Fish species have evolved in environments where H2O temperature mostly fluctuates between hours of the twenty-four hours, whereas the majority of cognition on their thermic biological science originates from research lab experiments at changeless temperatures. Hence, it is unsure whether fish acquire a benefit or incur a punishment when exposed to temperature fluctuations. Here, we measured the growing of Nile Tilapia Oreochromis niloticus L. ( 20mg-20g ) fed in little surplus during the hours of light, following their exposure to assorted thermic governments fluctuating around the thermic optimum for growing ( TA°opt = 30A°C ) , i.e. 27A°L:33A°N, 28.5A°L:31.5A°N, 30A°L:30A°N, 31.5A°L:28.5A°N and 33A°L:27A°N ( two reproductions per intervention, six raising hebdomads, growing controls at hebdomadal intervals ) . For each thermic government, a log-log theoretical account was constructed between the fish moisture organic structure mass ( WM ) and the specific growing rate ( SGR, % WM day-1 ) , i.e. log SGR = a + B log WM. The a and b coefficients of the five theoretical accounts were equated to the day-to-day thermic amplitude ( from -6A°C to +6A°C ) . For both parametric quantities, the ensuing theoretical accounts were 3rd order multinomials with a parabolic form. These theoretical accounts ( R2 & gt ; 0.99 ) indicated that juvenile O. niloticus benefit from day-to-day oscillations around TA°opt until a pivot size of circa 160 milligram, above which hovering temperatures impact negatively on their growing. Slower growing under fluctuating temperatures is accompanied by greater size dispersion. Functional hypotheses on how temperature fluctuations affect fish growing are discussed, together with their deductions in natural home grounds and aquaculture systems with different contexts of nutrient handiness. Methodological facets, particularly the necessity of comparing the pivot temperature of fluctuating governments to TA°opt are besides debated.
Key-words fluctuating thermic governments, size heterogeneousness, optimal temperature, Oreochromis niloticus
Temperature is presumptively the most directional environmental factor ( sensu Fry 1971 ) in heterothermal animate beings as it closely governs their physiology, growing, distribution country, home ground usage and behavior. Proportionally, aquatic animate beings can be more affected by alterations in temperature than their tellurian opposite numbers, because H2O has a greater specific heat than air. The bulk of animate being species have evolved under temperatures that fluctuate between old ages and seasons, but besides during the day-to-day rhythm. The amplitudes of day-to-day thermic fluctuations vary between latitudes, climes and seasons, but besides between home grounds, as they are reciprocally relative to depth and H2O current. Similarly, the huge bulk of aquaculture takes topographic point in Waterss that are non thermoregulated and undergo more or less marked day-to-day fluctuations. Quite paradoxically, while thermic fluctuations prevail in the wild and in most civilization systems, the majority of cognition on the thermic biological science of aquatic heterotherms originates from laboratory surveies under invariable or about changeless temperatures. Without proper proof, it is unsure whether consequences obtained at changeless temperatures can be extrapolated to fluctuating environments ( see Crawshaw, 1977 ) .
Previous comparings between the effects of stable and fluctuating thermic governments on fish have produced contrastive decisions. Some surveies demonstrated that the exposure to fluctuating thermic governments enhanced the fish tolerance to high temperatures ( mosquitofish Gambusia affinis ( Baird & A ; Girard ) , Otto, 1974 ; sheepshead minnow Cyprinodon variegatus variegatus Lacepede, Bennett & A ; Beittinger, 1997 ; cutthroat trout Oncorhynchus clarkii clarkii ( Richardson ) , Johnstone & A ; Rahel, 2003 ; spikedace Meda fulgida Girard, Carveth et al. , 2007 ) , whereas no difference was found in others ( Currie, 1995 ) . In several cases, day-to-day fluctuations were found to hike growing ( rainbow trout Oncorhynchus mykiss ( Walbaum ) , Hokanson et al. , 1977 ; striped bass Morone saxatilis ( Walbaum ) , Cox & A ; Coutant, 1981 ; brown trout Salmo trutta Linnaeus, Spigarelli et al. , 1982 ; silver salmon salmon Oncorhynchus kisutch ( Walbaum ) and goldfish Carassius auratus ( Linnaeus ) , Konstantinov et al. , 1990 ; spikedace, Carveth et al. , 2007 ) . In other surveies, fluctuating thermic governments were reported to bring forth slower growing or lower growing efficiencies than changeless temperatures ( Tahoe chump Catostomus tahoensis Gill & A ; Jordan, Vondracek et al. , 1982 ; Arctic charr Salvelinus alpinus ( Linnaeus ) , Lyytikainen & A ; Jobling, 1998, 1999 ; brown trout, Flodmark et al. , 2004 ) . Finally, in other species, no important consequence was documented ( Lahontan cutthroat trout, Dickerson & A ; Vinyard, 1999 ; Nipponese medaka Oryzias latipes ( Temminck & A ; Schlegel ) , Dhillon & A ; Fox, 2007 ; bluish Tilapia Oreochromis aureus, Steindachner Baras et al. , 2000 ) .
