Reproductive performance of commercial sheep flocks in South Island districts

Transcription

1 New Zealand Journal of Agricultural Research ISSN: (Print) (Online) Journal homepage: Reproductive performance of commercial sheep flocks in South Island districts R.W. Kelly To cite this article: R.W. Kelly (1982) Reproductive performance of commercial sheep flocks in South Island districts, New Zealand Journal of Agricultural Research, 25:2, , DOI: / To link to this article: Published online: 09 Feb Submit your article to this journal Article views: 142 Citing articles: 12 View citing articles Full Terms & Conditions of access and use can be found at

2 N.Z. Journal of Agricultural Research 25 (1982): Reproductive performance of commercial sheep flocks in South Island districts 1. Flock performance and sources of wastage between joining and tailing R. W. KELLY Invermay Agricultural Research Centre. Ministry of Agriculture and Fisheries. Private Bag. Mosgiel. New Zealand In predominantly 4-tooth ewes from 64 flocks on 21 commercial sheep farms in four South Island districts were studied to identify various sources of reproductive wastage between joining and tailing. Average flock size was 426 ewes (range ). of the Romney. Coopworth. Perendale. Borderdale. Drysdale. Half-bred. and first cross (e.g. Coopworth x Romney) breeds. Flock live weights during the prejoining-joining period ranged from 42.8 to 72.7 kg. with mean values for all flocks between times of weighing ranging from 55.1 to 56.5 kg % of ewes were marked in the first period of joining. Rates of return to service were low. with a mean of 11.8% of marked ewes returning at least once. Ovulation rates. recorded on an average of 93 (50-168) ewes in each flock. varied from 1.12 to 2.31 corpora lutea per ewe (mean 1.6\). Lambs born per ewe lambing also varied considerably (100.&-192.7%). witha mean of 141.2%. Barrenness levels ranged from 1.0 to 14.4% with a mean value of 4.0%. Lamb deaths averaged 9.7% oflambs born. In 44 flocks it was estimated that 8.7% of single-born and 10.6% of twin-born lambs died between birth and tailing. Percentage of lambs tailed per ewe present at joining ranged from 8 1.8% to 163.0%. with a mean of 120.4%. From a potential of 161 lambs per 100 ewes in 57 of the flocks. determined by ovulation rates. various sources of reproductive wastage calculated as eggs or lambs lost per 100 ewes were: ewes falling to mate. 0.6; ewes mated-failed to lamb. 5.5; ewe deaths mating-tailing. 1.7; partial failure of multiple ovulation. 18.8; and lambs dying birtlhailing Partial failure of multiple ovulation. predominantly ewes having two corpora lutea but giving birth to one lamb. was responsible for a major proportion of the wastage recorded. Of ewes with two corpora lutea 29.6% (range %) gave birth to one lamb. Keywords sheep; Ovis aries; Bovidae; ewes; lambs; lambing; animal reproduction; mamalian re~roduction; animal production; conception; fertilisation; ovulation; animal breedmg INTRODUCTION The factors limiting lamb tailing percentages. and the contribution of different sources of reproductive wastage (Chopping & Lindsay 1970) to the total loss between joining and tailing in commercial sheep flocks in New Zealand is inadequately documented. Many workers have used information obtained by return mail surveys or limited farm visits (e.g.. Quinlivan & Martin a. b, c), but such Receired 2 September 1981; rerisivn 27 October 1981 studies are of necessity based on records readily taken by farmers. so that information that requires specialised skills (measuring ovulation rate) or markedly increases the workload of the farmer (recording live weights and crayon marks during joining) is invariably absent. As a consequence, neonatal losses are emphasised, whereas losses during pregnancy and low lambing potentials due to low ovulation rates largely go unrecognised. Recent approaches to estimating reproductive wastage in commercial sheep flocks have highlighted the importance of measuring ovulation rate (Chopping & Lindsay 1970; Cutten 1975), which also

