NUTRITIONAL VALUE OF OPUNTIA FICUS-INDICA AS A RUMINANT FEED IN ETHIOPIA

Firew TEGEGNE

INTRODUCTION

Most tropical livestock production systems have low productivity due to low feed availability and quality, especially in drought-prone areas where the livestock sector regularly suffers large losses.

The problem is not limited to arid and semi-arid regions of Ethiopia. Declining crop yields accompanied by increasing requirements for food are forcing farmers of the central and northern highland regions to cultivate more land at the expense of grazing pasture and browses. These problems are exacerbated by poor management and financial limitations.

Research efforts to match seasonal fluctuations in feed supplies to needs include: treating crop residues; modifying agronomic practices; varietal selection; pasture improvement; supplementing with non-protein nitrogen (NPN); planting multi-purpose forages; conservation; and rumen manipulation.

However, there are financial limitations and inadequately qualified staff to carry out analytical work. If treated feeds were to be used, animal products would be at unaffordable prices (Nkhonjera, 1989). Availability of crop residues is limited to areas suitable for crop production. In some arid and semi-arid areas, improvement of pasture is restricted to sowing improved grasses on more fertile soils (Evans, 1982). The nitrogen (N) from NPN or legumes degrades very rapidly and there is a mismatch between the degradation of organic matter and N. Manipulation of the rumen ecosystem does not seem economically effective in extensive forms of farming (Leng, 1982). Silage making from low-nutritive-value tropical forages involves the risk of bad fermentation and needs more facilities (Jarrige et al., 1982). Consequently, there is a need to get N-source feeds that can immediately supply rumen degradable organic matter to serve as a link between NPN, forage legumes and crop residues.

Most experts recommend planting of trees and shrubs to provide standing feed resources so that herds and flocks can survive critical periods of shortage and prolonged drought. In screening plants for animal nutrition for drought-prone regions, the two most important criteria are drought tolerance and palatability for animals. However, adaptability of forage to marginal land, ease of propagation, persistency, dry matter (DM) yield, high digestibility (D), and N contents are also important. Opuntia meets all of these requirements. Most important, opuntia is suitable as a human food and has other miscellaneous uses. However, more information is needed concerning its nutritive value, its utilization for animal feed, management, establishment and its integration into pastoral and agropastoral systems.

The study of the potential and nutritive value of opuntia could contribute to the development of the livestock sector in dry regions of Ethiopia. This chapter reviews the feasibility and nutritional value of O. ficus-indica as a feed resource for farm animals in such areas.

ECOLOGICAL DISTRIBUTION AND UTILIZATION OF OPUNTIA IN ETHIOPIA

Opuntia ficus-indica (L.) Miller was taken to North Africa from Mexico in the sixteenth century (Pimienta, 1993) and introduced into Ethiopia at the end of the 19^th century (CFDP, 1994). It is widely distributed in the northern arid and semi-arid regions of Ethiopia. A survey indicated that about 30 520 ha (1.88% of the total area of the Tigray region) were covered with O. ficus-indica, 48.62% growing wild and 51.34% cultivated. It was also found in the hills of the Welo region, where the vegetation is severely degraded.

Since the 1960s, the fruit has been consumed by almost all domestic animals, and livestock totally depends on opuntia during the dry season. Planting of O. ficus-indica is common and extensive. Two Ethiopian organizations that play an important role in the expansion of cactus acreage are the Relief Society of Tigray (REST) and the Regional Natural Resource Conservation and Development Bureau. The Cactus Fruit Development Project (CFDP) has promoted the selection, production and distribution of cactus varieties, identification of diseases and design of erosion control measures as part of its strategies (CFDP, 1994).

NUTRITIONAL VALUE OF OPUNTIA FICUS-INDICA

The fruits of O. ficus-indica contain water (92%), carbohydrates (4-6%), protein (1-2%), minerals (1%) and a moderate amount of vitamins, mainly A and C (Cantwell, 1991, and Neri, 1991, cited by Pimienta, 1993). According to these figures, its fruits are high in carbohydrates (50-75% of DM) and moderate in protein content (12.5-25% of DM), minerals and vitamins.

