Trap. An/In. Hlth Prod. (I~8) ~ , 57-64

E F F E C T S O F S U P P L E M E N T A R Y F E E D I N G A R O U N D L A M B I N G T I M E O N T H E P R O D U C T I V I T Y O F

J A V A N E S E T H I N - T A I L E W E S

T. D. CHANIAC, O, A. NATASASMITA I and I. C. FLETCHER 2

Balai Penel'aian Ternak (Research Institute for Animal Production), PO Box 123, Bogor, Indonesia

SUMMARY

The extent to which ewe reproduction could be improved by supplementary feeding around lambing time was investigated with four groups each of 16 ewes which were run continuously with fertile rams until they had lambed three times. One group was fed at a maintenance level throughout and the other three groups were given supplementary feed during the last four week of each pregnancy only, or during the last four weeks of pregnancy plus the first six weeks after each lambing, or during the last four weeks of pregnancy plus the first 12 weeks after each lambing. The average body weight of lambs weaned at 13 weeks of age increased (P <0.001) progressively from 6.0 to 9.1kg with increasing duration of supplementary feeding. There were however no significant differences between treatment groups in mean interval between successive lambings (202 days), ovulation rate (2.26), litter size (1.58), lamb birth weight (1.8 kg), preweaning lamb mortality (37%) or number of lambs weaned per lambing (0.99).

INTRODUCTION

The average number of lambs born per breeding ewe per year from Indonesian sheep in village situations (Bell, Inounu and Subandriyo, 1983) represents barely half of their potential productivity expressed under experimen- tal conditions of unrestricted nutrition, adequate disease/parasite control and intensive mating management (Obst, Boyes and Chaniago, 1980). Differences in the average number of parturitions per breeding female per year (0-7 to 1-1 and 1.5 to 1-8 under village and experimental conditions respectively) appear to account for most of the disparity while differences in average litter size (1.3 to 1.9 and 1.6 to 2.1 under village and experimental conditions respectively) appear to be of lesser importance. Average ewe body weights of about 25 kg in village situations (Inounu, Bell, Sitorus and Subandriyo, 1983) compared with 40 kg or more under experimental conditions of unrestricted nutrition (Fletcher, Guna- wan, Hetzel, Bakrie, Yules and Chaniago, 1985) show clear evidence of nutritional differences between the two management situations. Extensive evid- ence of nutritional effects on reproduction in temperate sheep breeds (Gun, 1983) indicates that nutrition may well be one of the factors limiting the lamb production of Indonesian village sheep.

The limited capability of most Indonesian farmers to improve the nutrition of their sheep (Basuno and Petheram, 1984) precludes village adoption of the unrestricted feeding levels applied in previous investigations. The aim of the present experiment was to determine whether the productivity of ewes fed at a

t Faculty of Animal Husbandry, Bogor Agriculture University, Bogor, Indonesia. 2 Address for reprints: Department of Agriculture, Rosedale, SA5350, Australia.

57

58 CHANIACrO, NATASASMITA AND FLETCHER

maintenance level in a continuous mating system could be increased by more practicable feeding regimes which provided only limited periods of supplementary feeding around lambing time.

MATBR/AL~ AND METHODS

Young ewes of the Javanese thin-tail breed were purchased from local markets in West Java. The reproductive history of the ewes was unknown but the fact that they weighed only about 15 kg and had either none or only two permanent incisor teeth indicated that most if not all of them had probably never lambed. All animals were vaccinated against anthrax, foot-and-mouth disease, haemorrhagic septieaemia and elostridial disease and treated with a broad spectrum anthelmintic. They were maintained in animal houses in group pens and fed freshly cut elephant grass (Penniaetum purpureum) ad lib. with varying amounts of a commercial concentrate (Beef-kwik, Cargill) and unrestricted drinking water. Elephant grass was harvested after seven to nine weeks growth and contained about 9 NLl/kg metabolisable energy and 9 to 10% crude protein. The commercial concentrate contained about 10 MJ/kg metabolisable energy and 14% crude protein and included a multi vitamin/mineral mix. The ewes were held apart from rams and fed a common ration slightly in excess of their maintenance requirement for approximately 12 months before the beginning of the experiment.

Sixty-four ewes were allocated by stratified random sampling on the basis of body weight and number of permanent incisor teeth (two or four) to four treatment groups. One group (TO) was fed at a maintenance level throughout the experiment: this had been ascertained by trial-and-error feeding during the I2 month pre-experimental period as ad lib. grass with a daily supplement of 150 g concentrate. The other three groups were given in addition 300 g concentrate/head/day either during the last four weeks of each pregnancy only (T4), or during the last four weeks of pregnancy plus the first six weeks after each lambing (T10), or during the last four weeks of pregnancy plus the first 12 weeks after each lambing (T16). It was postulated on the basis of current knowledge of temperate sheep breeds that T4 could have favourable effects on lamb birth weight and ewe milk production. T10 an additional effect on milk production and a possible stimulation of ewe post-partum reproduction and T16 an enhancement of T10 with additional lamb growth from direct ingestion of concentrate by the lambs.

Ewes were run continuously with rams for two years or until the progeny from their third lambing had been weaned whichever came first. Supplementary feeding commenced four weeks before the initial joining and was continued for four, '10 and 16 weeks in the T4, TI0 and T16 groups respectively. Thereafter ewes were treated on an individual and not on a group basis because of the spread of conception dates which became progressively wider with each succes- sive lambing. Thus pre-partum supplementary feeding commenced 17 weeks after the last oestrus recorded for each ewe (assuming a 21-week gestation) and post-partum supplementary feeding was terminated for each ewe according to the recorded day of parturition.

Six fertile rams fitted with Sire-sire marking harnesses and crayons (Hortico) were used in a continuous two-week rotation between the mating pens and a resting pen. Oestrus detected from crayon marks on ewes' rumps was recorded

P E R I - P A R T U R I E NUTRITION OF EWES 59

daily. Ovulation at each conception was recorded by laparoscopy approximately four weeks after the last recorded oestrus which also allowed conception to be confirmed from visual appraisal of uterine distension. Records were taken at each lambing of lamb birth weight, sex, birth type and mortality before weaning at 13 weeks of age. All animals were weighed every week throughout the experiment.

Statistical methods were taken from Steel and Torrie (1960). Measurements taken on the ewes were assessed by analyses of variance using successive lambings as sub-units in a split plot in time experimental design. Measurements taken on different groups of lambs at each lambing were considered to be independent and these data were assessed by analyses of variance using successive lambings as a factor in a 4 (nutritional treatments) × 3 (lambings) factorial design. Variations in preweaning lamb mortality and weaning weight in relation to birth weight and birth type were tested by Chi-square or analyses of variance.

RESULTS

The average interval from four weeks before joining until weaning after the third lambing was 99 weeks. Mean body weights during the four periods when different nutritional treatments were applied are shown in Fig. 1. Periods not represented in the figure when all ewes were offered a common maintenance ration before the first lambing, between first and second lambings and between second and third lambings averaged nine, 13 and 13 weeks respectively. Mean body weights in all groups increased during the initial joining and at each lambing showed the same pattern of a sharp decrease associated with parturition followed by a further slow decline during lactation. Generally there were no significant differences in body weight among the three supplemented groups. A significant (P<0-05) difference of about 2kg between these three groups and the unsupplemented group was established before the first lambing and did not change significantly thereafter.

32-

30-

28-

mO 24 . Z

22-

20- -4

I I I I I I I I I

J

I I I

0 6 12 I I I I

0 6 12 -4 0 6 12 -4 0 6 12 -4

JOINING FIRST LAMBING SECOND LAMBING THIRD LAMBING

TIME RELATIVE TO INITIAL JOINING AND SUBSEQUENT LAMBING

(weeks)

PIG. 1. Ewe body weights during periods of supplementary feeding. (0-----0 TO, O O T4, A ~ A T10, C] ,,. D T16).

60 CHANIAGO, NATASASMITA AND FLETCHER

There were no significant differences between the three lambings in the mean number of lambs born or weaned per ewe lambing or in lamb mortality between birth and Weaning. Mean ovulation rates were significantly (P < 0-05) higher at the first (2.48). than at the second (2-16) or third (2.10) conception and mean lamb birth weights and weaning weights were significantly (P < 0.01 and P < 0.05 respectively) lower at the first (1.6 and 7.2 kg) than at the second (2-0 and 8.0 kg) or third (1.9 and 7.6 kg) lambing. There were however no significant interactions between lambing number and nutritional treatment in any of the parameters studied.

Variation between nutritional treatments totalled or averaged over the three lambings is summarised in Table I. The total number of lambings failed to reach the potential of 48 per treatment group mainly because of ewe mortality and to a lesser extent because of the few surviving ewes which failed to lamb three times but neither of these factors varied significantly between treatment groups. Nutritional treatment had no significant effect on lambing interval, ovulation rate, numbers of lambs born, lamb mortality or numbers of lambs weaned. Mean birth weights tended to be higher in the three supplemented than in the unsupple- merited group though the difference was not significant (P < 0.10). The only clear effect of nutritional treatment was a progressive increase (P < 0-001) in lamb growth rates and weaning weights with increasing periods of supplementary feeding.

It is clear from the data presented in Table I that the main constraints on productivity in all treatment groups were not ewe reproductive traits but a

TAm.~E I

Effects of different periods of supplementary feeding on lamb production over three successive lambings

Significance

Error TO T4 T1O I"16 MS P

Ewe numbers Joined 16 16 16 16 m Died 3 3 2 3 Failed to lamb 3 times 1 1 2 0 m

Total No. lambings 43 42 42 44 m Mean interlambing intervals

(days) First-second lambing 199 (15) 199 (14) 210 (15) 194 (15) 779 NS Second-third lambing 201 (12) 206 (12) 208 (12) 203 (13) 785 NS

Mean ovulation rate 2-10 (42) 2-35 (43) 2.39 (44) 2-20 (44) 0-23 NS Mean No~ lambs born/

ewe lambing 1.58 (43) 1.52 (42) 1.69 (42) 1-52 (44) 0-27 NS % lamb mortafity

(birth-weaning) 31 (68) 44 (64) 42 (71) 31 (67) 35 NS Mean No. lambs weaned/

e w e lambing 1-09 (43) 0.86 (42) 0.98 (42) 1.05 (44) 0-17 NS Lambs

Mean birth weight (kg) 1-7 (66) 1-9 (62) 1.8 (69) 2-0 (67) 0.41 NS Mean growth rate to

weaning (g/day) 45 (47) 56 (34) 66 (41) 75 (45) 626 <0-001 Mean weanin~ weight (kg) 6.0(47) 7-1 (36) 8,1 (41) 9-1 (46) 6-15 <0.001

Parentheses represent numbers contributing to each mean.

PERI-PARTURIENT NUTRITION OF EWES 61

TABLE 1I

Preweaning mortality and growth rate of surviving lambs in relation to birth weight and birth type

Birth weight class (kg) All birth

Birth type <1.0 1.0-1.4 1.5-1.9 2.0-2.4 >2.4 weights

% mortality Single 100 (2) 25 (4) 14 (14) 11 (26) 12 (24) 15 Twin 75 (4) 57 (28) 30 (46) 21 (33) 11 (9) 34 Triplet 100 (12) 79 (24) 40 (10) 0 (2) - - (0) 73 Quaduplct 100 (2) 100 (4) 100 (2) - - (0) - - (0) 100 All birth types 95 67 31 16 12 37

Mean growth rate (g/day) Single - - (0) 72 (3) 58 (12) 78 (23) 79 (37) 75 Twin 62 (1) 33 (12) 47 (32) 52 (26) 54 (8) 48 Triplet - - (0) 50 (5) 61 (6) 44 (2) - - (0) 54 All birth types 62 43 52 63 75 61

Parentheses represent numbers contributing to each mean.

combination of low birth weight, high preweaning mortality and low growth rate in the lambs. Relationships between these factors are shown in Table II. Data have been pooled over all treatments without serious bias because of the similar distributions of birth type 0C 2 = 5.53, df = 9, P > 0.70) and birth weight class ~2 = 15.79, df = 12, P > 0-20) between treatment groups. Over all birth weights lamb mortality increased (P < 0-001) with .increasing litter size and growth rates were higher (P < 0.001) from single than from multiple births. Over all birth types mortality decreased (P < 0.001) and growth rate increased (P < 0.001) with increasing birth weight.

DISCUSSION

The conditions of the present experiment maintained average non-pregnant ewe body weights within the range (22 to 26 kg) normally found in village situations (Inounu et al., 1983). No significant differences in ewe body weight developed between the T4, T10 and T16 groups although they were given supplementary feed for approximately 15%, 40% and 65% respectively of the 99-week duration of the experiment. Also the difference in body weight between these three groups and the unsupplemented group was established before the first lambing and did not change substantially thereafter. In view of the differences between treatment groups in lamb weaning weights discussed below and evidence that ewes of this breed can reach mean body weights in excess of 40 kg when there are no nutritional constraints (Fletcher et al., 1985) it is clear that the feed supplements offered during the experiment were utilised principally for progeny growth and not for increased maternal body weight.

Variations in productivity recorded between the three successive lambings were possibly physiological consequences of the fact that the ewes had been purchased when young and maintained without reproducing for approximately one year before the experiment commenced. Thus a higher mean ovulation rate may have been recorded at the first than at subsequent conceptions because this was the only observation taken when the ewes were not lactating and lower lamb birth weights and weaning weights may have been recorded at the first than at

62 CHANIAC.,-IO, NATASASMITA AND FLETCHER

subsequent lambings because this was for most ff not all ewes their malden parturition and lactation. Variation between lambings for whatever reason was relatively unimportant, however, since there were no significant interactions between lambing number and nutritional treatment.

There were no significant effects of supplementary feeding on ewe reproduc- tive traits. It should be noted, though, that ovulation rate and litter size were already relatively high at the maintenance feeding level. Subsequent investigation (Henniawati and Fletcher, 1986) has shown that ovulation rate and litter size in Indonesian sheep and goats can in fact be increased by improved nutrition around mating time but only when supplements were fed in much greater amounts and for longer periods than applied in the present experiment. On the other hand the average interlambing interval of only 200 days in unsupplemented ewes was as short as has been recorded previously in ewes fed unlimited amounts of high quality feed (Obst et al., 1980; Fletcher et al., 1985). The ability of Javanese thin-tail ewes to breed at frequent intervals thus does not appear to depend on high levels of nutrition and other factors such as disease or poor mating management are considered more likely to be responsible for the longer interlambing intervals commonly found in village sheep (Bell et al., 1983).

Lamb growth rates and weaning weights increased with increasing periods of supplementary feeding but they nevertheless remained well below the growth potential shown by this breed when there are no nutritional constraints (Fletcher, Chaniago and Obst, 1982, Fletcher et al., 1985). Increased lamb growth in response to supplementary feeding could only have been a consequence of increased ewe milk production in the "1"4 group. Lambs in the T10 group had access to the supplement for the first six weeks after birth but the nutrient intake of unweaned lambs during this period comes predominantly from milk (Munro, 1962). Only in the T16 group did supplementary feeding continue long enough to allow lambs the opportunity to ingest directly significant amounts of the supplement. If the only potential for improved productivity from supplementary feeding is through increased lamb growth then a limited amount of supplement might well be used more effectively if the relatively inefficient conversion of supplement to milk was bypassed and all supplement was instead fed directly to the lambs.

There remains, however, the problem of 30 to 40% preweaning lamb mortality which also applies in at least some village production systems (Basuno and Petheram, 1984). A highly significant inverse relationship between lamb birth weight and preweaning mortality though not necessarily causal suggests a possibility of improving lamb survival through increased birth weight. Lamb birth weights are influenced by ewe nutrition during the last third of pregnancy (Wallace, 1948). Although pre-partum supplementary feeding did not sig- nificantly affect lamb birth weight in the present investigation it remains open to question whether a larger amount or a longer period of supplementary feeding before lambing might improve productivity through increased lamb birth weights and reduced preweaning mortality.

It is concluded that ovultion rate, litter size and interlambing interval in Javanese thin-tail ewes at a maintenance feeding level approximating village conditions were not improved significantly by moderate levels of supplementary feeding around lambing time and the relatively high reproductive rates recorded at the maintenance feeding level indicate that nutrition may not be a primary constraint on reproduction in village animals. Further investigation is needed to

PERI-PARTURIENT NIYrRITION OF EWES 63

determine whether a response in lamb weaning weight to supplementary feeding could be improved if supplements were witheld from the ewes and fed only to growing lambs or whether higher levels or a longer period of supplementary feeding to the ewes before lambing could improve productivity through decreased lamb mortality associated with higher birth weights.

ACKNOWLEDGEMENTS

We thank Budiarsana and Maulana for care of the experimental animals and assistance with data collection. The work was supported in part by the Australian Development Assistance Bureau through the CSIRO Project for Animal Re- search and Development at Ciawi.

Accepted for publication November 1986

REFERENCES

BASUNO, E. & PETHERAM, R. J. (1984). A "village profile" for livestock component FSR in Java. Farming Systems Support Project, Networking Paper No. 4 (University of Florida).

BELL, M., INOUbrO, I. & SUeAJ~RP~O. (1983). Variability in reproductive performance of sheep and goats among village farms in West Java, Indonesia. Proceedings of the Vth World Conference Animal Production, 2, 823-824.

Pt.ETCHER, I. C., CHANIAGO, T. & OSST, J. M. (1982). Proceedings of the Australian Society of Animal Production, 14, 455-458.

FLErctmR, I. C., GUNAWAN, B., HETZEL, D. J. S., BAKmE, B., YATES, N. G. & CHAN~GO, T. (1985). Tropical Animal Health and Production, 17, 127-134.

GUN, R. G. (1983). The influence of nutrition on the reproductive performance of ewes. Sheep Production (Ed. William Haresign), Butte,worth, London, pp. 99-110.

HENNIAWATI & FI~TCHI~, I. C. (1986). Animal Reproduction Science, 12, 77-84. INOUWO, I., BELL, M., SrroRus, P. & SUBANDRIYO (1983). Preweaning and postweaning growth rate

and mature weights of female thin-tail sheep in three villages in West Java, Indonesia. Proceedings of the Vth World Conference on Animal Production, 2, 887-888.

Mtn~o, J. (1962). Animal Production, 4, 203-213. O~T, J. M., BOYES, T. & ~ G o , T. (1980). Proceedings of the Australian Society of Animal

Production, 13, 321-324. STEEL, R. G. D. & TORRIE, J. H. (1960). Principles and Procedures of Statistics McGraw-Hill, New

York. WAI.J..ACE, L. R. (1948). Journal of Agricultural Science, Cambridge, 38, 367-401.

EFFETS D'UNE ALIMENTATION COMPLEMENTAIRE AU MOMENT DE L'AGNELAGE SUR LA PRODUCTIVITE DES BREBIS JAVANAISES A QUEUE FINE

R~sum~---On a recherch6 Hnfluence d'une alimentation compl~mentaire, donn~e au moment de l'agnelage, sur la reproduction de 4 groupes de chacun 16 brebis; jusqu'a leur 3e parturition, ces brebis 6talent accompagntes de btliers fertiles. Un groupe a 6t6 maintenu en settle subsistance; les 3 antres groupes ont refu une alimentation complementaire: ou bien pendant les 4 derni~res semaines de leur gestation, ou bien pendant les 4 derni~res semaines de gestation plus les 6 premieres semaines apr~s l'agnelage, ou encore pendant les 4 derni~res semaines de gestation plus les 12 premieres semaines apr~s l'agnelage. Le poids corporel moyen des agneaux sevrts a 13 mois augmente progressivement (P<0,001) de 6,0 a 9,1 kg avec l'augmentation du temps de l'alimentation compl~mentaire. Cependant /l n'y a pas de differences significatives entre les groupes concernant rintervalle moyen entre agnelage (202 jours), le taux d'ovulation (2,26), la taille des port~es (1,58), le poids a la naissance (1,8 kg), la mortafit~ avant sevrage (37%) our le hombre d'agneaux sevr~s par port~e (o,~).

EFECTO DEL SUMINISTRO DE ALIMENTO SUPLEMENTARIO EN LA EPOCA DE PARICIONES SOBRE LA PRODUCrIVIDAD DE OVEJAS JAVANESAS DE COLA

DELGADA Resmmew--Se investig6 el efecto de la suplementaci6n sobre la reproducci6n, en cuatro grupos de 16 ovejas cada -,no, los cuales esmvieron acompafiados de moruecos f~rtiles hasta la tercera parici6n. Un

64 CHANIAGO, NATASASMITA AND FLETCHER

grupo se ,liment6 a nivel de mantenimiento a tray, s del experimento y los otros tres recibieron suplementaci6n durante las ~ltimas cuatro semanas de gestaci6n, o durante las dltimas cuatro semanas de prefiez, n~s las primeras seis semanas despues de cada paric~6n, o durante 1~" dltimas cuatro semanas de gestaci6n mds 12 semanas despues de e~_d~ parici6n. El peso promedio de los destetos a las 13 semanas de edad se increment6 (P < 0.001) progresivamente de 6.0 a 9.1 kg cuando el alimento suplementatio se di6 m~s tiempo. No hubo sin embargo diferencias significativas entre grupos tratados en intervalo de parto~ (202alias), ovulaci6n (2.26), tamafio de la camada (1.58), peso al nacer (1.8 kg), mortalidad antes del destete (37%) 6 n6mero de borregos destetados por camada (0.99).

BOOK REVIEW

Evaluaci6n de Pasturas con Animales--Alternativas Methodol6gicas. Editas por Carlos Lascano y Estenan Pizarro, 1986. The evaluation of pastures using animals--Papers presented at a Conference organised by CIAT, Peru, October, 1984. 290 pp.---short index. ISBN 84-8280-154-6.

This publication was produced by scientists belonging to the international network for the evaluation of pastures and forages in tropical America particu- larly in marginal regions where the soils are acid and of low fertility. The present publication is the fourth in the series; the first was entitled "A manual for the collection, preservation and characterization of tropical forage resources", the second entitled "A manual for the evaluation of forage germ plasm", the third entitled "Forage germ plasm grazed in small plots". The present publication has been produced both to give the fundamental concepts of pasture management and to act as a guide in ways to put into "effect. A total of 12 papers were given, many of them by employees of CIAT, although there are also contributions from Latin America, Australia and the USA. The papers deal with the design of experiments using animals, measurement of the responses of animals both in terms of gain in weight and production of milk and the effect of animals on the pasture. The papers have all been prepared by the top research workers in their fields. The book concludes with general recommendations concerning the evaluation of pasture with animals including aspects related to the design of experiments, the experimental area needed, the measurements that should be made on both the animal and the pasture and both the statistical and economic analysis of the data obtained--a valuable book for those working in this field and in particular those with an interest in Latin America.

A. J. Smith