Surely, portion of this fluctuation originated from echt differences between species or populations. On the other manus, the protocols that were used for proving the effects of day-to-day thermic fluctuations varied well. In some cases, fluctuating governments were compared to changeless temperatures matching to the upper and lower values of the fluctuation. In other cases, the changeless temperature that served as control stood at mid of the fluctuating thermic scope. Furthermore, the thermic governments under survey were seldom equated with the thermic optimum for growing ( hereafter TA°opt ) , which besides corresponds to the concluding thermal preferendum, i.e. the temperature at which fish placed in a thermic gradient finally congregate, if given clip ( Jobling, 1981 ) . During the growth, the ratio between gill surface and organic structure volume varies allometrically, so TA°opt is size-dependent in most fish species ( synthesis in Jobling, 1994 ) , and in general little fish prefer ( McCauley & A ; Huggins, 1979 ; Baras & A ; Nindaba, 1999 ; Hernandez et al. , 2002 ) and execute better at warmer temperatures than larger conspecifics ( Pedersen & A ; Jobling, 1989 ; Imsland et al. , 1996, 2006 ) . Cichlids are no exclusion to this general regulation ( Oreochromis mossambicus Peters, Mironova, 1976 ; Nile Tilapia O. niloticus ( Linnaeus ) , Melard, 1986 ; bluish Tilapia, Baras et al. , 2002 ) . The comprehensiveness of the growth-to-temperature response besides varies during the growth ( e.g. rock loach Barbatula barbatula ( Linnaeus ) , Elliott et al. , 1996 ; bluish Tilapia, Baras et al. , 2002 ) . Hence, the consequence of fluctuating temperatures on growing is besides expected to change during the growth of fishes, but grounds is enormously missing.
This survey aimed to prove the effects of changeless and hovering thermic governments on the endurance, growing and size heterogeneousness of the Nile Tilapia Oreochromis niloticus, and whether these effects were size-dependent ( size scope of 20 milligrams to 20 g ) . In reverberation to the aforesaid paragraph, the pivot temperature for the hovering governments was set every bit close as possible to TA°opt. By definition, TA°opt is the temperature that produces the fastest growing, but besides the value around which fluctuations in temperature usually do the smallest fluctuations of fish growing, unless the fluctuation itself produces an intrinsic positive or damaging consequence.
Material and methods
The fish were Oreochromis niloticus Maryut, from a confined population, which originated from the Marine Centre of Tajoura ( Lybia ) , and was transferred in 1999 to the Aquaculture Research Station of the Tunisian National Institute for Marine Sciences and Technologies at Bechima Gabes ( Turki & A ; KraA?em, 2002 ) . In each experiment of this survey, the fish were full siblings that were collected as hatchlings from the oral cavity of a mouthbrooding female. Water temperature in the broodfish armored combat vehicle, and rise uping installations prior to the experiments was 30A±1A°C, i.e. the really same mean temperature as during the experimental periods.
The survey aimed at proving the effects of controlled day-to-day oscillations of H2O temperature on the endurance, growing, and size heterogeneousness of immature Nile of different ages and sizes ( from circa 20 milligram to circa 20 g ) . Two day-to-day thermal amplitudes ( 3 and 6A°C ) were evaluated and compared with a control, where temperature was maintained every bit changeless as technically possible. In two fluctuating thermic governments, temperature was warmer during the twenty-four hours than during the dark, as it usually is in the natural state. In add-on to these “ natural ” governments, we besides evaluated two mirror interventions with the really same day-to-day amplitudes ( 3 or 6A°C ) , but where H2O temperature was warmer during the dark than during the twenty-four hours ( Figure 1 ) . This design enabled proving whether the consequence of day-to-day thermic amplitude was dependent on the yoke or decoupling of visible radiation and temperature. It besides permitted proving whether the variables under examination were dependent on the temperature at the clip of eating, since fish were fed entirely during the twenty-four hours in this survey ( see Rearing conditions ) . For all five thermic governments, two replicate groups were used. In the remainder of the text, each thermic intervention is designated as XA°L: YA°N, where Ten and Y are the temperatures during the hours of visible radiation ( L ) and dark ( N ) , severally.
The average day-to-day temperature in all groups was indistinguishable and set as 30A°C, which corresponds to the thermic optimum for growing ( TA°opt ) in immature O. niloticus Maryut ( Azaza et al. , 2008 ) . The maximal thermic amplitude that was evaluated here was 6A°C ( i.e. 27-33A°C ) , a value that is often observed in natural environments, every bit good as in civilization pools in tropical parts. No amplitude greater than 6A°C was evaluated, basically because it was of import that the maximal day-to-day temperature did non transcend the temperature at which nutrient consumption is maximum ( a few grades above TA°opt ; Brett, 1979 ; Jobling, 1997 ) , otherwise feeding diminutions steeply and affects growing straight, every bit good as indirectly, due to the debasement of H2O quality that ensues from the decay of uneaten nutrient. The temperature at which nutrient consumption is maximum is presently unknown for the Maryut population, but in position of old surveies on the thermic biological science of this population ( Azaza et al. , 2008 ) , it is surely non exceeded at 33A°C for fish runing from 20 milligrams to 20 g.