South African measurements of nutritional quality of Opuntia of 4% CP, 64% TDN, 1.4% Ca, 0.2% P and 0.1% Na were similar to Texas data (De Kock, 1980). He indicated that in contrast to fertilized opuntia plantations, the protein content of the ‘wild prickly pear’ was so low that mineral and protein supplements were necessary.

The effect of cultivar is illustrated by a comparative study conducted in Brazil of fodder cultivars for milk production: the CP contents of O. ficus-indica cvs Gigante and Redonda, and Nopalea cochenillifera cv. Miúda were: 4.83, 4.21 and 2.55%, and their CF contents 9.53, 8.63 and 5.14%, respectively. In vitro dry matter digestibility (IVDMD) was 77.37% for Miúda compared with 74.11 and 75.12 for Redonda and Gigante, respectively; mean milk production and milk fat were not significantly different among treatments (Ferreira-dos-Santos et al., 1990).

Cladode age is an important factor for nutritional value. Young cladodes of O. ficus-indica grown for commercial fruit production in Spain had 10.6-15.0% protein, while mature cladodes varied from 4.4 to 11.3% protein (Retamal et al., 1987b). Similarly, Gregory (1988, cited by CFDP, 1994) reported that as the age of O. ficus-indica increased from one to four years, the CP content decreased: 11.53, 5.74, 5.5 and 5.65%, respectively in the four years, with an average of 7.10%. Compared to mature 12-year-old cladodes, 2-year-old cladodes had substantially higher N, K and Mn, but lower Na, Ca and Fe. This was attributed to age and to higher metabolic activity of young cladodes (Nobel, 1983). Concentrations of 15.3% protein and 0.3% P were reported in commercial O. ficus-indica fruit plantations in California (Nobel, 1983). In contrast, the chlorenchyma contained 9.6% protein and 0.12% P for 5-year-old plantations and 7.8% protein and 0.09% P for 12-year-old Chilean opuntia plantations. Young cladodes had significantly higher N, K and Mn, but lower Na, Ca and Fe. Epstein (1972) suggested that Ca and Fe are not very mobile so that both would be expected to accumulate in older tissues (Retamal et al., 1987b).

In contrast, Gregory and Felker (1992) reported that O. ficus-indica had similar protein contents in all age classes. Their results are unusual, as young cladodes are generally of better nutritional quality than older cladodes, which is attributed to the thickening of the cuticle of older cladodes and the increase in thickness due to the expansion of the water-storage parenchyma (which is very slowly degradable) at the expense of cell contents (Rodríguez-Felix and Cantwell, 1988). The latter authors reported the following composition values per 100 g for cladodes harvested when about 20 cm long: 91.7 g water, 1.1 g protein, 0.2 g lipids, 1.3 g ash and 1.1 g CF (13.3, 2.4, 15.7 and 13.3% on a DM basis, respectively). It was observed that total carbohydrates increased considerably during cladode growth, while protein and CF contents decreased.

Season has a profound impact on the chemical composition of O. ficus-indica. According to Retamal et al. (1987b), the highest values of moisture content, free reducing sugars, starch and CP were detected in spring (92.5%; 103 mg/g DW; 226 mg/g DW; 14.8% respectively) in young cladodes, while at the end of the season, ash content, ether extractive, crude fibre and calorific content presented the highest values (29.8%; 36 mg/g DW, 144 888 KJ/kg, respectively). Highest concentrations of N, P and K occurred in winter, with Ca showing the opposite pattern (Esteban-Velasco and Gallardo-Lare, 1994).

Compared to most agronomic plants, chlorenchyma levels of Ca and Mg (5.3% and 2.5% of DM, respectively) in cacti tended to be higher and Na level (0.11% of DM) lower (Retamal et al., 1987b). They found that among the 11 elements tested, N was strongly correlated with the nutrient level and metabolic activity, where nocturnal acid accumulation (NAA) tended to be greater when the N level in the chlorenchyma was higher (r² = 0.39). In contrast, NAA was negatively correlated with chlorenchyma Na content (r² = 0.32).

