Early growth and survival rates of crossbred lambs ( Dorper x indigenous ) under semi-intensive management at Areka , Southern Ethiopia : Effects of non-genetic factors

Stagnant early growth and poor lamb survival are the major constraints of sheep production in Ethiopia. The aim of this study was to evaluate early growth and survival rate of crossbred lambs (Dorper × indigenous, Adilo) under semi-intensive conditions. Body weight (BW, kilograms) at 0-180 days, average daily gain (ADG, g/day), pre(0-90), post (90-180) and overall (0-180 days), and preweaning survival rate (PSR) of crossbred (50%) lambs (n=305) were evaluated. The non-genetic factors (season, parity, sex of lambs, litter size) were also determined. The least squares means (±SE) of BW at birth, 30, 60, 90, 120, 150 and 180 days of age were 2.6± 0.63, 5.6± 0.12, 8.7± 0.18, 11.6±0.23, 15.0±0.46, 17.2± 0.31, and 18.4±0.26 kg, respectively. The crosses were higher by 11.9% and 9.3% at 90 and 150 days, respectively, than indigenous Adilo lambs. There was clearly evident effect of season on body weights at various ages, preand post-weaning and overall ADG. Parity influenced weight at 30, 60, and 90 days, and ADG (pre, post and overall). The litter size (1.68 ± 0.6) consistently affected BW at all ages, pre-weaning and overall ADG. Weight at birth, 30 and 90 days, and ADG from 30-60, 60-90 and 120-150 days were affected by parity-by-litter size interaction. The PSR rate (90.2%) was influenced by all nongenetic factors except sex. The improvement in litter size, body weight, and survival represents potentially significant economic advantage of crossbred over local sheep. Managing dam age through replacement ewes, and improving nutrition and litter size would improve lamb survival and growth that enhances total lamb output per ewe per year.


INTRODUCTION
Small ruminant production is an integral part of mixed farming system of smallholder farmers in Southern Ethiopia.It plays a significant role in creating employment opportunities, income generation, capital reserve, and improving household nutrition (Kocho et al., 2011;Teklebrhan et al., 2014).Ethiopia's sheep population is estimated at 24 million heads categorized into 18 to 19 sheep breeds or populations (Gizaw et al., 2011).
Adilo sheep, named after routes to terminal market, is widely distributed in semi-arid to sub-humid areas of Central Southern Ethiopia.Generally, Adilo sheep is relatively productive than many other sheep populations or breeds of the region due to their higher prolific capacity (Deribe et al., 2014).In most parts of the country, the major objective of sheep rearing is for lamb production and is the main source of income for smallholder farmers (Kocho et al., 2011).As lamb marketing price depends mainly on the animal's body weight, improvements in lamb birth weight, pre-weaning and post-weaning growth rates are of interest (Gootwine et al., 2008).This could be achieved through optimization of growth (Deribe et al., 2014) and reproduction and survival traits (Gavojdian et al., 2013).To improve productivity of the indigenous breeds, Ethiopian Sheep and Goats Productivity Improvement Program (ESGPIP) imported tropically adapted sire breed (Dorper sheep).Dorper sheep breed is generally documented as one of the most popular mutton breeds in South Africa (Fourie et al., 2009).The Dorper rams can be used as terminal sire to produce lambs with high growth rates, better post-weaning feed efficiency, and superior carcass traits (Gavojdian et al., 2013).The advantage in growth and body composition of crossbred lambs over pure bred dam line lambs results from the additive contribution of the sire line and from the heterosis effect (Gootiwne et al., 2008).Evaluation of growth and survival performance of Dorper sheep and their crosses, under various and prevailing environment is essential before their wider use under smallholder management conditions.This study, therefore, was aimed at evaluating early growth performance and survival rate of crossbred (50%) lambs (Dorper ram × local, Adilo-ewes).It was hypothesized that non-genetic factors would not suppress growth and survival rate of crossbred lambs under semi-intensive management conditions.

The study site
This study was conducted at Areka research sub-site, Boloso district of Southern Ethiopia, 309 km South of Addis Ababa.The area is situated at 7° 06'N latitude, 37° 47'E longitude, and altitude of 1772 m above sea level.It was characterized by semi-arid to sub-humid agro-ecological zones.The sub-site is a microenvironment suitable for production of diverse grass and legume species.Major cultivated grasses adapted in the area include Chlorias gayana, Panicum maximum, Brachiaria brizantha ('desho' grass) and Pennisetum purpureum while major legumes are Medicago sativa, Vigna unguliguleta, Lablab purpureous and Sasbania sasban.

Data collection
Four Pure Dorper rams were introduced to Adilo ewes flock in 2012, end of rainy season.64 local Adilo ewes were categorized into 4 mating groups.The rams were 18 months old with a mean weight of 46 ± 3 kg.The rams were crossed with Adilo ewes (indigenous sheep) selected based on their reproductive traits, with initial body weight of 19.9 ± 2.30, reared under similar management conditions.The ewes were housed indoors during last trimester, early lactation period, and part of winter periods.Animals had ad-libitum access to a medium-quality Rhodes and Brachiaria hay, water, mineral and shade.Ewes were kept on a natural pasture with the average stocking rate of 6 to 8 heads ha -1 .During indoor housing, ewes received medium-quality Napier and Rhodes grass hays ad libitum, as well as 200 to 300 g concentrate per head, based on physiological status of ewes.Lambs were allowed to suckle their dams from birth until weaning (90 days).Supplementary concentrate feeds containing 120 g CP kg -1 DM on average was provided to all lambs.Within 24 h of parturition; date of birth, birth weight, litter size, sex of lamb and dam parity were recorded.All lambs were weighed at birth and then at weekly interval up to 180 days of age.Mortality of lambs at pre-weaning (birth until 90 days of age) was also recorded.

Data analysis
Data were subjected to the general linear model (GLM) procedure of SAS (SAS, 2008).In the analysis, all non-genetic factors were considered as fixed effects except the error term considered as random effect.Where F test declared significant, Tukey's test was used to compare least squared means (LSMs).The following linear model was employed to analyze the traits: where Yinjmo is the observation on production traits: birth, weight at various ages and average daily gain (ADG) (excludes birth weight category); pre-weaning survival rate (includes birth weight category): µ = is the overall mean; xn = is the fixed effect of n th lamb sex (n=male, female); pj = is the fixed effect of j th ewe parity (j = 1, 2, 3, 4); bm = is the fixed effect of m th type of birth of lamb (m=1=single, 1=twin, 3=triple); yb = the effect of m th birth weight of lamb {b=1 (≤ 2 kg), 2(2 to 2.5 kg) and 3 (≥ 2.6 kg)}; (px)jn = the interaction effect of parity and sex on body weights and ADGs; (pb)jm = is interaction effect of parity and litter size on body weights and ADGs; εnjmbo = effect of random error

Body weights at various ages
The non-genetic factors considered in this study were shown to affect body weights of crossbred lambs at birth and various ages (Table 1) and trends of weight changes (Figure 1).The least squares means and ±standard error *Corresponding author: deribeg2000@yahoo.com.
Season is one of the most important sources of variation on lamb birth weight and weight at various ages (Figure 1).Lambs born during the heavy rain season demonstrated higher birth weight (P<0.01) and 120 (P<0.001), and those born during the light and heavy rain seasons attained higher body weights at 30 (P<0.001), 90 (P< 0.001), 150 and 180 days (P<0.001) compared to the dry season.
Litter size (1.68 ± 0.6) influenced body weights of crossbred lambs at all ages consistently (Table 1).Single births showed consistently higher body weights over their twin and triplet counterparts.Likewise, twin births were higher compared to triplets at birth (P<0.001) and 30 days (P<0.001) while non-significant (P>0.05) in the remaining ages.There was significant effects of sex on body weights at birth (P<0.05),non-significant from 30 to 90 days (P>0.05) and the influence resumed after 90 days (weaning) (Table 1).Male lambs were superior over their female counterparts at birth, 120, 150 and 180 days.
Parity (dam age) due to differences in ewe weight influenced body weights of lambs at 30, 60 and 90 days (Table 1).Lambs born from the 2nd parity attained significantly higher (P<0.001)body weights at 30 days compared to those from the 1st, 3rd and 4th parities.Likewise, lambs from the 1st and 2nd parity demonstrated higher (P<0.001)60 day weight compared to 3rd and 4th parities.Similarly, lambs from 3rd and 4th parities attained significantly lower body weights at 90 day (P<0.01)and 120 days (P<0.05),respectively.There was parity-by-litter size interactions on birth weight (P<0.001),30 (P<0.01) and 90 days (P<0.05).Single births from 3rd and 4th parities exhibited higher weights over single births from the 1st and 2nd parities and multiple births from the 3rd and 4th parities.

Average daily gain
All fixed factors considered in this study were shown to influence average daily gain at some stages of lamb ages (Table 2).Pre-weaning, post weaning and overall ADG were affected all fixed factors except sex at pre-weaning and litter size at post weaning (Table 3).The least squares means and SE for ADG from birth to 30, 60, 90, 120, 150, and 180 days of age were 101±3.2,109±3.5, 98.6±3.1, 122±13.7,80.3±4.5, and 76.4±4.1 kg, respectively.The pre-weaning, post-weaning and overall ADG were 99.8±2.4,85.9±2.0 and 96.1±1.5 g/day, respectively.Lambs born during the heavy rain season had higher ADG at 30-60 days (P<0.01) and 90-120 days (P<0.05)compared to the dry season (Table 2).Conversely, lambs born during the dry season had higher ADG at 60 to 90 (P<0.01)compared to the heavy rain season.Likewise, lambs born during the dry and light rain seasons grew faster (P<0.05) from 150 to 180 days than those during the heavy rain season.The influence of season was also important on pre-weaning ADG (P<0.01),post-weaning ADG (P<0.05) and overall ADG (P<0.01).Litter size demonstrated variation on ADG at early ages of lambs: birth to 30, 30-60, 60-90 and 90-120 days (Table 2), pre-weaning, post weaning and overall ADG (Table 3).Single births were consistently heavier over their twin and triplet contemporaries from birth to 30 (P<0.001), 60-90 (P<0.001) and 90-120 days (P<0.05).
Similarly, single and triple births grew faster (P<0.001)than twin from 30-60 days.Single births showed faster (P<0.001)pre-weaning ADG in comparison to twin and triplets, while single and twin births had heavier (P<0.01)overall ADG compared to triplets.Litter size by parity showed significances from 30-60 (P<0.01),60-90 (P<0.05) and 120-150 days (P<0.01).Single births from 1st and 2nd parity were heavier over single birth of 3rd parity and twin and triplets of other combinations (data not presented).Single births showed faster pre-weaning growth (P<0.001)and overall ADG (P<0.01) over triplets.
Triplet births from the 1st parity were found to be the least (1.91±0.11kg) in birth weight compared with other combinations.
Except 30-60 (P<0.01),60-90 (P<0.01) and 90-120 days (P<0.05),where male lambs were superior over their female contemporaries, sex effect was not significant (P>0.05)(Table 2).Sex effect was significant on post-weaning and overall ADG (Table 3).Highly significant, but variable, effect of parity was observed on ADG of lambs (Table 2).It was significant on ADG at all ages except from 90-120 days.Lambs born from the 1st parity demonstrated faster ADG compared to the higher parities at birth to 30 and 30-60 days (P<0.001).Similarly, lambs born from dams of 2nd parity grew faster (P<0.001) compared to the 4th parity from 0-30 and 30-60 days.The higher ADG of lambs from the 1st parity at early ages was reversed from 120-150 and 150-180 days.At 60-90 days, lambs from 1st and 4th parity were heavier (P<0.001) over 2nd and 3rd parities.
Lambs born from the 4th parity from 120-150 days and 2nd and 3rd parities from 150-180 days were heavier (P<0.01) over those born from the lower parities (Table 3).The pre and post weaning, and overall ADG were consistently and significantly affected by parity.Parity by sex interaction showed a marginal difference (P<0.05) at birth while non-significant (P>0.05) in the remaining ages while it was significant on ADG from 120-150 days (P<0.05) and post weaning ADG (P<0.05).Male lambs from 1st and 2nd parities where superior over males from 3rd and 4th parities and female lambs from other combinations (data not presented).Similarly parity by type of birth was significant on body weights at birth (P<0.001),30 (P<0.01) and 90 days (P<0.05) while it was significant on ADG from 30-60 (P<0.01),60-90 (P<0.05) and 120-150 days (P<0.01).Single births from 2nd parity were higher over single births of other parties and twin

Effect of season of birth
The influence of season of birth on birth weight, weight at various ages and ADG has been well documented at station (Gootwine et al., 2008;Teklebrhan et al., 2014) and under smallholder management systems (Hassen et al., 2002;Deribe et al., 2014).Generally, higher birth weight and subsequent weights found in our study, during the heavy rainy season, concurs with the previous reports (Taye et al., 2010;Teklebrhan et al., 2014).Except preweaning growth rate, there was a general faster postweaning and overall ADG during the light and heavy rain season mainly due to feed availability both in quality and quantity.However, there were no clear cut trends of season on body weights and ADG's.This could be associated with seasonal nutrient fluctuations (Legesse, 2008) and feed selection behaviour of sheep (Deribe et al., 2014).The reported higher body weights and ADG of lambs during the heavy rain and part of light rain seasons is consistent with other reports (Teklebrhan et al., 2014).
On the other hand, the higher body weight and faster preweaning gain during early dry season at some ages is partly due to better body reserves of dams at the end of heavy rain season, and the associated higher milk yield.
It has been noted that during early dry season lambs born with higher weight and usually grew faster due to feed flushing of the dams during the heavy rain season (Deribe et al., 2014).The fluctuation of lamb weight (gain and loss) within and across seasons is a commonly reported event, and is attributed to heat stress during the dry season and disease and parasite infestation during the light and heavy rain seasons.Yilmaz et al. (2007) noted seasonal differences on weight of lambs due to differences in ambient temperature and maternal prenatal effects during gestation.According to Hassen et al. (2002) and Gbangboche et al. (2006), effects of heat stress affects feed intake, and consequently body weights among indigenous lambs in sub-humid ecological zones.On the other hand, quality of feed and variations in feed composition as well as milk yield of dams are factors that have frequently been reported to influence lamb weight during early growth period (Legesse, 2008;Yilmaz et al., 2007;Deribe et al., 2014).

Effect of litter size
Litter size consistently affected lamb growth in our study due to the higher prolific capacity of the sire (Gavojdian et al., 2013) as well as dam breeds (Deribe et al., 2014).The higher litter size or percentage of multiple births, twins (59.3%) and triplets (3.9%) may have resulted in lower body weights of individual lambs at birth but have improved the overall lamb outputs.That could be a reason for decreased body weights and ADG as litter size increased.The overall ADG (0 to 180 days of age) and lamb output found in our study is comparable with previous reports of crossbred lambs under similar management condition (Teklebrhan et al., 2014).Generally, single births attained higher body weights (0 to 180 days) and ADG (pre, post and overall) than twin and triplets.At the same time, single births from 1st and 2nd parities were higher by 20.2 to 62.3% compared with single births of the 4th parity and twin and triplet births from other parity-by-birth combinations.The higher body weights and overall ADG of single births is due to the fact that they were the sole consumers of their dam's milk while twins and triplets compete for limited milk of their dams (Tibbo, 2006;Taye et al., 2010;Deribe et al., 2014).

Effect of parity (dam age)
Differences in parity due to ewe age affected body weights of lambs mainly at early ages, pre-weaning, post weaning and Overall ADG.Generally higher parities (3rd and 4th) attained higher body weights and faster overall ADG compared with lower (1st and 2nd) parities.This results are consistent with reports of Teklebrhan et al. (2013) who have shown profound effect of parity on body weights of crossbred lambs (Dorper × Somali) in Eastern Ethiopia.The higher body weights and ADG (0 to 90 days) obtained from the lower parities, however, is that most ewes at lower parities gave birth of single lambs.Single births had advantage over their twin or triplet counterparts because they are the sole users of milk of their dams (Tibbo, 2006).This suggests that the individual weight of multiple births was lower compared to single although the total lamb output is higher from the prolific dams at higher parties.Dams from 3rd and 4th parities gave birth of large number of multiple births, which compete for limited milk of their dams, and it could be a possible reason for the reported lower body weights of lambs at some ages from 3rd and 4th parities.

Sex effects
The effect of sex on body weights has been variable.Majority of the findings indicated that male lambs had usually higher weight at birth and grew faster than females (Gardner et al., 2007;Deribe et al., 2014), while few have shown that there is less general effect of sex on body weights (Legesse, 2008).The higher weight and faster growth of male over female might be explained by the favour of sexual precocity articulated on body growth.This sexual dimorphism favours body growth in males than females.Hormonal differences between sexes and letters (a, b) within a trait in a column are different at indicated P value: NS non-significant (P>0.05):*P<0.05;** P< 0.01; ***P<0.001.

Table 1 .
Least squares means ±SE analysis of variance for body weights (kilograms) from birth to 180 days of age of crossbred lambs (Dorper × indigenous) as affected by sex, parity and litter size in Areka.

Table 3 .
Lease squared means ± SE of pre-weaning, post-weaning and overall ADG of crossbred lambs (Dorper-Adilo) as affected by season, sex, parity and litter size in Areka.
Within main effect and trait, means followed by different letters are different at indicated P level; NS, non-significant (P>0.05);*P<0.05;**P<0.01.

Table 4 .
Maximum-likelihood analysis of variance of pre-weaning survival rate of crossbred lambs (Dorper x indigenous) in Areka.

Table 5 .
Pre-weaning survival rate of crossbred lambs (Dorper x indigenous) as influenced by fixed factors and birth weight category in Areka.