Evaluation of Leucaena leucocephala and cactus (Opuntia sp.) as forages for growing rabbits

C A Ruiz-Feria^*#, S D Lukefahr^*## and P Felker^**

^*Department of Animal & Wildlife Sciences and ^**Center for Semi-Arid Forest Resources, Texas A&M, University-Kingsville, TX 78363, USA
#Present address: University of Arkansas, O-419 Poultry Science Center, Fayetteville, AR 72701, USA.
##Corresponding author: Steven D. Lukefahr, Department of Animal & Wildlife Sciences, Campus Box 156, Texas A&M University-Kingsville, TX 78363, USA; phone: (512)593-3699; FAX: (512)593-3788; e-mail: s-lukefahr@tamuk.edu.

Abstract

Leucaena leucocephala and cactus (Opuntia sp.) were evaluated as forages for rabbits based on growth, feed utilization, gastrointestinal tract (GIT) development, and carcass traits. A total of 64 purebred Altex and New Zealand White weanling rabbits were fed for 42 d. Experimental diets were 100, 90, 80 or 70% commercial pellets with corresponding levels of 0, 10, 20 or 30% leucaena leaves (dry matter basis), fed either with or without cactus offered ad libitum.

Weekly growth response slopes varied by diet but not breed. Control rabbits (100% pellets fed ad libitum with and without cactus) had improved average daily gains by 9.0 g/d (P<0.01), reached 1,800 g minimum market weight earlier by 6.5 d (P<0.05), and had heavier 42-d final weights by 371 g (P<0.01) than rabbits fed leucaena with and without cactus. In general, cactus did not significantly influence any of the traits studied. Pellet and gross (pellet plus forage) feed intake performance in control versus forage-fed pens depended on whether cactus was included in the diets (P<0.05). However, independent of the cactus effect, conversion of gross feed to total gains was improved in control versus forage-fed pens (P<0.01). Rabbits fed 10% leucaena (with and without cactus) had higher pellet and gross feed intake levels (P<0.01), whereas no differences existed between 20 and 30% leucaena diets. Rabbits on the 20 and 30% leucaena with cactus diets numerically consumed the least quantity of pellets per unit of body gain. Control animals had 2.0% higher carcass (P<0.05), 0.85% lower GIT (P<0.01), and 0.14% lower stomach (P=0.07) weights as percentages of pre-slaughter weight than animals fed 30% leucaena with and without cactus. Carcass and GIT traits were similar between control animals fed only pellets and those fed pellets and cactus (P>0.05). Rabbits fed 30% leucaena with cactus had 2.7% higher carcass weight (P<0.05) and tended to have 1.0% higher pelt weight (P<0.10) compared to those fed 30% leucaena without cactus. Breeds were not different for growth, GIT and carcass traits (P>0.05).

No signs of hair loss or discolored livers or urine due to possible mimosine toxicity were observed in 30% leucaena-fed rabbits. Hence, it is recommended for small-scale producers that the feeding level of leucaena (within a 0 to 30% range) should depend on economics in terms of realized feed cost savings in relation to growth response.

Key words: Rabbits, Leucaena leucocephala, cactus (Opuntia stricta), growth, feed utilization, carcass traits

Introduction

It is well recognized that malnutrition is a common problem for impoverished people in the lesser developing countries (LDC). For example, per caput animal protein consumption is about 13 g/d in the LDC compared to 60 g/d in developed countries (Sansoucy 1995). Malnutrition, coupled with limited land holdings, high unemployment, and increasing cereal grain costs, have attracted limited-resource farmers to rabbit meat production as an alternative agricultural enterprise in several of the LDC's (Owen 1981; Lukefahr and Cheeke 1991; Colin and Lebas 1996).

