Nutritional effects of supplementary feeding on maternal blood metabolites, cortisol, thyroid hormones levels and on outcome of pregnancy of dry season kidding Tswana goats

The nutritional effects on maternal blood metabolites levels and outcome of pregnancy in goats were investigated. The supplementary diet (106g/kg crude protein) was given to S animals at a rate of 400g/animal/d. Two weeks prior to and at parturition, S (Supplemented) does were heavier (P<0.05) and were in better (P<0.001) body condition than C (Control) animals.

Eighty-nine percent of S animals kidded including abortions, while the value was 79 for C animals, however the difference was not significant (P>0.05). Prolificacy was higher for S but was not significantly different (P>0.05; 1.93±0.17 versus 1.64±0.18kid/doe for S and C respectively). Percentage reproductive wastage was lower (P<0.05) in S than C group. Birth weights were similar between S and C animals (P>0.05; 2.70±0.11 and 2.60±0.13kg). A significant (P<0.05) difference was recorded on the concentrations of cholesterol, haematocrit, triiodothyronine (T₃) and tetraiodothyronine (T₄) (P<0.01) between the groups, S animals having high values than C animals.

It is discussed that infectious agents were not a prime cause of reproductive wastage. Kid viability at birth contributed to reproductive wastage and difference in T₃ and T₄ may be involved but not cortisol. Supplementary feeding of pregnant goats grazing natural pasture during the dry season can offset the detrimental effects of maternal nutritional stress and therefore reducing reproductive wastage.

Key words: Abortions; body condition; dry season; nutritional stress; reproductive wastage

Introduction

Tswana goat is a meat-type indigenous breed and was reported by Senyatso and Masilo (1996) to account for 80% of the national goats’ population. The need to improve productivity of Tswana goat is emphasized by the fact that this animal holds promise to increase smallholder incomes and household nutrition. Past government policies to improve small ruminant productivity included Financial Assistant Policy (FAP) to cover small ruminant projects. However, the anticipated increased productivity was never realised, mainly due to low level of husbandry and feeding management (Letsebe et al 1999)

Past studies (APRU 1987, Gray 1987, APRU 1989 and Madibela et al 2002) reported frequency of twinning in Tswana goat of 50% and this has been associated with rapid increases in goat numbers. In addition short gestation period means that meat production can be realized in a relatively shorter period of time. However, nutrition plays a major role in the attainment of genetic potential for reproduction and growth. Reproductive wastage in terms of abortion, fetal loss and neonatal loss undermine productivity of goats in Botswana. There are reports of abortion and foetal loss in goats elsewhere (Hussain et al 1996a, Hussain et al 1996b, Engeland et al 1997, Romero-M et al 1998, and Engeland et al 1998). In Botswana, abortion storms occur during the winter months, (Binta et al 1996) which coincide with the dry season. Though infectious causes have been found in aborting goats, these accounted for only 10% (Binta et al 1998a) and 23% (Binta et al 1998b) of the total number of animals aborting. According to Binta et al (1996), several infectious and non-infectious factors may be responsible for causing abortions in goats in Botswana, but the majority of the abortion cases have never been explained. Few cases of abortions are caused by infectious agents in Norwegian dairy goats, (Hussain et al 1996a, Hussain et al 1996b) and Engeland et al (1997) suggested that nutritional and environmental factors might be important. Ahmadu (1996) reported incidents of abortion in Zaria area of Northern Nigeria and attributed it to improper feeding of goats during pregnancy. Underfeeding leads to nutritional stress and may affect hormonal secretion associated with metabolisms. There is lack of information on the effects of nutrition on the extent of reproductive wastage, cortisol and thyroid hormones in Tswana goats in Botswana. Therefore present study investigated the nutritional effects on maternal blood metabolites, cortisol and thyroid hormones levels and on the outcome of pregnancy of dry season kidding Tswana goats.

Materials and methods

The study was conducted at Sebele Station which is situated at latitude of 24^o 33'S and longitude 25^o 57' E, at an altitude of 994 m a.s.l. The vegetation type of the area has previously been described by Madibela et al (2000). Grass quality and quantity fluctuate with season, being adequate in months of October-March (ONDJFM) and in short supply or low in quality in April-September (AMJJAS), when browse supplies most nutrient requirements of goats. Data for rainfall and temperature for Sebele were collected from Botswana Meteorology Services weather station located approximately 0.5km from the study site. Mean rainfall for the area is 500mm. Monthly average minimum and maximum temperature is 12.8 and 28.6^oC respectively.

