Abbreviations:

IMBW: Initial Mean Body Weight; FMBW: Final Mean Body Weight; FMBL: Final Mean Body Length; MBWG: Mean Body Weight Gain; SGR: Specific Growth Rate; GFY: Gross Fish Yield; FCR: Feed Conversion Ratio; FCF: Fulton’s Condition Factors; DSF: Daily Supplementary Feed; DO: Dissolved Oxygen; TWv: Total Water volume; TDS: Total Dissolved Solid; NFARRC: National Fishery and Other Aquatic Resources Research Center; NSF: Number of Stocked Fish; NDF: Number of Dead Fish

1 Introduction

In the view of cultured fish species, tilapia is one of the most commercially important and widely used fish in the global aquaculture production particularly in tropical and subtropical areas (Gjedrem, 2005; El-Sayed, 2006). Oreochromis niloticus is by far the most important tilapia species and distributed throughout the world (Bentsen et al., 1998; Pillay and Kutty, 2005; El-Sayed, 2006). The importance of Oreochromis niloticus seams from its biological reasons: fast growth, high food conversion ratio, readily accepting artificial feeds, ease of breeding in captivity, disease resistance, high fecundity; social reasons: good table food quality, good market price; and physical reasons: tolerant to a wide range of environmental conditions (El-Sayed, 2006; Ashagrie et al., 2008) Therefore, determination of different factors such as feed quality, stocking density and strain variations for cultured tilapia is essential to maximize its production, profitability and sustainability. Among these, selection of the best strains for efficient breeding program is crucial; not only to reach the production goal but also to reduce production cost, to improve disease resistance, utilisation of feed resources and product quality (Gjedrem, 1997). Even though Ethiopian aquaculture practice is growing and seams promising, there are no breeding centres that produce genetically improved fish seeds for local fish farmers and thus, fish seeds are collected from the nearby lake or lakes. Moreover, there are no research reports on the growth performance of different Oreochromis niloticus strains in pond culture. Hence,comparative data are not available on the growth performance and feed utilization efficiency on different Oreochromis niloticus strains in Ethiopia. The general objective of this study was therefore to compare the growth performance of different Oreochromis niloticus strains collected from various Ethiopian freshwater lakes (such as Lake Hawassa, Lake Ziway, Lake Koka and Lake Hora) in pond culture to enhance the production of tilapia aquaculture in the country.

2 Results
The mean value of some water quality parameters of the ponds were calculated and summarized in Table 1. The results revealed that all the ponds have similar water quality parameters and were not significantly (p > 0.05) different among triplicate. Moreover, variations in growth performance among triplicate for all strains were insignificant. Similarly, the initial body size of all the strains were no significantly (p > 0.05) different (Table 2). This implies, the initial body size of all the strains were homogeneous.

Table 1 The mean value of some water quality parameters measured during the experiment

Table 2 Growth parameters, feed conversion ratio, Fulton condition factor, gross fish yield and survival rate of four Oreochromis niloticus stains during experimental period

The different growth parameters for all the strains were calculated after 60 days of the experiment and were summarized in Table 2. The results revealed that the different growth parameters were significantly (P < 0.05) affected by strain difference. The highest and the lowest final mean body weight were observed in the Koka strain (64.04 g) and the Ziway strain (53.13 g), respectively. The Koka strain had significantly (p < 0.05) higher final mean body weight than the Hora and the Ziway strains, but no from the Hawassa strains. Similarly, mean body weight gain and specific growth rate exhibited the same trend. Specific growth rate deceased as rearing period increased (Figure 1). The final mean body length of all the strains also revealed significantly (p < 0.05) different among the strains (Table 2).

Figure 1 Mean values of specific growth rate for different Oreochromis niloticus strains at different sampling periods

The best mean values of feed conversion ratio (1.72) was achieved by the Koka strain while the poorest mean values of feed conversion ratio (1.76) was achieved by the Hawassa strains. In addition, the mean value of feed conversion ratio increases with time (Figure 2). There was also a direct relationship between Fulton condition factor and culturing period though out the experiment (Figure 3). However, statistical analysis revealed that mean values of feed conversion ratio and Fulton condition factor were not significantly different among all the strains (Table 2).

