A comparison of the effects of three water-circulation regimes on the aquaculture of bullfrog (Rana catesbeiana Shaw) tadpoles

A comparison of the effects of three water-circulation regimes on the aquaculture of bullfrog (Rana catesbeiana Shaw) tadpoles

Aquacultuve ELS EV 1ER 105-114 A comparison of the effects of three watercirculation regimes on the aquaculture of bullfrog (Rana catesbeiana Shaw) ...

788KB Sizes 0 Downloads 85 Views

Aquacultuve ELS EV 1ER


A comparison of the effects of three watercirculation regimes on the aquaculture of bullfrog (Rana catesbeiana Shaw) tadpoles Alejandro Flores-Nava”, Miguel A. Olvera-Novoa, Eucario Gasca-Leyva Centro de Investigacih y de Estudios Avanzados de1 IPN (CINVESTAV), Unidad Mhida, Apdo. Postal 73-CORLIEMEX, 97310 M&iah, Yucatdn, Mexico Accepted 2 July 1994

Abstract A 15-week experiment was conducted with bullfrog (Runa catesbeiana) tadpoles to determine the effects of three different water management strategies in rearing tanks: static, a daily exchange ( 1OO%/ day) and a recirculating system (900% exchange rate/day). Survival, growth rate and feeding performance of tadpoles stocked at 3.5 individuals/l were compared among the three treatments. The static system gave the best results in terms of survival (57.9%)) percent total weight gain ( 1803%)) daily weight gain (6.919 mg/day) and SGR (2.34%/day), all of which were statistically greater than the corresponding values from the other treatments. The lowest performance was found in tadpoles reared in the recirculating system. These differences are attributed to a supplementary source of nourishment provided by microalgae covering the bottom and walls of rearing tanks of the static system. In recirculating systems, flushing of feed could have been an important component contributing to inadequate nourishment. Keywords: Rana catesbiana; Water management: Growth / amphibians

1. Introduction Water quality, along with adequate nutrition, is recognized as one of the most important requisites for successful cultivation of any aquatic species (Alabaster and Lloyd, 1982; Boyd, 1990). In an aquacultural environment (i.e., a pond or a tank), water quality is expressed as a combination of several biotic and abiotic factors, which in turn are related * Corresponding


0044-8486/94/$07.00 0 1994 Elsevier Science B.V. All rights reserved SSDIOO44-8486(94)00176-6


A. Flares-Nava

et al. /Aquaculture

128 (1994) 105-114

to the intensity of culture and the management strategy employed. It is well documented that the more intensive an aquaculture system, the faster metabolites accumulate in the water, and therefore, the faster the rate of water turnover that should be maintained to provide a healthy environment for the cultured organisms (Muir, 1982). However, there are some species whose environmental requirements do not fit this generalization; this is the case with tadpoles of the bullfrog (Rana catesbeiuna) , a commercially important species. Wild frog tadpoles of many anuran species live in lentic waters, where the environmental conditions are often unstable. In captivity, tadpoles of the several species of the genus Runu, including R. cutesbeiunu, have been commonly cultured under semi-static, controlled conditions, with variable results (Cairns et al., 1967; Mohanty and Dash, 1986; Culley, 1991; Flores-Nava et al., 1992). The present study was carried out to test the performance of premetamorphic tadpoles of R. cutesbeiuna under three water management regimes by measuring rates of body growth and metamorphic change.