Thirty-seven Tswana goats (3-6 years of age) divided into two groups; (C; mean±SE 38.8±0.9kg, n=19) and (S; 39.8±0.9, n=18) were used in a trial that commenced in May 2000. The animals had previously kidded in January and kids were weaned at the end of April. The animals were cared for according to International Guidelines for Biomedical Research Involving Animals (CIOMS 1985). They were treated for internal parasites (Ivomec®, Logos Agvet; Republic of South Africa). A plunge dip was use for tick control (9%v/v chlorfenvinphos concentrate, Agricura, Republic of South Africa) only when infestation was observed. The supplementary diet has been used previously for breeding goats (Madibela et al 2002) Chemical composition of the supplementary feed and its estimated nutrient supply is shown on Table 1.

Table 1. Gross energy, chemical composition and estimated nutrient intake/animal/d from the supplemented feed at the start and two weeks prior to parturition
Chemical composition		Nutrient intake^†, g		Nutrient intake^‡, g
Nutrient	Amount	Supplemented	Control	Supplemented	Control
GE, MJ/kg DM	10.2	4.08	0	7.65	6.12
CP, g/kg DM	106	42.4	0	79.5	63.6
Ca, g/kg DM	4	1.6	0	3	2.4
P, g/kg DM	3	1.3	0	2.25	1.8
ADF, g/kg DM	244	97.6	0	183	146.4
GE = Gross energy; CP = Crude protein; Ca = calcium; P = phosphorous; ADF = acid detergent fibre ^†Estimated nutrient intake at the start of trial ^‡Estimated nutrient intake two weeks prior to parturition

Goats feed intake is about 4% of their body weight (Kadzere et al 1996); so at the start of the trial it was assumed that S animals were consuming 1.2kg from grazing while 0.4kg was provided by the supplement. C animals were assumed to be getting the whole of their 1.6 kg feed intake from grazing. The animals were accompanied by bucks during grazing (one buck to 25 does), as is the case in the communal farming system of Botswana. Due to the poor grazing conditions which lead to decline in body condition of the animals two weeks prior to parturition, feed for the S group (body condition score, BCS = 2.83 on a 1 to 5 scale, where 1=thin, 5=fat as per Karua 1990) was increased from about 400 to 750g/animal/d while the C group (condition score =1.89) were offered 600g/animal/d of feed to avoid unnecessary suffering by the animals. Data records included monthly BCS, live weights and birth weights. The numbers of does that delivered, fetal loss and kid mortality were recorded.

Blood was collected from five animals selected at random and sampled from each of the two treatments during the last two weeks of pregnancy. Each doe was bled by jugular vein puncture and blood collected in heparin coated and into plain vacutainers. The blood was centrifuged at 3000 x g for 10 minutes and plasma was harvested, stored at -20^oC until analysed for triiodothyronine (T₃), tetraiodothyronine (T₄) and cortisol. Serum was harvested from blood samples from plain tubes by allowing them to stand for one hour at room temperature. Serum was used in analysis for total protein, albumin, calcium, phosphorous, copper, zinc, urea, cholesterol, triglycerides.

Thyroid hormone concentrations were determined with commercial kits. T₃ and T₄ concentrations were measured in duplicate by solid phase ¹²⁵I radioimmunoassy according to the manufacturer’s instructions (Diagnostic Products Corporation, Los Angeles, CA, USA). The radioactivity of teraiodothyronine- and triiodothyronine-bound tubes was counted with a gamma counter, as previously described (Chopra 1972). Cortisol was determined using a solid phase technique GammaCoat ¹²⁵I cortisol radioimmunoassay kit (DiaSorin, Stillwater, Minnesita, USA). The procedure is based on the competitive binding principles and a gamma counter was used to estimate cortisol levels, as previously described (Rehbinder and Hau 2006).

Serum samples were taken to the National Veterinary Laboratory (Gaborone, Botswana) where they were analysed for copper, zinc (UV spectrophotometry), total protein, albumin, calcium, phosphorous, urea, cholesterol and triglycerides (Automatic chemical analyzer) according to methods described by Mushi et al (1998). Globulin was calculated as a difference between total protein and albumin. Haematocrit analysis was done using standard haematological procedures.

General linear models (GLM) procedure (SAS Institute Inc. 1990) was used to determine the effects of supplementation on body weight, condition score and blood parameters. The effects of supplementation on pregnancy rate were evaluated by frequency analysis, using a Chi-square test (SAS Institute Inc. 1990). The Fisher Exact Test was used when the number of animals in a treatment category was less than five (Steel and Torrie 1980). Goats that did not give birth were excluded from the analysis of body weight and BCS at two weeks prior to parturition and at parturition and from analysis for reproductive wastage. GLM procedure (SAS Institute Inc. 1990) was also used to test the effects of supplementation on reproductive wastage including and then excluding animals whose kids and/or fetuses died due to infectious cause. Effects of supplementation on birth weight of kid, and correlations between doe parturition weight and kid birth weight were tested. Because of low numbers of kids born as singles this test was also repeated without singles. One animal from the Control group died due to ingestion of plastic material and was found to be carrying twin fetuses, but was excluded from reproductive wastage analysis. One animal from the S group had an early abortion and only placental material was recovered but no infectious microorganism was isolated. This animal too was excluded from the analysis of body weight, BCS and reproductive wastage. Another animal from the S group sustained a fracture just before it gave birth and thus was not weighed at parturition. Reproductive wastage was defined as the number of fetuses that were aborted, born dead or kids that died during the first 24 hours, per doe.

Results

The percentages of animal in S and C treatment groups which either gave birth or aborted were similar (P>0.05; Table 2).

Table 2. Effect of supplementation on the outcome of mating and pregnancy of Tswana goats
Parameter	Supplemented	Control	SL
No of goats at mating`	18	19
No of goats pregnant¹	16	15²
Pregnancy rate, %	88.9	78.9	NS
No of goats aborted	1	2
No goats with dead foetuses	0	2
No of goats whose kids dead in first 24 hours	3	2
No of goats abortive microorganism isolated	0	1^†
Reproductive parameters	n=15	n=14
Prolificacy, kid/doe³	1.93±0.17	1.64±0.18	NS
Total Wastage, kid/doe⁴	0.27±0.18	0.71±0.19	NS
Percentage Total wastage, %⁵	10.0±0.1	41.6±0.1	*
Wastage due to non-infectious causes, kid/doe	0.21±0.18 (n= 14)	0.62±0.19 n=13)	NS
Percentage non-infectious wastage, %	8.4±0.1 (n= 14)	37.2±0.1 n=13)	*
¹Parturiation was used as a sign of animal having conceived since neither ultrasonic nor progesterone were used to verify pregnancy. ²One animal died and was found to be carrying twin fetuses. ³Number of kids born and fetuses aborted per doe kidding ⁴Total reproductive wastage including infectious and non-infectious causes ⁵Expresses as 4/3 x 100 ^†Toxoplasma gondii and S. bovis were isolated in this aborted material. SL = Significance level; NS = P>0.05; = P<0.05*

Of the two goats whose kids died within 24 hours in C group, the diagnosis was postnatal weakness leading to secondary septicaemia. However, for the three goats whose kids died within 24 hours in the S group, the diagnosis for one of them was purulent bronchopneumonia while the other two was inconclusive. One of the two goats that aborted in the C group had twin fetuses from which Toxoplasma gondii and Streptococcus bovis were isolated.

Prolificacy for S animals was similar (P>0.05) to that of C animals. Total reproductive wastage was not significantly (P>0.05) different between the two groups although it was lower for S than C animals. However, percentage total reproductive wastage was different (P<0.05; Table 2).

Birth weights were similar (P>0.05) between S and C animals. Multiples weighed lower (P<0.001) than singles. Multiples kids in both S and C animals weighed similar (P>0.05). Kids born as singles had similar (P>0.05) weights at birth for S and C. Does parturition weight had a significant effect (P<0.001) on kid birth weights (Table 3).

Table 3. Body weight and condition score of Tswana goats
Parameter	Supplemented	Control	SL
At mating	n=18	n=19
Body weight, kg	39.8±0.90	38.8±0.88	NS
Condition score¹	2.81±0.17	2.74±0.17	NS
At two weeks prior to parturition	n=15	n=14
Body weight, kg	43.9±1.3	39.1±1.4	*
Condition score	2.83±0.16	1.89±0.16	***
At parturition	n=14	n=14
Body weight, kg	40.1±1.3	35.1±1.3	*
¹Condition score: 1= thin, 5 = fat, according to Karua (1990) SL = Significance level; NS = P>0.05; = P<0.05; *** = P<0.001*

At mating, the two groups had similar body weight (P> 0.05) and body condition score (P> 0.05) for S and C animals (Table 3). However, two weeks prior to parturition S animals exhibited higher body weight (P<0.05) and high BCS (P<0.001) than C animals. At parturition S animals were still heavier (P<0.05) than C animals. Doe parturition weights of S (r=0.617, P<0.001) and C (r=0.550, P<0.05) were positively correlated to kid birth weight. However, when singles were removed from the analysis, doe parturition weight of S animals was highly correlated (r=0.756, P<0.001) to birth weight while correlation of doe parturition weight of C to birth weight was not significant (r=0.250, P>0.05).

Total protein, albumin, calcium, phosphorus, copper, zinc, urea, triglycerides and cortisol were similar (P>0.05) between the two groups (Table 4).

Table 4. Blood parameter of Tswana goats at two weeks prior to parturition
Variable	Supplemented	Control	SL
Total protein, g/l	66.6±3.39	61.0±3.39	NS
Albumin, g/l	40.8±1.46	37.2±1.46	NS
Globulin, g/l	25.8±2.01	25.2±2.01	NS
Calcium, mmol/l	2.36±0.09	2.26±0.09	NS
Phosphorous, mmol/l	1.98±0.14	1.76±0.14	NS
Copper, μmol/l	18.2±2.54	20.4±2.54	NS
Zinc, μmol/l	20.2±5.53	10.1±5.53	NS
Urea, mmol/l	7.44±0.42	7.23±0.42	NS
Cholesterol, mmol/l	1.65±0.14	1.14±0.14	*
Triglycerides, mmol/l	2.41±0.58	2.12±0.58	NS
Haematocrits	21.7±0.78	18.2±0.78	*
T_3, pmol/l	5.54±0.27	4.30±0.30	*
T_4, nmol/l	48.9±3.30	30.8±3.68	**
Cortisol, nmol/l	24.7±4.63	18.3±5.17	NS
T₃ = triiodothyronine; T₄ = tetraiodothyronine; SL = Significance level NS = P>0.05; = P<0.05; ** = P<0.01*

A significant difference between S and C animals was observed on cholesterol, haematocrits, T₃ (P<0.05) and T₄ (P<0.01).

Discussion

In Botswana, goats kidding during the later part of the dry season (August to September) would have been mated at the end of the wet season (April) when nutrition would be adequate for ovarian activities and this period also coincides with decrease in day length. Mating at this time means that the level of sexual activity is high and may result in multiple births. However, the rest of pregnancy goes through a period of shortage of nutrients especially at the last trimester when the fetus growth rate is high. As a result malnutrition may results in abortion in goats.

The insignificant and small difference between S and C animals in the proportion of goats that either gave birth or aborted (pregnancy rate) in the present study may be due to nutrition. Toxoplasma gondii has previously been associated with goats with a history of abortion in Botswana (Binta et al 1996, Binta et al 1998a). However, it is suggested that in the present study, infectious agents were not the prime cause of reproductive wastage. Previous data (Madibela et al 2002) with the same flock of animals showed that fertility and the proportion of animals that gave birth during the wet season was similar between the supplemented and control animals. If the ovarian activity of the animals in the present study was the same; the reduced pregnancy rate in C group would then be due to embryonic loss as the dry season progress.

The consistent twinning of 50% in Tswana goats (APRU 1987, APRU 1989, Madibela et al 2002) may be indicative of a genetic attribute of this breed. Prolificacy rate for S animals was close to 2 kids/doe and this may contribute substantially to farmer’s productivity. This indicates that if abortion and neonatal death are reduced, Tswana goats would have high productivity. When infectious causes were removed from the analysis, C animals still had high reproductive wastage. Lower body weight and BCS at two weeks prior to parturition and at parturition of C animals showed that when pregnant goats are undernourished, reproductive wastage would occur. Besides the problem of abortion due to undernutrition, neonates are weak and are susceptible to infection. In the present study kids from multiple births were weak at birth.

Lack of difference in kid birth weight indicate that the high dam liveweight and BCS of S animals did not benefit kid birth weights, which was consistent with findings of Madibela et al (2002). Doe parturition weight was positively correlated to birth weight of multiple born kids in S but not in C group. This is an unexpected finding since C animals were nutritionally stressed and they were expected to produce kids of low birth weight. This may be due to enhanced fetal growth as a result of stimulated placental growth and function when females are subjected to nutrient restriction at mid pregnancy (Robinson 1990, Wallace et al 1997, Perry et al 1999 and Robinson et al 1999).

Thyroid hormones concentrations are nutritionally sensitive (Rae et al 2002). In the present study the difference in the concentrations of thyroid hormones between the two groups may be associated with the difference in reproductive wastage. Manalu et al (1997) did not found difference in T₃ between aborted goats and dams bearing single and twin kids. T₃ was found to be high in non-pregnant, indicating that the increase in T₃ was not related to pregnancy (Manalu and Sumaryadi 1999). However, the reproductive physiology of non-pregnant animals and those that aborted is not the same, hence; it would be invalid to compare the two. Rae et al (2002) found that undernutrition in pregnant sheep results in low plasma T₃. The viability of neonatal kids contributed to reproductive wastage in the present study and according to Robinson (1990) there is a positive correlation between neonatal survival and plasma T₃ and T₄ concentrations. In the present study thyroid hormones of kids were not measured. However, Rae et al (2002) found that foetal T₃ and T₄ concentrations were low in response to maternal undernutrition which may affect gonadal development of offsprings.

The results of cortisol concentrations cast doubts to its direct involvement in reproductive wastage. Manalu et al (1997) found no difference in average serum cortisol level of aborted, single and twin bearing does. In addition, Manalu and Sumaryadi (1999) reported no relation between cortisol concentrations during pregnancy with lamb birth weight. In the present study birth weights in the two groups were similar thus maternal serum cortisol may not have contributed to reproductive wastage or kid viability. Romero-R et al (1998) found that maternal cortisol concentration for normal delivering goats to be 35.5nmol/l at peripartum, which is 21.5 and 41.9% higher than cortisol levels of S and C animals of the present study respectively. Those animals that aborted were found to have even higher values of cortisol at the time of abortion leading the authors to conclude that cortisol was associated with abortion.

Values of total protein in the present study were similar, but at the lower end of the normal range for goats. Although it has low affinity for thyroid hormones, albumin has high binding capacity due to its high concentration in plasma (Cunningham 1997). In the present study, the difference in albumin, albeit in small quantities, may have contributed to a significant difference in thyroid hormones between the two groups. Blood minerals did not contribute directly to differences in reproductive wastage. Gray et al (1990) reported calcium levels of supplemented goats that were similar to the value of S animals in the present study. Phosphorus was found to be higher while copper to be lower than values obtained by Gray et al (1990). Urea levels were found to be similar to values for goats in Botswana (Gray et al 1990). Protein content of pasture for foraging goats during periods similar to those of animals in the present study was reported by APRU (1983/4) to be 12g/kg. High protein in browse diet may have resulted in high serum urea. In addition, BCS of C animals were lower than those of S, indicating that animals’ nutrient requirements were partly met through mobilization of body reserves. Pambu-Gollah et al (2000) found that dry season kidding goats in South Africa were not able to maintain glucose homeostasis during pregnancy but their plasma urea levels were elevated during the last month of pregnancy.

In the present study, cholesterol level was higher in S group but it is not clear how the trend would have progressed over time. This is because according to Kaneko (1989), the net effect of thyroid hormones on cholesterol metabolism is to increase the rate of its catabolism by the liver, thereby lowering cholesterol. Even though triglyceride levels were similar between the two groups, C animals were under metabolic stress, which was consistent with their lower body condition at two weeks prior to parturition and at parturition. Therefore, the lower levels of cholesterol and triglycerides in C animals are consistent with the underlying negative energy balance of these animals.

Conclusions

Acknowledgements

The summary of this paper was presented at an International Conference of the British Society of Animal Science, University of York. England. [Madibela O R and Segwagwe B E V (2003) Nutritional effects on maternal blood metabolites and on outcome of pregnancy of dry season kidding Tswana goats. Paper No 83]

The authors would like to thank National Veterinary Laboratory for their technical input and A. Human (University of Pretoria, SA) for hormonal analysis. Prof. J J Robinson (Scottish School of Agriculture, Aberdeen, Scotland) is recognized for comments on the initial draft of this manuscript. This study was funded by Botswana’s Ministry of Agriculture.

References

Ahmadu B 1996 Abortions due to malnutrition in caprine in Zaria area of Northern Nigeria. Bulletin of Animal Health and Production in Africa 44:265-267

APRU (Animal Production Research Unit) 1983/4). Dietary selection and nutrition of goats. In: Livestock and Range Research in Botswana. Ministry of Agriculture. Government Printers. Gaborone 36-44pp

APRU (Animal Production Research Unit) 1987 Effect of kidding on goat production. In: Livestock and Range Research in Botswana. Ministry of Agriculture. Government Printers, Gaborone 73-76pp

APRU (Animal Production Research Unit) 1989 Effect of kidding on goat production. In: Livestock and Range Research in Botswana. Ministry of Agriculture. Government Printers, Gaborone 36-41pp

Binta M G, Mushi E Z and Adom E K 1996 Abortion in goats in Botswana. Bulletin of Animal Health and Production in Africa 44:205-208

Binta M G, Mush,i E Z, Raborokwe M, Monyame R K and Ndebele T R 1998a The prevalence in Botswana of antibodies to Toxoplasma gondii in goats with a history of abortion. Zimbabwe Veterinary Journal 29:47-50

Binta M G, Monyame R K, Ndebele R, Mushi E Z and Raborokwe M 1998b Antibodies to Chlamydia psittaci in goats in Botswana. Zimbabwe Veterinary Journal 29:91-94

Chopra I J 1972 A radioimmunoassay for measurement of thyroxin in extracted serum. Journal of Clinical Endocrinology and Metabolism 36:938-947

CIOMS (Council for International Organization of Medical Science) 1985 International Guidelines for Biomedical Research Involving Animals. Geneva http://www.cioms.ch/1985_texts_of_guidelines.htm

Cunningham J G 1997 Textbook of Veterinary Physiology. 2^nd edition. W B Saunders Company, Philadephia 137: 404-410pp

Engeland I V, Waldeland H, Andresen Ø and Tverdal A 1997 Foetal loss in dairy goats; an epidemiological study in 515 individual goats. Animal Reproduction Science 49:45-53

Engeland I V, Waldeland H, Andresen Ø, Løken T, Bjorkman C and Bjerkas I 1998 Foetal loss in dairy gaots: An epidemiological study in 22 herds. Small Ruminant Research 30:37-48

Gray R C 1987 Goat production by ATIP farmers during 1986 in Tutume District, Botswana. Agricultural Technology Improvement Project (ATIP) Progress Report. Ministry of Agriculture, Botswana and Mid-America International Agricultural Consortium

Gray R C, Koch B A, Mahabile W, Windsor R, Holmes A and Berg C 1990 Blood parameters of goats in Botswana. Tropical Agriculture 67:355-359

Hussain Q, Ropstad E and Andresen Ø 1996a Effects of type and quality of roughage and energy level on plasma progesterone levels in pregnant goats. Small Ruminant Research 21:113-120

Hussain Q, Waldeland H, Havrevoll Ø, Eik L O, Andresen Ø and England I V 1996b Effect of type of roughage and energy level on reproduction performance of pregnant goats. Small Ruminant Research 21:97-103

Kadzere C T, Llewelyn C A and Chivandi E 1996 Plasma progesterone, calcium, magnesium and zinc concentrations from oestrus synchronization to weaning in indigenous goats in Zimbabwe. Small Ruminant Research 24: 21-26

Kaneko J J 1989 Clinical biochemistry of domestic animals. 4^th edition. Academic Press Inc. San Diego 630-649pp

Karua S K 1990 Effects of management of feed availability, body weight and body condition of indigenous goats in Central region of Malawi. In: PANESA/ARNAB, Utilization of research result on forage and agricultural by-products materials as animals feed resources in Africa. Proceedings of the first joint workshop held in Lilongwe, Malawi 5-9 December, 367-379pp

Letsebe E, Seleka T B and Marumo D S 1999 Economic assessment of sheep and goats keeping in Botswana. Enhancing Sheep and Goats Production in Botswana. In: A A Aganga, R G Chabo, M Kgosimore, M Letso and U J Omphile (editors. Proceeding of the Sheep and Goats Workshop held at the Centre for In-service and Continuing Education, Botswana College of Agriculture, Gaborone, Botswana. April 19-23 1999. 61-78pp

Madibela O R, Boitumelo W S and Letso M 2000 Chemical composition and in vitro dry matter digestibility of parasitic plants (Tapinanthus lugardii, Erianthenum ngamicum, Viscum rotundifolium and Viscum verrucosum) in Botswana. Animal Feed Science and Technology 84: 97-106

Madibela O R, Mosimanyana B M, Boitumelo W S and Pelaelo T D 2002 Effect of supplementation on reproduction of wet season kidding Tswana goats. South African Journal of Animal Science 32:14-22

Manalu W and Sumaryadi M Y 1999 Correlation between birth weight and the concentrations of hormones and metabolites in maternal serum during pregnancy. Journal of Agricultural Science, Cambridge 133:227-234

Manalu W, Sumaryadi M Y and Kusumorini N 1997 Maternal serum concentrations of total triiodothyronine, tetraiodothyronine and cortisol in different status of pregnancy during late pregnancy in Ettawah-cross does. Asian-Australasian Journal of Animal Science 10: 385-390

Mushi E Z, Binta M G, Chabo R G, Isa J F W and Modisa L 1998 Serum biochemical values of farmed ostrich (Struthio camelus) in Botswana. Onderstepoort Journal of Veterinary Research 65:189-193

Pambu-Gollah R, Cronje P B and Casey N H 2000 An evaluation of the use of blood metabolite concentrations as indicators of nutritional status in free ranging indigenous goats. South African Journal of Animal Science 30:115-120

Perry V E A, Norman S T, Owen J A, Daniel R C W and Phillips N 1999 Low dietary protein during early pregnancy alters bovine placental development. Animal Reproduction Science 55:13-21

Rae M T, Rhind S M, Kyle C E, Miller D W and Brooks A N 2002 Maternal undernutrition alters triiodothyronine concentrations and pituitary response to GnRH in fetal sheep. Journal of Endocrinology 173:449-45

Rehbinder C, Hau J 2006 Quantification of cortizol, cortizol immunoreactive metabolites, and Immunoglobulin A in serum, saliva, urine and feces for noninvasive assessment of stress in reindeer. Canadian Journal of Veterinary Research 70: 151-154

Robinson J J 1990 Nutrition in the reproduction of farm animals. Nutrition Research Review 3:253-276

Robinson J J, Sinclair K D and McEvoy T G 1999 Nutritional effects on foetal growth. Animal Science 68:315-331

Romero-R C M, Lopez G and Luna-M M 1998 Abortion in goats associated with increased maternal cortisol. Small Ruminant Research 30:7-12

SAS Institute Inc. 1990. SAS User’s Guide. Statistics. SAS Institute INC Version 6 Cary, NC. USA.

Senyatso E K and Masilo B S 1996 Animal genetic resources in Botswana. In Animal Genetic Resources Information; No 17. Food and Agricultural Organisation and United Nations Environment Programme. Rome 57-68pp http://agtr.ilri.cgiar.org/Library/docs/agri17_96.pdf

Steel R G D and Torrie J H 1980 Principles and Procedures of statistics, a biometrical approach. 2^nd edition. McGraw-Hill Book Co. New York 504-506pp

Wallace J M, Aitken RP, Cheyne M A and Humblot P 1997 Pregnancy-specific protein B and progesterone concentrations in relation to nutritional regime, placental mass and pregnancy outcome in growing adolescent ewes carrying singleton fetuses. Journal of Reproduction and Fertility 109:53-58