3 176 N.Z. JOURNAL OF AGRICULTURAL RESEARCH, VOL. 25, 1982 permits losses from joining to lambing to be estimated. On-farm measurements of ovulation rate have been greatly facilitated by the use of laparoscopy, allowing large numbers of ewes to be examined in a short time with minimal interference to their subsequent performance (Kelly & Allison 1976 a). Using this and other techniques (Kelly et al. 1978), the reproductive performance of flocks from 21 commercial farms in the south of the South Island was studied between 1975 and This paper presents data on the ovulation rate, live weight, and mating and lambing performances of 64 flocks from these 21 farms, totalling ewes, and identifies the relative contribution of the various sources of reproductive wastage to total losses between joining. and tailing. o " ~METHVEN MATERIALS AND METHODS Conduct of the study The study ran from 1975 to Farmers willing to co-operate were selected in four South Island districts (Fig. I) to provide a range in lamb tailing percentages within districts. 21 farmers co-operated in the study; five from the Methven area were involved for 4 years, and of the remaining 16 from the other three areas, 13 were involved for 3 years, two for 2 years, and one for 1 year. One flock was studied on each farm in each year. The flocks were comprised predominantly of four-tooth ewes, with 20 farms supplying different animals each year-a total of 60 flocks. The remaining farm, studied in all four years, had only a flock of about 100 mixed-aged Romney ewes in which replacements had been selected for multiple birth-rank since the early 1950s. On this farm many ewes appeared in more than one year's record, but for the purpose of the analyses each year's perf6rmance has been regarded as being from a separate flock. Average size of the 64 flocks at joining was 426 ewes (range ). Th~ breeds of sheep on the 21 farms were Romney (7), Coopworth (4), Perendale (4), Borderdale (1), Drysdale (1), Half-bred (I), and first cross (e.g., Coopworth x Romney) (3). Farmers were asked to manage the study flocks in the 'normal' farm manner, although some preferential treatment may have occurred since the ewes had to be run as a separate flock durin~ joining and lambing to facilitate recording. All'ewes were individually identified with numbered eartags. The following information was recorded for each flock of ewes: Live weights A random sample of 100 ewes from each flock was weighed 2-3 weeks before both joining and lambing. Weighing was usually carried out 2-4 hours after yarding. Live weights were also recorded after an overnight fast in an average of 99 ewes (range ) from the draft retained for laparoscopy. In Fig. I - Localities of21 farms involved in the study and 1978 a random sample of 100 ewes from 37 flocks was also weighed following yarding at the end of the second period of joining. Mating records Rams were fitted with "Sire-Sine" harnesses and crayons, and crayon colours were changed on Days and (end of first and second period of joining, respectively). The day rams were first joined with each flock (varied from 10 to 28 April) was taken as Day O. Crayon marks on all ewes were recorded at the end of the second and third period of joining, and rams were removed at about Day 51. In addition, marked ewes were recorded at the first crayon change in flocks where marks were light. From the crayon information, the percentage of ewes marked, and returns to service, in each of the three periods of joining were calculated. Ovulation rate Ovulation rate-number of corpora lutea (CLs) per marked ewe was recorded by laparoscopy (Kelly & Allison 1976a) in a random sample of marked ewes. The operation was usually performed at the end of the first period of joining. However, this was not possible in several flocks, the extreme occasions being ol1e flock in which'laparoscopy was performed on Day 12 of joining, and another flock examined on Day 20. In 1975,50 marked ewes were laparoscoped per flock (five flocks) but in sufficient were laparoscoped to provide ewes with two

4 KELL y; REPRODUCTIVE PERFORMANCE OF SHEEP 177 TABLE I - Mean live weights, changes in mean live weights, and range between flocks during the period of prejoining-joining and before lambing. No. of flocks Mean Range Flock live weight at I. 2-3 weeks prejoining end Period I of joining end Period 2 of joiningt weeks prelambing Live weight changes \. to to to 3t to to 4t to t to tdata only for 1977 and TABLE 2 - Mean and range in values between flocks of performance during joining and ovulation rates. No. of flocks Mean Range Rams per IOQ ewes % ewes first marked in Period I Period Period mated, no mark recorded not marked Returns to Service Period Period Period Period Total returnst Ovulation rate per marked ewe tpercentage of ewes that returned to service (irrespective of whether once or twice) for the total flock. TABLE 3 - Mean and range in lambing performances between flocks. LB/EL, lambs born per ewe lambing; L T IEJ, lambs tailed per ewe joined. No. of flocks Mean Range ewes barren Barren ewes (% ewes joined LB/EL (%) Lamb deaths (% lambs born) LT/EJ (%) Lamb deaths-singles twins TABLE 4 - Sources of reproductive wastage Mean and range in components of reproductive wastage per 100 ewes from joining to tailing for 57 flocks with complete records. (Potential, i.e., ovulation rate) Ewes-failure to mate mated, failed to lamb deaths mating-lambing partial failure of multiple ovulations Lambs-deaths birth-tailing Eggs or lambs lost Mean Range (161.0) ( )