The periods of thermic passage were synchronised with the light rhythm and were programmed to last for two hours each ( 06:00-08:00 H and 18:00-20:00 H ) . This agenda permitted maintaining H2O temperature every bit changeless as possible during the 10-hour eating period ( 08:00-18:00 H ) . At first sight, thermic passages every bit steep as 3A°C per hr might look overdone, but they do non transcend the celerity of thermic passages of natural baby’s room home grounds on cheery yearss ( e.g. Baras & A ; Nindaba, 1999 ; Finlay et al. , 2000 ) .
This experimental design was implemented in two experiments ( continuance: 3 hebdomads each ) with fish of indistinguishable beginnings but of different sizes and ages. The first experiment started at mid of the first hebdomad of exogenic eating, after fish had to the full exhausted their yolk ( average moisture organic structure mass [ WM ] A± SD = 22.3 A± 2.3 milligram, average lopsidedness of -0.337, groups of 100 fish each ) . The 2nd experiment started with fish about 100 times larger ( average WM A±A SD = 2.27 A± 0.17 g, average lopsidedness of-0.046, groups of 30 fish each ) , which were about the same size as at the terminal of the first experiment. Wholly, these two experiments enabled proving the effects of fluctuating thermic governments over three orders of magnitude ( 1000:1 ratio ) for organic structure mass, which represents more than 50 % of the species size scope when plotting organic structure mass on a logarithmic graduated table ( i.e. Nile Tilapia seldom exceed 2 kilograms ) . In both experiments, fish were sedated with tricaine methanesulfonate ( 50 ppm ) , weighed separately to bring forth groups where average size and size heterogeneousness were every bit similar as possible. Thereafter, the groups of fish were indiscriminately allocated to the different thermic governments.
The five thermic governments were evaluated in five indoor H2O recirculating systems in the raising installations of the Aquaculture Research Station of Bechima Gabes ( Tunisia ) . Each system comprised two 85-L rise uping fish tank ( 35 x 80 ten 45 [ H ] centimeter ) , a 180-L reservoir armored combat vehicle for filtration and thermoregulation ( with a 2-kW thermostatic submergence warmer ) , and a 20-L biofilter with UV-sterilization lamp ( Oase, theoretical account Filtoclear UWC 9/11W ) . The H2O flow in the fish tank was 1.5-2.0 L min-1, and 2.0-4.5 L min-1 ( get down to stop of experiment ) , in the first and 2nd experiments, severally. Auxiliary aeration was provided by single air rocks to keep the O degree every bit near as possible to impregnation. Water pH and alkalinity averaged 7.5 and 105 milligrams CaCO3 L-1, severally. Day length was maintained at 12L:12D, with visible radiations on from 08:00 to 20:00 H ( light strength of 800 Lx at the surface of the H2O, as measured with a Digital Instrument LX-101 ) .
In the systems with a fluctuating thermic government, the scene of the thermoregulator was changed at 06:00 and 18:00 H in order to achieve the dedicated temperatures at 08:00 and 20:00 H, and therefore to suit the thermic government to the light rhythm ( Figure 1 ) . The experiments were conducted during wintertime ( November 2007 and January 2008 ) , when the air temperature inside the raising installation was cool plenty to enable the return of the temperature to the day-to-day lower limit within 2 hours. However, every twenty-four hours, it was verified that the fluctuations followed the desired forms, and, for illustration, little sums of cool H2O was added to chill down the system within clip on somewhat warmer yearss.
Fish were fed formulated provender, which were distributed with automatic feeders from 08:00 to 18:00 H, during the period when visible radiations were on and H2O temperature was changeless. The limitation of the feeding period to the hours of visible radiation was based on the common sense hypothesis that O. niloticus is a diurnal species, that feeds basically during the twenty-four hours ( even though juveniles can every bit profit from night-time eating ; Baras et al. , 1995 ) . This limitation was farther motivated by cut downing every bit much as possible the hazard of H2O debasement during the dark if, for any ground, fish consumed less nutrient than expected in a peculiar intervention. Food composing stood as 42 % protein, 8 % fat and 18.9 kJ.g-1 gross energy for the first experiment on little juveniles, and 36 % protein, 6 % fat and 17.7 kJ.g-1 gross energy for the 2nd experiment on larger fish. These composings were found to bring forth fast growing in O. niloticus Maryut ( Azaza et al. , 2005 ) . The diameter of pressed pellets was less than 0.25 millimeter, and 1.0-1.5 millimeter for the first and 2nd experiments, severally. Food rations were calculated after Melard ( 1986 ) in order that fish be fed in little surplus throughout. They were incremented every twenty-four hours, partially based from computations, partially from the observation of the sum of the uneaten nutrient at the terminal of each eating period. Uneaten nutrient was removed by syphoning during the hr following the terminal of nutrient distribution, ever with the aim of forestalling any debasement of H2O quality that might hold interfered with the echt consequence of day-to-day thermic fluctuations.