2. Materials and methods Experimental procedure Runu cutesbeiunu tadpoles were collected from a natural spawn in Morelos State, Central Mexico, placed in plastic bags filled with oxygen and transported to the aquaculture facilities of CINVESTAV in Merida, Yucatan, Southeast Mexico, where they were incubated. After a 2-week acclimation period, tadpoles in stage 25 (Gosner, 1960) were randomly distributed into nine indoor fiberglass tanks, each 1.0 X 1.0 X 0.55 m (water column= 20 cm; volume = 100 l), at a stocking rate of 3.5/l, using 24-h aerated tap water (chlorine before aeration= 0.18 ppm, total hardness = 365 mg/l as CaCO,). Experimental tanks were installed in a building with a translucent fiberglass roof, kept at ambient temperature and under natural photoperiod ( 13L: 11D). The experimental design consisted of three treatments, each with three replicates. Treatment 1 (T-l) consisted of a static system, where water was added to the tanks only as needed to replace loss from evaporation and syphoning (used for cleaning the tank bottom, see below) ; this varied from 3 to 5% of the total volume per day. Treatment 2 (T-2) involved virtually 100% replacement of the water, accomplished by syphoning out to the last few mm of water and then adding aerated fresh water. All siphoning, for cleaning (T1), or water exchange (T-2) was carried out daily at noon, so that the tadpoles in T-l and T-2 were subjected to a regular handling stress at midday, in order to stimulate metamorphosis as suggested by Horseman et al. ( 1976). The three tanks in treatment 3 (T-3) were part of a recirculating system that included settling tanks, biological filter, header tank and pump tank; the water flow was such that water in the system was exchanged at 900% per day. Tanks under T-l and T-2 were constantly aerated. Following the recommended feeding regime of Lopes-Lima and Agostinho ( 1988), the tadpoles were fed daily at a rate of 13% body weight with a moist dough prepared using the formula shown in Table 1. Feeding was carried out early in the morning placing the dough directly at the bottom of the tank in the center (the dough was stable for 3 + 0.5 h).

A. Flares-Nava et al. /Aquaculture

I28 (1994) 105-114


Table 1 Formulation and proximal composition of the moist dough used for tadpole feeding Ingredients (8 as fed) Whole powdered milk Boiled egg yolk Boiled potatoes Vitamin premix’

Composition (%) 15 25 58 2

Moisture Crude protein Ether extract Crude fibre Ash NFE

64.00 2.30 4.53 0.66 1.35 27.16

1 Tacon et al. (1983).

To provide protein of plant origin (Fontanello et al., 1985, 1988; Culley, 1991)) each tank was supplied daily at midday after water siphoning in T- 1 and T-2, with 1.5 liters of a multispecies microalgal culture from an outdoor tilapia tank. Combined phytoplankton density in the tanks fluctuated between 10 and 34 X lo4 cells/ml (Flores-Nava et al., 1992). Aeration and water circulation were stopped in all treatments for approximately 1 h at feeding time to allow for better grazing and food consumption by tadpoles. Every day, unconsumed food and feces were removed from the tanks by syphoning the bottom. Water quality parameters were monitored throughout the experiment. Dissolved oxygen and temperature (YSI 54) and pH (Pe Unicam 9409) were measured daily, and total ammonia and nitrite were determined every week following the methods recommended by APHA-AWWA ( 1985). Every 2 weeks, all the tadpoles from each tank were removed carefully with a dipnet which was passed over a paper towel to remove excess of water, and bulk weighed to the nearest 0.01 g and counted, adjusting the quantity of food offered accordingly. All tanks were inspected daily for metamorphosed frogs, which were recorded and removed.

Statistical methods The results, in terms of growth, survival, food conversion ratio, and duration of metamorphosis period, were analyzed statistically using a one-way analysis of variance. Mean differences between treatments were tested for significance (P
3. Results For experimental purposes, the metamorphosis of R. catesbeiana was divided into 15 2week periods, totaling 210 days. In order to avoid the stress and skin damage resulting from handling, weighing was suspended after the ninth period, when differences in growth rates had reached a significant level. Fig. 1 shows the average growth curves of bullfrog tadpoles for each of the three treatments. There was no significant difference in body weight over the first five sample periods (70 days), after which tadpoles in the static system increased their growth rate significantly


A. Flares-Nava et al. / Aquaculture 128 (1994) 105-l I4

SAMPLE (two-week periods) Fig. 1. Mean growth curves for the tadpoles subjected standard deviation.

to different water-circulation

regimes. Vertical bars are

compared to T-2 and T-3. Covariance analysis showed a significant difference between means and slopes for the treatment curves after nine sampling intervals. Fig. 2 shows the cumulative frequency (%) of metamorphosed frogs per week, over the 22-week period. The first metamorphosed individuals were collected from both the static




10 11 12 13 14 15 16 17 18 19 20 21 22 23

TIME (weeks) Fig. 2. Cumulative

percent frequency

(%) of metamorphosed

frogs during the experiment,

expressed per week.

A. Flares-Nava

et al. /Aquaculture

128 (1994) 105-114


Table 2 Survival, mean growth and feed utilization of bullfrog tadpoles under three different water-management regimes over 105-day experimental period Mean values’

Survival (%) Initial body weight (g) Final body weight (g) Weight gain ( % ) Weight gain (mg/day)3 Specific growth rate ( %/day)4 Feed intake (mg/day)5 Food conversion ratio6

Treatment T-l



-I se.’