Figure 2 Mean values of feed conversion ratio for different Oreochromis niloticus strain at different sampling periods

Figure 3 Mean values of Fulton condition factor for different Oreochromis niloticus strains at different sampling periods

The highest gross fish yield was achieved in the Koka strain (160.09 g/m³) followed by the Hawassa (149.08 g/m³) and the Hora strains (139.58 g/m³) while the Ziway strain had the lowest gross fish yield (131.04 g/m³). Statistical analysis revealed that the Koka strain had significantly higher gross fish yield than the Ziway and the Hora strains, but there not than the Hawassa strains (Table 2). Survival rate of all the strains was very high ranging from 96% to 100%.

As indicated by Figure 4 and Figure 5, at early life stage, change in length was greater than change in weight. Later, change in weight was greater than change in weight, implying that fish grow in weight than in length at late life stage. However, there was a strong relationship between length and weight of the fish for all strains (Table 3). With the exception of Ziway strain (3.37), all the strains had not significantly different from 3. The Koka strain had slightly higher 'r' value (93.5%) than the Hawassa strain (88.0%), the Ziway (87.7%) and the Hora strains (89.12%). This means that length and weight of the Koka strain had higher correlation than other strains, but statistically, there was no significant difference among strains in their 'r' value.
 

Figure 4 Changes in live body length (cm/fish) of different strains of Oreochromis niloticus at different sampling periods

Figure 5 Changes in live body weight (g/fish) of different strains of Oreochromis niloticus at different sampling periods

Table 3 Estimated parameters of length-weight relationship for different Oreochromis niloticus strains

3 Discussions
The growth performance of Oreochromis niloticus depends on genetic materials, food quality, energy content of the food, stocking density and environmental factors (Gjedrem, 1997; El-Sayed, 1999; Ashagrie et al., 2008). In this study growth performance of different juvenile Oreochromis niloticus strains collected from different Ethiopian freshwater lakes were compared under pond culture condition. At all experimental ponds, the different water quality parameters (such as DO, pH, TDS and temperature) measured during the study period were similar and remained in the favourable range for rearing Oreochromis niloticus fingerlings (Yi, 1998; Lei and Li, 2000; Xu et al., 2005; Azaza et al., 2008). However, the temperature was much less than that of the temperature ranges from 28℃ to 30℃ which are optimum for maximum growth of Oreochromis niloticus (Beamish, 1970; El-Sayed, 2006). Moreover, the initial mean body size of the fish was homogeneous. Thus, variation in growth performance and food conversion ratio of the fish for all the strains were not affected by external factors and initial body size among the strains. In other hand, variation in growth performance was caused by strain variation. Thus, in this study comparison of growth performance of different juvenile Oreochromis niloticus strains were made based on different growth parameters including mean body weight gain, specific growth rate and daily growth rate as indicator of growth, food conversion ratio as indicator of feed utilization efficiency, gross fish yield as indicator of production potential and length-weight relationships as indicator of growth condition and growth pattern of the fish.

The results of the present study revealed that the growth performance of juvenile Oreochromis niloticus affected by strain variation and were significantly different (p < 0.05). The Koka strain had the highest growth performance while the Ziway strain had the lowest growth performance. The body weight of the Koka strain was nearly 20.53%, 14.71% and 7.4% higher than that of the Ziway, the Hora and the Hawassa strains, respectively. These results were in line with the work of Abdel-Tawwab (2004), who worked on the growth performance of four different Oreochromis niloticus strains collected from four Egyptian lakes (Abbassan, Asswan, Manzalah and Maryut lakes). The author reported that growth performance is affected by strain differences. Ridha (2006) made a similar observation and reported that growth performance of Oreochromis niloticus is affected by strain differences. The present result is also in agreement with the finding of Eknath et al (1993) and Palada-de-Vera and Eknath (1993) who compared the growth performance of four African and four Asian Oreochromis niloticus strains at different environmental conditions. The authors found that the growth performance of the fish was significantly affected by strain variation in which the African strains had better growth performance than the Asian strains with the exception of Ghanaian strain. Ibrahim et al (2012) also confirmed that growth performance of Oreochromis niloticus is affected by strain variation. The authors reported that Abassa line showed a superior final body weight at harvest (28%) over the Kafer El Sheikh strain.

The SGR (2.59% to 2.73% /day) obtained in this study were comparable with the report of Middendorp (1995) for juvenile Oreochromis niloticus in pond culture ranging from 2.2% to 3.1% /day. Moreover, the specific growth rate observed in these study was higher than the specific growth rate that reported by Ashagrie et al (2008) (0.787% to 1.035% /day), Abdel-Tawwab (2004) (0.967% to 1.277% /day) and Al-Hafedh (1999) (0.49% to 0.75% /day) obtained for juvenile Oreochromis niloticus. Generally, the SGR of Koka strain was nearly 13.28%, 9.2% and 5.41% higher than that of the Ziway, the Hora and the Hawassa strains, respectively.