Under small-scale farm conditions in arid and tropical regions, rabbits can subsist on inexpensive diets based on forages (Lebas 1983; Lukefahr and Goldman 1985). For example, leaves from drought-resistant, leguminous trees (eg: Leucaena leucocephala and Gliricidia sepium) can provide the major source of dietary protein for rabbits (Raharjo et al 1986; Onwudike 1995), but may also contain toxic compounds (eg: mimosine in leucaena). Agricultural by-products (eg: sugar cane, cassava root meal and rice bran) can constitute the major dietary energy source (Ramchurn 1978; Lukefahr and Cheeke 1991; Le Thi Thu Ha et al 1996). In addition, drought-resistant, low-nitrogen demanding prickly pear cactus (Opuntia stricta var. stricta) (Photo 1) has been reported to have high dry matter digestibility (Gregory and Felker 1992) and also to be highly palatable to wild and domesticated rabbits (Hoffman et al 1993; Ruiz-Feria et al 1996).

Photo 1: The prickly pear cactus (Opuntia stricta var. stricta)

The experimental objective was to evaluate browse forages: leucaena (Leucaena leucocephala) and thornless prickly pear cactus (Opuntia stricta var. stricta) as partial replacements for a commercial pelleted diet for growing rabbits of two breeds.

Materials and Methods

Population and experimental diets

The experiment was conducted at the Texas A&M University-Kingsville (TAMUK) rabbit research facility from November 1995 to January 1996. The region (2736N, 9757W) is classified as semi-arid and subtropical. Several forage plant species are prevalent which are common to arid and tropical regions. The control diet, which was fed ad libitum, consisted of commercial pellets (Nutrena®, Cargill-Nutrena Feeds Division, Minneapolis, MN). This alfalfa-based ration was formulated to meet nutrient requirements of growing rabbits according to the recommendations of the National Research Council (1977). Experimental diets consisted of 90, 80 or 70% pellet restriction substituted on a DM basis by 10, 20 or 30% fresh leucaena leaves, respectively. The limit of 30% was chosen based on previous studies (Ramchurn 1978; Parigi-Bini et al 1984) that reported no mimosine toxicosis signs when leucaena was fed to rabbits at levels of 40% or less. Throughout the study, the restricted feeding levels of pellets were determined based on the average feed consumption of control rabbits on the previous day.

Diets were offered either without or with fresh cactus pads fed ad libitum. Gregory and Felker (1992) reported that the cultivar used in this experiment (accession #1270; Photo 1) had superior protein and higher phosphorus content compared to seven other thornless varieties. Based on a preliminary trial, this cultivar was observed to be more palatable to rabbits than O. ficus-indica or O. cochenillifera. The rationale for feeding cactus with leucaena was to increase total forage intake to improve rabbit growth indirectly through greater stimulation of GIT development. Also, although not analyzed in this experiment, pectin and mucilage (natural compounds found in cactus) may potentially reduce the absorption of mimosine, a toxic amino acid present in leucaena which retards growth (Rainer et al 1994). A secondary control diet consisted of 100% pellets and cactus. Hence, diets consisted of a 4 x 2 factorial arrangement of leucaena and cactus forages as diet treatments.

Leucaena leaves (<8 weeks old) were harvested daily from research plots at TAMUK, and were placed in rabbit forage feeders (Bass Equipment Company, Inc., Monett, MO) on the same day of collection. Freshly cut cactus pad strips (about 10 cm wide) were presented daily on a wire strand which was secured to the back of the inside of the pen, approximately 20 cm above the pen floor. Refuse weights of pellets, leucaena, and cactus were recorded daily. Chemical analysis of forage and pelleted feed duplicate samples was made according to the methods of the Association of Official Analytical Chemists (1984; Table 1). Water was supplied continuously via an automatic watering system.

Pen assignment and traits studied

A 42-d feeding trial was initiated with weanling kits that ranged from 30 to 34 d of age. Two commercial breeds: Altex and New Zealand White were involved in the study. The Altex is a new sire breed with a selection history which emphasized heavy 70-d body weights at marketing (Lukefahr et al 1996). Two rabbits of each breed were randomly assigned to each of two pens as replicates for each of eight diets. A total of 64 rabbits and 16 growing pens were involved. Individual body weights were recorded on a weekly basis to determine growth responses in relation to diet and breed. Other growth traits included initial weight (IW), 42-d final weight (FW), and average daily gain (ADG). Also recorded was the age of rabbits when the minimum 1,800 g market weight (AGE) was achieved. Animals that failed to have an AGE record by 42 d were maintained on the same assigned diet and weighed daily until this could be determined. On a dry matter basis, total pellet (PI) and gross feed intake (pellets plus forage, GFI) and refuse weights were recorded daily and summed weekly by pen. Pellet and gross feed conversion (PFC and GFC) were also determined, calculated as PI or GFI divided by the total pen weight gain involving initial and final body weights. Market weight uniformity (CV) was calculated on a within-pen basis and by breed as the coefficient of variation in FW.

Rabbits from control (100% pellets with and without cactus) and 30% leucaena-fed (with or without cactus) diets and both breeds were slaughtered for carcass trait evaluation. All were slaughtered at a minimum body weight of 1,800 g (AGE). Although no animals died during the 42-d growth trial, one control animal succumbed later before AGE could be recorded. Hence, 31 rabbits were slaughtered. Rabbits were humanely killed by sudden cervical dislocation (Journal of the American Veterinary Medical Association, 1986). Carcass traits included hot carcass (CY), pelt, and emptied gastrointestinal tract (GIT) weights, expressed as percentages of pre-slaughter weight (PSW). Pelt weight was measured to reflect possible hair loss as a sign of leucaena toxicosis. Hot carcass weight excluded the head, skin, blood, viscera, heart, lungs, liver, kidneys, and abdominal fat. The GIT components (stomach [STO], cecum [CEC], and small and large intestines [SI and LI]) were also emptied and weighed, and expressed as percentages of PSW.

Statistical methods

Individual weekly body weight data were subjected to least-squares statistical analyses (Harvey, 1990) according to the following mixed-model:

Y_ijklm = µ + D_i + p_ji + B_k + W_l + (DB)_ik + (DW)_il + (BW)_kl + _ijklm (1)

where Y_ijklm = observed value of a given dependent variable; µ = overall mean; D_i = fixed effect of the ith diet; p_ji = random effect of the jth pen replicate nested within the ith diet; B_k = fixed effect of the kth breed; W_l = fixed effect of the lth week; (DB)_ik = diet x breed interaction; (DW)_il = diet x week interaction; (BW)_kl = breed x week interaction, and _ijklm = the random error.  From ANOVA, least-squares means were subjected to polynomial regression procedures (Blouin and Saxton 1990) to obtain prediction equations for plotting weekly body weights (ie: growth response curves) according to cactus and leucaena levels and also breed. For remaining growth traits (IW, FW, ADG, and AGE), sources relating to weeks (W) were eliminated from model 1. Pen traits recorded on a weekly basis (PI, GFI, PFC, and GFC) were analyzed according to the following model:

Y_ijkl = µ + D_i + p_ji + W_k + (DW)_ik + _ijkl (2)

For remaining pen traits (total 1-42 d PI, GFI, PFC, and GFC), sources pertaining to p and W were eliminated from model 2. For CV, this same reduced model was used but with the addition of B and DB (defined in model 1). Carcass and GIT-related traits were analyzed according to model 1, but excluding sources involving W. Following these analyses, pre-slaughter weight (PSW) was added to the model as a linear covariate to determine if diet and breed differences in carcass and GIT traits were affected by PSW. In preliminary analyses, it was revealed that individual diet and breed class regression coefficients were similar (P>0.05) for all carcass and GIT traits, so pooled regression analyses were performed.

From ANOVA, least-squares diet and breed means for all rabbit and pen traits studied were compared using a pre-planned set of orthogonal contrasts (defined in Tables 2 and 4) and tested at the = 0.05 probability level. Diet and breed contrasts were tested by the random pen and error sources, respectively.

Results

Chemical composition of diets

Dry matter (DM) values for fresh leucaena and cactus were 28.6 and 5.7%, compared to 89.4% for the commercial pelleted diet (Table 1). Crude protein was highest for leucaena (22.9%), intermediate for pellets (16.1%), and lowest for cactus (9.3%). Cactus had lower crude fiber and a considerably higher ash content than either leucaena or pellets. Values for NDF and ADF and also ether extract were comparable.

Table 1. Chemical composition of leucaena, cactus and commercial pelletsa (% dry matter basis)
	Leucaena	Cactus	Pellets
Dry matter	28.6	5.7	89.4
Crude protein	22.9	9.3	16.1
Crude fiber	18.4	10.5	17.7
Ether extract	6.0	5.9	7.7
Ash	7.9	24.1	9.3
Nitrogen-free-extract	44.8	50.2	49.2
Neutral detergent fiber	36.2	32.8	38.2
Acid detergent fiber	21.3	19.5	20.3
aAverage of duplicate samples.

Growth response to diets and breed

The best-fit regression equation (R²=0.99) for predicting growth response (Y) was a second-order polynomial model with interaction, as follows:

Y = 784.6 + 3.592(L) - 0.1275(L²) - 18.01(C) + 3.566(L*C) - 0.1654(L²*C) + 313.8(W) - 9.590(L*W) + 0.2193(L²*W) - 22.66(C*W) + 7.609(L*C*W) - 0.2196(L²*C*W) (3)

where codes representing levels of leucaena (L=0, 10, 20 or 30), cactus (C=0 or 1), and weeks (W=0, 1, 2, 3, 4, 5 or 6) were used. A second-order interaction involving leucaena, cactus, and weekly levels (L²*C*W) was detected (P<0.001). Datum points generated from the above equation were plotted separately according to diet groups that included or excluded cactus (Figures 1 and 2). Overall, initial weights (IW) were similar across leucaena dietary groups, but subsequent growth rate was highest for control groups. Also, growth was largely a linear function of weeks (P<0.001). In contrast, 42-d final weights (FW) decreased at a gradually increasing rate from 0 to 30% leucaena levels when cactus was offered (Figure 1). Whereas without cactus, FW declined sharply from 0 to 10%, decreased moderately by 20%, but increased slightly by the 30% leucaena level (Figure 2).

Although Altex compared to NZW rabbits had numerically heavier body weights throughout the 6-week growth phase, the fitted least-squares growth response lines were parallel (P>0.05).

Growth traits

There were no statistical differences for IW due to diet or breed (Table 2). Control groups (defined as average performance of rabbits fed 100% pellets with and without cactus) had more rapid ADG by 9.0 g/d (P<0.01), earlier AGE by 6.5 d (P<0.05), and heavier FW by 371 g (P<0.01) than experimental groups fed leucaena (averaged across 10, 20, and 30% levels with and without cactus). The inclusion of cactus did not generally influence growth traits (P>0.05). Rabbits fed 10% leucaena with or without cactus tended (P<0.10) to have more rapid ADG by 3.9 g/d and attain AGE earlier by 5.3 d, but were not different in FW compared to the average of rabbits fed 20 and 30% leucaena with and without cactus. In comparing 20 versus 30% leucaena-fed rabbits (with and without cactus), no statistical differences were detected for ADG, AGE or FW. Interactions between the above described control, leucaena, and cactus group comparisons, and also between diet and breed, were never significant for growth traits. In addition, growth trait performance was similar for Altex and NZW breeds (P>0.05), although the former had numerically heavier IW and FW by 44 and 122 g.

Table 2. Least-squares diet and breed means and selected contrasts for growth traits in rabbits#
Item	IW, g	ADG, g	AGE, d	FW, g
Dietb
C	786	45.1	56.9	2,677
L10	745	34.4	63.6	2,192
L20	819	30.5	67.8	2,100
L30	765	31.9	65.8	2,103
Cc	753	41.3	59.0	2,489
L10c	816	39.1	58.3	2,459
L20c	751	36.5	64.4	2,285
L30c	752	32.9	66.9	2,133
SEM	55	2.6	3.3	141
Breed
Altex	795	37.4	62.3	2,366
(NZW)	751	35.5	63.3	2,244
SEM	28	1.3	1.7	70
Selected contrastc
1	-5	9.0**	-6.5*	371**
2	10	-2.0	1.4	-73
3	9	3.9‡	-5.3‡	170
4	27	1.1	-0.3	74
Altex-NZW	44	1.8	-1.1	122
#Trait codes: IW = initial body weight, g; ADG= average daily gain, g/d; AGE = age at 1,800 g body weight, d; FW = final body weight at the end of the 42-d growth trial, g. bDiet codes: C = 100% commercial rabbit pellets or 90%, 80% or 70% pellets plus 10, 20 or 30% leucaena leaves (L10, L20, L30) fed without or with cactus (Cc, L10c, L20c, L30c) fed ad libitum. cContrast comparisons: 1 = controls (C and Cc) versus remaining experimental forage diets; 2 = diets excluding versus including cactus; 3 = 10% leucaena versus average of 20 and 30% leucaena diets (with and without cactus); 4 = 20 versus 30% leucaena (with and without cactus). ‡Approached significance at P<0.10; *significant at P<0.05; ** significant at P<0.01.

Pen growth and feed utilization traits

For sake of brevity, weekly pellet and forage intake response by diet are not presented as figures. Basically, total pellet and forage intake increased curvilinearly from 0 to 6 weeks for all diets involved. Control animals consumed more pellets than those on leucaena diets, as expected. As the study progressed, pellet consumption rates became increasingly higher for rabbits fed 10 or 20% leucaena with cactus versus 10 or 20% leucaena without cactus. However, pellet consumption rates were quite similar for 30% leucaena diets with or without cactus.

Control pens had 5,985 and 3,249 g higher PI and GFI, and had 0.38 better GFC than pens of rabbits fed leucaena with or without cactus (P<0.01; Table 3). Cactus versus non-cactus fed rabbits tended (P=0.05) only to have improved PFC (average calculated PFC was 3.40 and 3.57, respectively). Pens of rabbits fed 10% leucaena with and without cactus had higher PI and GFI by 2,332 and 2,297 g than those fed 20 and 30% leucaena with and without cactus (P<0.01), although there were no differences in measures of feed conversion (P>0.05). Pen performances were statistically similar between 20 and 30% leucaena diets. However, an interaction between control and cactus comparisons was detected for PI and GFI (P<0.05). Specifically, PI and GFI performance in control and leucaena pens depended on whether cactus was excluded or included in the diet (Table 3). An additional contrast testing the effect of excluding or including cactus in control pens (means of 26,960 and 24,688; Table 3) revealed that only PI was higher in the former diet (P<0.05).

Table3. Least-squares diet means and selected contrasts for pen traits in growing rabbits#
Dietb
Item	PI, g	GFI, g	PFC	GFC	CV
C	26,960	26,960	3.58	3.58	6.6
L10	20,453	22,975	3.53	3.97	8.0
L20	19,047	21,665	3.73	4.24	6.2
L30	18,420	21,411	3.44	4.00	10.1
Cc	24,688	25,261	3.56	3.64	8.5
L10c	22,336	25,812	3.40	3.93	7.0
L20c	20,369	23,623	3.32	3.86	15.4
L30c	18,412	21,685	3.34	3.93	12.1
SEM	666	724	0.10	0.12	3.8
Selected contrast ##
1	5,985**	3,249**	0.11	-0.38**	-2.3
2	-231	-843	0.17‡	0.11	-3.0
3	2,332**	2,297**	0.01	-0.06	-3.5
4	1,292‡	1,096	0.14	0.09	-0.3
#Trait symbols: PI = total pellet intake, g; GFI = pellet and forage (leucaena and/or cactus) gross feed intake, g; PFC = pellet feed conversion (PI/total pen weight gain); GFC = gross feed conversion (GFI/total pen weight gain); CV = within-pen 42-d final weight uniformity (coefficient of variation by breed). Forage intake was calculated on a dry matter basis. ##Diet treatments and contrast comparisons are defined in Table 2. ‡Approached significance at P<0.10; *significant at P<0.05; ** significant at P<0.01.

The CV was not significantly different among diets or between breeds (Table 3). However, controls were numerically the most uniform (CV = 6.6%) while rabbits fed 20% leucaena plus cactus were the least uniform (CV = 15.4%). Also, Altex were numerically less uniform than NZW (CV = 10.4 versus 8.0%).

Carcass and gastrointestinal tract component traits

Control animals had 2.0% higher CY (P<0.05), 0.85% lower GIT (P<0.01), and 0.14% lower STO (approaching significance at P=0.07) than animals fed 30% leucaena with or without cactus (Tables 4 and 5). However, when CY was adjusted to a constant PSW (regression coefficient related to a 0.00271% change in CY per g increase in PSW; P<0.05), the difference due to diet became smaller and non-significant (2.0 to 0.5%). Conversely, when STO was similarly adjusted (regression coefficient related to a -0.00023% change in STO per g increase in PSW; P<0.05), the difference became larger and highly significant (-0.14 to -0.25%). In other words, controlling variation in PSW resulted in larger diet differences for STO. In the same comparison, although the PSW covariate was non-significant, larger diet differences also resulted for CEC and LI (Table 5).

Table 4. Least-squares diet and breed means and selected contrasts for carcass traits in rabbits, not adjusted (na) and adjusted (a) to a constant pre-slaughter weight basis#
	CY		Pelt		GIT
Item	na	a	na	a	na	a
Diet##
C	49.3	48.0	11.1	11.3	5.06	5.25
L30	46.1	46.9	10.1	10.0	5.96	5.85
Cc	49.6	49.3	10.8	10.9	5.19	5.24
L30c	48.8	49.5	11.1	11.0	5.98	5.87
SEM	0.6	0.7	0.2	0.3	0.28	0.32
Breed##
Altex	48.6	48.3	10.6	10.6	5.61	5.67
NZW	48.3	48.5	10.9	10.9	5.49	5.44
SEM	0.5	0.4	0.2	0.2	0.20	0.21
Selected contrast
(C+Cc)-(L30+L30c)/2	2.0*	0.5	0.4	0.6	-0.85**	-0.61**
C - Cc	-0.3	-1.2	0.3	0.4	-0.14	0.01
L30 - L30c	-2.7*	-2.7*	-1.0‡	-1.0‡	-0.02	-0.02
Altex - NZW	0.3	-0.2	-0.4	-0.3	0.12	0.23
#Trait abbreviations: CY = carcass yield (hot carcass weight), %; Pelt = fresh pelt weight, %; GIT = emptied gastrointestinal tract weight, %. All traits expressed as a percentage of preslaughter weight. ##Diets and breeds defined in Table 2. ‡Approached significance at P<0.10; *significant at P<0.05; ** significant at P<0.01.

In the second contrast shown in Tables 4 and 5, carcass and GIT trait performance was similar (P>0.05) between control groups (pellets versus pellets and cactus). The last contrast shown in the tables indicated that rabbits fed 30% leucaena without cactus had 2.7% lower CY (P<0.05) and tended to have 1.0% lower pelt weight (P<0.10) than those fed 30% leucaena and cactus. However, GIT and its component organs were not influenced by the inclusion of cactus in 30% leucaena-fed rabbits. Also, the preslaughter weight adjustment resulted in no change in 30% leucaena diet mean differences for all carcass and GIT traits. The effects of breed and diet x breed interaction were never important (P>0.05) for any carcass or GIT traits studied.

Table 5. Least-squaresdiet and breed means and selected contrasts for emptied gastrointestinal tract components in rabbits, not adjusted (na) and adjusted (a) to a constant pre-slaughter weight basis#
		STO		SI		CEC		LI
Item		na	a	na	a	na	a	na	a
Diet##
C		1.16	1.06	1.81	1.85	1.16	1.09	1.13	1.06
L30		1.33	1.39	2.08	2.06	1.28	1.32	1.15	1.19
Cc		1.11	1.09	1.82	1.83	1.20	1.18	1.11	1.09
L30c		1.22	1.27	2.13	2.11	1.38	1.41	1.14	1.18
SEM		0.06	0.05	0.18	0.15	0.08	0.07	0.06	0.06
Breed##
Altex		1.24	1.22	2.03	2.04	1.23	1.22	1.15	1.13
NZW		1.17	1.19	1.89	1.89	1.28	1.29	1.11	1.12
SEM		0.04	0.04	0.11	0.11	0.06	0.05	0.04	0.04
Selected contrast
(C+Cc) - (L30+L30c)/2		-0.14‡	-0.25**	-0.29	-0.25	-0.15	-0.23**	0.02	-0.11‡
C - Cc		0.04	-0.03	-0.01	0.02	-0.04	-0.09	0.02	-0.03
L30 - L30c		0.11	0.11	-0.05	-0.05	-0.09	-0.09	0.01	0.01
Altex - NZW		0.08	0.03	0.15	0.15	-0.04	-0.07	0.05	0.01
#Trait abbreviations: STO = stomach, %; SI = small intestine, %; CEC = cecum, %; LI = large intestine, %. All traits expressed as a percentage of preslaughter weight. ##Diets and breeds defined in Table 2. ‡Approached significance at P<0.10; **significant at P<0.01.

Discussion

Based on chemical analyses (Table 1), compared to leucaena, the lower crude protein content of cactus (9.3%) is in agreement with the 11.4% value reported by Gregory and Felker (1992). The high ash content of 24.1% for cactus is consistent with the value of 20.5% for Opuntia sp. reported by Ensminger et al (1990). Crude fiber, NFE, and NDF values of 10.5, 50.2 and 3.8 are consistent with values of 13.2, 59.2, and 31.6, respectively, reported by Ensminger et al (1990). Shoop et al (1977) determined that Opuntia sp. had 40% more soluble carbohydrates and 55% more hemi-cellulose than alfalfa hay.

The present crude protein value of 22.9% for leucaena is similar to that (21.9%) reported by Raharjo et al (1986) but lower than those (28.0 and 27.8%, respectively ) given by the National Academy of Sciences (NAS 1977) and Keir et al (1997). No signs of hair loss or dermatitis nor discolored livers or urine were observed. Also, no changes in pelt weight were detected in 30% leucaena diet comparisons (Table 4). In agreement, Parigi-Bini et al (1984) reported no signs of mimosine toxicosis in rabbits fed 30% leucaena meal. In rabbits fed 40% leucaena leaf meal, Harris et al (1981) noted reddish-brown colored urine. Onwudike (1995) reported dark-colored livers and reddish-brown urine, as well as degenerative changes in kidney and liver tissues, in rabbits fed fresh leucaena forage ad libitum.

Reduced growth performance in forage-fed rabbits may have been attributable to poorer diet quality and/or the adverse effects of toxic compounds, such as mimosine and trypsin and chymotrypsin inhibitors, and tannins present in leucaena (Mtenga and Laswai 1994). Onwudike (1995) similarly reported poorer growth in rabbits fed increasing levels (0 to 100%) of fresh leucaena. However, growth performance was generally improved at the 10 and 20% leucaena levels when cactus was included than when it was excluded in the diets, although this contrast is not shown in Table 2. However, pellet consumption also increased in 10 and 20% leucaena diets when cactus was offered. This result might be explained by one or more of the causative factors: complementary nutritional effect (ie: high protein level in leucaena and high soluble carbohydrate level in cactus), higher dry matter intake of pellets and forages, or mimosine absorption by pectin and mucilage in cactus. Possibly other factors exist as well. Pectin and other gel-forming polysaccharides apparently affect digestion and absorption of dietary protein and amino acids (Rainer et al 1994). Gallaher et al (1993) reported that compounds that increase the viscosity of intestinal contents may effectively reduce plasma cholesterol and that only moderate viscosity is necessary to achieve this effect. However, these beneficial factors associated with cactus, if real, may not have been as effective at the 30% leucaena level. Further research is needed to substantiate these nutritional inter-relationships.

The daily intake of leucaena forage was 15.0, 15.6, and 17.8 g of dry matter per rabbit 10, 20 and 30% leucaena diets, respectively (calculated from Table 3). When leucaena and cactus were combined, gross forage consumption increased by 38.0, 24.4, and 9.6% in 10, 20, and 30% leucaena diets, respectively. In addition, when only cactus was included with pellets, the daily dry matter intake of cactus per rabbit was 3.4 g (59.8 g as-fed basis), reflecting the high moisture content. In a preliminary trial, palatability was quite poor after cactus strips had been dried for 24 hours or longer. Certainly, the level of leucaena consumption suggests that leucaena was palatable (Table 3). As rabbits have nocturnal feeding habits, on many mornings during the study, some pellets remained in the feeders while large quantities of leucaena had been consumed. However, the high moisture contents of the forages (especially cactus) may have precluded the animal's ability to meet its total dry matter feed intake requirements. Basically, as the study progressed, control animals continued to grow more rapidly, their feed intake level increased concomitantly, contributing to proportionately larger body weight differences than in forage-fed animals.

Control animals had lower GIT weight percentage than 30% leucaena-fed animals (Table 4). Muir and Massaete (1991) and Ruiz-Feria et al (1996) reported an enlargement of the GIT in rabbits fed fresh forages (cactus and mesquite, Prosopis glandulosa) with pellets versus only pellets. In the present experiment, the emptied GIT components had consistently and numerically larger volume capacity in 30% leucaena-fed rabbits compared to control rabbits (Table 5). Fraga et al (1991) proposed that fresh forages stimulate stomach growth, which accounted for subsequent higher feed intake capacity compared to rabbits fed only pellets. Hoover and Heitmann (1972) reported higher cecal volume when diets contained a greater quantity of fiber. Pote et al (1980) reported improved feed conversion when pellets were restricted and supplemented with a variety of green forages compared to pellet-fed controls, but CY was not different.

Altex were heavier numerically than NZW throughout the experimental, post-weaning growth period, although statistical significance was not achieved. Previously, Ruiz-Feria et al (1996) reported that Altex had significantly heavier FW by 203 g than NZW. Breed differences were also not observed for carcass or GIT traits. However, Ruiz-Feria et al (1996) reported significantly larger GIT for Altex than for NZW, although CY was similar.

Conclusions

There is a need to identify suitable forages that will support low-cost meat rabbit production, especially in the lesser developing countries. While improved growth performances are often realized in rabbits typically fed commercial broiler or swine mash or pelleted diets (Lukefahr and Goldman 1987), subsistence farmers generally lack the capital and transportation means necessary to acquire commercial feeds. Of relevance, labor costs associated with the collection and feeding of forages is negligible because this activity is usually shared among family members. Also, extensive-scale producers often reduce commercial feed costs by feeding forages to increase profits. It is recommended that the feeding level of leucaena (within a 0 to 30% range) should depend on economics in terms of realized feed cost savings in relation to growth response.

References

Received 20 April 1998

Return to top