Using biopolymer blends for shrimp feedstuff microencapsulation — II: dissolution and floatability kinetics as selection criteria

Using biopolymer blends for shrimp feedstuff microencapsulation — II: dissolution and floatability kinetics as selection criteria

Food Research International 33 (2000) 119±124 www.elsevier.com/locate/foodres Using biopolymer blends for shrimp feedstu€ microencapsulation Ð II: d...

138KB Sizes 0 Downloads 16 Views

Food Research International 33 (2000) 119±124

www.elsevier.com/locate/foodres

Using biopolymer blends for shrimp feedstu€ microencapsulation Ð II: dissolution and ¯oatability kinetics as selection criteria R. Pedroza-Islas a,*, J. Alvarez-RamõÂrez b, E.J. Vernon-Carter b Departamento de IngenierõÂas, Universidad Iberoamericana. Prol. Reforma 880 Lomas de Santa Fe. 01210 MeÂxico, D.F., Mexico Departamento de IngenierõÂa de Procesos e HidraÂulica, Universidad AutoÂnoma Metropolitana-Iztapalapa. 09340 MeÂxico, D.F., Mexico a

b

Received 21 October 1999; accepted 21 December 1999

Abstract In this study, shrimp larvae diets were microencapsulated using as wall materials gum arabic, mesquite gum and maltodextrin at pH values of 4.0 and 8.0 and in a diet-to-wall material ratio of 1:2 and 1:3. The microencapsulated diets were then put in seawater in order to determine their dissolution and ¯oatability rates. The experimental dissolution data followed a ®rst-order kinetics model, whilst the experimental ¯oatability data followed a ®rst-order decay kinetics model. As a result, it was determined that the best microcapsules could be selected by using the characteristic dissolution and ¯oatability parameters as screening criteria, without having to carry out cumbersome bioassays with all the experimental diets. # 2000 Elsevier Science Ltd. All rights reserved. Keywords: Shrimp dissolution kinetics; Floatability kinetics; Selection criteria; Larvae; Microencapsulated diets

1. Introduction One of the main factors for determining the quality of arti®cial diets for crustacean larvae is their behaviour in aqueous systems. Crustacean larvae consume their diets continuously in the bulk of the aqueous media (Jones, Kumlu, Le-Vay & Fletcher, 1997; Jones, Yule & Holland, 1997; Medina, 1998). Therefore, it is important that the feedstu€ available to the larvae exhibit good buoyancy or ¯oatability, show slow sedimentation velocities, and display low dissolution and leaching rates, thereby retaining its integrity and nutritious content as well as being more cost ecient than live diets (Bengston, 1993). It is generally recognised that leaching of nutrients of arti®cial diets can be controlled by preparing them through microencapsulation (ME) or microbonding (MB) techniques (LoÂpez-Alvarado, Langdon, Teshima & Kanazawa, 1994). An important aspect for achieving this is the accurate determination of the bonding or encapsulation agents and the concentration level used (Karel, 1990; Pothakamury & BarbosaCanovas, 1995). * Corresponding author. Tel.: +52-5-267-4089; fax: +52-5-2674254. E-mail address: [email protected] (R. Pedroza-Islas).

The purpose of this work is to characterise the dissolution and ¯oatability behaviour, in a saline aqueous media, of a microencapsulated crustacean larvae diet made from di€erent biopolymer matrices in order to establish crossed-term selection criteria for identifying the best microcapsule treatments. This approach allows for the quick and easy identi®cation of those diets leading to ecient larvae growth and survival rates, without the need for bioassays with all the treatments. It should be stressed that these studies form part of a wider, integral program for evaluating the potential use of polymer microencapsulated diets for crustacean larvae, taking into account other parameters such as microcapsule morphology, particle size (Pedroza-Islas, Vernon-Carter, DuraÂn-Dominguez & Trejo-Martinez, 1999), lipid oxidation barrier properties and shelf life. 2. Material and methods 2.1. Crustacean larvae diet The shrimp larvae diet formulation to be encapsulated contained 60% protein, 7% lipids, 19% carbohydrates, as well as a vitamin and mineral premix (PuelloCruz, 1998; Tacon, 1987) and approximately 4 kcal/g of

0963-9969/00/$ - see front matter # 2000 Elsevier Science Ltd. All rights reserved. PII: S0963-9969(00)00015-6