59.90” 0.049” 0.920a 1803’ 6.919” 2.34” 28.127” 4.08a

51.24a 0.04P 0.70+ 1446ab 5.229”b 2.17” 20.672” 3.95”

52.51” 0.052a 0.563b 988b 4.063b 1.87b 22.474” 5.84”

5.93 0.00 0.08 152.01 0.61 0.09 2.51 0.65

’ Values with the same superscripts are not significantly different (P < 0.05).

’ Standard error, calculated from the mean-square for error of the ANOVA. 3 WC (mg/day) = (C fortnightly individual WG) /time (days). 4 SGR (%/day) = lOO(log, final body wt-log, initial body wt) /time (days). a FI (mg/day) = lOOO(Cfortnightly individual food fed)/time (days). 6 FCR = feed offered (wet weight)/wet tadpole weight gain.

and the recirculated systems 56 days after stocking. After this, T- 1 showed a consistent and significantly greater number of metamorphosed frogs compared with both T-2 and T-3. A maximum was observed between days 140 and 154 after stocking (weeks 20 to 23), a period during which more than 50% of the tadpoles inthis treatment metamorphosed. T-3 yielded the lowest number of metamorphosed organisms, which accounted for less than 16% of the tadpoles reared in the treatment (Fig. 3). Although not significant (Duncan’s multiple range test), overall survival (Table 2) appeared highest in T- 1 (59.9%)) intermediate in T-2 (57.2%) and lowest in T-3 (52.6%). Tadpoles in T-l showed the highest values in final body weight, weight gain (% and mg/ day), and specific growth rate (SGR), which were all significantly different from corresponding values in T-3 (Table 2). As water-exchange rate increased, there was a-tendency for growth and metamorphic rate to decrease. Tadpoles in the lOO%/day water-renewed system showed intermediate values. Apparent food conversion ratio (feed offered/weight gained, on a wet-weight basis) was higher in tadpoles reared under T-3 (5.84:l.O) than those from the static and the lOO%/ day water-exchanged systems (4.08 and 3.95:1.0, respectively), but the Galues were not statistically different among the treatments. Water quality parameters in the different treatments showed similar values during the experimental period (Table 3). There were no differences in mean midday temperature between treatments; temperatures varied between 26 and 30°C with an overall mean ( f s.e.m.) of 27.81+ 0.08”C. Minimum dissolved oxygen concentrations were observed in the static system toward the end of the experiment, with a more-oscillating trend throughout the experimental period. T-2 and T-3 showed a more stable pattern with maximum values observed in T-3. Mean pH values were lowest in T-2 and highest in T-3 during the experimental period.

A. Flares-Nava


et al. /Aquaculture 128 (1994) 105-114

T-3 -


~..___ -___I




PERCENTAGE Fig. 3. Relative percentages for each treatment. Table 3 Water-quality means)


of metamorphosed



(ME), unmetamorphosed

during the experimental

(UM) and dead (DE) tadpoles obtained

period (values are means + standard error of the





Temperature ( “C) Dissolved oxygen (mg/l) PH Total ammonia (mgll) Nitrite (mg/l)

27.81 kO.08 5.10*0.37 8.15-Lr0.16 1.223 + 0.79 0.805 k 0.11

27.81 iO.08 5.3OkO.21 8.04 k 0.09 0.967 k 0.75 0.145 * 0.05

27.8lkO.08 6.45 + 0.31 8.34*0.15 0.129 f 0.02 0.277 AO.09

Total ammonia (TAN) showed similar trends among treatments. After an initial peak in both T- 1 and T-2, levels of TAN were low in all the treatments although slightly higher in the static system. There were no direct toxic effects detected in the treatments. Nitrite concentrations were also higher in T-l; the lowest mean concentration was registered for T-2, which was slightly lower than that for T-3.

