RUMINAL DIGESTION AND CHEMICAL COMPOSITION OF NEW GENOTYPES OF BUFFELGRASS (Cenchrus ciliaris L.) UNDER IRRIGATION AND FERTILIZATION

Guillermo Juan García-Dessommes, Roque Gonzalo Ramírez-Lozano, Rocío Morales-Rodríguez and Graciela García-Díaz

Guillermo Juan García-Dessommes. Doctor in Food Sciences, Instituto Nacional de Investigaciones Forestales, Agrícolas y Pecuarias (INIFAP). Researcher, INIFAP, Mexico. Address: INIFIAP, Apartado Postal 3, General Terán, N. L: CP 67400. México. e-mail: garcia.guillermo@inifap.gob.mx

Roque Gonzalo Ramírez-Lozano. Ph.D. in Ruminant Nutrition. Professor-Researcher, Universidad Autónoma de Nuevo León (UANL), Mexico. e-mail: roqramir@fcb.uanl.mx

Rocío Morales Rodríguez. Doctoral Candidate, UANL, Mexico.

Graciela García-Díaz. Doctor in Biotechnology, UANL, Mexico. en Ciencias en Biotecnología, UANL, Mexico.

SUMMARY

The study was conducted with the aim of evaluating and comparing the total dry matter production (TDMP), chemical composition and effective degradability of dry matter (EDDM), crude protein (EDCP) and neutral detergent fiber (EDNDF) of the Nueces hybrid and five new genotypes (PI 1, PI 2, PI 3, PI 4, PI 5) of buffelgrass. Grasses were irrigated and fertilized with 100kg·ha^-1 of urea-N, and hand harvested on June 5, 2001 at Nuevo Leon, Mexico. All grasses were established on a completely randomized design with three replicates. The TDMP was significantly different among genotypes. The crude protein content and cell wall components (cellulose, hemicellulose, and lignin) were significantly different among grasses. Moreover, EDDM, EDCP, and EDNDF were significantly different among genotypes. The P, Na, Cu and Zn contents, in all grasses, were insufficient to meet growth requirements of beef cattle. Data of dry matter production and nutritional dynamics suggest that the new genotypes PI 3 and PI 4 could be considered good sources of nutrients for grazing ruminants in Northeastern Mexico.

DIGESTIÓN RUMINAL Y COMPOSICIÓN QUÍMICA DE NUEVOS GENOTIPOS DE PASTO BUFFEL (Cenchrus ciliaris L.) BAJO IRRIGACIÓN Y FERTILIZACIÓN

El estudio se llevó a cabo con el objetivo de evaluar y comparar el contenido nutritivo y degradabilidad efectiva de la materia seca (DEMS), proteína cruda (DEPC) y pared celular (DEFDN) del híbrido Nueces y cinco nuevas líneas (PI 1, PI 2, PI 3, PI 4, PI 5) de pasto buffel en el noreste de México. Todos los pastos se establecieron usando un diseño completamente al azar con tres repeticiones. Fueron regados y fertilizados con 100kg·ha^-1 de úrea-N y cosechados manualmente el 5 de junio de 2000 en Nuevo León, México. La producción de materia seca fue significativamente diferente entre los pastos. El contenido de proteína cruda, pared celular y sus componentes (celulosa, hemicelulosa y lignina) fueron significativamente diferentes entre los pastos evaluados. Asimismo, DEMS, DEPC y DEFDN fueron significativamente diferentes entre pastos. Los contenidos de P, Na, Cu y Zn no fueron suficientes para satisfacer los requerimientos del ganado de carne en crecimiento. Datos de producción de materia seca y dinámica nutricional sugieren que los nuevos genotipos PI 3 y PI 4 pueden ser considerados como buenas fuentes de nutrientes para el ganado en pastoreo en regiones del noreste de México.

