JH Hoon, MJ Herselman, M van Heerden & AP Pretorius

Grootfontein Agricultural Development Institute, P/Bag X529, Middelburg, 5900



Ewe productivity is the most important factor that determines the profitability of small stock farming. The most important components of ewe productivity are the total weight of lamb produced, body weight at culling and in the case of wool producing sheep the amount and quality (mainly fibre diameter) of the wool. From 65 to 88% of the total income from wool sheep is derived from mutton, while it is even higher in the case of mutton sheep. This emphasises the importance of increasing the kilograms of meat produced per hectare in order to increase ewe productivity and thus profitability. Probably the easiest way to do it is by increasing the weaning percentage of the ewes and the growth rate of the lambs. Although weaning percentage can be enhanced by increasing the conception rate of ewes and the number of multiple births, the best option is probably to decrease the lamb mortality rate. Most lamb mortalities (± 80%) occur in the period just before birth until seven days after birth. Research has shown that nearly 80% of these mortalities are related to the nutrition of the ewe during the last weeks before lambing and the first weeks after lambing (Seymour, 1998). The amount of colostrum available at lambing and the milk production of the ewe after lambing are important factors influencing lamb survival and growth rate of lambs. The supply of sufficient protein during late pregnancy and lactation influences the quantity as well as quality of milk produced. Some research results suggested that the supply of a high level of bypass protein (rumen undegradable protein) is essential to increase the colostrum and milk production of ewes (Hinch et al., 1996). Several local studies have been done on the effect of supplemenation of small stock with protein sources with different levels of rumen degradability. Most of these studies were, however, done on non-reproducing animals (Bekker, 1996). The high cost associated with supplementation of reproducing ewes, especially with bypass protein sources, necessitate further investigations into this matter. The aim of this study was therefore to investigate the effect of bypass protein supplementation during late pregnancy and lactation on the reproductive performance of Merino sheep on Eastern mixed karoo veld.


Material and Methods

 The study was carried out over a period of two years (1998-1999) under natural veld conditions (Eastern mixed karoo veld) at Grootfontein A.D.I. near Middelburg in the Eastern Cape. In 1998, 351 Merino ewes were randomly divided into three equal groups and grazed camps comparable in size and vegetation type. The three groups were rotated every two weeks between camps to eliminate any possible camp effects. Treatment 1 received no supplementation (Control), Treatment 2 received a urea-based supplement (Urea) and Treatment 3 a bypass protein-based supplement (Bypass). Supplementary feeding was supplied from four weeks before lambing until eight weeks after lambing. Intakes of supplements were limited to 300 and 400 g/sheep/day for pregnant and lactating ewes respectively. In 1998, soybean oilcakemeal (OCM) and sunflower OCM were used as sources of rumen undegradable protein (UDP), while feed grade urea was used as a source of rumen degradable protein (RDP). Both supplementary diets contained the same amount of total energy and protein (N x 6.25), namely 9.7 MJ/kg ME and 16% respectively, with a RDP:UDP ratio of 3.8:1 for the Urea supplement and 1.6:1 for the Bypass supplement.


Supplementary diets used in 1998:

            Urea               Bypass

Maize meal           75.6%       57.9%

Cottonseed OCM                   15.0%

Sunflower OCM                     5.0%

Molasses meal         5.0%        5.0%

Salt                 15.0%       15.0%

Feed grade urea       3.3%        1.0%

Slaked lime           1.0%        1.0%

Feed grade sulphur    0.1%        0.1%


The experiment was repeated in 1999 with 375 ewes. All the ewes were allocated to the same treatment groups as the previous year and replacement ewes were randomly divided among the groups. In 1999, a commercial protein concentrate for small stock was used as a source of UDP. The energy and protein contents of both supplements were 10.0 MJ/kg ME and 18% respectively with a RDP:UDP ratio of 3.9:1 for the Urea supplement and 1.7:1 for the Bypass supplement.


