the intake of food supplied during each feeding was not evaluated, only total consumption per day, so it was not possible to determine the time of day with the highest food intake. In this study, HI in all treatments was not significantly affected by feeding frequency, and PFI was significantly higher in juveniles from the 5MDT treatment than in juveniles fed only once a day, regardless of feeding time (1M and 1T). Low PFI values suggest that during feed restriction, fish use stored fat to meet the energy requirements for growth (Baloi et al. 2016), which would explain why in the present study fish fed only once a day (1M and 1T) showed lower PFI values than fish fed 5 times a day (5MDT), since after having had insufficient food, they were possibly unable to store as much peritoneal fat as fish in the 5MDT treatment. Other studies on marine fish have also reported that PFI is affected by feeding frequency, the higher the feeding frequency, the higher the PFI values (Mizanur and Bai 2014, Baloi et al. 2016, Guo et al. 2018).
From the results obtained in the present study, it can be concluded that, under the same experimental conditions, the optimal feeding frequency for pre-fattening C. viridis juveniles is 3 times a day within 6-hour intervals. These results will contribute to the development of farming biotechnology aiming to obtain adequate growth and feeding efficiency for this species, thus enabling operational cost optimization by using only the required feed, in addition to preventing organic waste accumulation due to unconsumed food.
