Evaluation of tree lucerne (Chamaecytisus palmensis) dried leaves as a substitution for concentrate mixture on biological performance and socioeconomic of Washera sheep fed on desho grass hay

Abstract Background The experiment was conducted to evaluate the effect of tree lucerne dried leaves (TLDL) as a substituting supplement for noug seed cake, and wheat bran concentrate mixtures (CM) on feed intake, digestibility, growth, carcass characteristics and economic benefits using Washera sheep fed desho grass hay as a basal diet. Methods Twenty‐five male Washera sheep with an initial body weight of 16 ± 4.01 kg (mean ± SD) were used for both growth and digestibility trials which lasted for 90 and 10 days, respectively The experiment was conducted using a randomized complete block design having five blocks with five treatments (T1 = 0% TLDL + 100% CM; T2 = 25% TLDL + 75% CM; T3 = 50% TLDL + 50% CM; T4 = 75% TLDL + 25% CM and T5 = 100% TLDL + 0% CM). Results Significant (p < .05) differences were observed among treatments in total dry matter (DM), nutrients and metabolizable energy (ME) intake; as the level of TLDL increases total DM and nutrient intake decreased (p < .05) except neutral detergent fibre intake (NDFI) and acid detergent fibre intake (ADFI). Digestibility of DM, organic matter (OM) and crude protein (CP) were significantly reduced (p < .001) among the treatments increasing the substitution level of TLDL for local concentrate mixture, but non‐significant for fibre fractions digestibility. Similarly, body weight, average daily gain and feed conversion efficiency were significantly different (p < .01) and highest in T1 than other treatments, whereas lowest values of the same parameters were observed for T5. Non‐significant difference among treatments was observed in almost all of the carcass characteristics except for rib‐eye area that showed significant difference among treatment and was highest in T1 and T2 (low level of TLDL inclusion). The economic analysis showed that supplementation TLDL was economically feasible in which the net return of the treatments were 31.66 (T1), 30.68 (T2) and 30.34 (T3), 27.88 (T4) and 26.32 (T5) USD (United States Dollar). Conclusions It was concluded that TLDL could be used as alternative feed source by replacing concentrate mixture up to 75% (T4) which enhanced nutrient intake, digestibility and growth performance of Washera sheep. Finally, TLDL supplementation is recommended as replacement of CM up to 75% for its biological and economic feasibility.

grown in Northwest Ethiopia (Getnet, 1998) and has high protein source and it can be an alternative crude protein (CP) supplement for concentrate mixtures. The latter is usually are inaccessible and unaffordable by the majority of smallholder farmers in Ethiopia. Getu et al. (2012) reported that tree lucerne foliage has a 22% CP which is higher than the CP content in wheat bran and lower than the CP content in noug seed cake (Guizotia abyssinica). Because of its CP content tree, lucerne could be included as supplement feed.
Smallholder farmers in the highlands of Ethiopia are currently participating on the Tree lucerne cultivation both in the backyard and natural resource conservation areas for purposes of livestock feed and environmental management (Tadesse, 2016). The utilization of these fodder species as livestock feed, however, is low due to inadequate information particularly about the supplementary value of these feeds.
The supplementary potential of a feed resource is chosen based on both in terms of availability and biological response. In the current study biological variables or animal response variables such as nutrient intake, digestibility, growth and carcass characteristics as well as economic parameters were considered due to the fact that these variables are more important for tropical feed evaluation. Moreover, it was difficult to consider other variables because of shortage of facilities and financial limitations. Therefore, the objectives of this study were to evaluate tree lucerne (Chamaecytisus palmensis) dried leaves as a substitution for concentrate mixture (wheat bran and noug seed cake) supplement on intake, digestibility, growth and carcass characteristics of Washera sheep fed on desho grass hay as a basal diet.