5 178 N.Z. JOURNAL OF AGRICULTURAL RESEARCH, VOL. 25, 1982 ovulations or at least 80 ewe records in flocks with low ovulation rates (i.e., :$ 130%). This latter procedure ensured that there was a 95% chance of the ovulation rate of the sample of ewes being within 10% of the true value for the flock (Kelly et al. 1978) Lambing records At lambing, all ewes laparoscoped in 1975 and 1976, and only the laparoscoped ewes with multiple ovulations in 1977 and 1978, were fitted with neck tags, and individual lambing performances (date of birth, number of lambs born) were recorded by the farmer. From this information actual losses from mating to lambing in multiple ovulating ewes were calculated. Lamb deaths, number of lambs tailed, and ewe deaths were noted for each flock. In 44 flocks the farmers attempted to raddle-mark at birth all multiple-born lambs, to permit birthrank identification of dead lambs. Birthrank was estimated in lambs dying before raddle marking. To calculate percentage of lamb deaths within birthranks it was assumed that only single and twin births contributed to lambs born per ewe lambing (LB/EL), so allowing calculation of the number oflambs born within these birthranks. At tailing (Iambs 2-3 weeks old) the number of barren ewes was recorded, with tag numbers being noted in the majority of flocks. The number of ewes present in each flock was checked throughout joining, at the pre-lamb weighing and at tailing. Small losses that were unaccountable by ewe deaths occurred between joining and lambing. It has been assumed that these losses were due to a loss of eartags, and hence all calculations were based on the number of ewes at joining that could be accounted for at tailing (i.e., number of ewes present at tailing plus number of ewe deaths since joining). Analyses of data Data are presented as mean performance and range in values for each variable for all flocks examined. These values include variation from year to year as well as amongst flocks within years. From the information on mean live weight ofthe flocks before and during joining, and just before lambing, mean live weight changes for each flock over the prejoining-joining period and fromjoining to pre lambing were calculated. The components of reproductive wastage from joining to tailing per 100 ewes present at joining were estimated for each flock as detailed by Kelly et ai. (1978), 'who discussed errors associated with these records. RESULTS The number of flocks, and means and ranges between flocks for each variable are presented in Tables Live weight (5kg interval) 1 I Fig. 2 - distribution of mean live weights for 64 flocks at about the end of Period 1 of joining (grouped into 5 kg weight intervals) r- r-- r-.--- o I I I n Mean weight change (kg) Fig. 3 - Distribution of mean live weight change within 63 flocks from 2 to 3 weeks prejoining to about the end of Period I of joining (grouped into 2 kg weight intervals). Live weight There was little variation in the overall mean live weight for all flocks between the times of weighing (Table I). Similarly, within each time of weighing, between years' variation was small, the greatest difference for all flocks being at the measurement 2-3 weeks before lambing, when overall mean live weights ranged from 53.4 kg in 1976 to 57.8 kg in These results contrast with the substantial differences between individual flocks in their mean live weight and change in live weight between times of weighing (Table I). At the four times of weighing the range in live weights between flocks was 24.9, 27.9, 29.9, and 31.9 kg. The distribution of flocks over the fasted live weight range recorded at about the end of -

6 16 ~8 E :l Z 4 O~-L-L-L~L-~~-L~~~~~~~~_ o Marked ewes returning to service (DID) Fig. 4 - Distribution of mean total rates of return to service (0/0) for 63 flocks (grouped into 2% intervals) ~8 ;;::: 0 6 ~4 E ~2 O~-L-L~~L-L-~-L~~ L-~~~ 1 20 lao Mean ovulation rates (NO. CLs per marked ewe) Fig. 5 -Distribution of mean ovulation rates of 64 flocks (grouped at 0.1 CLs per ewe intervals) ~ E :l Z4 O~~ ~~L-~ L--L L--L L-~ o >18 Lamb deaths Ilambs born (%) Fig. 6 - Distribution of percentage of lambs dying from birth to tailing of lambs born for 58 flocks (grouped at 2% intervals). KELLY: REPRODUCTIVE PERFORMANCE OF SHEEP 179 the first period of joining (time of laparoscopy) is. shown in Fig. 2. For each flock the change in mean live weight between successive weighings ranged from losses of up to 9.5 kg to gains of up to 11.5 kg. The change in mean live weight