The distinctive features of cacti: shallow root systems, leaves modified into spines, and shading of photosynthetic organs; could affect mineral relations. The shallow root enables them to accumulate elements from the upper part of the soil and shading results in accumulation of certain elements (Nobel, 1977). An important feature common to most cacti is the relatively high levels of Ca which may represent accumulation of calcium oxalate (Nobel, 1983).

Flachowsky and Yami (1985) studied composition, D and feed intake of O. ficus-indica by Ogaden sheep, where 70-75% (DM basis) was total carbohydrate and about 20% was crude ash. They indicated an apparent D of OM of 70.9%, corresponding to 35 and 467 energy feed units (cattle)/kg (72 MJ ME) in fresh material and DM, respectively. CP was 4.5-5.5% of DM, less than maintenance requirement.

Given free choice, rams preferred chopped fresh cactus to chopped dried cactus or whole fresh cactus (Flachowsky and Yami, 1985). In conditions where water is not a limiting factor for animal production, it could be difficult for animals to take enough fresh cactus to meet their requirements, as a level of water exceeding 780 g/kg fresh forage is claimed to have a detrimental effect on voluntary intake (John and Ulyatt, 1987, cited by Minson, 1990a). Fortunately, this effect may be small and of no disadvantage in arid and semi-arid areas where water is limiting for animal production.

In an experiment replacing alfalfa hay with O. ficus-indica cladodes as supplementary summer forage for milk goats, 50 goats were grazed on indigenous pasture alone (control); with lucerne hay (LH) ad libitum; and with three LH + cladodes (C) combinations (85% LH+15% C; 79% LH+21% C; 66% LH+34% C). Milk production increased by 55.4, 93.8, 103.6 and 12%, respectively, compared to the control (p<0.05) (Azócar and Rojo, 1991).

Gregory and Felker (1992) reported Opuntia to be high in moisture content (94.26%) and high in in vitro D (about 75%). Most workers have suggested that Opuntia is low in its CP (4%) content and P (0.2%) contents, and have recommended that supplements should be given to meet the requirements of animals (De Kock, 1980; Hanselka and Paschal, 1990). It is moderate in energy content measured as digestible nutrients, and high in water, vitamin A, fibre and ash (Hanselka and Paschal, 1990). Fortunately, there are ways of improving it. The application of low rates of N increases the percentage of CP significantly. It was proposed that high N treatment (224 kg/ha every two years) is needed to meet the requirements for lactating cows. Application of P (112 kg/ha) also doubles P content, which is normally low in O. ficus-indica (González and Everitt, 1990 cited by Pimienta, 1993).

Anti-nutritional characteristics, such as spines, may affect nutritional value by limiting palatability and digestibility and so utilization efficiency. The common method for removing the spines is burning. A device has been designed to mechanically remove the spines (Carmorlinga-Sales et al., 1993). Another method is use of a chaff cutter (De Kock, 1980).

Given that O. ficus-indica is a Crassulacean Acid Metabolism (CAM) plant, the organic acid content varies during the day. Teles et al. (1984) found that levels of malonic, malic and citric acids in materials collected at 18:00 were: traces, 0.95 and 0.31 mg/g, respectively. In similar material collected at 06:00, concentrations were 0.36, 9.85 and 1.78 mg/g, respectively. The pH ranged from 5.2 in the evening to 4.4 in early morning, and the percentage of malic acid varied from >0.5% at 08:00 to <0.1% at 16:00 (Cantwell, 1991, and Neri, 1991, cited by Pimienta, 1993). However, the effect of organic acid variation during the day has not been studied with opuntia.

Nutritional changes after harvest have been noted, though not explained. Neri (1991, cited by Pimienta, 1993) observed reduction in the content of both total and reducing sugars, and an increase in pH and protein content. In production systems where water is not limiting, storing opuntia increases DM so that animals can consume more of it to meet their requirements. The increase in its protein content is more important and needs investigation.

ANALYSIS OF ETHIOPIAN OPUNTIA

Samples were taken from opuntia plants grown in a greenhouse on sandy soil, with no fertilization, representative of tropical poor soils in which Opuntia ficus-indica usually grows.