The Koka strain had the highest growth performance with slightly lower feed conversion ratio value than the other strains. This result is in agreement with the work of Abdel-Tawwab (2004) who reported that the lowest feed conversion ratio is obtained in the Aswan strain that had the highest growth performance. The feed conversion ratio that ranges from 1.72 to 1.76 obtained in this study was consistence with the range of 1.45 to 2.40 for juvenile Oreochromis niloticus, reported by Yi (1998) in cage culture. However, the present result was much less than the feed conversion ratio (3.15 to 4.86) reported by Al-Hafedh (1999). In contrast, it was slightly higher than the feed conversion ratio that ranges from 1.01 to 1.6 reported by Diana et al (2004) in pond culture. Thus, it is possible to say that different strains have different feed conversion ratio at different environmental conditions.

The results of the present study also showed that fish survival for all stations was reasonably good ranging from 96% to 100%. This result is in line with the work of Ridha (2006) who observed that the survival rate of juvenile Oreochromis niloticus was relatively high for all strains (97.4% to 100%) in tank culture. El-Sayed (2002) also observed that the survival rate of Oreochromis niloticus fry was between 90% and 100%.

The results of this study also revealed that gross fish yield was affected by strain differences. The maximum gross fish yield was obtained from the Koka strain (160.09 g/m³) followed by the Hawassa strain (149.08 g/m³). The lowest gross fish yield was obtained from the Ziway strain (131.04 g/m³). The Koka strain had 22.17% higher gross fish yield than that of the Ziway strain. This result is in agreement with the work of Ridha (2006) who found that the strain with the highest growth performance had the highest gross fish yield (45.4 kg/m³) and the lowest feed conversion ratio (1.27) and the poorest growth performance had the lowest gross fish yield (30.4 kg/m³) and the highest feed conversion ratio (1.55).

In this study, length-weight relationship of all the juvenile Oreochromis niloticus strains showed a strong relationship (r ≥ 88% for all strains). This implies that the body length and body weight of the fish was proportional and has good body condition. Fulton's conditional factor (1.58 to 1.82) obtained in this study was comparable with the average value of 1.46 in Lake Ziway and 1.86 in Lake Chamo reported by Admassu & Ahlgren (Admassu and Ahlgren, 2000). Except Ziway strain, the growth exponent 'b' of length-weight relationship obtained in this study was nearly 3 that indicates an isomeric growth pattern. This means that the fish shape is consistent. The Ziway strain (3.78), however, had allotropic growth pattern (Costa and Araújo, 2003; Olurin and Aderibigbe, 2006).

Generally, the growth performance of the Koka strain was ranked first followed by the Hawassa strain. The Ziway strain had the lowest growth performance. The results of this study clearly indicated that the Ziway strain had the poorest growth performance among the four strains tested, which is supported by the work of Admassu and Ahlgren (2000) who reported that the growth performance of juvenile Oreochromis niloticus strain in Lake Ziway showed the lowest growth performance than found in Lake Chamo and Lake Langano in their natural habitat using otolith analysis.

4 Conclusions and Recommendations
The results of the study clearly demonstrated that the growth performance of different Oreochromis niloticus strains collected from Ethiopian freshwater lakes was significantly different in which the Koka strain had the highest growth performance and feed utilization efficiency than the other strains tested. The poorest growth performance was observed in the Ziway strain. It is worth to say that the best growth performed strain (Koka strain) had best feed utilization efficiency and had the highest gross fish yield at harvest. This implies that growth performance of Oreochromis niloticus is strain dependent. Thus, it is very important to select the correct strain to enhancement Oreochromis niloticus aquaculture production. Therefore, it is recommended that Oreochromis niloticus hatcheries should use Koka strain than other strains tested for fry propagation at least for similar climate condition. However, further research may still need to compare more strains at different life stages as well as at different environmental conditions with varieties of feed to evaluate the genetic potential of the strains. Thus, the expected genetic differences between strains can be used for establishment of base population that could be utilized for the development of fish culture and breeding program.