4. Discussion The present work began with the hypothesis that, under similar environmental conditions, better water quality would improve the growth and development of bullfrog tadpoles. However, it is clear from the results that the best water quality (0,, TAN, nitrate) was recorded in T-3, but the best growth and development were in T- 1, with the less desirable

A. Flares-Nava etal. /Aquaculture

128 (1994) 105-114


water quality. Lowest mortality and highest growth rates were observed in the static system, while the reverse was true in the constant recirculation system (Fig. 3). Several factors determine the growth performance of tadpoles. First it is important to point out that the stocking density used in this trial was relatively high as compared with other studies with Rana catesbeiana tadpoles (Lopes-Lima and Agostinho, 1988, 1992; Culley, 199 1) . When tadpoles are reared in conditions of more than one per liter, there may be a negative effect of density (Just0 et al., 1985; Fontanello et al., 1988; Garcia and Banco, 1992). Although stocking density of 3.5 animals per liter is routinely used at CINVESTAV, the culture conditions typically employed are different from those described in the present work, especially regarding nutritional aspects (Flores-Nava et al., 1992). In the present study, T-3 tanks were supplied with water under pressure via a perforated PVC pipe, which, as in many commercial operations, is used to provide aeration through stirring. However, in nature, bullfrog tadpoles occur in lentic, non-disturbed waters. Circulating water could adversely affect their performance, for example, by requiring higher metabolic energy to swim and maintain their position in the water column. Water flow in the culture system is one of the main factors to be controlled, including flow velocity, associated noise, and type of agitation or stirring. Another important consideration is feeding efficiency. The food conversion ratio did not differ significantly between treatments. However, SGR and weight gain per day were significantly higher in tadpoles reared under static conditions. In addition to the relationship between energy expenditure and water circulation, the tadpoles in T-2 and T-3 had less access to food, since both dough particles (with attached bacteria), and microalgae added to the tanks were flushed out with the water exchange. In T-l unconsumed feed particles remain on the bottom together with feces, probably creating an important source of protein. Anuran tadpoles use several alternative feeding strategies, including filter-feeding (Wassersug, 1970; Seale and Beckvar, 1980; Viertel, 1992), grazing (Dickman, 1968) and coprophagy (Steinwascher, 1978). In culture conditions, tadpoles rely mostly on supplementary food, but depending on the management practice, they may also filter-feed and practice coprophagy, especially if the supplementary food does not meet their nutritional requirements. Culley ( 1991) states that growth and development is enhanced in bullfrog tadpoles that have access to aquatic microflora, in addition to supplemental diets. Even though all tadpoles in this trial were provided with a dense volume of microalgae, the l-h break during which both aeration and recirculation were stopped to allow for better feeding, may not have been enough for the tadpoles to make use of the algae, which were then flushed out when circulation was restarted. In T- 1 a microflora proliferated, covering the bottom and walls of the culture tanks, thus providing an additional and more balanced food source than was available in T-2 and T-3 (see Table 4). Evidently water flow in T-2 and T-3 tanks did not provide a suitable environment for settling and proliferation of microalgae. It has been reported that tadpoles are better suspension feeders when they are in the early metamorphic phase, and as they approach stages 42-46 of Gosner they become surface grazers and tend to rely more on detritus (Seale and Beckvar, 1980). At the beginning of the experiment, the addition of phytoplankton to the tanks apparently supplemented the nutritional requirements of the tadpoles. When they approached more advanced stages they


A. Flares-Nava

et al. /Aqua&hue

128 (1994) 105-114

Table 4 Presence of benthic microalgae species on the bottoms of rearing tanks under the different water-exchange regimes (*scarce; **abundaqt;.***very abundant) Species


Cosmarium sp. Fragilaria sp. Navicula sp. Nitichia sp. Scenedesmus sp. Chlorella sp.

