Trop. Anita.HlthProd. (1995)27, 175-185

INFLUENCE OF SEASON AND HOUSING ON OVARIAN A C T I V I T Y O F I N D I G E N O U S G O A T S IN Z I M B A B W E C. A. LLEWELYN1'3, J. S. OGAA2 and M. J. OBWOLO2 1Department of Animal Science and ZFaculty of Veterinary Science, Universityof Zimbabwe, Mount Pleasant MP 167, Harare, Zimbabwe SUMMARY

Progesterone profiles were monitored in goats housed in single (n ~ 9) or group (n = 14) pens during winter (JJA) and spring (SON). Normal cycles (n = 97) were <<.30 days. Extended cycles (n = 45) were > 30 days and, except for one cycle with a persistent corpus luteum, had periovulatory periods of 10 to 20 days (n ~ 29) or averaging 65"1 days in length (n = 15), mostly characterised by recurrent oestrus and/or occasional transient rises in progesterone. The proportion of normal cycles occurring in winter was 87"5% (28/32) and 77"7% (42/54) for goats in single and group pens respectively, falling to 62"5% (15/24) and 37"5% (12/32) respectively in spring. The distribution o f normal vs extended cycles according to season was significant (P <0"05, single; P < 0,001 group pens). Goats housed communally experienced a greater fall in the percentage of normal cycles in spring, possibly due to increased stress associated with group feeding. Within each season, however, housing per se did not influence the distribution of normal vs extended cycles. For normal cycles, Harvey's Analysis of Variance showed that season was significantly associated with length of the periovulatory period (3"99 days (JJA) vs 5"79 days (SON); P
Tropical goats are reported to undergo ovarian activity throughout the year (Devendra and Burns, 1983). Seasonal patterns in the distribution of kidding are thought to be associated primarily with changes in the availability and nutritional quality of forage (Amoah and Gelaye, 1990), or restricted access to bucks (Chemineau, 1983). In Zimbabwe, seasonal changes in kidding rates have been attributed to improved conception rates ("flushing") following the increased availability of grazing after the onset of the rains in November. In the high-veld area of Mashonaland there is a sharp peak in the kidding rate in April/May at the end of the rainy season although the majority of kiddings are spread over the spring and early summer months from August to December (Hale, 1986). This corresponds to higher conception rates during the dry, winter season when the quality and quantity of forage is at its lowest. It is possible that in addition to factors such as nutrition and suckling (Llewelyn et ai., 1992) ovarian activity is regulated to some extent by season. Present address: c/o Croyland House, Dry Drayton, Cambridge CB3 8BU, England.

175

176

LLEWELYN E T AL.

A preliminary study was conducted to investigate ovarian activity throughout the year in a group of Mashonaland goats of the East African Dwarf phcnotype housed indoors and fed ad libitum (Llewelyn et al., 1993a). Ovarian activity was highest during the cool, dry winter months of June to August, declined during the hotter spring and summer months of September to February and rose again at the end of the rainy season. These results were not conclusive, since the decline in ovarian activity in spring coincided with moving the goats from single pens into group pens and thus might have been due to stress as the goats established a social hierarchy. In addition, the goats improved in body condition throughout the experiment. The aim of the present study was to investigate the effect of season on ovarian activity in goats housed either in single or group pens, under conditions of controlled body condition. MATERIALS AND METHODS

Animals Twenty-three full-mouthed, parous does were kept at the University Farm, latitude 17°5'S. in a covered barn with open sides from 1 June 1989 until 30 November 1989. To compensate for the slight reduction in lighting due to the solid roof, natural daylight was supplemented by overhead artificial lights from 06.00h to 16.00 h. All goats were dewormed, vaccinated against pulpy kidney and foot-andmouth disease, and acclimatised to their housing at the start of the experiment. Fourteen does were housed communally, divided between pens (n = 8, and n = 6) of 4 × 3 m. Nine goats were kept individually in wire-mesh pens of 2 × 1 m. Each goat received 0-75kg/day of good quality Katambora hay plus 125 to 500g/day of a concentrate supplement (Llewelyn et al., 1992) fed individually to maintain body condition scores between 2-5 (good) and 3"0 (very good) (Honhold et al., 1989). Blood was sampled 3 times a week by jugular venepuncture with vthylene-diaminetetra acetic acid as anti-coagulant (7"5 raM). Samples were transported on ice and plasma centrifuged within 2 hours of collection and stored at -20°C until assayed. Progesterone assay Progesterone was measured directly in 50 pl plasma aliquots by radioimmunoassay, as described by Llewelyn et al. (1993a). The minimum detectable level was 0.4 ng/ml with low, medium and high quality control pools giving mean (4-s.e.m.) values of 0-74 4- 0.04 ng/ml, 2-31 4- 0-52 ng/ml and 4"47 4- 0" 13 ng/ml respectively. Analysis of cycle length Plasma progesterone profiles were used to divide the cycle into 4 periods (days); 1) post-ovulatory rise from basal to >t 2-0ng/ml, 2) plateau period (_progesterone > 2.0 ng/ml), 3) fall from > 2.0 ng]ml to basal and 4) periow.flatory period when progesterone concentrations remained basal, including transient rises to ~< 1 ng/ml. The luteal phase was the sum of periods 1, 2 and 3 and cycle length the sum of 1, 2, 3 and 4. Cycles of ~ 30 days approximated a normal distribution curve and were classified as normal (N). In N cycles period 4 was 1 to 9 days duration. The remaining cycles ( > 30 days) were classified as extended (E). E cycles were divided into 3 groups: those in which period 4 was extended to 10 to 21 days or >21 days, and those which had a luteal phase exceeding 40 days.