5 Materials and Methods
5.1 Description of experimental site
The experiment was conducted in National Fishery and Other Aquatic Resources Research Center (NFARRC), Ethiopia. The research center is situated to southwest direction of Addis Ababa, the capital city of Ethiopia at 24 km. The climatic zone is classified as mid-altitude zone (2 240 m above sea level).The mean annual temperature of the area is 21℃ with an average annual rainfall of 890 mm (Ashenafi and Eshutu, 2004).

5.2 Experimental design and juvenile Oreochromis niloticus strain collection
The experiment was designed to investigate the effects of strain variation on growth performance, survival rate, feed conversion ratio, growth pattern and gross fish yield. For this purpose, three similar sizes (10 m × 10 m × 1.5 m) of ponds were selected randomly. All the three ponds were cleaned and left for one week for complete drying. Then, each pond was partitioned into four parts using nets. All the ponds were filled with water at 100 cm water. Following pond preparation, healthy juvenile Oreochromis niloticus strains of mixed-sex were collected from four Ethiopian lakes (Hawassa, Ziway, Koka and Hora Lakes) by using a 50 m length and 2.5 m width beach sieve mesh whose stretched length was 20 mm. Immediately after captured, appropriate size of juvenile Oreochromis niloticus was screened for other species as well as for large and small size of the same species by hand picking. Polyethylene bags containing approximately 25 litres of water were used to transport about 200 individuals of juveniles Oreochromis niloticus (Ashagrie et al., 2008). Depending on the distance between the lake and the research center, the fish were provided pure oxygen more than once using oxygen cylinder. At the study site, the polyethylene bag containing the fish was immersed into the pond for about 20 minutes to acclimatize the fish to the new water condition. Moreover, the polyethylene bag was tied off and allowed a flow of water into the bag to make an equilibrium condition. The fish then swim and move to the pond. Finally, all the strains of the fish were stocked into acclimatization ponds for three weeks until fish become more active and stopped mass mortality due to stress during transportation. Dead and weaken fish were removed daily. Then the fish having an average body length 9.18 cm and weight 12.48 g were transferred to the experimental ponds.

5.3 Feeding and feed supplements
The experimental fish were fed twice a day at 10:00 and 17:00 hours with the feed produced by Akaki feed factory when the daily surface water temperature is warmer (Tran-Duy et al., 2008). The feeding rate was 3% of the body weight of the fish per day throughout the experiment (Abdel-Tawwab, 2004). The nutritional composition of the diet used for the experiment was 89.69% dry matter which contains 92.18% organic matters and 7.82% ash. The ash also contains Phosphorus (0.08%), nitrogen (2.2%), Calcium (0.19%), Magnesium (0.32%), Iron (0.355%), Potassium (1.09%) and sodium (79.9‰). About 23% of the organic matter is crud proteins. The amount of the feed was adjusted once in two weeks based on the body weight of the fish. Thus the amount of daily supplementary feed (DSF) was calculated using the average body weight (ABW), the total number of the fish (N) and the feeding rate per day (FR /d) using DSF = ABW × N × FR /d.

5.4 Data collection
5.4.1 Water quality parameters
During the experiment, water quality parameters such as temperature, dissolved oxygen (DO) concentration and pH were measured daily using Hydrolab, Model "Multi 340I/SET". Total dissolved solid particles (TDS), conductivity and salinity were also measured once in two weeks interval with the same instrument.

5.4.2 Sampling of fish and measuring of growth parameters
50% of fish were sampled every two weeks for their body weight and body length measurement. Length and weight of the fish were measured using ruler and digital weight balance (Ohaus portable balance), respectively. Mortality of the fish was also registered throughout the experiment. Sexual maturity and other effects were also registered for all strains.

5.4.3 Data analysis
Based on the data collected during the experiment, growth performance, and feed conversion ratio Futon condition factor, gross fish yield and rate of survival were calculated as described by Ridha (2006) as follows:

SGR (%/day) = 〔(lnFMBW － lnIMBW)/dã€•× 100
FCR = FI (g)/ MBWG (g)
FCF = FMBW / FMBL³ × 100
GFY = FMBW/TWv
Survival rate (%) = (NSF － NDF/NSF) × 100

Where: - FBW and IBW are the final and initial mean body weight, respectively
- d is the time interval in days during the study period
- FI is amount of feed intake (g) in dry weight basis
- MBWG and FMBL are mean body weight gain in gram (g) and final mean body length fish (cm), respectively
- TWv is the total water volume used for culturing fish in meter cube (m³)
- NSF and NDF are Number of stocked and dead fish during, respectively

5.5 Statistical analysis
Mean growth performance parameters, survival and feed utilization were analyzed for all strains at different sampling periods using analysis of variance (General Linear Model). T-test and Turkey test were used to identify the mean value that causes a significant difference for the analysis of variance. Statistical significance was determined at p < 0.05. All calculation was performed using Minitab 15 and SPSS 14 versions of statistical software.