Aquariums were searched for dead fish twice a twenty-four hours, before and after the period of nutrient distribution. Water temperature in the reservoir armored combat vehicle of each recirculating system was recorded automatically every hr, and dissolved O twice a twenty-four hours ( 08:00 and 18:00 H ) with a digital thermo-oxymeter ( WTW, MIQ/C184, truth of 0.1 A°C and 0.1 mg O2.L-1 ) . Entire ammonium and nitrite concentrations were measured with a spectrophotometer on the yearss of fish measuring ( D8, D15 and D22 ) . Thereafter, H2O was added to counterbalance for vaporization during the past raising hebdomad, so the values of H2O quality that were measured here are somewhat pessimistic since they were measured on the twenty-four hours when the H2O volume in the recirculating system was lowest.
Each experiment lasted for three hebdomads, with hebdomadal controls. On the forenoon of the control yearss, fish were captured with a dipnet, sedated with 50 ppm tricaine methanesulfonate, weighed ( nearest 0.1 milligram and nearest 0.01 g for the first and 2nd experiment, severally ) , and returned to their fish tank. The exclusive difference between the two experiments resided in that all subsisters ( n a‰¤ 30 ) were measured at each control in the 2nd experiment, whereas in the first experiment, 30 fish were indiscriminately sampled from each fish tank. Food distribution on the control twenty-four hours was suspended during the forenoon, and resumed in the early afternoon, at least 3 hours after the last fish were controlled, in order to understate the effects of managing on nutrient consumption.
Survival rates were compared between groups with chi-square analysis. The consequence of the thermic government on growing was tested with one-way analyses of discrepancy ( ANOVA ) and Scheffe post-hoc trials for comparings of agencies. Kruskal-Wallis and Mann-Whitney U-tests were used wheresoever parametric analyses could non be applied ( between-treatment comparings of coefficient of fluctuation of wet organic structure mass, lopsidedness coefficients, lower limit and maximal organic structure mass within each group of fish ) . For each raising hebdomad, the mean specific growing rate ( SGR ) was calculated as SGR ( % WM d-1 ) = 100 ( Ln WM2 – Ln WM1 ) ( t2-t1 ) -1, where WM2 and WM1 are the average moisture organic structure mass at times t2 and t1, severally. Here, clip was expressed by mention to the figure of feeding yearss, therefore 7 yearss for the first hebdomad of each experiment, and 6.5 yearss for the following two hebdomads of each experiment. For each thermic government, a SGR-to-WM theoretical account was produced by simple additive arrested development analysis ( after a logarithmic transmutation of the two variables ) , utilizing the information from both experiments ( 6 hebdomads and two groups per thermic government ) . Null hypotheses were rejected at P & lt ; 0.05.
In both experiments, the experimental scenes were good respected. Fish size at the start of the experiments did non differ significantly between groups ( average WM of 22.1 to 22.4 milligram, and 2.25 to 2.30 g ) , nor did the coefficient of fluctuation of the distributions of moisture organic structure mass ( CV WM of 10.0 to 11.3 % , and 6.1 to 8.4 % ) and the lopsidedness coefficients ( from -0.60 to -0.06, and -0.38 to +0.41 ) . Temperature ne’er deviated from the experimental marks by more than 0.05A°C on norm, and by more than 0.3A°C duly, and thermic passages ever took topographic point within 2.0 A±0.2 hours ( Table I ) . Similarly, O degrees ne’er dropped below 70 % impregnation, and the concentration of nitrogen-bearing compounds remained really low in all interventions, during each hebdomad of the two experiments ( Table I ) . In both experiments, endurance on D22 was high and did non differ significantly between replicates or interventions ( i.e. 92-95 % , and 90.0-93.3 % in the first and 2nd experiments ; chi-square of 1.372 and 1.087, P=0.9980 and P=0.9992, severally ) .
At the terminal of the first experiment ( D22 ) , fish size was relative to the temperature during the hours of visible radiation, when nutrient was distributed. However, between-treatment differences were tenuous ( average WM of 1637 to 1802 milligram ) and non important, due to strong size heterogeneousness within groups ( average CV WM of 31.1 % ; Fig. 2 ) . Kruskal Wallis trials revealed no important intervention consequence on the CV WM ( H=5.24 ) or skewness coefficients ( H=1.29 ) , and on the sizes of the smallest ( WMmin ; H=4.58 ) and largest persons ( WMmax ; H=0.19 ) . No individual important difference between replicates or interventions was observed for any of the aforesaid variables on D8 and D15 either. This was partially due to the fact that the CV WM soared in all groups during the first raising hebdomad ( on norm from 10.6 to 32.3 % ) , and remained high thenceforth. Skewness besides increased in all groups during the experiment, but basically during the 2nd and 3rd raising hebdomads ( average lopsidedness of -0.35, -0.24, -0.03 and +0.55, on D1, D8, D15 and D22, severally ) .