for 63 flocks from 2-3 weeks before joining to about the end of Period 1 of joining, an interval of approximately 5 weeks, is shown in Fig. 3. This figure underestimates (probably by up to 4 kg per flock) the true change in live weights between these two weighings, as the weight measured at about the end of Period 1 of joining followed an overnight fast, whereas all other weighings were made within a few hours of removal from; pasture. Mating performance 86.1 % of ewes were marked in the first period of joining (Table 2). Of the ewes 'that were not marked in the first period of joining, most were marked in the second period of joining, with very few (0.3%) being first marked in the third period. On average 6.7% of ewes were recorded as being not marked at all, but lambing records indicated that only 0.4% of ewes in the flock from the not-marked category failed to lamb. As a consequence the remaining 6.3% have been noted as mated, no mark recorded (Table 2). It seems reasonable to assume that the majority of these 'mated, no mark recorded' were in fact mated in the first period of joining; harder crayons were used during this period and resultedin lighter marks. A more accurate indication of the mean percentage of ewes marked in Period 1 would therefore be 92.4% ( ). Mean rates of return to service from Period 1 to Period 2 of joining were 9.7% (Table 2). 1.3% of marked ewes from Period 1 of joining returned in Period 3. Overall, 11.8% of marked ewes in all flocks returned to service at least once. The range in total rate of return to service was large (Fig. 4), with 9 flocks (14%) having total rates of return greater than 18%. In the extreme case, where 27.8% of ewes returned to service during the period of joining, subsequent ram inspection identified a high incidence (-40%) of Brucella ovis infection. Zero rates of return to service occurred when the number of ewes marked in a period was very small. Ovulation rate The mean number of marked ewes laparoscoped per flock was 93 (range 5~168). The mean ovulation rate was 1.61 (Table 2), but varied by a factor of 2 between flocks (Fig. 5), with values ranging from 1.12 to 2.31 CLs per ewe. Lambing performance Lambing data are summarised in Table 3. Barrenness percentages ranged between flocks from 1.0 to 14.4%, with only two flocks having levels exceeding 10% (12.2 and 14.4%). The mean value was 4.0%.

7 180 N.Z. JOURNAL OF AGRICULTURAL RESEARCH, VOL. 25, 1982 Lambs born per ewe lambing (LB/EL) varied considerably between flocks ( %), with a mean value of 141%. The distribution of lamb losses between flocks is shown in Fig. 6. Lamb deaths averaged 9.7% of lambs born, with the maximum loss of 23.1% recorded in the flock with the highest lambing performance. High death rates in other flocks were commonly associated with adverse weather conditions (e.g., snow storms). In the 44 flocks in which farmers attempted to record birthrank of dead lambs, it was estimated that 8.7% of single-born and 10.6% of twin-born lambs died. The percentage of lambs tailed per ewe joined (LT/EJ) showed a twofold range ( %) with a mean value of 120.4%. The distribution of percentages between flocks is shown in Fig. 7. Reproductive wastage The components of reproductive wastage are summarised in Table 4 for the 57 flocks in which records are available for all sources of wastage. From a mean potential of lambs per 100 ewes (mean ovulation rate for the 57 flocks), 26.6 were lost between joining and lambing, the majority (18.8 = 71%) through failure of ewes with multiple ovulations to carry all ova through to lambs born-termed partial failure of multiple ovulation. For the 61 flocks in which more than 10 laparoscoped ewes with two CLs lambed (mean number of ewes = 42), a mean of 29.6% of ewes gave birth to only one lamb. The distribution ofthese two partial wastage estimates is given in Fig. 8. Partial failure of multiple ovulation contributed between 2.3 and 41.5 potential lambs lost per 100 ewes, and partial failure was recorded in % of ewes with two CLs. From birth to tailing a further 13.1 lambs died per 100 ewes. These losses from joining to tailing total to a mean of39. 7 of the potential lambs, and give a tailing percentage of 121.3%. This differs by 0.9% from the mean value for all flocks (120.4%), since not all flocks contributed to the calculation of the wastage components. 