Acknowledgement
We would like to express our appreciate to all staffs of the National Fishery and Other Aquatic Resources Research Centre (NFARRC) and Ziway Fisheries Resource Centre who collaborated in different ways while we were conducting our experiment. We would also indebted to the Norwegian Government State Educational Loan Fund.

Author's Contribution
Professor Gjoen, H.MAB contributes substantial contribution for this paper. It is impossible to conduct this experiment and prepare such kinds of manuscript. His contribution on experimental design, data analysis and interpretation and writing up of the manuscript was unlimited.

Reference
Abdel-Tawwab M., 2004, Comparative study of growth performance and feed utilization of four local strains of Nile tilapia, (Oreochromis niloticus L), collected from different location in Egypt, In: Bolivar R., Mair G. and Fitzsimmons.K.(eds.), The 6th International Symposium on Tilapia in Aquaculture, Manila Philippines, pp.510-517  

Admassu D., and Ahlgren I., 2000, Growth of juvenile tilapia (Oreochromis niloticus) from Lakes Ziway, Langeno and Chamo (Ethiopian rift valley) based on otolith microincrement analysis, Ecology of Freshwater Fish, 9: 127-137
http://dx.doi.org/10.1111/j.1600-0633.2000.eff090301.x

Al-Hafedh Y.S., 1999, Effects of dietary protein on growth and body composition of Nile tilapia, Oreochromis niloticus (L., 1758), Aquaculture Research, 30: 385-393
http://dx.doi.org/10.1046/j.1365-2109.1999.00343.x

Ashagrie G., Getahun A., and Mengistou S., 2008, Effect of stocking density on the growth performance and yield of Nile tilapia [Oreochromis niloticus (L., 1758)] in a cage culture system in Lake Kuriftu, Ethiopia, Aquaculture Research, 39: 1450-1460
http://dx.doi.org/10.1111/j.1365-2109.2008.02021.x

Ashenafi H., and Eshutu Y., 2004, Study on gastrointestinal helminths of local chickens in central Ethiopia, Revue Méd. Vét, 10: 504-507
http://dx.doi.org/10.1023/B:TROP.0000042852.82477.af

Azaza M.S., Dhraïef M.N., and Kraïem M.M., 2008, Effects of water temperature on growth and sex ratio of juvenile Nile tilapia (Oreochromis niloticus) reared in geothermal waters in southern Tunisia, Journal of Thermal Biology, 33: 98-105
http://dx.doi.org/10.1016/j.jtherbio.2007.05.007

Beamish F.W.H., 1970, Influence of temperature and salinity acclimation on temperature preferenda of the euryhaline fish Tilapia nilotica, Journal of the Fisheries Research Board of Canada, 27: 1209-1214
http://dx.doi.org/10.1139/f70-143

Bentsen H.B., Eknath A.E., Palada-De Vera M.S., Danting J.C., Bolivar H.L., Reyes R.A., Dionisio E.E., Longalong F.M., Circa A.V., Tayamen M.M., and Gjerde B., 1998, Genetic improvement of farmed tilapias: growth performance in a complete diallel cross experiment with eight strains of Oreochromis niloticus, Aquaculture, 160: 145-173
http://dx.doi.org/10.1016/S0044-8486(97)00230-5

Costa M.R.D., and Araújo F.G., 2003, Length-weight relationship and condition factor of Micropogonias furnieri (Desmarest) (Perciformes, Sciaenidae) in the Sepetiba Bay, Rio de Janeiro State, Brazil, Revista Brasileira de Zoologia, 20: 685-690
http://dx.doi.org/10.1590/S0101-81752003000400022

Diana J.S., Yi Y., and Lin C.K., 2004, Stocking densities and fertilization regimes for Nile tilapia (Oreochromis niloticus) production in ponds with supplemental feeding, In: Bolivar R., Mair G., and Fitzsimmons K.(eds.), The 6th International Symposium on Tilapia in Aquaculture, Manila, Philippines, pp.467-499