120

R. Pedroza-Islas et al. / Food Research International 33 (2000) 119±124

digestible energy, as suggested for Litopenaeus schmitti (Gaxiola, 1991). 2.2. Biopolymers for wall material The following wall materials were selected: gum arabic (Acacia senegal) (GA) ``Spray-gum'' from Colloids Naturels (Marseilles, France), mesquite gum (Prosopis juli¯ora) (MG) collected from the Mexican State of San Luis PotosõÂ, Mexico; and maltodextrin with a dextrose equivalent of 10 (MD) (Arancia, S.A. de C.V. MeÂxico). All three biopolymers were combined following a simplex centroid experimental design consisting of three components (Hare, 1974), giving rise to ten treatments (Table 1). 2.3. Microcapsule formation The crustacean larvae diet was encapsulated with the di€erent biopolymer treatments shown in Table 1, at pH values of 4.0 and 8.0 and wall material±feedstu€ ratios of 2:1 and 3:1, respectively, resulting in the four experiments shown in Table 2. A detailed description of the microcapsule formation is given in a previous paper (Pedroza-Islas et al., 1999). 2.4. Microcapsule evaluation 2.4.1. Dissolution Microcapsule dissolution was determined in a sea water aqueous system (30 ppt, pH 8.0). An amount of 0.3 g of each microcapsule treatment was weighed and put into six test tubes, into which 25 ml of seawater was added. The test tubes were then put into a warm water bath at 28‹1 C. This temperature corresponds to that maintained on commercial farms raising crustacean larvae. A test tube of each treatment was withdrawn from the water bath at 15, 30, 60, 120, 180 and 240 min, and then vacuum ®ltered (Whatman No. 40). The ®ltrate

was dried at 60 C until constant weight was achieved (Jayaram & Shetty, 1981). The amount of microcapsule dissolution was calculated based on material balance. All determinations were done in duplicate, using 25 ml seawater as a blank control. 2.4.2. Floatability Two grams of each microcapsule treatment were placed in the surface of 10 ml of seawater contained in a test tube. Immediately afterwards, the test tube was placed in a Bausch and Lomb Spectronic 20 spectrophotometer. Readings were taken every minute until a transmittance of less than 10% at 395 nm was reached. 2.4.3. Dissolution kinetics model Fig. 1 shows typical experimental data of dissolution kinetics. Dissolution of microcapsules is a rather complicated process, consisting, it appears, of a sequence of bonding-layer fracture which generates partial or total leaching of nutrients. The shape of the dissolution Table 2 Characteristics and code number of the four experiments Experiment code number

PH

Wall material ± feedstu€ ratio

E-421 E-431 E-821 E-831

4.0 4.0 8.0 8.0

2:1 3:1 2:1 3:1

Table 1 Experimental design for using biopolymer blends as microcapsule wall material Treatment

Biopolymer

Code number

Gum arabic

Mesquite gum

Maltodextrin

GA100 MG100 MD100 GA50-MG50 GA50-MD50 MG50-MD50 GA33-MG33-MD33 GA66-MG17-MD17 GA17-MG66-MD17 GA17-MG17-MD66

100 0 0 50 50 0 33.33 66 17 17

0 100 0 50 0 50 33.33 17 66 17

0 0 100 0 50 50 33.33 17 17 66

Fig. 1. Typical curve of dissolution kinetics of the microencapsulated diets.

R. Pedroza-Islas et al. / Food Research International 33 (2000) 119±124

121

curves in Fig. 1 demonstrates that the dissolution process is not composed by a single stage; in fact, although the curves increase monotonically, they display several (up to three) in¯exion points. This, in turn may, de®ne the structure of the bonding matrix. One of our objectives in this study is to quantify the dissolution rate, so that it can be used as a parameter to screen microcapsule formulations. As discussed above, despite the complexity of the dissolution process, we will use a rather simple empirical model that retains the main characteristics of the procedure. The dissolution curves show the following behaviour: (a) initial zero dissolution rate, (b) monotonic increase, and (c) steadystate value. These characteristics suggest a ®rst ± order kinetics model (Wagner, 1969): dC ˆ …Cs ÿ C†=d ; C…t ˆ 0† ˆ 0 dt

…1†

where C is the concentration of the solute at time t; Cs is the steady-state (equilibrium) of the solute at the experimental temperature, and d is the dissolution time constant. Notice that Kd ˆ 1=d corresponds to the ®rstorder rate constant. To ®t the experimental dissolution data, we use the integral form of the model (1):   C…t† ˆ Cs 1 ÿ eÿ1=d