Four branches (A, B, C and D), as groups, each with three cladodes, as age groups, and their fruits (f) were separated. The cladodes on each branch were labelled as young (y), middle-aged (m) or old (o). They were six months, one year and two years old, respectively.

Dry matter, ash and mineral content determinations

Dry matter content was determined by drying chopped samples for four days in an oven set at 80°C. Ash content was determined by incinerating dried samples at 500°C until a greyish-white colour was attained. The solution for mineral determination was prepared as stated by Retamal et al. (1987b) except that the solution for Ca, Mg and K analysis was further diluted with distilled water (1:100) making the final dilution factor 1:1000. The concentration of Ca, Mg, K and Na in the solution was determined by atomic absorption spectrometry and the concentration of P was determined spectrophotometrically. The result for each element was calculated from the respective standard graphs (MAFF, 1986). Crude protein (CP), crude fibre (CF) and ether extract (EE) were determined by the proximate analysis method (MAFF, 1986). Nitrogen-free extract (NFE) was calculated as the DM not accounted for by the sum of CP, CF, EE and ash (NFE = DM -CP -CF -EE -ash) (Van Soest, 1982). A bioassay was performed using the faeces liquor technique (El Shaer et al., 1987) and used for in vitro dry matter digestibility (IVDMD) assay. IVDMD was calculated as: IVDMD = (A - (B - C))/A, where: A = dry weight of sample; B = dry weight of residue after digestion; and C = dry weight of reagent blank.

The mean proportional weight loss of the triplicates or duplicates for each sample was recorded as the IVDMD (Omed et al., 1989). Data were analysed by ANOVA (General Linear Model, GLM) test and the significance of difference between means detected using Fisher’s least significance difference (LSD) test.

The relationship of chemical composition data with IVDMD was performed by simple linear regressions, and significance of correlation by ANOVA. For comparison, appropriate multiple regression equations using combinations of CP, CF, NFE, EE and ash as an independent variable and IVDMD as a dependent variable were used.

Mineral composition

The mineral composition of samples is summarized in Table 44. There was significant age effect on Ca, Mg and Na contents and a highly significant effect on P content. Age did not affect K content. It is well established that tropical legumes, tropical grasses and other roughages are low in minerals, particularly P (Fleming, 1973; Minson, 1988). The P content (Table 44) of the present samples was low in comparison to temperate pasture grasses (McDonald et al., 1995). Older cladodes had lower P contents than younger cladodes and fruits, which was in accordance with most previous results (De Kock, 1980; Nobel, 1983; Hanselka and Paschal, 1990; Gregory and Felker, 1992). All the results were within the range of 0.02 to 0.58% reported for 586 tropical grasses, whose mean was 0.22% (Minson, 1990b). In addition, all the P values were above the recommended level (0.17%) for cattle weighing 450 kg and gaining 0.5 kg/day (NRC, 1968).

O. ficus-indica has been reported to be high in Ca content (Nobel, 1977; De Kock, 1980; Retamal et al., 1987b). The values obtained disagree with this (Table 44). This may be due to the young age of the samples having allowed less accumulation of calcium oxalate (Nobel, 1977). Fruits had significantly lower Ca content than other parts and Ca content of young cladodes was higher (but not significant; p>0.05) than either middle-aged or old cladodes. The significantly higher Ca content found in young cladodes also disagreed with other reports (Epstein, 1972; Nobel, 1983; Retamal et al., 1987b), but the difference was small. Fruits had lower Ca content than cladodes, explained in part by the low mobility of Ca (Epstein, 1972).

Ca content of 390 tropical grasses varied from 0.14 to 1.46% (Minson, 1990b), a range containing most of the values obtained. All samples contained sufficient Ca to meet the required 0.17% recommended by NRC (1968). Most samples were within the high range of temperate pasture grasses (>0.6%) (McDonald et al., 1995).