DIGESTÃO RUMINAL E COMPOSIÇÃO QUÍMICA DE NOVOS GENOTIPOS DE CAPIM BUFFEL (Cenchrus ciliaris L.) SOB IRRIGAÇÃO E FERTILIZAÇÃO

RESUMO

O estudo se realizou com o objetivo de avaliar e comparar o conteúdo nutritivo e degradabilidade efetiva da matéria seca (DEMS), proteína crua (DEPC) e parede celular (DEFDN) do híbrido Nozes e cinco novas linhas (PI 1, PI 2, PI 3, PI 4, PI 5) de capim buffel no noroeste do México. Todos os capins se estabeleceram usando um desenho completamente aleatório com três repetições. Foram regados e fertilizados com 100 kg·ha-1 de uréia-N e recolhidos manualmente em 5 de junho de 2000 em Nuevo León, México. A produção de matéria seca foi significativamente diferente entre os capins. O conteúdo de proteína crua, parede celular e seus componentes (celulosa, hemicelulose e lignina) foram significativamente diferentes entre os capins avaliados. Assim mesmo, DEMS, DEPC e DEFDN foram significativamente diferentes entre capins. Os conteúdos de P, Na, Cu e Zn não foram suficientes para satisfazer os requerimentos do gado de carne em crescimento. Dados de produção de matéria seca e dinâmica nutricional sugerem que os novos genótipos PI 3 e PI 4 podem ser considerados como boas fontes de nutrientes para o gado em pastoreio em regiões do noroeste do México.

KEYWORDS / Buffelgrass / Chemical Composition / Nutrient Digestibility / Ruminal Digestion /

Common buffelgrass (Cenchrus ciliaris L.) is widely disseminated in semiarid regions of Texas and Northeastern Mexico; however, seasonality of rainfall and low temperatures are the major influences on its nutritional quality (Ramírez et al., 2003b). Previous studies carried out by García-Dessommes et al. (2003a, b) have reported that five new genotypes of buffelgrass, produced without irrigation, were less sensitive to environmental factors, yielding more dry matter and CP content than common buffelgrass and the Nueces hybrid.

Among forages, crude protein (CP) levels are well correlated with many desirable plant components like digestibility, vitamins, Ca and P. However, all these decline to deficient levels at about the same time, and CP serves as a reliable measure of overall nutritional quality (Ganskoop and Bohnert, 2001). Furhtermore, CP values of a number of grasses showed lower mean values (11.5%) than legumes (17.0%) collected in different regions of the world (Minson, 1990), whereas tropical grasses have lower CP than temperate grasses (means= 10.0 and 12.9%, respectively; Minson, 1992). Several factors affect CP in grasses. One of the most important ones is the N₂ content in soils, and it has been shown that N fertilization in temperate grasses increases CP (Whitehead, 2000). However, N fertilization did not affect dry matter or cell wall digestibility of beef cattle grazing temperate grasses (Puoli et al., 1991).

This study was carried out at the Experimental Station "General Terán", Instituto Nacional de Investigaciones Forestales, Agrícolas y Pecuarias (INIFAP) and the Universidad Autónoma de Nuevo León (UANL). General Terán, N.L., México, is at 25°18’N and 99°35’W, at an altitude of 332masl. The climate is typically semitropical and semi-arid, with a warm summer. The main and most common type of vegetation is known as the Tamaulipan Thorn scrub or subtropical Thorn scrub woodlands. The dominant soils are deep, dark-gray, lime-clay Vertisoles, which are the result of alluvial processes. These soils are characterized by high calcium carbonate (pH= 7.5-8.5) and relatively low organic matter content. Annual mean temperature is 22.4°C and the annual rainfall average is 784mm (INIFAP, 1991).

Five strains of buffelgrass PI-307622 (PI 1), PI-409252 (PI 2), PI-409375 (PI 3), PI-409443 (PI 4) and PI-409460 (PI 5), as well as the hybrid buffelgrass Nueces, which was considered as a reference grass with high nutritional quality, were established in an experiment under a completely randomized design with three replicates. The experimental plots consisted of rows 5m long with 0.8m between rows. In order to achieve a uniform grass growth, all grasses were cut prior to the experiment, on March 1, 2001. The grasses were irrigated four times during the experiment, on March 15, April 15, May 1, and May 15 to avoid any water stress in the plants. On March 15 the experimental plot was fertilized with the equivalent of 100kg·ha^-1 of urea-N. Plants reached full blossom by June 5, and were then harvested by hand to 0.15m above ground. Partial dry matter was determined in an oven at 55ºC for 72h. Then samples were ground in a Wiley mill (1mm screen) and stored in plastic containers for further analyses.

Samples were analyzed for dry matter (DM), organic matter (OM), crude protein (CP), (AOAC, 1997), neutral detergent fiber (NDF) and acid detergent fiber (ADF), Van Soest et al. (1991) and acid detergent lignin (ADL), (AOAC, 1997). Hemicellulose (NDF-ADF) and cellulose (ADF-ADL) were estimated by difference. Estimation of insoluble N₂ in NDF (INNDF) and insoluble N₂ in acid detergent fiber (INADF), which corresponds to the non-degraded N₂ was performed by procedures of Van Soest et al. (1991) and the slowly degraded N₂ associated to the cell wall components was calculated as INNDF minus INADF (Krishnamoorthy et al., 1982).