Size heterogeneousness and lopsidedness besides increased quickly during the 2nd experiment, as CV WM passed from an norm of 7.4 to 25.1 % , and lopsidedness from -0.03 to +0.40 during the first raising hebdomad. At this minute ( D8 ) , there was no important difference between groups as respects fish size ( Figure 3 ) and size dispersion ( CV WM [ Kruskal Wallis trial, H=0.27 ] , lopsidedness [ H=1.64 ] , Wmmin [ H=1.19 ] and Wmmax [ H=0.96 ] ) . By contrast, at the terminal of the 2nd raising hebdomad ( D15 ) , fish raised under strongly fluctuating temperatures ( 33A°L:27A°N and 27A°L:33A°N ) were significantly smaller than those raised at changeless or somewhat oscillatory temperatures. These differences amplified during the 3rd raising hebdomad, and on D22 fish raised under 27A°L:33A°N and 33A°L:27A°N averaged no more than 11.1 and 12.8 g, while the others averaged 15.4-17.7 g ( Fig. 3 ) . Between-treatment differences were besides conspicuous for the smallest and largest persons within each group, thereby bespeaking that all size categories were negatively affected by pronounced thermic oscillations ( i.e. WMmin of 8.00-8.31 g versus 8.62-10.43 g ; Kruskal Wallis trial, H=6.87, p=0.0322 ; WMmax of 20.30-23.08 versus 23.08-24.70 g ; Kruskal Wallis trial, H=6.12, p=0.0468 ) . The CV WM and lopsidedness of size distributions on D22 were significantly higher when fish were raised under strongly fluctuating temperatures than under more stable thermic governments ( CV WM of 24.0-28.1 % versus 20.5-23.6 % ; Kruskal Wallis trial, H=6.55 ; p=0.0379 ; lopsidedness of 0.55 to +2.25 versus -0.29 to +0.62 ; Kruskal Wallis trial, H=7.28 ; p=0.0262 ) .
SGR-to-WM relationships were calculated for all five thermic governments, and compared in order to find the advantage or punishment for a fish of a peculiar size to be exposed to chair or strong day-to-day thermic oscillations ( Table II ) . The greater the day-to-day thermic amplitude, the higher the intercept and the steeper the incline of the log-log relationship. The intercepts and inclines of these growing theoretical accounts were equated with the amplitude of the day-to-day thermic oscillations, and for both parametric quantities, consistent parabolic relationships were found ( Figure 4 ) . These theoretical accounts clearly indicate that little juveniles exposed to temperatures hovering about TA°opt acquire a growing advantage over those raised at changeless TA°opt. However, because of the greater abruptness of the inclines under fluctuating temperatures, this advantage vanishes quickly and above a pivot size of circa 160 milligram, fish exposed to hovering temperatures incur a growing punishment, the badness of which increases with fish size ( and with the amplitude of the thermic oscillation ) .
In both experiments and for all thermic governments, the presence of little sums of wasted nutrient in each armored combat vehicle at the terminal of all rampant yearss indicated that fish had so been fed in surplus throughout. Despite fish were fed in surplus, the concentrations of nitrogen-bearing compounds remained low, and extremely satisfactory by mention to the recommendations for rise uping Nile Tilapia ( Ballarin & A ; Hatton, 1979 ) . Furthermore, the measurings of H2O quality were done on a hebdomadal footing, prior to adding clean H2O, so they give a pessimistic position of the existent H2O quality during each rise uping hebdomad. On all raising yearss, the O degrees at the start of the feeding period were close to impregnation, and they ne’er dropped values that were found to impact nutrient consumption or nutrient transition in O. niloticus ( 40-50 % impregnation ; C. Melard, O. Plunus & A ; E. Baras, unpublished information ) . The adequateness of the experimental conditions is farther attested by the low fish mortality in all experimental groups ( & lt ; 0.5 % d-1 ) , and by the fact that growing was fast, by mention to old surveies on the Maryut population ( Azaza et al. , 2008 ) , but besides by mention to other populations of O. niloticus raised in similar experimental conditions ( e.g. Manzala population, Melard, 1986 ) . The average diurnal or nocturnal temperatures ne’er deviated from the mark temperatures by more than 0.05 A°C, and the inclines of thermic passages did non go from the targeted values by more than 0.03A°C h-1. Wholly, these observations support the thought that the consequences documented in this survey reflected the intrinsic effects of temperature oscillations on the growing of O. niloticus, instead than disagreements between experimental armored combat vehicles, or side effects ensuing from treatment-dependent changes of H2O quality. The similarity between the consequences observed under mirror governments of indistinguishable day-to-day thermal amplitudes suggests that the yoke between visible radiation and temperature has no major importance on the growing of O. niloticus. A similar decision was drawn in a survey on O. aureus where mirror regimes fluctuating between 28 and 35A°C were evaluated ( Baras et al. , 2000 ) , thereby bespeaking that this characteristic is shared by several fish species, at least among the tilapiines, but this is non systematical in fish ( see the contrasting responses of Theragra chalcogramma and Anoploma fimbria, Sogard & A ; Olla, 1998 ) .