O. ficus-indica has been reported as high in Mg content (Retamal et al., 1987b). Mg content of these samples was high and significantly (p<0.05) increased with age. All the values were within the range reported by Minson (1990b). In addition, the results were above the 0.11% Mg level recommended by the ARC (1965). Though there is less likelihood for Mg to be deficient, as most tropical grasses and legumes have enough of it (Norton, 1982), these results showed that Opuntia had a sufficiency of Mg.

Table 44. Mean mineral composition (% of DM) of fruits and cladodes of Opuntia ficus-indica

Notes: (1) Different superscripts indicate significantly (p<0.05) different means. (2) ns = Non-significant.

The low K content of the older cladodes (Table 44) may reflect the high metabolic rate of fruits and younger cladodes (Nobel, 1983). Retamal et al. (1987b) observed that younger cladodes had substantially higher K content, which was not found in this study.

The Na content of both fruits and cladodes was very low (Table 44) as reported by De Kock (1980) and Retamal et al. (1987b). Retamal et al. (1987b) reported that younger cladodes had lower Na contents, which was not observed in the results reported here (Table 44). The values indicate that the low Na content of cacti was probably due to the low genetic capacity for accumulation, low requirements for growth or low availability in the soil (Norton, 1982; Retamal et al., 1987b). The latter authors reported that Na content was negatively correlated with nocturnal acid accumulation (NAA), confirming the above claim.

It is firmly established that tropical plants have low Na contents (Fleming, 1973), though its deficiency is related to particular species (Minson, 1990a). The results were within the range of Na contents typically found in tropical grasses, i.e. 0.01 to 1.8%. All the samples contained less than 0.08%, which is the recommended level (ARC, 1965). However, in the arid and semi-arid areas, salinity of drinking water may be high (McDowell, 1985), which could compensate for any deficiency.

Chemical composition

Both fruits and cladodes had low DM contents (9.17%), with the lowest values observed in young cladodes (Table 45). Average ash percentage of the DM was 8.67%. CP declined with age (r = -0.79) in the cladodes, though the pattern was inconsistent. Fruits and young cladodes had significantly (p<0.05) higher CP content than middle-aged and old cladodes (Table 45) while there was no significant differences between fruits and young, middle-aged and old cladodes. Young cladodes had the lowest mean CF content (Table 45). CF content was negatively correlated with CP contents (r = -0.33), and NFE contents (r = -0.53). However, the differences between CF contents were not significant at the 0.05 level. NFE was positively correlated with age (r = 0.64), and negatively correlated with EE (r = -0.42) and ash (r = -0.77).

In vitro dry matter digestibility

Average IVDMD (Table 45) was highest for fruits (p<0.01), followed by young cladodes, and significantly declined with age in the older cladodes. IVDMD was negatively correlated with age (r = -0.95), and NFE (r = -0.80), and positively correlated with CP (r = 0.76) and ash contents (r = 0.73). A relationship between IVDMD and chemical composition, including age was calculated: IVDMD = 74.1 - (4.12 × Age) - (0.009 × CP) + (0.482 × CF) - (0.91 × EE) + (0.989 × ash)

(r²= 0.93; p<0.001).

Table 45. Mean in vitro dry matter digestibility (IVDMD), estimated digestible energy (DE) and total digestible nutrient (TDN) contents and chemical composition of fruits and cladodes of Opuntia ficus-indica

Key: DM = dry matter. IVDMD = in vitro DM digestibility. DE = digestible energy. TDN = total digestible nutrient. CP = crude protein. CF = crude fibre. NFE = nitrogen-free extract

Notes: (1) Different superscripts indicate significantly (p<0.05) different means. (2) ns = Non-significant.