OVARIAN ACTIVITY IN GOATS IN ZIMBABWE

177

Oestrns detection Two vasectomised males were kept within audible and olfactory range of the does and used for oestrus detection at 07.00 h and 15.00 hours. The males were alternated between morning and afternoon observation periods in order to minimise possible variations in oestrus detection due to male preferences (Llewelyn e t al., 1987). Does in group pens were observed for 20 minutes after introduction of the billy after which the does kept in single pens were hand-teased. The presence or absence of tail wagging was recorded separately as an indicator of oestrous behaviour (Llewelyn e t al., 1993b). Onset of oestrus was deemed to have occurred when the doe was served by the male and duration of oestrus estimated as the interval between first and last coitus. If a doe was served again after an interval of 3 successive observation periods (> 48 hours) without either showing signs of oestrous behaviour or having been served, this was recorded as a separate oestrus. In this study, the term oestrus detection refers to the presence or absence of oestrus within the periovulatory period 4 of each cycle. The oestrus detection rate ( % ) = sum of periovulatory periods in which oestrus was detected/total number of periovulatory periods recorded x 100. Seasons Winter comprised the cool, dry months of June, July and August (JJA), and spring the hot months of September, October and November (SON) at the onset of the rainy season. When alloting cycles to their respective seasons, those spanning both seasons were assigned to JJA if the mid-point of the luteal phase, period 2, fell on or before 31 August and to SON if the mid-point of period 2 fell on or after 1 September. Statistical analysis The association between season (JJA v s SON) and the distribution o f N vs E cycles was analysed separately for goats in single and in group pens, using the chi-square distribution. Likewise, within each season the distribution o f N vs E cycles according to housing (single vs group pen) was tested for significance. For normal oestrous cycles, Harvey's Least Squares Method (LSM) Analysis of Variance (1987), Model 3, with animals nested within pens, was used to test the significance of season (the primary factor) and housing (secondary factor) on various parameters of the cycle. No significant season x housing interaction was found so this was omitted from the model. Season (1 d.f.), housing (I d.f.) and animal x pen interaction (20 d.f.) were the main sources of variation. Season (JJA vs SON) was tested against the residual variation (74 d.f.) for significance, whilst housing (single vs group pen) was tested against the animal x pen interaction to provide a more robust test of significance. RESULTS

Animals Mean (±s.e.m.) weights and corresponding body condition scores were 34.1 ± 1.3 kg and 2.9 ± 0"08 respectively at the beginning of the study, and averaged 3 I" I ± 1"1 kg and 2.7 ± 0-04 at the end. These changes were not statistically significant.

Plasma progesterone profiles Figure 1 gives examples of plasma progesterone profiles from 8 goats throughout winter and spring. Some individuals tended to exhibit a higher proportion of cycles

178

LLEWELYN E r AL.

with extended pcriovulatory periods than others, but in general these cycles became more common in the spring. This trend was observed both for goats housed in single (Fig. la,b,c) or in group (Fig. ld,e,f,g,h) pens. The distribution of N vs E cycles was significantly associated with season, and was more pronounced for goats housed in a group (Table I). Within each season however, the distribution of N vs E cycles was not significantly associated with housing (single vs group pens). Extended cycles E cycles (n = 45) comprised a normal luteal phase followed by a prolonged a

b 4 w

w

¢

IM Z O w w

Ira

e

0

.

.

•

"

~

f

w

0 0

4

ck

0

,

.

;

v

,

4

'g

0

h

0

"'

1/7 W inter

1/8

1/g DATE

1/10

""

1111

" 1112

spr t ng

Flo. 1. Plasma progesterone profiles from individual goats housed in single (a,b,c) or group (d,o,f,g,h) pens throughout winter (JJA) and spring (SON). y = occurrence of oestrus+

179

OVARIAN ACTIVITY IN GOATS IN ZIMBABWE TABLE I Distribution of normal vs extended cycles according to season for indigenous does housed in single pens (n = 9) or group pens (n ffi 14) Extended cycles

Normal cycles

Periovulatory period

C.L. 1

(days)

(days)

10-20

> 21

> 40

No. of cycles

Single pens JJA 28 (87.5%) SON t5 (62-5%) Group pens JJA 42 (77-0%) SON 12 (37-5%)

I 4

3 5

0 0

32 24

11 13

1 6

0 1

54 32

Total

29

15

I

142

97 (68-3%)

l C.L. persistent corpus luteurn. Association between season (JJA vs SON) and incidence of normal vs extended cycles: a) single pens X2 = 4.81, P < 0-05; b) group pens X2 = 13"95, P < 0.001

(l d.f.). Association between housing (single vs group) and incidence of normal vs extended cycles i) winter X2 1.25(n.s.): ii) spring X2gg 3.44 (n.s.).

periovulatory period 4, with the exception of one cycle which had a persistent corpus luteum. The luteal phase of the latter had reached 47 days by the conclusion of the experiment. In 29 cycles, period 4 was 10 to 20 days in duration with an oestrus detection rate of 79% (23/29). The mean (4- s.e.m.) interval from corpus luteum regression to first service was either 4.3 4- 1-3 (n = 4) days or 16 4- 1.0 (n = 9) days when oestrus occurred only once during period 4. When oestrus occurred more than once within period 4, the mean (4-s.e.m.) inter-oestrus interval was 8"4+0.96 days (n = 10). Transient rises in progesterone took place 10.6-4-0-77 (n = 9) days from the start of period 4 with oestrus often occurring before and/or after the rise (see Fig. 1b,g,h). In 15 cycles, the mean (4-s.e.m.) length of period 4 was 65-6 4-5-1 days, with an oestrus detection rate of 87% (13/15). In 3 cycles oestrus occurred only once within TABL~II Analysis of variance (Harvey's Least Square means+ s.e.m.) for main effects of season and housing on components of the oestrous cycles and mean plateau progesterone ( P#) concentration in indigenous does Luteal phase length (days) Main effect of season JJA 18-004-0.29 SON 17.25 4- 0"37 Main effect of pen Single 16.931 :[:0.38 Group 18.324-0.35

Periovulatory period (days)

Overall cycle length (days)

Mean P4 cone. (ng/ml)

3-99~ -1-0 - 4 1 5'79 4- 0.48

22-004-0-33 23.05 4- 0,46

3-994-0-41 3-50 4- 0-22

4.31 4-0-55 5.484-0.49

21.242 4-0.40 23-814-0.39

3.63 4-0.25 3-86±0-23

I p < 0.01, 2p< 0"001 for comparisons within season or pen.

180

LLEWELYN ET AL. TABLE III

Analysis of variance (Harvey's Least Square means 4- s.e.m.) for main effect of season and pen on oestrous detection, timing of onset of oestrus and oestrus duration in indigenous does i

Oestrus detection (% O"10-2) Main effect of season JJA 0.872 :i: 0.06 SON 0.55 4- 0-09 Main effect of pen Single 0-70 4- 0-06 Group 0.75 4- 0-06

i

Time from basal to onset of oestros (days)

Duration of oestrus (days)

1.454-0-19 1.57 4- 0.27

1.941 4-0.16 1.13 4- 0.25

1-32 4- 0-23 1-7l 4- 0.22

1.42 4- 0.18 1-65 4- 0-18

IP<0.05, 2P<0.01 for main effect of season or pen.

period (4), with oestrus taking place on 2 separate occasions (n = 6 cycles) or t> 3 times (n ffi 4 cycles) in the remainder. Within period (4) 7 transient rises in progesterone were recorded with oestrus before and/or after the rise. Main effect of season and housing on normal cycles (Analysis of variance)

Season was significantly associated (P < 0-001) with the length of the periovulatory period 4, which was approximately 2 days longer in SON than in JJA, but not with luteal phase length, or overall cycle length (Table II). Mean progesterone concentrations during period 2 of the luteal phase were reduced in spring, but this was not significant (Table II). The lengthening of the periovulatory period of the cycle in spring (SON) was accompanied by a significant decrease in oestrus detection rate (P<0.01) and in oestrus duration (P< 0.05), with no apparent change in the timing of onset of oestrus (Table III). Housing was not significantly associated with periovulatory period length (Table II), oestrus duration, oestrus detection rate or interval to onset of oestrus, although the latter appeared to increase in goats housed in a group (Table III). In contrast, housing was significantly associated with duration of the luteal phase (P<0.01) and overall cycle length (P<0.001), which was about 2-5 days longer for goats in group pens (Table II). This seemed to be due to delayed luteolysis, with Harvey's least significant means for period 3 being 2-864-0.19 and 3.404-0.19 days for goats in single and group pens respectively (P < 0-01). Mean progesterone concentrations during period 2 of the luteal phase were not influenced by housing (Table II). DISCUSSION

The present study indicates that indigenous Mashonaland goats, housed inside and fed a maintenance diet of veld hay and concentrates, undergo limited seasonal changes in ovarian function. Cyclical ovarian activity was maximal during the cool, dry winter months ( J J A ) , and minimal in the ensuring hot, spring months (SON) at the start of the rainy season, confirming the results of a preliminary one year trial (Llewelyn et al., 1993a). Although body condition and suckling are probably the major factors regulating fertility under field conditions (Llewelyn et al., 1992), the present findings suggest that season plays a contributory role in determining the level of ovarian activity. A seasonal increase in ovarian activity around the start of the dry season (Llewelyn et al.,

OVARIAN ACTIVITY IN GOATS IN ZIMBABWE

181

1993a) would be associated with a higher incidence of conceptions between April and August and a preponderance of kiddings between August and December, as reported by Hale (1986). In this investigation 30% of cycles were classified as extended, as reported previously (Llewelyn et al., 1993a). These had a normal luteal phase followed by a periovulatory period exceeding 10 days; the majority (66-6%) of which were 10 to 20 days in duration with the rest averaging 65 days. The high rate of oestrus detection during the latter periods (79% and 87% respectively) showed that follicular development was continuing. Multiple heats were frequent, with inter-oestrus intervals of around 8 days in length, often punctuated by transient rises in progesterone to around 1 ng/ml. These profiles, termed "short oestrous cycles" by other workers, have been ascribed to ovulatory failure (BonDurant et al., 1981) or ovulation followed by premature corpus luteum regression (Battye et al., 1988). In temperate breeds, which have a well-defined breeding season following a period of anoestrus, short oestrous cycles are confined to the transitional phase at the start of the breeding season (Camp et aL, 1983; Ottet aL, 1980), when the frequency of pulsatile LH secretion is increasing to the level necessary to sustain regular ovarian activity (Chemineau et al., 1986). It is evident from this and other studies (Prasad and Bhattacharyya, 1979; Eiamvitayakorn et al., 1988; Skea et al., 1990; Llewelyn et al., 1993a), that short oestrous cycles are common throughout the year in non-temperate breeds, constituting around 10 to 20% of all cycles observed. In this study and that of Llewelyn et al. (1993a) it appeared that certain does were more inclined than others to exhibit this pattern of ovarian activity. Conceivably, continued exposure to the buck may be stimulating ovarian activity in does which would otherwise be experiencing "light" anoestrus. It would be interesting to compare conception rates following exposure to entire bucks to determine whether the incidence of short cycles bears any relationship to subsequent fertility. In this experiment, the oestrus detection rate during the periovulatory period of normal cycles fell from 85% in winter to only 57% in spring (Harvey's LSM: P<0.01). Does which were not served during the periovulatory period also failed to exhibit tail wagging, a behavioural sign closely correlated with oestrus (Llewelyn et al., 1993b). The fall in oestrus detection in spring therefore appeared to result from an increased incidence of silent ovulations, rather than a reduction in male libido. Nevertheless, the present results do not exclude the possibility that a fall in male libido in spring caused either by repeated use of the same 2 males as teasers throughout the experiment, or in response to increasing daylength and temperature (Roca et al., 1991), contributed to the reduction in oestrus detection. Oestrus duration in normal cycles was reduced from 1-9 days in winter to 1-1 days in spring (Harvey's LSM; P < 0.05). In West African Dwaffgoats seasonal differences in oestrus length were also reported by Akusa and Egbunike (1990), hut in their study oestrus increased in duration during the rainy season. It is not known how environmental changes trigger variations in oestrus length but presumably the mechanism involves modulation of the responsiveness of target receptor cells in higher brain centres to oestradiol. In normal cycles, the interval between corpus luteum regression and onset of oestrus remained unchanged irrespective of season, averaging 1-5 days. In contrast, the duration of the periovulatory period increased from 3.99 days in winter to 5.79 days (Harvey's LSM; P < 0"001) in spring. Since the time of ovulation was not ascertained, this increase could be due to either delayed ovulation or retarded development of the secretory function of the corpus luteum. Mean luteal progesterone concentrations

182

LLEWELYNET A L

were not reduced in the spring, but overall levels were low and fell within the range 3 to 4ng/ml. These values are 50% lower than those measured in temperate goats using a similar RIA method (Llewelyn et al., 1987) which may simply reflect between° breed differences in ovulation rate or corpus luteum function. On the other hand, low luteal progesterone concentrations following conception have been linked with increased embryonic mortality in sheep (Ashworth et al., 1989) and cows (Shelton et al., 1990). Recent studies in cows indicate that low plasma progesterone concentrations are associated with an enhanced luteolytic mechanism, as evidenced by an increased ability of the endometrium to synthesize PGF2a in response to oxytocin which could predispose some animals to early embryo loss (Lamming and Mann, 1993). In addition to the lengthening of the periovulatory period observed in normal cycles, the proportion of extended cycles, with periovulatory periods exceeding 10 days, also rose in the spring. These findings, suggestive of follicular dysfunction, are consistent with a decrease in the frequency of pulsatile LH release (Chemineau et al., 1986) and may be linked to changes in photoperiod. Angora goats, which are seasonal breeders in temperate zones, also exhibit a definite winter breeding season when kept in this region (Hofmeyr et al., 1965) even though the decline in daylength between the summer and winter solstice is only 3 hours. It is arguable that a change in photoperiod of this magnitude would not be sufficient to entrain seasonal changes in ovarian activity and that other factors such as temperature and humidity may act in concert with photoperiod to modulate the hypothalamic pulse generator and regulate LH release. The mechanism could involve secretion of endogenous opioid peptides (EOPs) which have been found to suppress LH secretion, particularly during the follicular phase of the oestrous cycle in sheep and cows (Cosgrove et al., 1993). Differences in EOP secretion in response to environmental factors might account for the large variation in the pattern of ovarian activity between individuals observed in this and the previous study (Llewelyn et al., 1993a). Responsiveness to season appeared to be influenced by housing in this experiment, as indicated by the much stronger association between season and distribution o f N vs E cycles for goats kept in group as opposed to single pens. Stress induced secretion of adrenal glucocorticoids may have impaired follicular development, thereby exacerbating the responsiveness to season in goats housed in groups (Moberg, 1991). The latter were probably subjected to greater stress at feeding time because they were removed one by one from their pen to consume their allotted concentrate rations. This is known to increase frustration and raise nor-adrenaline levels (Friend, 1991). Goats housed in groups had significantly longer cycles than goats housed in single pens (P <0.001). This was due to a corresponding extension of the luteal phase by about 2 days, apparently resulting from an increase in the time span of corpus luteum regression. As discussed above, goats housed in communal pens may have been subjected to greater levels of stress due to the feeding protocol employed. It is conceivable that increased stress levels interfered with oestradiol secretion by developing follicles, thus delaying the onset of pulsatile PGF2a secretion and functional luteolysis (Hughes et al., 1987). ACKNOWLEDGEMENTS

This work was supported by a research grant from the University of Zimbabwe. Dr Llewelyn was in receipt of a Fellowship from the Beit Memorial Trust (UK) sponsored through the Royal (Dick) School of Veterinary Studies (University of Edinburgh)

OVARIANACTIVITYIN GOATSIN ZIMBABWE

183

and received additional financial support from the Wellcome Trust and the International Atomic Energy Agency. We thank Dr P. Tawonezwi for statistical advice, Mr G. Mufandaedza for excellent technical assistance, and staff of the Department of Animal Science and the University Farm for their assistance. Accepted for publication October 1994 REFERENCES AsmvoxrH, C. J., SALES,D. I. & WILMtrr,I. (1989). Evidence of an association between the survival of erabryce and the preovulatory plasma progesterone concentration in the ewe. Journal of Repro~wtion and Fertility, 87, 23-32. AKusA,M. O. & EOBUNm~,G. N. (1990). Effects on oestrus duration of West African Dwarf goats. Small Ruminant Research, 3, 413--418. A~O^H, E. H. & GELXYE,S. (1990). Reproductive performance of female goats in south Pacific countries: Small Ruminant Research, 3, 257-267. B^1"rYe, K. M., F^UtCLOUOH,It. J., C~ERON, A. W. N. & TRotn~eN, A. O. (1988). Evidence for prostaglandin involvement in early luteal regression of the super-ovulated nanny goat (Capra hircus). Journal of Reproduction and Fertility, 84, 425-430. BONDUItAN'r,R. H., DApa~, B. J., MUNRO,C. J., STAB~VeLIYr,G. H. & W^NO, P. (1981). Photoperiod induction of fertile ocstrus and changes in LH and progesterone concentrations in yearling dairy goats (Capra hircus). Journal of Reproduction and Fertility, 63, 1-9. C ~ , J. C., WILVT,D. E., HOWARD,P. K., SrU^RT, L. D. & CH~RAeORrV, P. K. (1983). Ovarian activity during normal and abnormal length oestrous cycles in the goat. Biology of Reproduction, 28, 673681. CHEMrNV~u,P. (1983). Effect on oestrus and ovulation of exposing creole goats to the male at three times of the year. Journal of Reproduction and Fertility, 67, 65-67. CHEMINeAU,P., NORMAm',E., RAV^ULT,J. P. & TmMONIER,J. (1986). Induction and persistence of pituitary and ovarian activity in the out-of-season lactating dairy goat after a treatment combining a skeleton photoperiod, meiatonin and the male effect. Journal of Reproduction and Fertility, 78, 497-504. CosoRov~, J. R., DE Resin, F. & FOXCROFr,G. R. (1993). Opioidergic pathways in animal reproduction: Their role and effects of their pharmacological control. Animal Reproduction Science, 33, 373-392. DeVeNnXA,C. & BURNS,M. (1983). Reproductive performance. In: Goat Production in the Tropics. 2nd ed. Commonwealth Agricultural Bureaux. Farnham Royal, pp 74-89. E~MVlT^Y^KORN,J., Ric3oa, E. M., GARCI^,B. R. & Ape.LO,C. L. (1988). A study of the aberrant oestrous cycles in goats (Capra hircus). In: 1lth Congress on Animal Reproduction and Artificial Insemination. University College Dublin. Ireland, June 26 to 30, 1988. Vol 2. Brief Communications. FPJE~, D. H. (1991). Behvioural aspects of stress. Journal of Dairy Science, 74, 292-303. HALE,D. (1986). Systems of production and productivity of goats in three communal areas in Zimbabwe. In: Proceedings of the Workshop on the Improvement of Small Ruminants in Eastern and Southern Africa. (Eds K. O. Adenige and J. A. Kategele). Joint OAU/IBAR/IDRC workshop, 18 to 22 August, Nalrobi, pp 181-193. HARVEY,W. R. (1987). User's guide for least square and maximum fikelihood computer program. Ohio State University, Columbus, Ohio. HOFMEYR,H. S., JOUSERT,D. M., BADDENHORST,F. J. (3. & STETN,G. J. VAND. (1965) Adaptation of sheep and goats to a sub-tropicalenvironment. Proceedings of the South African Society of Animal Production, 4, 191-195. HONHOLD,N., PETIT,H. & HALLIWELL,R. W. (1989). Condition scoring scheme for Small East African goats in Zimbabwe. Tropical Animal Health and Production, 21, 121-127. Huos~s, 1". L., Vu.t~-G-ODOy,A., KESNER,J. & FOOWELL,R. L. (1987). Destruction of bovine ovarian follicles: effects on the pnisat~c release of luteinizing hormone and prostaglandin F2o~induced luteat regression. Biology of Reproduction, 36, 523-529. LAMMmO,G. E. & M x ~ , (3. E. (1993). Progesterone concentration affects the development of the luteolytic mechanism in the cow. Journal of Reproduction and Fertility. Abstract Series No 11. (no 8), p 9. LLBW~VNC. A., Luctass, A. G., MUNRO,C. D. & PERmE,J. (1987). Effects of 7".congolense on the ocstrous cycle of the goat. British Veterinary Journal, 143, 423-431. LLeWeLYS,C. A., Ooxx, J. S. & OSWOLO,M. J. (1992). Plasma progesterone concentrations during pregnancy and pseudopregnancy and onset of ovarian activitypost partum in indigenous goats in Zimbabwe. Tropical Animal Health and Production. 24, 242-250. LLEWeL~, C. A., OGAA,J. S. & OSWOLO,M. J. (1993a). Plasma progesterone profiles and variation in cyclic

LLEWELYN ET AL.

184

ovarian a~tivity throughout the year in indigenous goats in Zimbabwe. Animal Reproduction Science, 30, 301-311. LLeW~L~, C. A , I~tRIE, J., L ~ A. G. & M u ~ o , C. D. (1993b). Oestms in the British White goat: Timing of plasma luteinizing hormone surge and changes in behavioural and vaginal traits in relationship to onset of oestrus. British Veterinary Journal, 149, 171-182. Moe~ao, G. P. (1991). How behavioural stress disrupts the endocrine control of reproduction in domestic runfinants. ?ournai of Dairy Science, 74, 304-311. Orr, K. S., Nu~3N, D. R, & H[xos, J. E. (1980). Effeet of presence of male on initiation of oestrous cycle activity of goats. Theriogenology, 13, 183-190. PltASAD,S. P. & BHATACHAltYYA,N. K. (1979). Oestrous cycle length and behaviour in different seasons in Barbari nannies. Indian Journal of Animal Science, 49, 1058-1062. ROCA,T., MART[N~Z,E., V^s~uez, J. M., Rutz, S. & CoY, P. (1991). Influenceof season on testicle size and libido in male goats from the mediterranean area. Animal Production, 52, 317-321. SHeLTOS,K., G^YE~leDe ASR~U,M. F., HL~X~, M. G., P^~.~sos, T. J. & LAMMiNg,G. E. (1990). Luteal inadequacy during the early luteal phase of sub-fertile cows. Journal of Reproduction aad Fertility,

~o, 1-1o. S~eA, I., L ~ A , D., S~eA, R., NeUO~SAUeR,S. & M^'rrI~WMA~, R. W. (I990). Aspects of the breeding ~3~le of Galla goats at Marimanti. In: A study of Galla Goat Production in a Semi-arid Environment in Kenya. Report prepared for the Embu-Meru-Isiolo Goat and Sheep Project, Marimanti, February 1990. INFLUENCE DE LA SAISON ET DU LOGEMENT SUR L'ACTIVITE OVARIENNE DES CHEVRES INDIGENES AU ZIMBABWE R~mn6---Les profils de la progest6rone ont 6t~ survis sur des ch~vres log~-s en enclos soit individuellement (n = 9), soit en groupe (n ~ffi14), durant l'hiver (JJA) et le printemps (SON). Les cycles normaux (n = 97) ont 6t~ ~; ~ 30 jours. Les cycles prolong~ (n ~=45) ont ~t~ sup~rieurs ~ 30 jours. A l'exeeption d'un seul cas avec persistance du corpus luteum, ils ont pr~eent~ des p~riodes pr&ovulatoires de I0 ~ 20 j (n ffi 29), ou d'une dur~e moyenne de 65, I j, c a ~ le plus souvent par un uestrus r~urrent et/ou des mont~es transitoires et occasionnelles de la progesterone. La proportion des cycles normaux survenant en biver a 6t6 de 87,5 p. 100 (28/32) et de 77,7 p. 100 (42/54) respectivement pour les ch~vres log~s en enclos individuel ou en groupe. Elle est tomb~e respectivement ~ 62,5 p. 100 (15/24) et ~ 37,5 p.100 (l 2/32) au printemps. Ida distribution des cycles dits normaux face aux cycles prolong~s en fonction de la saison s'est r6v61~.esignificative (/'<0,05 pour le Iogement individuel, et P < 0,001 pour les enclos de groupe). I ~ eh~vres Iog~es collectivement oat present6 une chute plus 61ev6edu pourcentage des cycles normaux au printemps, peut~tre due ~ un accroissement du stress, ussoci6 ~ l'alimentationen gruupe. Cependant,/t l'int~,rieur de chaque saison, le Iogement en sol n'a pas influenc6 la distribution du factenr cycle normal/cycle prolong6. Pour les cycles normaux, l'analyse de la variance de Harvey a montr~ que la saison 6fair associ~e de fa~on significafive i la Ionguenr de la p~riode p~ri-ovulatoire: 3,99j (JJA) contre 5,79 (SON) avec P<0,001; taux de d~teetion de l'oestrus de 87 p. I00 (JJA) contre 55 p. I00 (SON) avee P < 0,01, et duroc de l'oestrus de 1,94 j (JJA) contre 1,13 j (SON) ave: P <0,05. A Finverse, la dur~ de la phase future n'a pas ~t~ affectb: par la saison, mais s'est trouv~e signiflcativementassoci~e avec le mode de Iogement (16,93 jours, en enclos individuel contre 18,32jours en enclos de groupe pour P < 0,01). La rbluction de l'activit~ ovarienne observ~e au printemps peut refl~ter une rbluction saisonni~re de la fertilitY, ~ventuellement li~e ~ l'~l~vation de la temperature et ~ l'accrois~ment de la photop~riode.

INFLUENCIA DE LA ESTACION Y EL ALOJAMIENTO SOBRE LA ACTIVIDAD OVARICA DE CABRAS INDIGENAS EN ZIMBABWE Resamen~Se registraron los niveles de progesterona de cabras alojadas individualment(n = 9) o en grupos (n ffi 14) en inviemo (JJA) y e n primavera (SON). Los ciclos nurmales (n = 97) fueron de ~ 30 dius. Los ciclos alargados (n = 45) fueron de > 30 d[us y, excepto pot un ciclo con un cuerpo Idteo persistente, los periodos preovulatorios fueron de 10 a 20 dlas (n = 29) o tuvieron un promedio de 65.1 dlas de duraci6n (n ~ 15), c a ~ o s e la mayoria por estro recurrente y/o pur ir~rementos transitorios oc~tsionales de progesterona. La propor~bn de cielos normales durante el invierno fue de 87.5% (28/32) y 77-7% (42/54) para cabrus alojadas individualmentey e n grupos respectivamente, disminuyendo a un 62.5% (15/24) y 37.5% (12/32) respeetivamente en primavera. La distribu~bn de delos normales w alargados de acuerdo alas estadones rue significativo (P < 0"05, individuales; P < 0.001, en grupos). Las eabras alojadas en grupos experimentaron tma calda mayor en el porcentaje de ¢iclos normales en primavera, posiblemente debido al mayor estr~

OVARIAN ACTIVITY IN GOATS IN ZIMBABWE

185

de la alimentacibn en grupo. Dentro de cada estaci6n, sin embargo, la estabulacibn per se no influy6 en la distribucibn cicios normales vs ciclos alargados. En los ciclos normales el an~lisis de varianza de Harvey mostr6 que la estaci6n estaba significativamente asociada con la duraci6n del periodo periovulatorio (3-99 dias JJA vs 5-79 dias SON; P < 0,001), la tasa de deteccibn del estro (87% JJA vs 55% SON; P < 0.01) y la duracibn del estro (1.94 dias JJA vs 1.13 dias SON; P < 0"05). Por el contrario, la duracibn de la fase hitea no se rib afectada por la estaci6n, aunque si estuvo significativamente asociada con el tipo de alojamiento (16.93 dias (cubiculos individuales) vs 18.32 dias (en grupo); P <0"01). La reducci6n en la actividad o v ~ c a observada en primavera podria reflejar una reducci6n e n la fertilidad estacional, posiblemente ligada al incremento de temperatura y al fotoperiodo.

UNIVERSITY OF EDINBURGH CENTRE FOR TROPICAL VETERINARY MEDICINE DIPLOMA]MSC POSTGRADUATE COURSES The following courses are available: i. Tropical Veterinary Medicine This course provides training in preventive medicine for veterinarians who will be concerned with planning and implementing the control of animal diseases at district, regional, national or international levels. It is designed to improve their ability to identify and assess the importance of such diseases in their own countries and to select and apply appropriate control measures. ii. Tropical Veterinary Science This course trains veterinarians to establish and direct veterinary diagnosis laboratories. It includes the organization of field investigations and surveys related to animals disease, particularly in tropical and sub-tropical countries and much hands-on training in laboratory processes. iii. Tropical Animal Production and Health A course taught in conjunction with the Institute of Ecology and Resource Management. This course is designed to meet the needs and in-service requirements of a wide range of animal related occupations. The overall aim of the course is to achieve a balanced and commonsense approach to livestock development in the tropics. A choice of modules is available. iv. Sustainable Development in the Tropics A course taught in conjunction with the Institute of Ecology and Resource Management. Tropical environments are more fragile than temperature ones. Students taking this course will learn how resources should be used in order to ensure that development programmes can be sustained in the long term. Flexibility is ensured by offering a wide range of modules. v. Tropical Animal and Crop Production A course taught in conjunction with the Institute of Ecology and Resource Management. This course is designed to provide students with an integrated approach to agricultural developments in the tropics. A wide choice of modules is on offer to ensure that students can tailor their courses to their specific needs. Further information can be obtained from Professor David W Taylor, Director of the Centre for Tropical Veterinary Medicine, Easter Bush, Roslin EH25 9RG, Midlothian, Scotland.