Supplementary diets used in 1999:


               Urea                Bypass

Maize meal              75.0%      49.6%

Protein Concentrate                39.6%

Molasses meal            5.0%

Salt                    15.0%      10.0%

Feed grade urea          3.9%                           

Slaked lime              1.0%       1.0%

Feed grade sulphur       0.1%


The following data were recorded for both years: birth weight, 42 day weight and weaning weight of lambs, number of lambs born and weaned and lamb mortality rate. Milk composition of five ewes from each group was determined every two weeks for the full lactation period in 1998, while milk production of each ewe was recorded three weeks after lambing in 1999. The method of test-milking over a four hour sampling period after intramuscular injection of the ewes with 1 ml oxytocin (Bencini, 1995) was used to obtain milk samples for the determination of milk production and composition. In 2000, all the ewes were scanned six weeks after the end of the mating period by means of an ultrasound scanner. This was done in order to determine the conception percentage of the ewes of the different groups and establish any possible carry-over effects of the different supplements supplied during the previous years.

Statistical analyses on reproduction data were performed by using the FREQ procedures of SAS and tested for significance by chi-square. Statistical analyses of body weight of lambs were performed using the GLM procedures of SAS. The model included sex, year, treatment, age of dam, status (birth, 42 day and weaning) and age of lamb and a year x treatment interaction. The effects were tested against the residual error, while least-squares means were separated using the PDIFF option of SAS (Littell et al., 1991). The milk production data recorded in 1999 were also analysed with the GLM procedures, but the model only included treatment, age of ewe and number of suckling lambs. Milk composition data (only collected in 1998) were analysed as a split-plot, repeated measures design. The model included treatment and week of lactation. The treatment effect was tested using animal within treatment as error term, while least-squares means were separated as mentioned before.


Results and discussion

The reproduction data of the ewes in 1999 and 2000, that received the same treatments as in 1998 and 1999 respectively, are presented in Table 1.


Table 1. Reproduction data of ewes





Ewes lambed/ewes mated (%)




Lambs born/ewes mated (%)




Lambs weaned/ewes mated (%)




Kg lamb weaned/ewe mated

17.46 ± 1.93

21.90 ± 1.98

20.34 ± 1.88





% Ewes - not pregnant




% Ewes – pregnant (single)




% Ewes – pregnant (twins)




Conception percentage (%)




Scanning percentage (%)





The reproduction data of the ewes of the different treatments did not differ for any of the calculated traits (P>0.05) in 1999. The conception and scanning percentages of the ewes of the different groups also did not differ (P>0.05) in 2000. This suggests that no carry-over effect, with regard to improved reproduction rate owing to supplementation during the previous lambing season, could be observed. Although some of the reproduction data of the ewes, especially in 1999, indicated differences between the control and treatment groups, factors such as group size, variation, etc., could have an effect on the results obtained and should be taken into consideration before any conclusions based on these data are drawn.

The growth data of the lambs, as well as lamb mortality rates, of both years (1998 and 1999) are presented in Table 2.


Table 2. Growth data of the lambs





Birth weight (kg)

4.23 ± 0.12a

4.40 ± 0.12b

4.27 ± 0.12

42-day weight (kg)

13.23 ± 0.62a

15.02 ± 0.63b

15.01 ± 0.62b

Weaning weight (kg)

23.49 ± 0.94a

24.33 ± 0.94b

23.79 ± 0.93

Mortality rate (%)




ab   Values with different superscripts differ significantly (P<0.05)

The data of the lambs (1998 and 1999) indicate that the birth weight and weaning weight of the lambs of the Urea group was higher (P<0.05) than that of the Control group. The 42-day weight of the lambs of the Control group was lower (P<0.05) than that of the Urea and Bypass groups, which was probably influenced by the differences in total milk production of the ewes (as indicated in Table 3). The lamb mortality rate also did not differ significantly (P>0.05).

The milk composition data of the five ewes in each group that were milked over the full lactation period in 1998, as well as the average daily milk production of all the ewes that have lambed in 1999, are presented in Table 3.


Table 3. Milk composition and milk production data





Milk composition (1998)

Fat (%)

7.04 ± 0.33

 6.73 ± 0.33a

7.95 ± 0.34b

Protein (%)

4.33 ± 0.10

4.40 ± 0.10

4.54 ± 0.11

Lactose (%)

4.86 ± 0.10a

4.53 ± 0.10b

4.86 ± 0.10a

Milk production (1999)

Daily milk production (ml/ewe)

1034 ± 45a

1404 ± 44b

1352 ± 43b

ab   Values with different superscripts differ significantly (P<0.05)

The milk composition data (1998) indicated that the fat content of the Bypass group was higher (P<0.05) than that of the Urea group, while the lactose content of both the Control and Bypass groups were higher (P<0.05) than the Urea group. The milk production of the ewes (1999) of the Control group was lower (P<0.05) than that of the Urea and Bypass groups. This is reflected in the significant differences that was observed in the 42-day weight of the lambs, where the Control group was significantly lower than the two treatment groups. Due to the fact that the energy value of milk is closely related to its fat content, this factor must be taken into account, together with total quantity of milk, in assessing the contribution of milk to lamb growth. In this experiment the differences in milk composition seem, however, to be too small to have a significant effect on lamb growth and the quantity of milk was probably a more important factor in determining growth rate, especially up until the 42-day weight measurement.

The average milk production over the lactation period (Week 2 to 14), as determined in 1998 with the pooled values of the five sheep in each group, is illustrated in Figure 1.

Milk production reached a peak from two to six weeks after parturition (highest at six weeks) and then declined. Twelve weeks into the lactation period, the milk production was less than 50% of the peak milk production. These results are in correspondence with the results obtained by Doney et al., (1979).

The effect of the age of the ewes on their average daily milk production as determined in 1999 from the data of all the ewes that lambed, is illustrated in Figure 2.

The highest milk production was recorded for the three to five year old ewes (highest for four year old ewes), with lower production rates for the ewes that lambed for the first time, as well as the older ewes. 



No conclusive evidence as to the effect of supplementation on the productive and reproductive performance of late pregnant and lactating Merino ewes on Eastern mixed karoo veld could be found in this study. It is, however, evident that supplementation with bypass protein-based supplements had no advantage over supplementation with urea-based supplements. To determine the economic advantages of protein supplementation of ewes during late pregnancy and lactation on a wider scale, on-farm trials are currently conducted in all the major small stock producing areas of South Africa (sheep and goats). The results from these trials, which will be carried out over at least three lambing seasons, will probably supply more accurate, scientifically-based information with regard to the economic viability of supplementing ewes during late pregnancy and lactation under different veld types and conditions.



Bekker, M.J., 1996. Proteïen- en energie-aanvulling aan nie-produserende wolskape op natuurlike weiding op die Hoëveld van Mpumalanga. Ph.D-thesis, University of Stellenbosch.

Bencini, R., 1995. Use of intramuscular oxytocin injections to measure milk output in nondiary sheep, and its effect on milk composition. Australian Journal of Experimental Agriculture, Vol. 35, 563-565.

Doney, J.M., Peart, J.N., Smith, W.F. & Louda, F., 1979. A consideration of techniques for estimation of milk yield by suckled sheep and a comparison of estimates obtained by two methods in relation to the effect of breed, level of production and stage of lactation. Journal of Agricultural Science, Vol. 92, 123-132.

Hinch, G.N., Lynch, J.J., Nolan, J.V., Leng, R.A., Bindon, B.M. & Piper, L.R., 1996. Supplementation of high fecundity Border Leicester x Merino ewes with a high protein feed: its effect on lamb survival. Australian Journal of Experimental Agriculture, Vol. 36, 129-136

Littell, R.C., Freud, R.J., & Spector, P.C., 1991. SAS System for Linear Models. Third Edition. SAS Institute Inc. Cary, NC.

Seymour, M., 1998. Farming Ahead, May 1998, No. 77, 77-78



Karoo Agric Vol 4 (1)

Article source: Grootfontein Agricultural Development Institute -