Mineral content was estimated by incinerating the samples in a muffle oven at 550ºC during 4h. Ashes were digested in a solution containing HCl and HNO₃, using the wet digestion technique (Díaz-Romeau and Hunter, 1978). Concentrations of Ca, Na, K, Mg, Cu, Fe, Zn, Mn and Mo were estimated using an atomic absorption spectrophotometer with an air/acetylene flame. The P content was determined in a colorimeter (AOAC, 1997).

The rate and extent of DM, CP and NDF digestibility in grasses were measured using the nylon bag technique. Four rumen fistulated PelibueyxRambouillet sheep (45.2 ±2.3kg, BW) were used to incubate bags (5x10cm, 53µm pore size) that contained ground (4g) samples of each grass and suspended in the ventral part of the rumen of each sheep. With the purpose to sustain adequate rumen microbial activity throughout the experiment, sheep were fed alfalfa hay ad libitum. For each grass, bags were incubated at 4, 8, 12, 24, 36, and 48h. Upon removal from the rumen; bags were washed in cold running water. Zero time disappearance was obtained by washing non-incubated bags (0h) washed in a similar fashion. All bags were dried at 60ºC in an oven during 48h. Weight loss of DM, CP and NDF was recorded. Disappearance of DM, CP, or NDF for each incubation time was calculated as

Disappearances of DM, CP, or NDF for each incubation time were used to estimate the digestion characteristics of DM, CP and NDF using the equation (Ørskov and McDonald, 1979)

p = a+b (1-e^-ct)

The nonlinear parameters a, b, and c as well as the effective degradability of DM (EDDM) or CP (EDCP) or NDF (EDNDF) = (a+b)c/(c+k)(e^-(ct)LT), were calculated using the Neway computer program (McDonald, 1981). Here, k: estimated rate of outflow from the rumen, and LT: lag time. The EDDM, EDCP and EDNDF values were estimated assuming a rumen outflow rate of 2.0%/h. Digestible dry matter was calculated as TDMxEDDM and DCP was calculated as TDMxEDCP. Data of chemical composition, a, b, c, EDDM, EDCP, EDNDF, DDM and DCP were statistically analyzed using a one-way analysis of variance. Means were separated (P<0.05) using the least significant difference technique. (Snedecor and Cochran, 1980).

Total dry matter production (TDMP) was significantly different among genotypes. PI 3 yielded more than other grasses (Table I), including Nueces, which has been recognized as a high producer of dry matter. As it would be expected, irrigation and fertilization led to a higher TDMP production than that of the same grasses (mean of 4.7ton·ha^-1 in two cuts) growing in the same area, but collected at different dates and without irrigation or fertilization (García-Dessommes et al., 2003a, b).

The CP content was also significantly different among genotypes. PI 5 had the highest and PI 1 the lowest CP values. The mean CP for all genotypes of 8.7% is ~2% above the minimum (7%) required to sustain rumen functionality in beef cattle (NRC, 1996). Related studies carried out in summer and autumn 1999 (García-Dessommes et al., 2003a, b) using the same genotypes growing in the area but without irrigation or fertilization, had lower CP contents (average of 7.1%). Therefore, N fertilization in this study increased the CP content in all grasses. This effect has already been reported by Minson (1992) and by Whitehead (2000), who found that the CP content of grasses depends on soil N availability and that N fertilization increases it.

The INNDF, INADF, and INNDF-INADF fractions were significantly different among grasses. The non-available N for ruminants estimated by the INADF fraction was higher (mean value of all grasses= 24.8%) than previous reports (3-15%; Van Soest, 1994) in temperate grasses. Buffelgrass PI 3 had the largest fully digestible portion of the NDF (100-INNDF= 31.3%) and PI 5 the lowest (26.2%). This fraction comes from true proteins, which in grasses constitute the enzymatic machinery that is rapidly degraded in the rumen and transformed in microbial protein (Van Soest, 1994).

The cell wall (NDF) content and its components (ADF, cellulose, and hemicellulose) were significantly different among grasses (Table I). PI 5 resulted with the highest NDF and PI 3 with the lowest. The irrigation and fertilization of buffelgrass plants with urea-N, evaluated in this study had about the same NDF content than that the same plants, but collected in summer and autumn 1999 and autumn 2000 (García et al., 2003a, b; Morales-Rodríguez, 2003) and without irrigation and fertilization. Even though cellulose and hemicellulose are degraded by the same rumen bacteria (Ruminococcus flavefaciens, R. albus and Fibrobacter succinogens), cellulose duplicated hemicellulose content in all grasses (Table I). The hybrid Nueces resulted with the lowest ADL and PI 5 with the highest. Irrigation and fertilization with urea-N lowered ADL (mean= 4.3 ±0.4%) compared with 5.5 and 7.0% obtained in previous studies carried out with the same grasses, but on different collection dates (García et al., 2003a, b; Morales-Rodríguez, 2003).

The fraction a (lost during the bag washing process) of the DM, CP, and NDF was significantly different among grasses (Table II). The fraction b (slowly degraded in the rumen of sheep) of the DM was not different (P>0.05); however, significant differences were obtained for CP and NDF. Degradation rate (c, %·h^-1) was significantly different only in DM and CP. The EDDM, EDNDF, and EDCP were significantly different among all grasses (Table II). The EDCP overall mean was estimated to be 47.3% and the potential available CP (100-INADF) was 75.2%. The difference between these two values (27.9%) corresponds to the value obtained by difference (INNDF-INADF), which is slowly available to rumen microbes, but could be fully digested in the abomasums (Van Soest, 1994). Digestion carried out in abomasums and the small intestine is also a vital process for the ruminant (Merchen and Bourquin, 1995).

With the exception of Na and Zn all minerals studied were different (P<0.05) among grasses (Table III). Ca, K, Mg, Fe and Mn contents were higher in irrigated and fertilized grasses than the same grasses without irrigation or fertilization, and collected at different dates (García-Dessommes et al., 2003a, b). In this study, all grasses had sufficient Ca to meet growing beef cattle requirements (4.5g·kg^-1 DM; McDowell, 2003), but the amount of P in all cases was insufficient for growing beef cattle needs (3.0g·kg^-1 DM; McDowell, 2003). García-Dessommes et al. (2003a, b) and Morales-Rodríguez (2003) also found low P content in new genotypes of buffelgrass growing in the same area. Thus, cattle grazing these grasses must be supplemented with P. Growing beef cattle needs about 1.0g·kg^-1 DM of Mg (McDowell, 2003). In this study, all genotypes had sufficient Mg content to satisfy the requirements of this nutrient, as was that of K, which met the requirements for growing beef cattle (6.0g·kg^-1 DM; McDowell, 2003). Growing beef cattle requires 0.6g·kg^-1 DM of Na in their diets (McDowell, 2003); in this study all grasses had insufficient Na and can be considered as Na non-accumulators because they contain less than 2g·kg^-1 DM of Na (Youssef, 1988). Furthermore, high K content in the evaluated grasses could reduce Na absorption of cattle feeding on them, because it has been reported that an elevated dietary K may decrease ruminal concentration and absorption of Na in steers (Spears, 1994). However, Na deficiencies can be alleviated by supplementing common salt.

Cattle consuming the evaluated grasses must be supplemented with Cu, as the latter contained insufficient amounts to meet growing beef cattle requirements (10mg·kg^-1; McDowell, 2003). Low Cu concentrations were also reported in cultivated grasses growing in semiarid regions of Northeastern Mexico (Ramírez et al., 2002a, b, 2005). A low Cu in the evaluated grasses may be caused by the high pH (7.5-8.5) in the soils (Spears, 1994) in these regions. Growing beef cattle requires about 50g·kg^-1 of Fe in the DM of their diets (McDowell, 2003). In this study, all grasses had adequate Fe amounts to meet such requirements. Similar findings were reported in previous studies (Ramírez et al., 2002a, b, 2003a, b, 2005) carried out in the same region. Iron deficiency seldom occurs in grazing ruminants due to generally adequate forage concentrations and contaminants of plants by soil (McDowell, 2003). All the grasses studied had sufficient amounts of Mn to meet requirements of growing beef cattle (20g·kg^-1 of DM; McDowell, 2003). Although Mn deficiency for ruminants under grazing conditions has been reported in USA and other countries (McDowell, 1985), with effects on skeletal development and reproductive performance, doubt has been expressed whether this deficiency arises under field conditions in Mexico. In the report by McDowell (1985), contrary to our findings the Zn content was significantly different among all grasses within seasons; in the present study, Zn content was marginally deficient to meet growing beef cattle requirements (30g·kg^-1 DM; McDowell, 2003). Similar findings were reported by García-Dessommes et al. (2003a, b), who evaluated the Zn content of the same grasses collected at different dates, but without irrigation or fertilization. High levels of Ca, found in this study, may increase the dietary Zn requirements (Underwood and Suttle, 1999) and therefore, supplemental Zn is required.

The grasses evaluated in the present study yielded more dry matter than they did when collected without irrigation or fertilization and at different dates. The same pattern was observed in CP, EDDM, Ca, K, Mg, Na, Fe and Mn. However, NDF, ADF, cellulose, hemicellulose, INNDF, INADF, P, Cu and Zn remained the same. Lower values were obtained in lignin, EDCP and EDNDF. Concentrations of P, Na, Cu, and Zn in all grasses were insufficient to meet growing beef cattle requirements. The genotype PI 4 resulted in higher dry matter production, CP and EDCP. In general, all genotypes had nutritional qualities comparable to the hybrid buffelgrass Nueces, which in this study was used as a reference grass with good nutritional quality. Cattle grazing these grasses must be supplemented with P, Na, Cu, and Zn.

1. Akin DE, Chesson A (1990) Lignification as the major factor limiting forage feeding value especially in warm conditions. In Proc. XVI Int. Grassland Cong., Vol. III. Association Francaise pour la Production Fourragere. Versailles, France. pp. 1753-1760.        [ Links ]

2. AOAC (1997) Official Methods of Analysis. 17^th ed. Association of Official Agricultural Chemists. Washington DC, USA.         [ Links ]

3. Díaz-Romeau RA, Hunter P (1978) Metodología para el muestreo de suelos y tejidos de investigación en invernadero. Mimeo. CATIE. Turrialba, Costa Rica. 26 pp.        [ Links ]

4. Ganskoop D, Bohnert R (2001) Nutritional dynamics of 7 northern Great Basin grasses. J. Range Manag. 54: 640-647.        [ Links ]

5. García-Dessommes GJ, Ramírez-Lozano RG, Foroughbakhch R., Morales-Rodríguez R, García-Díaz G (2003a) Ruminal digestion and chemical composition of new genotypes of buffelgrass (Cenchrus ciliaris L.). Interciencia 28: 220-224.        [ Links ]

6. García-Dessommes GJ, Ramírez-Lozano RG, Foroughbakhch R, Morales-Rodríguez R, García-Díaz G (2003b) Valor nutricional y digestión ruminal de cinco líneas apomíticas del pasto buffelgrass (Cenchrus ciliaris L.). Técnica Pecuaria en México 41: 209-218.        [ Links ]

7. INFAP (1991) Memorias Segunda Reunión Científica Forestal y Agropecuaria del Estado de Nuevo León. Centro de Investigación Regional del Noreste, Instituto Nacional de Investigaciones Forestales, Agrícolas y Pecuarias. Mexico. 48 pp.        [ Links ]

8. Krishnamoorthy UC, Muscato TV, Sniffen CJ, Van Soest PJ (1982) Nitrogen fractions in selected feedstuffs. J. Dairy Sci. 65: 217-220.        [ Links ]

9. McDonald I (1981) A revised model for estimation of protein degradability in the rumen. J. Agr. Sci. 96: 251-252.        [ Links ]

10. McDowell LR (1985) Nutrition of Grazing Ruminants in Warm Climates. Academic Press. New York, USA. 56 pp.         [ Links ]

11. McDowell LR (2003) Minerals in Animal and Human Nutrition. 2^nd ed. Elsevier. Netherland. pp. 26-86.        [ Links ]

12. Merchen NR, Bourquin LD (1995) Processes of digestion and factors influencing digestion of forage based diets by ruminants. In Fahey GC Jr. (Ed.) Forage Quality, Evaluation, and Utilization. University of Nebraska. Lincoln, NE, USA. pp. 564-612.        [ Links ]

13. Minson DJ (1990) Forage in Ruminant Nutrition. Academic Press. San Diego, CA, USA. 188 pp.        [ Links ]

14. Minson DJ (1992) Composición química y valor nutritivo de gramíneas tropicales. In Skerman PJ, Riveros F (Eds.) Gramíneas Tropicales. Colección Producción y Protección Vegetal. FAO. Roma, Italia. 166 pp.        [ Links ]

15. Morales-Rodríguez R (2003) Producción de materia seca y digestibilidad in situ del forraje de 86 genotipos del pasto buffel (Cenchrus ciliaris). Tesis. Universidad Autónoma de Nuevo León. San Nicolás de los Garza, NL, Mexico. 56 pp.        [ Links ]

16. NRC (1996) Nutrient Requirements of Beef Cattle. 7^th ed. National Academy Press. Washington, DC, USA. 76 pp.        [ Links ]

17. Ørskov ER, McDonald I (1979) The estimation of protein degradability in the rumen from incubation measurements weighed according to rate of passage. J. Agr. Sci. 92: 499-503.        [ Links ]

18. Puoli JR, Jung GA, Reid RL (1991) Effects of nitrogen and sulfur on digestion and nutritive quality of warm-season grass hays for cattle and sheep. J. Anim. Sci. 69: 843-852.        [ Links ]

19. Ramírez RG, Foroughbackhch R, González-Rodríguez H, García-Castillo CG, Alba-Ávila J, Háuad LA (2002a) Variación estacional del contenido mineral en la planta completa, hojas y tallos del zacate buffel común. (Cenchrus ciliaris L.). Livestock Res. Rural Devel. 14: 22-31.        [ Links ]

20. Ramírez-Lozano RG, González-Rodríguez H, García-Dessommes G (2002b) Chemical composition and rumen digestion of forage from kleingrass (Panicum coloratum). Interciencia 27:705-709.        [ Links ]

21. Ramírez RG, González-Rodríguez H, García-Dessommes G (2003a) Nutrient digestion of common bermudagrass (Cynodon dactylon L.) Pers. growing in northeastern Mexico. J. Appl. Anim. Res. 23: 93-102.        [ Links ]

22. Ramírez RG, García-Dessommes G, González Rodríguez H (2003b) Valor nutritivo y digestión ruminal del zacate buffel común (Cenchrus ciliaris L.). Pastos y Forrajes 26: 149-158.        [ Links ]

23. Ramírez RG, González-Rodríguez H, García-Dessommes G, Morales-Rodríguez GR (2005) Seasonal Trends in the Chemical Composition and Digestion of Dichanthium annulatum (Forssk.) Stapf. J. Appl. Anim. Res. 28: 35-40.        [ Links ]

24. Snedecor GW, Cochran WG (1980) Statistical Methods. 7^th ed. Iowa State University Press. Iowa USA. pp: 215-233.        [ Links ]

25. Spears JW (1994) Minerals in forages: In Fahey CG Jr (Ed.) Forage Quality, Evaluation and Utilization. National Conference on Forage Quality. University of Nebraska. Lincoln, NE, USA. pp: 281-311.         [ Links ]

26. Underwood EJ, Suttle NF (1999) The Mineral Nutrition of Livestock. 3^rd ed. CABI. Wallingford, UK. 600 pp.        [ Links ]

27. Van Soest PJ (1994) Nutritional Ecology of the Ruminant. 2^nd ed. Comstock - Cornell University Press Ithaca, NY, USA. pp. 122-139.        [ Links ]

28. Van Soest PJ, Robertson JB, Lewis BA (1991) Methods of dietary fiber, neutral detergent fiber and nonstarch polysaccharides in relation to animal nutrition. J. Dairy Sci. 74: 3583-3597.        [ Links ]

29. Youssef FG (1988) Some factors affecting the mineral profiles of tropical grasses. Outlook Agric. 17: 104-111.        [ Links ]

30. Whitehead DC (2000) Nutrient Elements in Grasslands. Soil-Plant-Animal relationships. CABI. Wallingford, UK. 135 pp.        [ Links ] ]]> 1 III 1753-1760 2 AOAC 1997 17 3 1978 26 4 2001 54 640-647 5 2003 a 28 220-224 6 2003 b 41 209-218 7 INFAP 1991 48 8 1982 65 217-220 9 1981 96 251-252 10 1985 56 11 2003 2 26-86 12 564-612 13 1990 188 14 166 15 2003 56 16 NRC 1996 7 76 17 1979 92 499-503 18 1991 69 843-852 19 2002 a 14 22-31 20 2002 b 27 705-709 21 2003 a 23 93-102 22 2003 b 26 149-158 23 2005 28 35-40 24 1980 7 215-233 25 281-311 26 1999 3 600 27 1994 2 122-139 28 1991 74 3583-3597 29 1988 17 104-111 30 2000 135