Yet, two peculiar facets of the experimental design might be pointed out as possible artifacts. At first, the fish used in this survey had been raised at about changeless temperatures around TA°opt prior to the experiments. It can non be ruled out that the absence of acclimatization to fluctuating temperatures might hold been responsible for the slower growing under strong day-to-day thermic oscillations. However, if the deficiency of acclimatization had been the key to the differences observed in this survey, so the differences would hold been expected to surge during the first rise uping hebdomad of the experiment, which was non the instance in any of the two experiments of this survey. It can non be purely ruled out that the deficiency of acclimatization to fluctuating temperatures impacted on fish growing, but if it did, the consequence was undistinguished or highly transeunt.
The 2nd possible artifact refers to the fact that nutrient was distributed entirely during the hours of visible radiation during this survey. The absence of night-time eating was calculated and aimed at continuing H2O quality, the debasement of which might hold impacted on fish growing to a greater extent than the variable under survey, particularly in a context where fish were fed in little surplus. A direct effect of this design was that fish from different interventions were fed at different temperatures ( i.e. 27, 28.5, 30, 31.5 and 33A°C ) . It is ill-famed that the growing of fish raised at changeless temperatures is slower if the temperature departs from TA°opt ( e.g. Jobling, 1994 ) . Hence, before debating the possible effects of temperature fluctuations on fish growing, it was necessary proving whether the growing rates observed at the five thermic governments under survey was similar to those that would hold been expected under changeless 27, 28.5, 30, 31.5 and 33A°C. Expected growing were modelled from the survey by Azaza et Al. ( 2008 ) who raised O. niloticus Maryut under changeless temperatures runing from 22 to 34A°C. Get downing from the natural information of this survey, it was possible bring forthing a general theoretical account of growing ( SGR, % d-1 ) against moisture organic structure mass ( WM, g ) and H2O temperature ( TA° , A°C ) for fish fed in little surplus, i.e. Log ( 1+SGR ) = -18.65+20.17 Log TA° – 3.16 ( Log TA° ) 3 – 0.15 ( Log WM ) ( Log TA° ) 2. In order to ease the comparings between the survey by Azaza et Al. ( 2008 ) and this survey, all informations were expressed by mention to a growing at changeless 30A°C ( i.e. the exclusive experimental conditions that were common to the two surveies ) . Fish growing at changeless 30A°C in the survey by Azaza et Al. ( 2008 ) was somewhat slower than here, so expected growing rates were adjusted to allow straightforward comparings. The anticipations of this theoretical account are shown in Figure 5, together with the values observed under temperatures fluctuating about 30A°C in this survey. In both experiments of this survey, angle raised under strongly fluctuating thermic governments attained sizes that departed well from those that would hold been expected if their growing had been entirely governed by the temperature during the period of nutrient distribution.
This analysis further emphasizes that the growing of O. niloticus is less influenced by the temperature at the clip of feeding than by the mean day-to-day temperature ( which was indistinguishable in all interventions here ) and by the amplitude of the day-to-day thermic oscillation. It is likely that this inclination is non restricted to O. niloticus. The determination that the mean day-to-day temperature has a prevailing consequence on growing ( notwithstanding the echt consequence of thermic fluctuations ) might lend to account for the contrasting decisions of surveies where thermic governments hovering between two utmost temperatures were compared to constant governments at these two extreme temperatures ( Figure 6 ) . With such protocols, it can be predicted on a conceptual footing that the growing under hovering temperatures would be intermediate if the two utmost temperatures were either below or above TA°opt ( Figure 6.C and 6.F ) . By contrast, the growing under the oscillating government would be faster than under the two changeless temperatures if these stood apart from TA°opt ( Figure 6.I ) . If TA°opt stood in between one changeless temperature and the pivot temperature of the oscillating government, so the growing at the oscillating government would similar to the aforementioned changeless temperature, and faster than at the other changeless government ( Figure 6.L ) . This functional account does non overlook the possibility that fluctuating governments have a genuine ( positive or negative ) impact on growing. However, it suggests that this consequence might be masked or amplified depending on the thermic scope under survey and its place by mention to TA°opt. Figure 6 provides, on a conceptual footing, illustrations of such possible confusions between negative and impersonal effects ( Figure 6.C versus 6.G ) , or between impersonal and positive effects of day-to-day oscillations ( Figure 6.E versus 6.L ) . In some instances, positive and negative echt effects of thermic oscillations might even bring forth growing public presentations that look most similar ( Figure 6.E versus 6.P ) . A practical illustration of such possible confusion can be found in Baras et Al. ( 2000 ) , where juvenile O. aureus ( 12-1500 milligram ) raised under 35A°L:28A°N or 28A°L:35A°N grew at a similar rate as under changeless 35A°C. At first sight, this suggests a impersonal consequence of hovering temperatures on the growing of this species. However, the average TA°opt over the 12-1500 milligram size scope in this species was subsequently found to average circa 31.5A°C ( Baras et al. , 2002 ) , therefore the value of the pivot temperature of the hovering governments in the survey by Baras et Al. ( 2000 ) . If the day-to-day thermic fluctuation had no impact on the growing of O. aureus, fish raised under thermic governments hovering around 31.5A°C should hold grown faster than at changeless 35A°C, whereas their growing was somewhat slower. All in all, the growing punishment for populating under a 7A°C day-to-day oscillation amounted to no less than 13.5 % in this species, but this punishment passed unnoticed before the temperatures under survey were equated to TA°opt.