12, r--- r-- r-- - r-- r Lambs tailed/ewe joined ("!o) Fig. 7 - Distribution of lamb tailing percentages (LT/EJ) for 61 flocks (grouped at 10% intervals). ~ ~O~~~~~~~~~~~-- ' Q).0 E :::l Z Lambs lost per 100 ewes because of partial failure of multiple ovulation 60 DISCUSSION The aim of the study was to examine the factors limiting lamb tailing percentages in commercial sheep flocks, including sources of reproductive wastage. Twenty-one farms contributed 64 flocks to the study over a period of four years. The range in lamb tailing percentages between flocks was similar to that recorded by the New Zealand Meat and Wool Boards' Economic Service over the same period for comparable South Island fattening-breeding and intensive fattening farms, although the mean tailing percentages were 1{}-15% greater. With lamb tailing percentages ranging from 81.8 to 163.0% it was not surprising to also record considerable variation between flocks in mean live 0~O----lLO-L-2~O~~3-0~-4~O~--5~0~~60 Ewes with 1 lamb born of ewes with 2 CLs that lamb (%) Fig. 8 - Distribution of measurements of loss caused by partial failure of multiple ovulation: lambs lost per 100 ewes (57 flocks) and percentage of ewes with I lamb bom of ewes with 2 CLs that lamb (61 flocks). weight ~nd ovulation rate, because of the well documented associations between live weight and lambing performance (Coop 1962; Hight & Jury 1973; and many others-reviewed by Allison & Kelly 1978) and ovulation rate (Kelly & Allison 1976b; Kelly 1980). The inter-relationships between

8 the various measurements taken during the course of this study will be critically examined in a subsequent paper. Mating performances indicated that the majority (- 80%) of the ewes in the flocks conceived during the first period of joining, and therefore that lambing would be concentrated over three weeks. These high conception rates with flock mating are consistent with the high fertilisation rates (83-98%) recorded in Romney ewes of similar age (2~ years old) by Quinlivan et al. (1966). The number of ewes failing to mate was extremely low, so this source of wastage does not represent a major source of reproductive loss. These results contrast with the performances recorded in some North Island flocks (Kelly & Knight 1979) in which 17% of ewes were not first marked until the last period of joining, consequently having only one opportunity to conceive 'fith resultant higher levels of barrenness. The mean values for rates of return to service compare favourably with those recorded by Allison (l975a, b; 1977) following flock mating. Returns to service in successive periods ( ) represent losses due to failure of insemination or fertilisation. or early embryonic mortality. The mean rates of return to service for successive periods were low. and within the levels generally accepted as normal. Excessive rates of return to service were associated with a high incidence of Brucella ovis infection in the rams, or in ewes returning during Period 2-3 of joining with small numbers of ewes marked in Pe'riQd 2 viz. maximum record of 50% was for two ewes in Period 2, one of which was re-marked in Period 3. Despite the obvious difficulty in determining the exact incidence oflong intervals (greater than 17 days) between returns to service. arising from recording raddle marks only at the end of each period of joining, the mean value of 1.3% for Period 1-3 returns indicates it is probably low. This suggests that complete failure of pregnancy beyond about Day 12 post mating, which leads to long return to service intervals (Edey 1967; Sawyer & Knight 1975), is a minor source of reproductive loss. The incidence of barrenness (mean value 4.0%) was similar to that of 3.8% recorded by Quinlivan & Martin (l971c) in commercial Romney flocks. For the 847 ewes that were barren and in which tag numbers were recorded, 11.2% were not marked during joining, 55.3% were marked once only (usually in Period 1) and 33.5% had returned to service at least once. In comparison, Quinlivan & Martin (l971b) reported that of barren ewes in stud flocks 8.3% were not marked, 42.3% were marked once only, and 49.3% were repeat breeders. The higher percentage of repeat breeders in the latter study is undoubtedly the result of differences in mating management, namely flock mating in commercial flocks where groups of rams are joined with groups KELL y: REPRODUCTIVE PERFORMANCE OF SHEEP 181 of ewes compared with single sire mating in stud flocks. With flock mating, individual ram infertility is largely overcome and conception rates for the flock approach that of the higher fertility rams, as suggested by the results from running vasectomised rams with entire rams at joining (Fowler & Jenkins 1970). Lamb death rates (mean value 9.7%) Were slightly lower than the % recorded by' Quinlivan & Martin (1971a, b), and substantially lower than the levels of 19% which have been reported under hill country conditions (Hight & Jury 1969; Dalton et al. 1980). The relative contribution of the various sources of reproductive wastage to