Eknath A.E., Tayamen M.M., Palada-De Vera M.S., Danting J.C., Reyes R.A., Dionisio E.E., Capili J.B., Bolivar H.L., Abella T.A., Circa A.V., Bentsen H.B., Gjerde B., Gjedrem T., and Pullin R.S.V., 1993, Genetic improvement of farmed tilapias: the growth performance of eight strains of Oreochromis niloticus tested in different farm environments, Aquaculture, 111: 171-188
http://dx.doi.org/10.1016/0044-8486(93)90035-W

El-Sayed A.F.M., 1999, Alternative dietary protein sources for farmed tilapia, Oreochromis spp, Aquaculture, 179: 149-168
http://dx.doi.org/10.1016/S0044-8486(99)00159-3

El-Sayed A.F.M., 2002, Effects of stocking density and feeding levels on growth and feed efficiency of Nile tilapia (Oreochromis niloticus) fry, Aquaculture Research, 33: 621-626
http://dx.doi.org/10.1046/j.1365-2109.2002.00700.x

El-Sayed A.F.M., eds., 2006, Tilapia culture, CABI Publishing Wallingford Oxfordshire, UK, pp.1-45
http://dx.doi.org/10.1079/9780851990149.0001

Gjedrem T., 1997, Selective breeding to improve aquaculture production, World Aquaculture, 28: 33-45

Gjedrem T., 2005, Status and scope of aquaculture. In: Gjedrem.T. (ed.), Selection and breeding program in aquaculture, Springer, Berlin, New York, pp.1-7
http://dx.doi.org/10.1007/1-4020-3342-7_1

Ibrahim N.A., Zaid M.Y.A., Khaw H.L., El-Naggar G.O., and Ponzoni R.W., 2012, Relative performance of two Nile tilapia (Oreochromis niloticus) strains in Egypt: The Abbassa selection line and the Kafr El Sheikh commercial strain, Aquaculture Research, 43:1-10
http://dx.doi.org/10.1111/j.1365-2109.2012.03240.x

Lei S., and Li D., 2000, Effect of temperature on energy budget of Taiwanese red tilapia hybrid (Oreochromis niloticus × Oreochromis mossambicus), YingYong ShengTai XueBao, 11: 618-620 PMid:11767691

Middendorp A.J., 1995, Pond farming of Nile tilapia (Oreochromis niloticus) in northern Cameroon. Feeding combinations of cottonseed cake and brewery waste in fingerling culture, hand-sexed male monosex culture, and mixed culture with police-fish, Clarias gariepinus, Aquaculture Research, 26: 715-722
http://dx.doi.org/10.1111/j.1365-2109.1995.tb00863.x

Olurin K.B., and Aderibigbe O.A., 2006, Length-weight relationship and condition factor of pond reared juvenile Oreochromis niloticus, World journal of Zoology, 2: 82-85

Palada-De Vera M.S., and Eknath A.E., 1993, Predictability of individual growth rates in tilapia, Aquaculture Research, 111: 147-158
http://dx.doi.org/10.1016/0044-8486(93)90033-U

Pillay T.V.R., and Kutty M.N., eds., 2005, Aquaculture principles and practices, UK, Blackwell Publishing, Oxford, pp.400-415

Ridha M.T., 2006, Comparative study of growth performance of three strains of Nile tilapia, (Oreochromis niloticus L.) at two stocking densities, Aquaculture Research, 37: 172-179
http://dx.doi.org/10.1111/j.1365-2109.2005.01415.x

Tran-Duy A., Schrama J.W., Van Dam A.A., and Verreth J.a.J., 2008, Effects of oxygen concentration and body weight on maximum feed intake, growth and hematological parameters of Nile tilapia (Oreochromis niloticus), Aquaculture Research, 275: 152-162
http://dx.doi.org/10.1016/j.aquaculture.2007.12.024

Xu J.Y., Miao X.W., Lu Y., and Cui S.R., 2005, Behavioural response of tilapia (Oreochromis niloticus) to acute ammonia stress monitored by computer vision, J. Zhejiang Univ. Sci. B, 6: 812-816
http://dx.doi.org/10.1631/jzus.2005.B0812

Yi Y., 1998, A bioenergetics growth model for Nile tilapia (Oreochromis niloticus) based on limiting nutrients and fish standing crop in fertilized ponds, Aquacultural Engineering, 18: 157-173
http://dx.doi.org/10.1016/S0144-8609(98)00028-4