Until this survey, the variable effects of fluctuating temperatures had been basically attributed to the extent of the amplitude, i.e. with moderate amplitudes holding a positive consequence, whereas higher amplitudes had a negative consequence on growing ( Konstantinov et al. , 1987, 1990 ; Meeuwig et al. , 2004 ; Dong & A ; Dong, 2006 ) . The conceptual attack illustrated in Figure 6 every bit good as this illustration from the writers ‘ informations clearly indicate that a dependable statement on the positive, impersonal or negative consequence of thermic oscillations can barely be derived from a straightforward comparing between growing rates under hovering and changeless extreme temperatures. Similarly, any protocol proving the effects of fluctuating thermic governments of different amplitudes around a peculiar pivot temperature can bring forth contrasting responses depending on whether the pivot temperature is colder, warmer or shut to TA°opt, as already expounded by Jobling ( 1997 ) . These statements strengthen the protocol that was developed in this survey, where the pivot temperature was selected every bit near as possible to the TA°opt of the size scope under survey. They besides emphasize the trouble of pulling meaningful comparings with other surveies on fluctuating temperatures where growing at the changeless pivot temperature was non evaluated or equated with TA°opt.
Size-dependent effects of thermic oscillations
This survey demonstrated that the effects of thermic oscillations around TA°opt were size-dependent. The modeling attack ( Figure 4 ) indicated that juvenile O. niloticus benefit from hovering governments until a pivot size of 160 milligrams, above which hovering temperatures start impacting negatively on their growing. The absence of important between-treatment differences during the first experiment of this survey can be accounted for by the fact that the geometric average size of fish during this experiment was close to the pivot size, whereas in the 2nd experiment, when fish were above the pivot size, growing was significantly dependent on the amplitude of day-to-day thermic fluctuations. The parabolic form of the theoretical accounts between the day-to-day amplitude of the thermic oscillation and the intercept or incline of the growing theoretical accounts is consistent with the well admitted parabolic nature of the relationships between temperature and growing in fish ( Jobling, 1994 ) .
As a affair of fact, no between-treatment difference was observed at the terminal of the first rise uping hebdomad of the 2nd experiment, despite fish were 10 times larger than the pivot size at the start of this experiment. It is suggested that the consequence of fluctuating temperatures during this period was masked by the colony of laterality hierarchies. It is frequent that laterality hierarchies settle or be reinstalled shortly after fish have been rearranged into groups, particularly when group size is low and fish are homogeneous in size, as was the instance in this survey. The colony of a laterality hierarchy is by and large accompanied by heterogenous nutrient consumption and growing, by and large to the hurt of the smallest fish, thereby ensuing in slower growing, greater size dispersion and increased lopsidedness. All three features were observed at the terminal of the first hebdomad of the 2nd experiment. Thereafter, size heterogeneousness remained about stable and lopsidedness increased somewhat, and growing rebounded, as it is often the instance after the hierarchy has settled. However, size heterogeneousness and lopsidedness were higher, and growing was slower among the groups that were raised under strongly fluctuating temperatures.
A functional account behind the greater size dispersion under fluctuating temperatures, lies in that fish exposed to fluctuating governments spend a greater portion of the day-to-day rhythm at temperatures that are farther from TA°opt. There is now a turning figure of groundss that size dispersion is lower among fish raised at temperatures near to TA°opt than at colder or heater temperatures ( for O. niloticus, see Azaza et al. , 2008 ) . The functional hypothesis behind this phenomenon invokes the possibility that TA°opt varies between persons, for illustration if the gill surface varies between fishes of indistinguishable sizes. If this hypothesis is valid, size heterogeneousness is expected to increase if the raising temperature departs from TA°opt. By virtuousness of the parabolic form of the relationship between growing and temperature, size heterogeneousness is expected to lift quicker if the raising temperature is warmer than TA°opt, than if it were cooler than TA°opt ( E. Baras & A ; M. Daffe , personal communicating ) . A similar account might use for thermic governments fluctuating about TA°opt, since the going of the raising temperature from TA°opt additions with the amplitude of the day-to-day thermic fluctuation. Additionally, it can be put frontward that non all fish of a peculiar size every bit tolerate strong day-to-day thermic fluctuations. Both hypotheses remain to be tested by experimentation, but in both instances, a formal presentation would necessitate cloning protocols. In add-on or instead to these physiological factors, it can non be excluded either that fluctuating temperatures impacted on the behavioral registry, and exacerbated laterality hierarchies through enhanced aggressiveness. Recent experiments on two Tilapia species ( O. niloticus and Sarotherodon melanotheron ) so provided grounds that aggressiveness was higher, and could bring forth higher mortality when fish were raised at warm temperatures ( E. Baras & A ; M.S. Azaza, unpublished information ) .