…2†

The ®tness parameters are Cs and d . Notice that Eq. (2) is non-linear, therefore, we used standard non-linear, least-square parameter estimation methods (Press, Flannery, Teukolsky & Vetterling, 1986). To start the non-linear estimation procedure, initial guess for Cs and d were read from the experimental data curves. For instance, the initial estimates Cs ˆ 80% and d ˆ 60 min were taken for the experimental data in Fig. 1. Then, using these preliminary estimates, the non-linear estimation procedure begun and iterated until a least square error was obtained. 2.4.4. Floatability kinetics model Fig. 2 shows typical experimental data of ¯oatability kinetics. Due to the failures in the bonding agents, induced by water absorption, microcapsules no longer stay suspended; rather, they can collapse and sediment in the bottom. Of course, microcapsule sedimentation has an adverse e€ect on the e€ective availability of nutrients for larvae growth. The ¯oatability process seems to be simpler than the dissolution process. In fact, the ¯oatability (measured as a percentage of transmittance) starts at 100% and decreases monotonically to zero over long periods of time. A ®rst-order decay rate model is proposed as follows:

Fig. 2. Typical curve of ¯oatability kinetics of the microencapsulated diets.

df ˆ ÿf=f ; f…t ˆ 0† ˆ 100% dt

…3†

where f is the ¯oatability and f is the ¯otation timeconstant. Parameter  can be seen as the mean time a microcapsule is retained in the bulk of the ¯uid before it sediments to the bottom. The integral version of Eq. (3) becomes: f…t† ˆ 100eÿt=f

…4†

This equation was used to ®t the experimental ¯oatability data following the same procedure as in the dissolution kinetic case. The ®tness parameter was the ¯otation time-constant f . 2.5. Statistical analysis The characteristic dissolution and ¯oatability time constants were subjected to an ANOVA, using a complete random model with a factorial arrangement of 10 blends, 2 pH levels and 2 wall material-feedstu€ ratios. 3. Results and discussion 3.1. Dissolution kinetics The dissolution rates of the microcapsules made with the di€erent biopolymer treatments reached a steady-state

122

R. Pedroza-Islas et al. / Food Research International 33 (2000) 119±124

dissolution of 80±90% at 240 min in seawater. Given the similarity of the steady-state dissolution values it was necessary to adopt another parameter as microcapsule selection criterion. The dissolution time constant d varied from 11.60 to 108 min, providing an adequate means for di€erentiating the microcapsule treatments in terms of a characteristic dissolution time. Standard commercial feeding practices for crustacean larvae are to dose the diets 3±5 times daily and to change 50% of the tank water volume per day, in order to eliminate dissolved feed components and larval faeces (Amjad & Jones, 1992; Kurmaly, Jones, Yule & East, 1989; Medina, 1998). However, these feeding practices disregard the nature and stability of the diets as well as the feeding requirements of the larvae, and tend to be related to the personal convenience of the sta€ in charge of feeding (Tacon, 1987). An unstable diet that leaches out rapidly a€ects larval growth and survival, since it not only diminishes diet availability, but also increases water fouling and bacterial production (Amjad & Jones, 1989). One method to avoid this problem is to increase the diet feeding frequency and to diminish the diet dosage. This diculty can also avoided by matching the diet's characteristic dissolution time with larvae feeding habits, utilising this knowledge to establish a feeding program. Larvae, having an underdeveloped digestive system, require a ®nite amount of time for consuming a diet. It is estimated that digestion, from intake to evacuation, occurs from 12 to 20 min. By establishing as 60 min the minimum dissolution characteristic time for selecting the most stable microcapsules, we are providing ample time for the larvae to achieve an adequate intake of the diet with an expected nutrient content. From the initial 40 treatments established by the experimental design, only 14 complied with the screening criteria. The di€erences found in the microcapsules characteristic dissolution times can be related to the nature and con®guration of the biopolymers making up the wall. The longest characteristic dissolution time (108.7 min) was exhibited by treatment GA66-MG17MD17, experiment E-831. At pH 8.0 GA and MG have a fairly extended con®guration due to the ionisation of their glucuronic acid end groups (Glicksman, 1983; Vernon-Carter & Sherman, 1980), whereas MD exhibits a more compact con®guration (see Table 3). Both GA and MG, having a signi®cant amount of protein covalently attached to the main polysaccharide structure (Islam, Phillips, Sljivo, Snowden & Williams, 1997; Vernon-Carter, Pedroza-Islas & Beristain, 1998), tend to adsorb rather strongly at the diet±water interface, projecting the largest portion of their molecules into the aqueous phase. Here, entanglement and interpenetration of adjacent layers takes place, entrapping in their ambit the smaller and more compact MD molecules. Furthermore, new arriving molecules di€using to the interface entangle with the already adsorbed molecules