Table 46. Mean DM and chemical composition of Opuntia ficus-indica fruits and cladodes from 16 locations in Ethiopia (% on DM basis)

Key: DM = dry matter. CP = crude protein. CF = crude fibre. EE = ether extract. NFE = nitrogen-free extract

CHEMICAL COMPOSITION

CP content

Opuntia ficus-indica was reported to be low in CP content (De Kock, 1980; Glanze and Wernger, 1981; Flacowsky and Yami, 1985; Ferreira-dos-Santos et al., 1990). In contrast, some authors reported Opuntia ficus-indica as a moderate CP source (Nobel, 1983; Retamal et al., 1987b; Rodríguez-Felix and Cantwell, 1988; Cantwell, 1991, and Neri, 1991, cited in Pimienta, 1993). The results obtained (Table 45) agreed with the last-named authors. However, most of their samples were from cultivated plantations, while the opuntia used for this study was treated as a wild plant. Age and conditions of cultivation may explain the difference (De Kock, 1980; Retamal et al., 1987a; Hanselka and Paschal, 1990).

As is the case in most plants, age significantly affected CP content. The mean CP contents of all the fruits and cladodes of all ages (grand mean = 11.61%) were greater than the average CP content of all fibrous crop residues (6.1%) (Kossila, 1984) and tropical grass samples (7.7%) reported by Butterworth (1967) or the 10.6% of Minson (1990b). However, it was less than the average CP content of 340 tropical legumes: 17.2% reported by Minson (1988) or 16.7% reported by Minson (1990b), while most were comparable to the average CP content (13.3%) of 470 temperate grasses (Minson, 1990b). All values were above the level (6-7%) reported as the limit to microbial activity, and thus productivity and feed utilisation efficiency (Minson, 1990b).

Crude fibre content

CF content is usually taken as a negative index of feed quality (Van Soest, 1982). In this study, Opuntia ficus-indica was extremely low in CF. Similar results were previously reported by Rodríguez-Felix and Cantwell (1988) and Ferreira-dos-Santos et al. (1990).

As plants mature there is a significant increase in CF content (Van Soest, 1982). In cacti, however, there were no significant differences in CF among age groups (p>0.05). Rodríguez-Felix and Cantwell (1988) even reported a decrease in CF in older cladodes, suggesting that the significant decrease in IVDMD values of older cladodes (Table 45) was not due to the increase CF content.

All the samples reported here were below the range of CF contents determined for either tropical legumes (12.4 to 43.4%, with a mean of 30.6%) and tropical grasses, with a mean of 33.4% (Butterworth, 1967). They were lower than the mean CF content of temperate grasses (20.0%) and temperate legumes (25.3%) reported by Norton (1982).

Nitrogen-free extract content

The NFE content, which represents the highly digestible carbohydrates (Van Soest, 1982), of all the samples was relatively high (Table 45). The high NFE values of the older cladodes indicated that they had the highest soluble cell contents. The increase in NFE with age (r = 0.64) agrees with the observation that total carbohydrates increased during cladode development (Rodríguez-Felix and Cantwell, 1988), which could, to some extent, buffer the decline in IVDMD as cladodes get older (Radojevics et al., 1994). The negative correlation between NFE content and IVDMD (r = -0.80) might be due to changes related to other factors.

In vitro dry matter digestibility

Low digestible energy and protein contents are the two most important features of a diet that imposes physical restriction on feed intake (Van Soest, 1982). Consequently, energy and protein are usually given first consideration in any feeding system, and thus there is a real need for a digestible feed resource (Yilala, 1989).

The data in Table 45 showed that Opuntia ficus-indica was highly digestible, agreeing with the values reported by Ferreria-dos-Santos et al. (1990). Although there were relatively small differences in CP and CF contents between fruits and young cladodes, the IVDMD was significantly higher for fruits. Their high digestibility was attributed, in part, to the translocation of soluble carbohydrates (Norton, 1982). Younger cladodes were more digestible than middle-aged and old cladodes. This seemed to be related to the lower CP contents of older cladodes (r = 0.76). However, none of the CP contents was below 6-7% - the limiting level for microbial growth (Minson, 1990b) - or below the DMD/CP ratio (>10:1) that was noted as limiting for microbial synthesis and fermentation conditions (Hogan, 1982).

It was less likely that CF content of old cladodes had significantly affected their digestibilities. This suggestion was confirmed by the extremely low CF contents (Table 45), which had no correlation with age (r = -0.04). When compared with other grasses and legume forages, it might be argued that Opuntia ficus-indica with such low CF content had a lower IVDMD than might be expected. The degree of lignification was also unlikely to cause significant reduction in D because non-legume dicotyledenous plants, to which Opuntia belongs, are chiefly unlignified and have a high cell wall recovery (Van Soest, 1982). The extremely low CF content might, however, have caused a high rate of digestion and affected digestibility due to acid accumulation in the bottles, which is difficult to buffer (Van Soest, 1982).

A proportion of the decline in digestibility values for the old cladodes could be associated with the indigestible cutin, which prevents microbial attack (Monson et al., 1972) Cutin is present in the cuticle of cacti (Hanna et al., 1973; Uden, 1984). Differences exist in the ability of cuticle to crack under stress (Hanna and Akin, 1978), which has not been investigated in Opuntia ficus-indica.

C₄ plants are photosynthetically more efficient than C₃ plants, but they exhibit low nutritive value (Van Soest, 1982). The morphological characteristics (Norton, 1982); temperature of growth (Minson, 1990a); the well-developed, more slowly degradable, parenchyma sheaths of C₄ plants (Akin, 1982); and the few mesophyll cells (Van Soest, 1982) might limit the digestibilities of fruits and cladodes. However, any impact of these must be small, as the samples were highly digestible (Table 45). These high IVDMD values were related to the high cell contents, which are roughly represented by nitrogen-free extract (NFE) contents and low CF contents (Table 45) (Van Soest, 1982).

Regression analyses of IVDMD against separate chemical composition data (CP, CF, NFE, EE and ash) confirmed that CF and EE contents are not related to digestibility (r2 = 0.0%), although combination had a highly significant (p<0.001) effect. IVDMD was best predicted by regression including age (r² = 93.6%).

The IVDMD of almost all the fruits and the cladodes were above the mean values reported for tropical grasses (30-75%, with a mean of 54%) (Minson and McLeod (1970) in Minson, 1988), temperate grasses (45-85%, with a mean of 67%), tropical legumes (36.0 to 69.3%, with a mean of 54%) and temperate legumes (mean of 60.7%) (Minson, 1988). None of the IVDMD values was below the digestibility level recommended for different ruminants kept for different production purposes. For example, for higher performance levels of larger animals, forage digestibility over 66% is required

(Burns, 1982); a lactating beef cow producing 10 kg milk/day requires forage of 67% D, and a cow producing 5 kg milk/day, high yielders of Ethiopian indigenous breeds, requires 53% D (Burns, 1982). Thus, Opuntia ficus-indica can be a feasible forage in the tropics where even applying N to grasses does not appear to improve D (Minson, 1973).

Higher IVDMD is obtained by drying samples at 100°C for one hour followed by a moderate temperature of 70°C (Burns, 1981). However temperatures above 80°C causes thermo-chemical degradation of non-structural parts. Content of water-soluble carbohydrates (WSC), in vitro digestibility (IVD) and percentage of nitrogen insoluble in neutral detergent are affected most by drying temperature. Thus, prolonged heating at high temperature promotes loss of sugar through the Mailliard reaction. The reaction is favoured by high temperature, moisture content and soluble carbohydrates in the plant material: all these requirements were met for opuntia. Oven-drying at high temperature can also increase structural constituents. Therefore, Mailliard products were produced and structural constituents increased, limiting digestion as they are totally unavailable or very slowly degradable (Van Soest, 1982).

CONCLUSIONS

O. ficus-indica was moderate in CP, high in Ca, normal in Mg and low in Na, K and P contents in relation to ruminant requirements from a diet, and similar to common temperate or tropical grasses and legumes. It was highly digestible. Opuntia ficus-indica may serve as a link between crop residues, legume forages and NPN sources by supplying readily available organic matter.

Extremely high water content may affect total DM intake by animals, especially during wet seasons and where water is not a limiting factor for animal production. Therefore, research must gear to silage production in combination with coarse crop residues.

This study has evaluated some feed quality parameters at one point in time. For any true evaluation and in order to incorporate Opuntia ficus-indica into feeding systems, its effect on animal performance must be investigated. Likewise, further work on its combination with other feeds is needed.