The grounds for why the consequence of thermic oscillations shifted from a benefit to a disadvantage during the growth of O. niloticus, remain vague as good. However, it should be reminded that the thermic scope under rating in this survey remained changeless ( 27-33A°C ) , whereas TA°opt and the thermic comfort scope likely shifted towards somewhat cooler temperatures as fish grew bigger, by analogy to surveies on other cichlids. For illustration, TA°opt in Oreochromis aureus was found to diminish by circa 1A°C for each 10-fold addition of the moisture organic structure mass ( Baras et al. , 2002 ) . Hence, the warmest temperatures, which were presumptively closer to the TA°opt of little juveniles, became progressively unsuitable, and potentially nerve-racking to angle of increasing size. It can non be excluded either that oscillations of indistinguishable amplitudes increasingly became progressively unsuitable to angle of increasing size.
Stress parametric quantities were non measured in this survey, so we can non claim that fish exposed to greater day-to-day thermic amplitudes were more stressed than others. However, it is deserving retrieving that disconnected thermic dazes induce emphasis in fish ( Reaves et al. , 1968 ; Bevelhimer & A ; Bennett, 2000 ) . Furthermore, a recent survey on the sea Cucumis sativus Apostichopus japonicus ( Echinodermata ) provided grounds that persons exposed to strongly fluctuating temperatures around TA°opt ( 18A°C ) exhibited a lower hexokinase activity but a higher ace oxide dismutase activity, which suggests an oxidative emphasis ( Dong et al. , 2008 ) . Dong et Al. ( op. cit. ) besides reported a higher degree of heat-shock proteins ( HSP70 ) among the animate beings that were temporarily exposed to the warmest temperatures, thereby proposing a greater degree of protein harm that might hold accounted for their slower growing. There is presently really small counterpart information on fish raised under fluctuating temperatures ( salmonids: Thomas et al. , 1986 ; Shrimpton et al. , 2007 ) , so it can merely be speculated that these mechanisms might hold intervened here.
About independently from the physiological mechanisms that govern the tolerance of fish of assorted sizes to H2O temperature, analogues can be drawn between this experimental survey and field surveies on the ontogenetic fluctuations of home ground usage by fish species. It is frequent that immature, little fishes with low swimming capacities but high TA°opt live in shallow home grounds with slow H2O speed, which undergo strong day-to-day thermic fluctuations on cheery yearss. By contrast, larger fishes tend to busy deeper or faster fluxing home grounds, which are thermally more stable. A series of factors, including predation force per unit area, ontogenetic fluctuations of O conveyance and thermic preferenda have been invoked to account for these size-dependent fluctuations in home ground usage and day-to-day migrations ( e.g. Baras & A ; Nindaba, 1999 ) . In position of the consequences of this survey, which show that little fish tolerate thermic fluctuations better than larger / older conspecifics, it can non be excluded either that the hunt for environments that are more thermally stable be an extra factor that fosters the displacement towards deeper and faster fluxing home grounds, which provide such features.
Synthesis, decisions, positions
This survey did non open up the field of fluctuating thermic governments, but it provided significant grounds, on a conceptual footing, that no sound decision can be drawn on whether changeless or fluctuating temperatures impact on fish growing unless the temperatures under survey have been equated with the thermic optimum for growing. It besides highlighted that the life in thermally stable or fluctuating environments can hold a cost or convey a benefit, depending on fish size and age, and that the growing benefit brought approximately by environments that offer a greater thermic stableness might be among the keys that force fish of increasing size to deeper and faster fluxing home grounds. The theoretical accounts provided in this survey likely echo to a figure of fish species, at least as respects their general nature ( i.e. immature fish being more tolerant than big fish to tag day-to-day thermic fluctuations ) . However, informations on how fluctuating temperatures impact on fish of different sizes are still enormously missing for other fish species, from other taxa or latitudes.
As respects aquaculture, the information provided in this survey indicates clearly that the growing observed under about changeless temperatures can non be extrapolated to state of affairss where fish would undergo pronounced day-to-day fluctuations. It should besides be reminded that fish the differences observed in this survey referred to a context where fish were fed in little surplus. If the life under fluctuating temperatures is more energy-demanding than under more stable thermic governments ( see LyytkaA?nen & A ; Jobling, 1998 ) , the growing punishment might be more terrible in existent raising conditions, where the eating degree by and large lies in between the optimum and the maximal nutrient rations to avoid nutrient wastage. Future surveies should take at look intoing this issue in contexts of different nutrient handiness. In add-on to their involvement for aquaculture, these surveies would besides lend to document the several advantages or drawbacks of life in extremely fluctuating thermic environments with contrasting trophic degrees.
This survey benefited from the bilateral coaction between Tunisia ( INSTM ) and France ( IRD ) , under the protections of the Gallic Ministry of Foreign Affairs. We want to thank Mrs S. Kalboussi and K. Elebdelli for proficient aid throughout the period of this survey. Thankss are besides given to M. Dominique Caseau for look intoing the English.