Table 3 Microcapsule characteristic dissolution times (min) for the di€erent experiments and treatments Treatment code number

GA100 MG100 MD100 GA50-MG50 GA50-MD50 MG50-MD50 GA33-MG33-MD33 GA66-MG17-MD17 GA17-MG66-MD17 GA17-MG17-MD66

Experiment code number E-421

E-431

E-821

E-831

65.78 99.11 51.9 18.0 44.7 21.5 54.3 68.4 76.4 67.8

24.03 24.15 17.6 15.9 14.9 13.5 17.4 23.0 19.9 15.28

47.77 11.60 15.8 21.7 17.2 39.8 33.7 41.9 42.1 84.2

95.97 87.62 65.89 87.6 90.9 67.8 59.3 108.7 47.3 43.37

at the monodies, forming multilayers. The multilayers are held together by interactions arising from hydrogen bonding, protein±protein interactions and the sheer physical entanglement of the biopolymers molecules' moieties. The outcome is a thick, coherent, hydrated ®lm covering the diet. In contrast, treatment GA17-MG17-MD66, experiment E-421, showed the second lowest characteristic dissolution time among the selected microcapsules. Here, both GA and MG, at pH 4.0 exhibit a very contracted, spherical shape, due to the low ionisation degree su€ered. The relatively large concentration and size of the MD molecules, in contrast with those of GA and MG, favours their di€usion into the interface. This gives an interface dominated by MD molecules, which are very poorly adsorbed due to the lack of hydrophobic groups, with some interspersed GA and MG molecules. The GA and MG molecules interact poorly among themselves, since the entanglement of their adjacent layers is limited due to the compactness of their con®guration. Furthermore, as polymer±polymer interactions are favoured over water±polymer interactions the hydration of the adsorbed layer does not achieve its full potential. The result is a rather incoherent ®lm, in which the GA and MG molecules strive to attain interfacial sites, displacing the MD molecules from the interface; furthermore, they are but weakly held in the multilayers due to lack of interaction among the adjacent adsorbed molecules. 3.2. Floatability kinetics As in the case of the dissolution kinetics, it was found that a ¯otation time constant was more useful than the ¯oatability rate in describing the microcapsule behaviour. The ¯otation time constant or characteristic ¯oatability time of the di€erent treatments (Table 4) varied between 1.55 and 11.11 min. It was decided to establish as a discrimination criterion for microcapsule

R. Pedroza-Islas et al. / Food Research International 33 (2000) 119±124

selection the characteristic that ¯oatability time should not be less than 5 min. This time was considered more than sucient for our purposes, as it was arrived at in a static water column, whereas in actual culture practices the aqueous system is gently aerated by 1 cm diameter air bubbles (Kurmaly et al., 1989) which provide extra buoyancy to the microcapsules. This way we insure that the diet is available for consumption for the larvae given their slow metabolism. The treatment showing the highest characteristic ¯oatability time (11.11 min) was GA33-MG33-MD33. As was the case with the explanation given for the characteristic dissolution time for treatment GA66-MG17-MD17, it may be concluded that the diet is enshrouded by thick, coherent, hydrated adsorbed multilayers of the three biopolymers, all intimately intermeshed. The high degree of hydration of the multilayers minimises the density di€erence between the aqueous phase and the dispersed phase, and also has a damping e€ect against sedimentation. On the other hand, treatment GA50-MD50 had a characteristic ¯oatability time of 5.07 min. Here, the GA molecules are quite extended whereas the MD molecules are considerably more compact and of a smaller size, di€using more rapidly to the interface. As the GA molecules arrive at the interface, they displace the MD molecules, which tend to form part of the multilayers. However, the interactions between the GA molecules and the MD molecules are feeble, resulting in more compact, dense multilayers, causing a more rapid precipitation of the microcapsules than that occurring with treatment GA33-MG33-MD33. In this fashion, using both criteria for screening the di€erent microcapsule formulations, only 5 treatments complied with the speci®cations; these, shown in Table 5, all correspond to experiment E-831. The ANOVA analysis of the characteristic dissolution and ¯oatability time constants is reported in Tables 6 and 7, respectively. These results support the methodology for selecting the best microcapsule treatments Table 4 Microcapsule characteristic ¯oatability times (min) for the di€erent experiments and treatments Treatment code number

GA100 MG100 MD100 GA50-MG50 GA50-MD50 MG50-MD50 GA33-MG33-MD33 GA66-MG17-MD17 GA17-MG66-MD17 GA17-MG17-MD66

Experiment code number E-421

E-431

E-821

E-831

4.6 4.8 3.1 6.2 6.7 1.9 2.86 2.85 3.23 3.18

7.3 4.8 2.8 5.7 4.3 5.8 4.9 5.8 6.2 4.6

3.7 4.4 3.0 7.3 3.5 2.3 1.5 2.2 1.9 4.6

7.1 5.7 1.7 4.2 5.1 4.4 11.1 5.9 5.3 3.9

123

via ¯oatability and dissolution time constants, since pH, biopolymer blend and wall material-feedstu€ ratio have signi®cant e€ects on the ®nal characteristics of the treatments. Finally, a bioassay with Litopenaeus vannamei larvae was carried out using a diet complying with our selection criteria (GA66-MG17-MD17), a rejected diet (GA50-MG50, pH 4) and a live control diet consisting Table 5 Selected microcapsules after screening Treatment code number

Experiment Characteristic Characteristic code dissolution time ¯oatability time number (min) (min)

GA100 MG100 GA50-MD50 GA33-MG33-MD33 GA66-MG17-MD17

E-831 E-831 E-831 E-831 E-831

96.0 87.6 90.9 59.3 108.7

7.1 5.7 5.1 11.1 5.9

Table 6 Analysis of variance for dissolution time constant Source of variation Main e€ects Aa Bb Cc Interactions AB AC BC ABC Residual Total (corrected) a b c

d.f.

Mean square

9 1 1

740.9059 6155.4878 3.5955

9 9 1 9 40 79

562.475 983.245 30929.899 788.611 2.8654625

F-ratio

Sig. level

258.564 2148.166 1.255

0.0000 0.0000 0.2693

196.295 343.137 10794.034 275.213

0.0000 0.0000 0.0000 0.0000

A: wall material composition. B: pH. C: wall material±feedstu€ ratio.

Table 7 Analysis of variance for ¯oatability time constant Source of variation Main e€ects Aa Bb Cc Interactions AB AC BC ABC Residual Total (corrected) a b c

d.f.

Mean square

F-ratio

Sig. Level

9 1 1

7.665716 0.320045 53.824805

258.564 2148.166 1.255

0.0000 0.0093 0.0000

9 9 1 9 40 79

2.148823 11.219205 2.760245 4.905223 0.0436250

49.257 257.174 63.272 112.441

0.0000 0.0000 0.0000 0.0000

A: wall material composition. B: pH. C: wall material±feedstu€ ratio.

124

R. Pedroza-Islas et al. / Food Research International 33 (2000) 119±124

of Artemia nauplii, Chaetoceros ceratosporus and Tetraselmis chuii. The following parameters were evaluated: survival percentage, quality index percentage and development index. A comparison of the selected diet and the live diet did not show a signi®cant di€erence in any the parameters; however, the results were signi®cantly di€erent in the case of the rejected diet. A full account of these results, which go beyond the scope of this paper, will be published separately. 4. Conclusions Given the above results, we can conclude that the setting characteristic dissolution and ¯oatability times as selection criteria for shrimp larvae microencapsulated diets allows us to determine the best potential treatments without having to realise cumbersome and timeconsuming bioassays with all the experimental diets. However, this screening procedure must be considered within an integral evaluation of the microencapsulated diets so that these arti®cial diets may contribute e€ectively to lowering costs while matching the nutritional value of live feeds. Acknowledgements The authors wish to extend their appreciation to the Consejo Nacional de Ciencia y TecnologõÂa (CONACyT) for partially ®nancing this research under agreement 25153-B. References Amjad, S., & Jones, D. A. (1989). (Abstract) A comparison of the stability of some arti®cial feeds used in penaeid larval culture. J. World Aquacult. Soc., 20, 12A. Amjad, S., & Jones, D. A. (1992). An evaluation of arti®cial larval diets used in the culture of penaeid shrimp larvae Penaeus monodon (Fabricius). Pakistan J. Zool, 24(2), 135±142. Bengston, D. A. (1993). A comprehensive program for the evaluation of arti®cial diets. J. World Aquacult. Soc., 24(2), 285±293. Gaxiola, C. G. (1991). Requerimientos nutricionales en postlarvas de Penaeus schmitti: relaciones proteõÂna/energõÂa y proteõÂna animal/energõÂa. M.Sc. Thesis. Universidad de la Habana, Cuba. Glicksman, M. (1983). Gum arabic (Gum Acacia). In M. Glicksman, Food hydrocolloids, Vol. II. (pp. 7±29). Boca Raton, FL: CRC Press.

Hare, L. B. (1974). Mixture designs applied to food formulation. Food Technology, 3, 50±72. Islam, A. M., Phillips, G. O., Sljivo, A., Snowden, M. J., & Williams, P. A. (1997). A review of recent developments on the regulatory, structural and functional aspects of gum arabic. Food Hydrocolloids, 11, 493±505. Jayaram, M. G., & Shetty, H. P. C. (1981). Formulation, processing and water stability of two new pelleted ®sh feeds. Aquaculture, 23, 355±359. Jones, D. A., Kumlu, M., Le-Vay, L., & Fletcher, J. (1997a). The digestive physiology of herbivorous, omnivorous and carnivorous crustacean larvae: a review. Aquaculture, 155, 285±295. Jones, D. A., Yule, A. & Holland, D. L. (1997). Larval nutrition. In: D. Conklin, L. R. D'Abramo, & D. M. Akiyama, Crustacean nutrition Adv. World Aquacult. Vol. 6. Baton Rouge, LA: World Aquaculture Soc. Karel, M. (1990). Encapsulation and controlled release of food components. In: Biotechnology and food process engineering. Schwartezberg, H. G and Rao, M. A. Eds. IFT Basic Symposium Series. Marcel Dekker, USA. p. 277±293. Kurmaly, K., Jones, D. A., Yule, A. B., & East, J. (1989). Comparative analysis of the growth and survival of Penaeus monodon (Fabricius) larvae, from protozoea 1 to Postlarva 1, on live feeds, arti®cial diets and on combination of both. Aquaculture, 81, 27±45. LoÂpez-Alvarado, J., Langdon, C. J., Teshima, S., & Kanazawa, A. (1994). E€ects of coating and encapsulation of crystalline amino acids on leaching in larval feeds. Aquaculture, 122, 335±346. Medina, R. C. (1998). DisenÄo y evaluacioÂn de un alimento nebulizado para la larvicultura de camaroÂn blanco, Litopenaeus vannamei. M. Sc. thesis. Centro de InvestigacioÂn Cientõ®ca y de EducacioÂn Superior de Ensenada. Baja California. Mexico. Pedroza-Islas, R., Vernon-Carter, E. J., DuraÂn-DomõÂnguez, C., & Trejo-MartõÂnez, S. (1999). Using biopolymer blends for shrimp feedstu€ microencapsulation. I. Particle size, morphology and microstructure of microcapsules. Food Research International, 32, 367±374. Pothakamury, U. R., & Barbosa-Canovas, G. V. (1995). Fundamental aspects of controlled release in foods. Trends Food Sci. Technol, 6, 397±406. Press, W. P., Flannery, B. P., Teukolsky, S. A., & Vetterling, W. T. (1986). Numerical recipies: the art of science computing. Cambridge: Cambridge University Press. Puello-Cruz, A. C. (1998). Requerimientos nutricionales de las larvas de camaroÂn. Notas del curso Internacional de las larvas de camaroÂn. ASA, USB, UANL, CIAD-MazatlaÂn. Tacon, A. G. J. (1987). The nutrition and feeding of farmed ®sh and shrimp. A training manual. FAO Field Doc. 2/E.GCP/RLA/075/ITA. Vernon-Carter, E. J., & Sherman, P. (1980). Rheological properties and applications of mesquite tree (Prosopis juli¯ora) gum. 1. Rheological properties of aqueous mesquite gum solutions. J. Texture Studies, 11, 339±350. Vernon-Carter, E. J., Pedroza-Islas, R., & BeristaõÂn, C. I. (1998). Stability of Capsicum annuum oleoresin-in-water emulsions containing Prosopis and Acacia gums. J. Texture Studies, 29, 553±567. Wagner, J. G. (1969). Interpretation of percent dissolved-time plots derived from in vitro testing of conventional tablets and capsules. J. Pharm. Sci., 58, 1253±1257.