*** ** *** * * *



* *

* *

relied more on the bottom than other surfaces, and the benthic flora and detritus became more important. Therefore, those reared in the static system would have access to more fed than those in the more dynamic water systems. With respect to the length of the metamorphic period, in the static system the peak number of new froglets was observed between days 154 and 165 after stocking. This is an unusually long interval, especially for tropical environments. Culley ( 1991) reports 10 weeks as the normal time for metamorphosis for bullfrog tadpoles under experimental conditions in temperate areas. Lopes-Lima and Agostinho ( 1992) state that under normal experimental conditions in Brazil (ambient temperature above 26°C)) the tadpole phase for Ram catesbeiana lasts between 90 and 120 days. Flores-Nava et al. ( 1992) used the same stocking density as in the present trial, in a static system ( < 3% water exchange per day) and a 40% protein diet, in the same laboratory in Merida, and they found that 75% of the tadpoles had metamorphosed by day 88 after hatching. The abnormally long period before metamorphosis is evidently related to nutrition. The feed provided in the present trial corresponds to that commonly used in Mexico for maintaining bullfrog tadpoles in earthen fishponds. This diet provides only 10% of required dietary proteins, which, including the amino acid profile, may not be adequate. Chim and GalIassini ( 1988) observed a direct relationship between the protein level in the diet and the growth rate and development of bullfrog tadpoles. Additionally, Fontanello et al. ( 1988) and De Souza et al.’ (1988) found that balanced diets with 50% animal and 50% plant proteins resulted’in better growth rates of bullfrog tadpoles than an entire animal protein source. This has also been reported by Mohanty and Dash ( 1986) in culture systems that lack natural food. In tropical environments, bullfrog tadpoles normally reach metamorphosis when they are relatively small in size and weight. In the present study, stage 46 of Gosner was reached at ari average mass.of 3.5 g, resulting in new froglets with a’mean weight of 1.5 g. Similar results are reported from Cuba (Garcia and Banco, 1992)) and Mexico (Flores-Nava et al., 1992). This could be due to a physiological response to tropical conditions, resulting in an acceleration of metabolism. Flares-Nava (in prep.)’ found that newly metamorphosed froglets weigh between < 1 g to 5 g, when they are fed a 40% protein diet and have access to dense blooms of microalgae. In general terms, water quality parameters remained,at acceljtable levels except possibly for ,me values of total ammonia and nitrite in the static tanks. Unfortunately no data were found in the literature about the toxicity of these nitrogen compounds to ranid tadpoles.

A. Flows-Nava et al. /Aquaculture 128 (1994) 105-114


However, as aminotelic organisms, tadpoles have very effective mechanisms to eliminate this metabolic product (Houillon, 1975). Moreover, the performance of tadpoles in T-l did not seem to have been affected by the high concentrations of these compounds, suggesting that they may be somewhat tolerant of these toxic metabolites. It can be concluded that water exchange affects the condition of tadpoles, with a direct effect in keeping the benthic flora from establishing and food particles from settling, thus restricting the availability of food resources. It is recommended to f&her compare static and water-exchange systems, but to provide a high-protein diet.

Acknowledgements We would like to acknowledge the invaluable help of Mrs. Elizabeth Real in water chemistry analyses and Mrs. Fanny Merino for her support in microalgae identification. Special thanks are due to Dr. John Frazier and Dr. Bryan M. Davies for their suggestions and correction of the manuscript. The present work was supported by CONACyT-MEXICO research grant 1746-A.

References Alabaster, J.S. and Lloyd, R., 1982. Water Quality Criteria for Freshwater Fish. Butterworth, London,. 361 pp. APHA-AWWA, 1985. Standard Methods for the Examination of Water and Wastewater, 16th ed. American Public Health Association, Washington, DC, 1268 pp. Boyd, C.E., 1990. Water Quality in Ponds for Aquaculture. Alabama Agricultural Experiment Station, Auburn University, AL, 482 pp. Cairns, A.M., Bock, J.W. and Bock, F.G., 1967. Leopard frogs raised in partially controlled conditions. Nature, 213: 191-193. Chim, F.M.R. and Gallassini, F.G., 1988. Varia$es no ganho de peso de girinos de r?&ouro-gigante (Rana catesbeiana Shaw, 1802) alimentados corn ravoes de diferentes niveis proteicos (20, 30 e 40% PB). Anais do 6”Encontro National de Ranicultura, 7-12 de agosto de 1988, Rio de Janeiro. [email protected] dos Ranicultores de Rio de Janeiro, Brazil, pp. 125-132. Culley, D.D., 1991. Bullfrog culture. In: C.E. Nash (Editor), Production of Aquatic Animals. World Animal Science, Vol. C4. Elsevier, Amsterdam, pp. 185-205. De Souza, M.E., Agostinho, C.A., Lopes-Lima, S. and Nogneira, F.C., 1988. Ensaios de alimentago II: influ&cia da terra, proteina de origem animal ou vegetal no desempenho da r%touro, Rana catesbeiana, Shaw 1802. Anais do 6” Encontro National de Ranicultura, 7-12 de agosto de 1988, Rio de Janeiro. [email protected] dos Ranicultores de Rio de Janeiro, Brazil, pp. 197-200. Dickman, M., 1968. The effect of grazing by tadpoles on the structure of a periphyton community. Ecology, 49(6): 1188-1190. Duncan, D.B., 1955. Multiple range and multipleF-tests. Biometrics, 11: 142. Flores-Nava, A., Gasca, E. and Gil, E., 1992. Resultados preliminares de1 cultivo intensivo de mna toro Rum catesbeianu en el CINVESTAV-M&ida, Mexico. An. VII Encontro National dos Ranicultores, 6-9 Avril, Rio de Janeiro. [email protected] dos Ranicultores de Rio de Janeiro, pp. 71-83. Fontanello, D., Arruda, S.H., Mandelli Jr., J., Justo, C.L., Penteado. L.A. and De Campos, B.E.S., 1985. Influencia de la proteina de origem animal e vegetal no desenvolvimento ponderal de girinos de Rana cafesbeiana Shaw, 1802, criados em ranario experimental. Bol. Inst. Pesca, 12(2): 43-47. Fontanello, D., Arruda, S.H., Mandelli Jr., J., De Campos, B.E.S., Penteado, L.A. and Prandi, D.M., 1988. Developmental control of bullfrog tadpole (Rum catesbeiunu Shaw, 1802) for commercial cultivation: population density, protein quality in the ration and localization. Bol. Inst. Pesca, 15( 1): 19-24.