total losses that occur between joining and tailing highlights the overall importance of partial wastage-ewes having fewer lambs than eggs shed at mating. This source of wastage accounted for approximately half of the mean total losses recorded. In order to examine whether changes may occur during joining in the magnitude of this source of wastage, partial losses in laparoscoped ewes were related to day oflambing in While it is recognised that variation in length of gestation would disguise any daily changes during joining, such an approach should yield information on trends. Mean partial failure in ewes with two CLs in the 15 flocks examined ranged from 10.5 to 42.9% (mean number of ewes with two CLs per flock = 41). Overall mean lambing day for the ewes with two CLs that had two lambs was 9.6 and for the ewes that had two CLs and one lamb it was 9.7, with animals in each category being similarly distributed throughout the period oflambing. Therefore there is no evidence to suggest that partial wastage rates change during joining.. Kelly (1980) reviewed information on partial wastage in the ewe, and concluded that there were negligable losses after about Day 30 of pregnancy. Age, genotype, nutrition, hormonal conditions, site of ovulation, and environment have all been reported to influence embryonic mortality. However, in many studies only extreme experimental treatments have been shown to influence the degree of wastage. Such extemes are unlikely to occur on farms over the period of joining. Nevertheless, a large range of partial wastage in ewes with two CLs was recorded between flocks (Fig. 8). Whether this range in losses represents the true underlying variability between flocks, or is an artefact of sampling error (performance of only 42 ewes from the total flock examined) cannot be ascertained from the present data. It was apparent from 20 farms in which at least two separate flocks were studied over different years that there was no consistency between flocks in amount of partial wastage in ewes with two CLs. Recent evidence

9 182 N.Z. JOURNAL OF AGRICULTURAL RESEARCH, VOL. 25, 1982 (Hanrahan, pers. comm.) suggests that repeatability of embryo survival in ewes with multiple ovulations is low. There are marked dissimilarities in the extent of the various sources of reproductive wastage recorded between the present study and that of Knight et al. (1975) in Western Australian sheep flocks. They recorded mean total wastage rates per 100 ewes of 51 potential lambs, with losses from mated ewes failing to lamb contributing 45% of this total. In contrast partial failure of multiple ovulation was the major source of wastage in this study (47% of total) with mated ewes failing to lamb accounting for only 14% of the mean total losses. The mean total losses due to ewe barrenness and deaths between mating and lambing was 7.8 potential lambs per 100 ewes (20% of total). Flocks studied by Knight et al. (1975) had a mean ovulation rate of 1.15, compared with 1.61 in this study. With a low incidence of multiple ovulations it is not unexpected that partial failure is relatively less important as a source of wastage. This contrast in results between the two studies emphasises the importance of determining ovulation rate in investigations of the sources of reproductive wastage in sheep flocks. Studies not incorporating this measurement have invariably over-emphasised the importance of ewe barrenness and lamb mortality (e.g., Connors 1971; Quinlivan 1977). In conclusion, with the aim towards flocks with high lambing performance, it is evident from these results that partial wastage will contribute significantly to the total reproductive losses. Much of this source of loss may be unavoidable because of the elimination of unfit genotypes (Bishop 1964: Long & Williams 1980). However, there may also be a component which could at least be minimised through modifications in management. Unfortunately there is a paucity of studies on factors that may influence partial wastage, the main effort in embryonic mortality being directed toward losses that occur in single ovulating ewes. Results yielded from the latter type of study may not be applicable to the partial wastage phenomenon. Acknowledgments Valuable and essential assistance in all facets of the field work was given by Fann Advisory and Sheep and Beef Officers of the Ministry of Agriculture and Fisheries at Ashburton, Timaru, Oamaru, Dunedin, Alexandra, and Gore. Technical support was provided by staff of the Anim~1 Producti~n Section, and assistance with preparation of this manuscnpt by Drs G. Montgomery, A. Allison, and Mr P. Johnstone, Invennay. Finally, I am extremely grateful for the effort, perserverance, and willing assistance provided by the farmers who collaborated in this work. REFERENCES Allison, A. J. 1975a: Flock mating in sheep. 