| Experimental feeds preparation
Desho grass (Pennisetum pedicellatum) was harvested and dried. The grass hay was manually chopped to the size of approximately 5-10 cm in order to facilitate feed intake and minimize wastage then stored under shade to maintain its quality. The sheep were offered ad libitum at 20% refusal rate adjusted every week and provided with free access to clean water throughout the experimental period. Tree lucerne (Chamaecytisus palmensis) leaves were harvested from the plantation grown in natural resource conservation area and dried under the shade (air dried). The supplement ratio was made based on their crude protein contribution for growing lamb weight 10-20 kg CP needs range from 16.3% to 26.4% per day recommended by Ensminger (2002).
Noug seed cake (NSC) and wheat bran (WB) were purchased from Bahir Dar city oil making factory and floor industry respectively. Diet formulation from these feed ingredients such as TLDL, NSC and WB was done based on the iso-nitrogenous basis in crude protein content (Ranjhan, 2001). The NSC and WB concentrate mixture was 1:2 ratio.

| Experimental sheep and their management
Twenty-five yearling male Washera sheep with similar initial body weight 16 ± 4.01 kg (mean ± SD) were purchased from local market. The experimental animals were ear tagged for identification and quarantined for 15 days in well-ventilated sheep barn equipped with feeding and watering troughs. Sheep were vaccinated against sheep Conclusions: It was concluded that TLDL could be used as alternative feed source by replacing concentrate mixture up to 75% (T4) which enhanced nutrient intake, digestibility and growth performance of Washera sheep. Finally, TLDL supplementation is recommended as replacement of CM up to 75% for its biological and economic feasibility.

K E Y W O R D S
daily weight gain, digestibility, fodder shrub, ovine, tree lucerne pox, anthrax and ovine pasteurellosis and injected with multivitamin before commencement of actual experimental period.

| Experimental design and treatments
The experiment was conducted in a randomized complete block design (RCBD) with five treatments and five replications. The sheep were divided into five blocks based on their initial body weight which was determined by overnight fasting (to avoid the effect of feed consumed on the weight of animals) and randomly assigned to one of five dietary treatments by using lottery method. The dietary treatments used in the study were presented in Table 1.

| Digestibility trial
Digestibility trial was conducted before growth trial. Each sheep were fitted with faecal collection bags for 3 days of acclimatization period followed by a total collection of faeces for 7 consecutive days. Fresh faeces were collected and recorded daily in the morning throughout the digestibility trial. From daily faecal output of each sheep, 20% of the representative samples were taken and frozen at −20ºC, and pooled over the collection period for each sheep.
At the end of the collection period, each sample was taken out from the freeze and allowed to thaw. After thawed, the sample was thoroughly mixed sub-sample taken and dried at 60ºC for 72 hr. Then grounded into 1 mm sieve size and stored in plastic bags and taken to laboratory for chemical analysis. The apparent digestibility coefficients of different nutrients were calculated by using the following equation: Where: DM = dry matter; DMI = dry matter intake.

| Growth trial
The 90 days of growth trial was conducted next to digestibility trial.
Feed offered and refusals for each sheep were collected and recorded throughout the experimental period to determine feed intake. Daily feed intake was measured as differences between offered and refused. The sheep were offered ad libitum desho grass hay, at 20% (Hozza et al., 2014;Tekliye et al., 2018) refusal throughout the experimental period. Samples of feed offered were collected per batch, whereas samples of refusal were taken from each sheep daily and pooled per treatment over the experimental period and stored in plastic bags used for chemical analysis.

| Body weight change and feed conversion efficiency
Body weights were taken every 10 days after overnight fasting before feed provision. Average daily body weight gains (ADBWG) was calculated as the difference between final and initial live weight divided by the number of feeding days. According to Brown et al. (2001), the feed conversion efficiency (FCE) was calculated as unit of body weight gain per unit of feed consumed by the formula;

| Carcass parameters evaluation
At the end of the growth trial, all experimental sheep were slaughtered after overnight fasting to determine the effects of treatment feeds on carcass parameters. Prior to slaughter, body weight was measured to determine the slaughter body weight of each sheep.
Empty body weight (EBW) was calculated as the difference between slaughter weight and gut content.
The hot carcass weight (HCW) was estimated after removing the weight of the head, thorax, abdominal, skin and pelvic cavity contents. Dressing percentage was calculated as the proportion of hot carcass weight to slaughter body weight. To evaluate the ribeye area, the average of the right and the left side measurement at the 11th and the 12th rib was taken. Total edible offal components (TEOC) were calculated as the total sum of the edible offal components. Total non-edible offal components (TNEOC) were calculated as the total sum of the non-edible offal components. All procedures followed were in accordance with the ethical standards of the responsible committee on animal experimentation.

| Chemical composition analysis
The feeds offered, refused and faeces samples of the experiment were analysed for DM, OM, CP and Ash contents follow-

| Statistical analysis
The data collected on feed intake, digestibility, body weight and carcass were subjected to analysis of variance (ANOVA) using the

| Partial budget analysis
The partial budget analysis was performed using the procedure of Upton (1979). It involved the variable cost of purchasing sheep, feeds and selling price of sheep. Total return (TR) was determined by subtracting the selling and purchasing price of sheep. The net income (NI) was calculated as subtracting total variable cost (TVC) from the total return (TR).
The change in net income (∆NI) was calculated as the difference between changes in total return (∆TR) and the change in total variable costs (∆TVC) and calculated as ∆NI = ∆TR−∆TVC. The marginal rate of return (MRR) that measures the increase in net income (∆NI) associated with each additional unit of expenditure (∆TVC) was calculated as MRR = (∆NI/∆TVC). The flow chart of current study is shown in Figure 1.

| Chemical composition of experimental feeds
The chemical composition of experimental feed and refusals are presented in Table 2. In the current study, the CP content of offered desho grass hay and tree lucerne dried leaves (TLDL) were 11.40 and 23.96%, respectively. As expected, higher fibres were recorded in desho grass hay than other feed ingredients. Similarly, the leftover of grass hay was higher in fibres than the offered, but less in crude protein content.

| Dry matter and nutrients intake
Significant (p < .05) difference were observed among treatments except DMI% body weight, total NDFI and total ADFI of the daily nutrients intake (Table 3); as the level of TLDL increase in the diet of the TDM, ME, TOM and TCP intake were in decreasing order.
The highest nutrient intake (813 g/day) was recorded in sheep assigned in T1 and lowest (740 g/day) in T5 (sole TLDL supplement).

F I G U R E 1 Flow chart of the study
In the current study, there was no significant (p > .05) differences in nutrient intake among the TLDL-supplemented groups. All treatments group had similar (p > .05) NDF and ADF intake, but high level (103 g/day) of lignin consumed in the sole TLDL assigned sheep. The ADL intake significantly (p < .05) increased as the level of TLDL increased; Sheep fed sole TLDL had higher consumption than the groups fed non-TLDL diet. A trend of total DM intake is shown in Figure 2.

| Apparent dry matter and nutrient digestibility
The apparent dry matter and nutrients digestibility percentage of experimental feeds are shown in Table 4. Like the daily dry matter and nutrients intake, digestibility percentage of dry matter and OM and CP nutrients significantly (p < .05) decreased as the level of TLDL increased in the diet. For these nutrients, the higher nutrients digestibility was for T1 and lowest for T5. Non-significant (p > .05) values were observed for NDF and ADF values.

| Body weight and feed conversion efficiency
Significant (p < .05) on body weight change (BWG), average daily gain (ADG) and feed conversion efficiency (FCE) were recorded and presented in Table 5

| Correlation among nutrients intake, digestibility and daily weight gain of Washera sheep
Dry matter intake was positively and strongly correlated (p < .0001) with OM intake as well as OM digestibility (Table 6).
DMI was also positively moderately correlated (p < .01) with

| Carcass characteristics
Non-significant (p > .05) differences were observed among treatments in most of the carcass characteristics of Washera sheep fed desho grass hay basal diet supplemented with different levels of TLDL and CM mixture except for rib-eye area (Table 7).
Sheep fed on T1 and T2 relatively had better (12 cm 2 ) rib-eye area (REA) than the sole TLDL-supplemented group (T5). Nonsignificant (p > .05) difference recorded among the groups supplemented with higher levels of TLDL which are in the T3, T4 and T5. (Table 8). In the current study, rib eye muscle and ribs were significant (p < .05) with the higher value for lower TDL supplemented groups (Table 8).

| Non-edible offal components
The non-edible offal components of Washera sheep fed desho grass hay as a basal diet and supplemented with tree lucerne dried leaves and concentrate mixtures at different proportion are presented in

| Partial budget analysis
The partial budget analysis results of Washera sheep fed desho grass hay as a basal diet and supplemented with TLDL and concentrate mixtures in different proportion are presented in
The CP content of tree lucerne dried leaves was in the range of values (21.6%-27.8%) as reported by Dereje (2014) but higher than (16.6%) the reports of Meron (2016). The NDF, ADF and ADL content of Tree Lucerne leaves were in line with the previous study (Getu et al., 2012). The DM, OM, CP, NDF, ADF and ADL content of the Desho grass hay refusals were almost similar among treatments.
This might be due to the feeding habits of sheep which selected the most nutritious parts of the grass such as leaves, shoots and left relatively less edible parts like stems of the grass (Alemu, 2016 (Solomon, 1992). The metabolizable energy (ME) requirement for a 10 kg live weight sheep has been documented to be 2 MJ/day for maintenance and 2.7-4.5 MJ/day ME for 50-150 g/day gain (Chesworth, 1992).

| Dry matter and nutrients intake
In the current study, ME in the feed was in line with the finding of

| Apparent dry matter nutrient digestibility
Regarding digestibility of nutrients, there was lower digestibility of DM, OM and CP in sheep fed only tree lucerne dried leaves compared with all the treatments which might be due to relatively lower energy and higher fibre contents of the feed. It has been stated that there would be variation in DM and OM apparent digestibility nutrients in a given diet that arise from variation related to the age of the animals, level of feeding, feed nutrient content and ration composition (Ranjhan, 2001).
In the current finding, the DM digestibility was in line with the value which ranges from 0.67% to 0.52% (Dereje, 2014)

TA B L E 11
Partial budget analysis the current finding was in agreement to the result of McRae and Armstrong (1969) who reported that supplementation had little or no effect on digestibility of NDF.

| Body weight change and feed conversion efficiency
In the present study, insignificant results for initial and final body weight recorded were similar to the reports of Desta et al. (2017) for Abergele rams in a study conducted to substitute local concentrate mix with dried mulberry (Morus indica L.) leaf meal. The body weight change, average daily gain (ADG) and feed conversion efficiency values were in agreement with the findings of Mekonnen et al. (2017) for a research that focused on the effect of Mulberry Leaf meal supplementation for Ethiopian Highland sheep.
The lower body weight of sheep in T5 indicated that tree lucerne dried leaves provided lowest metabolizable energy (5.2 MJ/ day) mainly for maintenance requirement of sheep rather than body weight gain. This was supported by Gatenby (2002) who stated that the lowest energy at which the sheep do not lose weight is between 8 and 10 MJ/kg per head/day. In addition, the minimum protein level required for maintenance is about 8% CP in the DM and the most productive animals such as rapidly growing lambs and lactating ewes need about 11% CP (McDonald et al., 2010). The lower weight gain of sheep supplemented with solely tree lucerne leaves could be due to the inefficient digestibility of TLDL associated with the high fibre content than concentrate mix (Dereje, 2014).
In the current study, the mean daily body weight gain was relatively comparable with reports of Awet and Solomon (2009)

| Correlation among nutrients intake, digestibility and daily body weight gain of Washera sheep
Sheep fed on higher protein diet consumed more total DM than sheep kept on low CP diet (Fluharty et al., 1999). This is in agreement with the results reported by Aschalew and Getachew (2013) for Farta sheep. The positive associations between total DM and CP intakes were also reported by Awet and Solomon (2009) for Afar sheep. The NDF and ADF intake were negatively and non-significantly correlated with the digestibility of DM, OM and CP. This implies that a high intake of fibre fractions reduced the digestibility of DM, OM as well as CP, which in turn reduces intake of feed (Martens, 2006).

| Carcass parameters
The relatively higher SW, EBW and HCW in T1 than other treatments might be due to the fact that concentrate mix can easily be digestible and higher nutrients consumption by sheep in T1 which was consistent with ADG and feed conversion efficiency of sheep in the treatment. Similar finding was reported by Desta et al. (2017) for Abergele rams supplemented with substitution of dried mulberry (Morus indica L.) leaf meal for concentrate mix. In the present study, the SW values were relatively heavier than the results of Dereje (2014) and Tekliye et al. (2018) 17.2-20.6 and 17.52-22.60 kg for local and Farta sheep, respectively, supplemented with Tree Lucerne Sesbania sesban fodder leaves. Hot carcass weight was relatively comparable with reported values of 5.6-9.6 and 9.75-11 kg for Arsi-Bale and local sheep reported by Abebe and Yoseph (2015) and Hagos (2015) (William et al., 2003). The rib-eye area muscles were consistent with the results of Alemayehu et al. (2015) and Gebregiorgis et al. (2017) fed Leucaena leucocephala leaves and pods and wheat bran for Centralhighland sheep. Higher rib-eye muscles area was observed in T1 12.6 cm 2 compared with other treatments. This indicates that the nutritional quality of supplementation would enable sheep to develop better muscling than the other group (Amare, 2007). This might be due to the higher nutrients provided by the supplements with CM and higher slaughter weight. The rib-eye area (REA) was relatively comparable with the result of Girma and Mengistu (2017)

| Main carcass components
Rib and rib-eye muscles were relatively higher groups in T1 than other treatments. This might be due to sheep in this treatment could have consumed high quality of feed and easily digestible that contain protein and energy (Kouakouet al., 1997) who reported that the weight of some visceral organs is affected by the level of nutrition.

| Edible offal components
Liver, Kidney, Tongue and Omaso abomasum were higher in sheep in T1 and lower T4 and T5. This might be due to high metabolic activities of the organs and high intake of energy and protein from concentrate supplemented sheep in T1. The non-significant differences in most of the alimentary canals could be linked with organs being early maturing and consequently being little affected by dietary treatments (Hag & Shargi, 1996).
Total edible offal components and total usable products were relatively higher in T1 (3.3 kg) sheep fed with concentrate mix only compared with others, even though non-significant among the treatments. In agreement with previous studies (Abebe et al., 2010;Hailu et al., 2014;Hirut et al., 2011;Meron, 2016)

| Non-edible offal components
Among non-edible offal components, skin weight of T1 (2.5 kg) was higher than T5 which might be due to the better subcutaneous layer fat deposition due to better feeding regime sheep supplemented with concentrate mixture and better growth of hair (Lawrence & Fowler, 1998).

| Partial budget analysis
The net return of supplemented sheep in the current study was relatively higher compared to the results reported by previous workers (Melese et al., 2014;Tesfaye, 2008) which was in the range of 52.00-82.16 and 60.10-153.20 ETB/head for different sheep breeds fed various basal and supplement diets. The observed difference in net return might be due to the variations in purchasing and selling price of sheep. The sheep in T5, however, gained lower BW as a result of lower nutrient (CP) and energy with higher fibre content intake that consequently resulted in a lower net return. The MRR result achieved in the present study was higher than 0.9-1.9 and 1.2-2.0 reported by Melese et al. (2014) and Tesfaye (2008) for Washera and Arsi-Bale sheep, respectively, fed different basal diet and supplements. This might be due to variations in sheep breeds and differences in basal diet and supplements used in different experiments.
As a limitation of this study, only few common variables were considered in this study due to constraints like laboratory and other inputs.

| CON CLUS ION
From the present study, it could be concluded that substitution of tree lucerne dried leaves by conventional concentrate mix (NSC & WB) up to 75% inclusion because it satisfies the nutrient requirement of fattening sheep which were biologically efficient and economically profitable. For commercial producers, however, up to 50% inclusion of TLDL in the diet of growing local sheep could be recommended. Therefore, TLDL could sustainably serve as alternative CP supplements to the roughage feed stuffs as a basal diet for ruminant animals in the smallholder farming systems where concentrate mix is not affordable and accessible.

S TU DY A R E A
The study was conducted at the College of Agriculture and Environmental Sciences, Bahir Dar University, Ethiopia. The area is geographically located at 11° 37′ N and 37° 28′ E with an elevation of 1912 meter above sea level. The average daily minimum and maximum temperatures were 7 and 29 • C respectively. The average annual rainfall ranges were from 1,430 to 1,520 mm.

ACK N OWLED G M ENTS
The first author acknowledges Bahir Dar University College of Agriculture and Environmental Science for allowing her to pursue this research through her MSc study.

CO N FLI C T O F I NTE R E S T
The authors declare that they have no conflict of interest. Writing-review & editing.

PE E R R E V I E W
The peer review history for this article is available at https://publo ns.com/publo n/10.1002/vms3.376.