A. Flores-Nava etal. /Aquaculture 128 (1994) 105-114

Garcia, C. and Banco, M., 1992. El ranario Nina Bonita, una experiencia para el cultivo intensivo de la rana toro Rana catesbeiunn en Cuba. An. VIIEncontro National dos Ranicultores, 6-9 Avril, Rio de Janeiro. [email protected] dos Ranicultores de Rio de Janeiro, pp. 84-90. Gosner, K.L., 1960. A simplified table for staging amtran embryos and larvae with notes on identification. Herpetologica, 16: 183-190. Horseman, N.D., Meier, A.H. and Culley, D.D., Jr., 1976. Daily variations in the effects of disturbance on growth, fattening, and metamorphosis in the bullfrog (Ram cutesbeiana) tadpole. J. Exp. Zool., 198: 353-358. Houillon, C., 1975. Embriologfa. Ediciones Omega, S.A., Barcelona, 183 pp. Justo, CL., Penteado, L.A., Fontanello, D., Soares, H.A., Mandelli, H. and Campos, B.E., 1985. Ganho de peso de girinos de Rana catesbeiuna Shaw, 1802 em [email protected] intensiva sob diferentes densidades populacionais. Bol. Inst. PescaBrasil, 12(3): 31-37. Lopes-Lima, S. and Agostinho, C.A., 1988. A [email protected] de Ras. Publicaciones Globo Rural, Rio de Janeiro, Brazil, 187 pp. Lopes-Lima, S. and Agostinho, C.A., 1992. A Tecnologia de Criacao de Ras. Universidade Federal de Viposa, Imprenta Universitaria, Vicosa, Brazil, 168 pp. Mohanty, S.N. and Dash, M.C., 1986. Effects of diet and aeration on the growth and metamorphosis of Ram tigrina tadpoles. Aquaculture, 51: 89-96. Muir, J.F., 1982. Recirculated water systems in aquaculture. In: J.F. Muir and R.J. Roberts (Editors), Recent Advances in Aquaculture, Vol. 1. Westview Press, Boulder, CO, pp. 357-446. Scale, D.B. and Beckvar, N., 1980. The comparative ability of anuran larvae (Genera: Hyla, Bufo and Runa) to ingest suspended blue-green algae. Copeia, 3: 495-503. Sokal, R.R. and Rohlf, F.J., 1979. Biometrla. H. Blume Ediciones, Madrid, 832 pp. Steinwascher, K., 1978. The effect of coprophagy on the growth of Ram cutesbeiana tadpoles. Copeia, 1978( 1): 130-134. Tacon, A.G.J., Stafford, E.A. and Edwards, C.A., 1983. A preliminary investigation of the nutritive value of three terrestrial lumbricid worms for rainbow trout. Aquaculture, 35: 187-199. Viertel, B., 1992. Functional response of suspension feeding anuran larvae to different particle sizes at low concentrations (Amphibia). Hydrobiologia, 234: 151-173. Wassersug, R., 1970. The mechanism of ultraplanktonic entrapment in amuan larvae. J. Morphol., 137: 279-288. Zar, J.H., 1984. Biostatistical Analysis, 2nd ed. Prentice-Hall Inc., Englewood Cliffs, NJ, 718 pp.