1. Effect of number of ewes joined per ram on mating behaviour and fertility. New Zealand journal of agricultural research 18: b: Ewe and ram fertility in commercial flocks mated with differing numbers of ewes per ram. New Zealand journal of experimental agriculture 3: : Flock mating in sheep. 2. Effect ofnumbel of ewes per ram on mating behaviour and fertility of two-tooth and mixed-age Romney ewes run together. New Zealand journal of agricultural research 20: Allison, A. J.; Kelly, R. W. 1978: Some effects of live weight and breed of ewe on fertility and fecundity. Proceedings of the Sheep and Beef Cattle Society of the N.Z. Veterinary Association, 8th seminar: Bishop, M. W. H. 1964: Paternal contribution to embryonic death. Journal of reproduction andfertility 7: Chopping, M. H.; Lindsay, D. R. 1970: An approach to the study of reproductive wastage in sheep. Proceedings of the Australian Society of Animal Production 8: Connors, R. W. 1971: Sheep fertility services. Wool technology and sheep breeding 181 I): Coop, I. E. 1962: Liveweight-productivity relationships in sheep. I. Liveweight and reproduction. New Zealandjournal of agricultural research 5: Cutten, I. N. 1975: A technique for the detennination of ovulation rate and reproductive wastage on an individual ewe or flock basis. Agricultural record 2: Dalton, D. C.; Knight, T. W.; Johnson, D. L. 1980: Lamb survival in sheep breeds on New Zealand hill country. New Zealand journal of agricultural research 23: Edey, T. N. 1%7: Early \!mbryonic death and subsequent cycle length in the ewe. Journal of reproduction and fertility /3: Fowler, D. G.; Jenkins, L. D. 1970: The effectoffertility of the ram group on the reproductive perfonnance of a flock. Proceedings of the Australian Society of Animal Production 8: Hight, G. K.; Jury, K. E. 1969: Lamb mortality in hill country flocks. Proceedings of the New Zealand Society of Animal Production 29: : Hill country sheep production. IV. Ewe live weights and the relationship oflive weight and fertility in Romney and Border Leicester x Romney ewes. New Zealandjournal of agricultural research 16: Kelly, R. W. 1980: Components of reproductive wastage in sheep. Proceedings of the Sheep and Beef Cattle Society of the New Zealand Veterinary Association, 10th seminar: Kelly, R. W.; Allison, A. J. 1976a: Measurement ofovulation rates by laparoscopy and effects on reproductive perfonnance. Proceedings of the New Zealand Society of Animal Production 36: b: Returns to service, embryonic mortality and lambing perfonnance of ewes with one and two OVUlations. In: Tomes, G. J.; Robertson, D. E.; Lightfoot, R. J. eds. Sheep breeding. Proceedings of the 1976 internatio'nal congress. Western Australian Institute of Technology. p

10 Kelly, R. W.; Allison, A. J.; Johnstone, P. D. 1978: A system of identifying factors limiting the reproductive performance of commercial sheep flocks. Proceedings of the New Zealand Society of Animal Production 38: Kelly, R. W.; Knight, T. W. 1979: Sources of reproductive failure in commercial sheep flocks. Proceedings of the Ruakurafarmers conference 31: Knight, T. W.; Oldham, C. M.; Smith, J. F.; Lindsay, D. R. 1975: Studies of ovine infertility in agricultural regions in Western Australia: analysis ofreproductive wastage. Australian journal of experimental agriculture and animal husbandry 15: Long, Susan E.; Williams, C. V. 1980: Frequency of chromosomal abnormalities in early embryos of the domestic sheep (Ovis aries). lournal of reproduction andfertility 58: Quinlivan, T. D. 1977: Investigation and correction of poor reproductive performance in sheep--method. Proceedings of the New Zealand Veterinary Association's 7th Sheep Society seminar: KELLY: REPRODUCTIVE PERFORMANCE OF SHEEP 183 Quinlivan, T. D.; Martin, C. A. 1971a, b, c: Survey observations on the reproductive performance of both Romney stud and commercial flocks throughout New Zealand. a. I. National Romney stud performance b. II. Lambing data from an intensive survey in stud flocks c. III. National commercial flock performance. New Zealand journal of agricultural research 14: a , b ,c Quinlivan, T. D.; Martin, C. A.; Taylor, W. B.;Cairney, I. M. 1966: Estimates of pre- and perinatal mortality in the New Zealand Romney Marsh ewe. 1. Pre- and perinatal mortality in those ewes that conceived to one service. lournal of reproduction andfertility II: Sawyer, G. J.; Knight, T. W. 1975: Detection of embryonic death and observations on resorption and subsequent fertility in the ewe. Australian journal of experimental agriculture and animal husbandry 15: