• No results found

Lamb rearing performance in highly fecund sheep : [this thesis is presented in partial fulfillment of the requirements for the degree of Doctor of Philosophy at Massey University, Palmerston North, New Zealand]

N/A
N/A
Protected

Academic year: 2020

Share "Lamb rearing performance in highly fecund sheep : [this thesis is presented in partial fulfillment of the requirements for the degree of Doctor of Philosophy at Massey University, Palmerston North, New Zealand]"

Copied!
233
0
0

Loading.... (view fulltext now)

Full text

(1)Copyright is owned by the Author of the thesis. Permission is given for a copy to be downloaded by an individual for the purpose of research and private study only. The thesis may not be reproduced elsewhere without the permission of the Author..

(2) Lamb Rearing Performance in Highly Fecund Sheep. J ulie Marie Everett-Hincks. 2004.

(3) This thesis is presented in partial fulfillment of the requirements for the degree of Doctor of Philosophy at Massey University, Palmerston North, New Zealand. The Massey University Animal Ethics Committee has approved the studies involving animal manipulations. I. have prepared this thesis and it is a record of I. my. own work.. took the photos of the sheep, unless otherwise specified..

(4) Abstract This thesis i nvestigates ewe and lamb behaviour, genetics and environmental effects to determi ne whether lamb reari ng petformance can be i mproved i n high l y fecund heep. The studies were carried out under commercial pastoral farm ing conditions. High performing sheep farmers were surveyed to identi fy management and performance practices that differentiate farms with high and low lamb reari ng ucces . Farmers agreed that motheri ng abi l ity was the most i m portant factor affecti ng l amb urvi val and considered lamb survi val to be the most important trait affecti ng fann profit. The survey identified the Coopwotth breed as the predomi nant breed of high lambi ng percentage flocks. Heritabi l ity estimates were derived for lamb survi val (h 2= 0. 1 6), ewe maternal behaviour score (h 2= 0.05 ) and l i tter survi val (h 2= 0.00) in a Coopworth flock that had been selected for i mproved maternal abi l i ty for nearly 30 year . Maternal genetic variation in the Coopworth flock was low for lamb and maternal traits and suggests that farmers must con ider the environment and management technique to improve lamb survival . A greater proportion of the vatiation i n triplet survi val was attri buted to environmental effects than it was for twi ns, therefore tri plets require more care.. Management and environ mental factors investigated i n this study affected ewe and lamb attachment behaviour in larger l i tters. A relationship was found between pregnant ewe physiology and maternal behaviour. Ewes that mai ntai n body condi tion in late pregnancy and have lower plasma �-hydroxybutyrate levels were more recepti ve to the demands of their li tter as these ewes had higher MBS. The effect of maternal nutri tion in late pregnancy and at lambing was explored further. Feeding levels did not affect the majori ty of maternal behaviours i nvestigated. However triplet lamb behaviour was affected by maternal nutrition in late pregnancy and lambs born to poorly fed ewes were less l i kely to tand, locate thei r dam ' s udder and fol low their movi ng dam after separation at tagging. Triplet lamb survi val was simi lar to twin lamb survi val when pasture al lowance wa not restricted in late pregnancy.. Ewe behaviour was affected by breed, selection l i ne and l i tter size. Ewes that high bleated less and showed less flocki ng behaviour, in the arena test and at tagging had.

(5) greater lamb reari ng performance at weani ng. Ewes that were quick to contact their lambs after separation at tagging weaned greater lamb weights ( weight weaned increased I Okg per ewe from MBS3 to M B S4). Lamb behaviour was not affected by breed and was not explai ned by bi rthweight. Lambs that stood and sucked from their dam withi n ten mi nutes from tagging had an i mproved chance of survi val to weani ng (survi val i ncreased 5-fold and 3-fold respectively). Lamb behaviour has a sign i ficant role in ewe-lamb attachment in large l i tters. In particular lamb bleati ng represents need and attracts the dam when separated. The l amb bleats more if i ts dam has a lower M B S and it w i l l bleat more i f it stands quickly. The dam wi l l high bleat when she has lower MBS and cannot locate her lamb or litter.. Lamb remi ng success for ewes with l arger l itters is determi ned by lamb behaviour and the lamb ' s interaction with its dam. Tri plet lambs can achieve survi val rates simi lar to twi ns, providing the maternal environment and lamb genetics support appropriate ewe­ lamb attachment behaviours.. ii.

(6) Acknowledgements I have real l y enjoyed the past three years and I have many people to thank for that. I am extremely grateful to my two Supervisors, Professor Hugh B lair and Professor Kevin Stafford . Thanks for your tremendous support, encouragement and "pearls of wisdom". I couldn ' t have asked for a better team of S upervisors.. I am also extremely grateful to Nicolas Lopez-Vi l l alobos. Without Nicolas my stats knowledge wouldn't be what it is today. Nicolas taught me to question everythi ng, don ' t accept 'anythi ng' and "get to know your data first".. Very special thanks to al l of the farmers that have contributed to this research project. Thanks to Bi l l , Si mon and Pascale Carthew for al lowi ng me to i nvestigate their Coopwo1th flock in the 2000 lambing season. Thanks to Robin Hi l son and Gary McLennan for giving me the opportuni ty to i n vestigate Fi nn and Texel behaviour i n the 200 1 lambing season. Thanks to Graham Poole, Manager of Massey Uni versity' s Keebles farm, for helping with 2002 trial work.. Thanks to AI lai n Scott for looki ng after us ( postgrads). Thanks to Luc Doornheege ( from the Netherlands) for working long hours in the 200 I lambi ng sea on. Thank you Barbara Gal lagher and Elaine Patton ( from Ireland ) for helpi ng with the arena behaviour tests. Thanks to Dean Burnham, Paul Kenyon and Steve Morri for al lowing the behaviour investigation to be i ncorporated i nto the trial: "The feedi ng requirements of high fecundity ewes". Thanks to Trevor Cook, from Manawatu Vet Services. I would also l i ke to thank M i ke Hogan ( I V ABS) for managing the lamb post mortems. A very special thank you to AGMARDT for personal financial upport made available. through a 3 year stipend. Thanks to IV ABS and C lutha Vets for fi nancial assistance. The research projects were jointly funded by WooiPro ( Meat and Wool Innovation), Massey Uni versity Ri verside Fund and IV ABS .. To my husband, fami l y and friends, thanks for your love and support.. iii.

(7) iv.

(8) Table of Contents Abstract Acknowledgements. iii. Introduction. 1. CHAPTER 1. 5. Literature R eview. 5. I . I Introduction. 6. 1 .2 Background. 6. 1 .3 Behaviour. 7. 1 .4 Selection. 13. 1 .5 Environment and Management. 15. CHAPTER 2. 19. Lamb rearing success on high performing sheep farms. 19. 2. 1 Summary. 20. 2.2 Introduction. 20. 2.3 Materials and Methods. 21. 2.4 Results. 22. 2.5 Discussion. 26. 2.6 Concl usions. 28. CHAPTER 3. 29. The effect of ewe maternal behaviour score on lamb and litter survival. 29. 3. 1 Summary. 30. 3.2 I ntroduction. 31. 3.3 Materials and Methods. 32. 3.3.1. Background. 3.3.2 Animals and Measurements. 32 32 V.

(9) 3.3.3 Statistical Analyses 3.4 Resu lts. 33 34. 3.4. 1 Environ mental Effects on M B S and LIS. 34. 3.4.2 Environ mental Effects on Lamb Survi val. 36. 3.4.3 Genetic Parameters for MBS and LIS. 38. 3.4.4 Genetic Parameters for Lamb Survi val. 38. 3.5 Di scussion. 39. 3.6 Concl usions. 45. CHAPTER 4. 47. The effect of pregnant ewe physiology on maternal behaviour score and lamb production. 47. 4. 1 Summary. 48. 4.2 Introduction. 49. 4.3 Materials and Methods. 51. 4.3. 1 Background. 51. 4.3.2 Ani mals and Measurements. 51. 4.3.3 Statistical Analyses. 53 54. 4.4 Results 4.4. 1 Physiological Status of the Pregnant Ewe. 54. 4.4.2 Physiological Effects on Dam M B S. 55. 4.4.3 Dam Production. 55. 4.5 Discussion. 57. 4.6 Impl ications. 60. CHAPTER. 5. 63. Lambing Performance of Finns and Texels. 63. 5.0 Preamble. 64. 5. 1 Summary. 64. 5.2 Introduction. 65. 5.3 Materials and Methods. 66. vi. 5.3. 1 Ani mals and Measurements. 66. 5.3.2 Ani mal Envi ronment. 68.

(10) 5.3.3 Stati stical Analyses 5.4 Resu lts. 68 69. 5.4. 1 Ewe Pre-l ambing Performance. 69. 5 .4.2 Ewe Lamb Reari ng Performance. 70. 5 .4.3 Lamb Performance to Wean ing. 71. 5.5 Discussion. 73. 5.6 Conclusions. 74. CHAPTER 6. 77. The effect of ewe and lamb post- parturient behaviour on lamb survival to weaning: A comparative study of Finns and Texels. 77. 6. 1 Summary. 78. 6.2 Introduction. 78. 6.3 Materials and Methods. 80. 6.3. 1 Animals and Measurements. 80. 6.3.2 Statistical Analyses. 83. 6.4 Resu lts. 85. 6.4. 1 Ewe and Lamb Behaviour. 85. 6.4.2 Relationship Between Behaviour and Lamb Survi val. 91. 6.5 Discussion. 94. 6.6 Concl usions. 97. 6.7 Appendix: Ewe Post-Parturient Behaviour. 99. 6.8 Appendi x : Lam b Production Traits. 1 00. 6.9 Appendi x : S i ngle Texel Dams and their Lambs: Natural ver u A i ted Birth.. 1 02. 6.9. 1 Ewe Production Trai ts. 1 02. 6.9.2 Lamb Production Traits. 1 02. 6.9.3 Lamb and Ewe Behavi our. 1 03. CHAPTER 7. 1 07. The effect of Finnish Land race and Texel ewe and lamb behaviour after tagging on lamb survival to weaning. 1 07. 7. 1 Summary. 1 08. 7.2 I ntroduction. 1 09 vii.

(11) 7.3 Materials and Methods. Ill. 7.3. 1 Animals. Ill. 7.3.2 Taggi ng. Ill. 7.3.3 Behavi our. 1 12. 7.3.4 Lamb Performance. 1 13. 7.3.5 Statistical Analyses. 1 13. 7.4 Results. 1 16. 7.4. 1 Taggi ng Behaviour. 1 16. 7.4.2 Post Taggi ng Behaviour and Lamb Pe1formance. 1 20. 7.4.3 Post Taggi ng Behaviour and Lam b Survi val. 121. 7.5 Di scussion 7.5. 1 Relationship between post-parturient and taggi ng behaviours. 1 23 1 25. 7.6 Conclusion. 1 27. CHAPTER 8. 1 29. An investigation of the relationship between ewe arena behaviour and lamb rearing performance. 1 29. 8. 1 Summary. 1 30. 8.2 Introduction. 131. 8.3 Materials and Methods. 1 33. 8.3. 1 Ani mals and Measurements. 1 33. 8.3.2 Behaviour Tests. 1 34. 8.3.3 Statistical Analyses. 1 36. 8.4 Results. 1 37. 8.4. 1 Arena Test Behaviour. 1 37. 8.4.2 Maternal Behaviour. 1 40. 8.4.3 Lamb Reari ng Pe1formance. 1 42. 8.5 Discussion. 1 44. 8 .6 Conclusion. 1 48. 8.7 Appendix. 1 49. viii.

(12) CHAPTER 9. 151. The effect of pasture allowance d uring pregnancy on maternal behaviour and lamb rearing performance in highly fecund ewes. 151. 9. 1 Summary. 1 52. 9.2 Introduction. 1 53. 9.3 Materials and Methods. 1 55. 9.3 . 1 Ani mals and Measurements. 1 55. 9.3.2 Arena Test Behaviour. 1 57. 9.3.3 Maternal Behaviour. 1 59. 9.3 .4 Lamb Behaviour. 1 60. 9.3.5 S tatistical Analyses. 161. 9.4 Resu lts. 1 63. 9.4. 1 Arena Test Behaviour. 1 63. 9.4.2 Maternal Behaviour. 1 65. 9.4.3 Dam Performance. 1 67. 9.4.4 Lamb Behaviour. 171. 9.4.5 Lamb Performance. 1 78. 9.5 Discussion. 181. 9.6 Implications for Industry. 1 85. CHAPTER 1 0. 1 87. General Discussion. 1 87. References. 1 97. ix.

(13) List of Tables Table 2. 1 Summary description of farms surveyed. . .... . ........... . ..................... .. . .. .... ........... 23. Table 2.2 The i mportance of factors affecting lamb survi val accordi ng to respondents from High and Low farm groups . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .... 24 Table 3 . 1 The maternal behaviour scores at tagging and the number of ani mals used i n this study. . ........ . ........................................................................................................ 32. Table 3.2 The effect o f age o f dam o n maternal behaviour score and l itter survi val ( least squares mean ± standard error) . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .......................... 35 Table 3.3 The effect of li tter size at birth on maternal behaviour score and l i tter survi val ( least squares mean ± standard error). .. .............................................................. . .. 35 .. .... Table 3.4 The effect of age of dam, sex of lamb and birth year of lamb on l amb survi val from birth to weani ng (least squares mean ± standard error). .. .. ....... ....... .. . .. ................ 37. Table 3.5 Estimated variance components for maternal behaviour score ( M BS ) and litter surv i val ( LI S ) ( standard error i ncl uded). . .... .. .................................. .. ...... .. .. ................... 38. Table 3 .6 Estimated variance components for lamb survi val from birth t o weaning (standard error i ncluded ). ..... 39. .......................... 52. . . . ....................... . ........... . . . . . . . . ..... ...... . ......... . . .... .............. .. Table 4. 1 The maternal behaviour scores at tagging and the number of dams with recorded M B S i n this study. ...... .. ............................................... .. ....... .. Table 4.2 Pregnancy measurements by breed group ( Coopworth and EFCoop ewes) ( least squares mean ± standard error). .. ............. . ...................... .. ................................. 54. Table 4.3 The effect of MBS on Coopworth and EFCoop ewes with twin and triplet li tters (least squares mean ± standard error). .. . ........................ .................................. 55. Table 4.4 The effect of ewe breed and litter size on li tter survi val ( least square mean ± standard error). . . . . . ........................................ . .. .. ............ . . . . . . . . . . ...................... .. .............. 56. Table 4.5 The effect of maternal behaviour score on dam l itter surv i val from tagging to weani ng ( least squares mean ± standard error). . ............ . . . . . . .................... .. ................. 56. Table 4.6 The effect of maternal behaviour score on weight of lamb weaned (kg) ( least squares mean ± standard error). . .......... . . .. .. ............. ....................... ............................. 57. Table 5. 1 The effect of ewe breed on pre-lambi ng performance ( least squares mean ± standard error). . ....................... . . . . .......... .................................................... ... ........ .. ....... 70. Table 5.2 The effect of ewe breed on lamb reari ng performance (least squares mean ± standard error). . ...... ..... .. ....... ............... .. .. .................................................. . . .. .. .... ............ 71. Table 5.3 The effect of lamb breed on performance to weaning ( least squares mean ± standard error) X. . ...................................... .. ...... . .. .. .......................................................... 72.

(14) Table 6. 1 Number of animals i ncluded in anal yses. . .................... .. ................................... Table 6.2 Defi nitions of ewe and lamb behaviours recorded after birth . . ... .... . 80 82. ........... ........ Table 6.3 Description of the maternal behaviour score used i n thjs study after parturition . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . ........ . .. . . . . . . . . . . . . . . . . ... . . . . . . . 82 Table 6.4 Lambing day according to litter size, time between the birth of fi rst and second born lambs for twi nn i ng ewes and observation period for Fi nn and Texel ewes i ncluded in this study ( least squares mean ± standard error). 85. . ......................... Table 6.5 The effect of breed on ewe and lamb behaviour after birth (Wi lcoxon mean ± standard error) .. ................................................. ... . ... .. ................ . . . . . .......... ... 87. .................. Table 6.6 Lamb and ewe bleati ng behaviour after birth (least squares mean ± standard error). ................. .. ..... .. ........ .. . .. .... ................... . .. .... ..................................... .. .................... . 90 .... Table 6.7 The effect of ewe breed and litter size on maternal behaviour score after birth ( least squares mean ± standard error) . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 90 Table 6.8 The effect of behaviour after parturition on twin lamb survi val to weani ng (Wi lcoxon . m ean ± standard error). . ........... .. ................................... ... .......................... .91. Table 6.9 The effect of ewe groom i ng ti me on lamb survival to weaning ( least squares mean ± standard eiTor) . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 93 Table 6. 1 0 The effect of lamb bleating frequency on lamb survi val to weaning ( least squares mean ± standard error). . ..... .................. ... . . .................................... .. ................. 93. Table 6. 1 1 The effect of ewe M B S on lamb survi val to weani ng ( least squares mean ± standard error). . ............................... .................................................................. . . ... ..... 93. Table 6.7. 1 Ewe behaviour fol lowing parturition ( least squares mean ± standard en·or) . ........................................................ .......................................................................... 99. Table 6.8. 1 Performance of Fi nn and Texel lambs observed after birth (least squares mean ± standard error). . .......................................................................................... I 00. Table 6.8.2 The effect of behaviour after parturition on si ngle lamb survi val to weani ng (Wi lcoxon mean± standard error). . .... .. ............... ........ .. .. ............................ ........... � ..... 101. Table 6.9. 1 Ewe l i veweight and l i veweight change from mating through to lambing ( least squares mean ± standard error). ................................... .. .................. :. . . ............ I 02. Table 6.9.2 Performance of lambs borh natural ly and to assisted ewes ( least squares mean ± standard error). . ...................................................................................... .. .... I 03. Table 6.9.3 Proportion of ewes and lambs that did the fol lowing behaviour wi thi n 30 minutes from bi rth . .. ...................................... .. ....... .. ..... .. ................... .. .. .. .............. .. ......... 1 03. xi.

(15) Table 6.9.4 Ti me taken for the ewe to make Iamb contact, the lamb to stand, locate and suck from its ewe ' s udder ( least squares mean±standard error) . . .. ...... . . .... ............. I 03. Table 6.9.5 Ewe and lamb bleating frequency (least squares mean±standard error) .. I 04 Table 6.9.6 Ewe behaviour after pmtlllition (least squares mean±standard etTor) . . . . . I 05 Table 7. 1 Number of ani mals i ncluded in anal yses . . .................. .. .. .... . . . . . .. .. . . . . . . . . . . . . . . ... I l l Table 7.2 Taggi ng dates and ti mes according to breed (least squares mean±standard error). ................. . . . . . . ................. . ................ ....... . . ....... . . . . . . . . ...................................... 1 12. Table 7.3 Description of the maternal behaviour scori ng sy tern at taggi ng . . . . . . . . . . . . . . . . 1 1 3 Table 7.4 Defi nitions of ewe and lamb behaviours recorded after taggi ng . .. . . . . . . . . . . . . . . . 1 1 3 Table 7.5 The effect of breed on ewe M B S after tagging (least squares mean±standard error) . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 1 6 Table 7.6 The effect of breed on ewe and Iamb behaviour after bi rth (Wi lcoxon mean± standard error) . .. . . . . . . . . . . . . . . . . . . .. .. . . . . . . . . . . . . . . . . ...................... ....................... . . . . . . . . . . . . . . . . . I 1 6 Table 7.7 The effect of breed and li tter size on ewe and lamb bleating behaviour after taggi ng ( least squares mean ±standard error) . ..................................................... 1 1 9 Table 7.8 Regression coefficient esti mates, odds ratios and 95% confidence i ntervals for the odds ratios and probabil ity levels from logistic regression analysi s of all lambs and their chance of survi val to weaning . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 22 Table 7.9 Phenotypic corre lations between post-parturient and post-taggi ng behaviours observed in this study (phenotypic correlation ±standard error) . . . . . . . . . . . . . . . . . . . . . . . . . 1 25 Table 8. 1 The maternal behaviour scores at taggi ng and the number of ani mals used i n this study (proportion o f ani mals i n dam selection l i ne with M B S ) . . . . . . . . . . . . . . . . . . . . . 1 34 Table 8.2 The effect of dam selection l i ne on movement within the arena (least squares mean±standard error). Means within dam selection line having different superscri pts are di fferent ( P<0.05) . . .... ....................... ....................... ... . . . . . . . . . . . . . . . . . 1 38 Table 8.3 The effect of dam selection l i ne on social behaviour withi n the arena ( least squares mean±standard error). Means within dam selection l i ne having di fferent superscripts are di fferent ( P<0.05) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 40 Table 8.4 The effect of dam selection l i ne on lamb reari ng petformance (least squares mean±standard error). Means within dam selection l i ne havi ng di fferent superscripts are di fferent ( P<O.OS) . . ......... . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . 1 43 Table 8.5 The effect of li tter size at bitth and whether a ewe does or does not enter Zone 2 in the arena behaviour test, on li tter survi val at taggi ng (least squares mean± standard error) . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . .......... ... . ... ................... .. . . . . . . . . . . . . . . . . . . . . . . . . 1 43 xii.

(16) Table 8 .7 .1 Correlation coefficients between measures of arena and maternal behaviour and lamb production traits for dam selection li nes. . ... ... ............................. .. 1 49. .............. Table 9.1 Number of ewes i ncluded i n statistical anal ysi s according to post-grazi ng pasture sward height, block and li tter size at birth. . . ....................................... .. 1 55. ........ Table 9.2 Desctiption of the maternal behaviour score ( M B S ) used i n this study . . .. ..... 159. Table 9.3 The effect of pasture al lowance and foetal number on movement within the arena ( least squares mean ± standard error). Means withi n pasture sward height and withi n litter size having di fferent letter superscripts are di fferent (P<O. I O) 164 . .. Table 9.4 The effect of pasture al lowance and foetal number on social behaviour within the arena ( least squares mean ± standard error). Means across pasture sward heights and across li tter size having different letter superscripts are di fferent ( P<O. I 0). ............................... . .................................................................................. 165. Table 9.5 The effect of pasture allowance and li tter size on M B S and ewe tagging behaviour ( least squares mean ± standard error). Means across pasture sward height and across litter size having di fferent letter superscripts are di fferent (P<O. l 0) . . . . . .................. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .... 1 67 Table 9.6 The effect of pasture al lowance and litter size at bi rth on ewe l i veweight and growth ( least squares mean ± standard error). Means within pasture sward height and withi n litter size having di fferent letter superscripts are di fferent (P<O. I O) 168 . .. Table 9.7 The effect of pasture al lowance and l itter size on ewe l amb reari ng performance ( least squares mean ± standard error). Means across pasture ward height and across l itter size having di fferent letter superscripts are different (P<O. l 0). .................... . . ... ....... .... ............ ........ .. ............................. ..... . ... ......... ...... .... ......... 170. Table 9.8 Correlations between ewe behaviour and performance traits across al l sheep regardless of li tter size at birth . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . 171 Table 9.9 The effect of pasture al lowance and litter size on lamb behaviour in the five­ m i nute observation period immediately fol lowi ng taggi ng (weighted least squares mean ± standard error). Propmtions across pasture sward height having di fferent letter superscri pts are di fferent (P<O. I 0). ...... . . . . . ..................................................... 1 72. Table 9.10 The effect of pasture al lowance (sward height) on lamb behaviour ti mes after taggi ng for twi ns and tri plets (Wi lcoxon mean ± standard error). Times are presented as minutes. .. ................................ .. . . ............................ .. .......................... .. ..... 178. xiii.

(17) Table 9. 1 1 The effect of pasture al lowance and litter size on individual lamb petformance ( least squares mean± standard error). Means across pasture sward height having di fferent letter superscri pts are di fferent (P<O. I 0) . . ....................... 1 79 Table 9. 1 2 Conelations between lamb behaviour and performance trai ts across al l l i tter si zes at birth. xiv. . ....... . . . .................................................. . . . ....... . . .................................. 1 80.

(18) List of Figures Figure I Some factors i nfl uenci ng lamb survival and ewe lamb reari ng abi l i ty . . . ... ........ 3 Figure 2. 1 Major sheep breeds on respondents farms . ................................................... 23 Figure 2.2 Percentage of post parturient lamb losses by l itter size and H igh, Low farm (standard error included) . .... ....... ........ ....... .............................. ...... ... . ...................... 24 Figure 2.3 Mean feedi ng levels for di fferent ewe groups at different stages of pregnancy for all the farms surveyed ( Feeding level: I =below mai ntenance; 2=mai ntenance and 3=above mai ntenance) . ............. ..................................... ................................... 25 Figure 2.4 Mean shepherding frequency for sheep raising si ngles, twins and tri plets on H igh and Low farms (standard error incl uded) . ...................................................... 26 Figure 3. 1 The effect of maternal behaviour score ( M B S ) on litter survi val ( L I S ) ( least square mean and standard errors) . ............. ............................................ ... ............. 36 Figure 3.2 The effect of maternal behaviour score ( M B S ) on lamb urvi val from birth to weani ng ( least squares mean ±standard error). . . . . . . . . . . ... ......................................... 37 Figure 3.3 The genetic pathways for maternal behaviour score ( M BS), l i tter urvi val (LIS) and lamb survi val to weani ng ( LAS) ............. ................................... ............. 40 Figure 5. 1 Post mortem resu l ts for Finn and Texel lambs . . ............................................ 72 Figure 6. 1 The effect of breed of ewe on the time between twin lamb births . ............... 86 Figure 6.2 The effect of breed on the proportion of si ngle and twin lambs to have ewe contact wi thi n ten mi nutes from bitth . .................................................................... 87 Figure 6.3 The effect of breed on the proportion of si ngle and twin lambs to stand within thi rty mi nutes from birth . . ............................................................................ 88 Figure 6.4 The effect of breed on the proportion of single and twin lambs to locate their ewe's udder thi rty m i nutes after birth . .................................................................... 88 Figure 6.5 The effect of breed on the proportion of si ngle and twin lambs to suck from thei r ewe's udder thi rty minutes after bi rth . ............................................................ 88 Figure 6.6 The effect of ewe breed on time spent groomi ng twin lambs . ...................... 89 Figure 6.7 The effect of ewe-lamb contact ti me on twi n lamb survi val to weani ng ....... 92 Figure 6.8 The effect of twin lamb stand ti me on lamb survi val to weaning . ................ 92 Figure 6.9. The effect of time to locate ewe's udder on twin lamb survi val to weaning . ..................................................................... ............................................................. 92. Figure 7. 1 The effect of breed on the proportion of si ngle and twin lambs to have ewe contact after taggi ng ........................ ...................................................................... 1 1 8 Figure 7.2 The effect of breed on the proportion of si ngle and twin lambs to stand . . . . 1 1 8 XV.

(19) Figure 7.3 The effect of breed on the proportion of single and twin lambs to suck ..... 1 1 8 Figure 7.4 The effect of breed on the proportion of single and twi n lambs to fol low their moving dam wi thi n ten minutes after tagging . . . . . . . . . . . . . ..... . . . . . . . . . . . . . . . . . . . . . . . . . ........... 1 1 9 Figure 7.5. The effect of M B S within litter size on weight of lamb weaned per ewe (least squares mean ± standard error) . .. .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . ... .. . . . . . . . . . . . . . . . 1 2 1 Figure 7.6. Behaviour con fl icts for the dam when her litter is bei ng ear-tagged (where D= Dam, L l and L2 =Dam' s Litter, L= Flock Lamb, E=Fiock Ewe) . .. . . . . . . . . . . . . . . . 1 23 Figure 8. 1 . Arena configuration for arena behaviour test. . . . . . . . . . . . . . . . .. . . . . . . ... . . . . . . . . . . .. . . . . . . 1 35 Figure 8.2. The effect of litter size at birth and dam selection l i ne on maximum distance from the person (standard errors i ncluded ) . ... . ..... .. . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . . . . . .. . . 1 39 Figure 8.3 Proportion of time spent High IGF, Control and Low IGF ewes i n arena zone. . ............................................................................................................. . . . . . . . . .. 141. Figure 8.4 Proportion of H igh IGF, Control and Low IGF ewes with M BS . . . . . . . . . . . . . . . . 1 4 1 Figure 9. 1 Pasture sward height treatments and brief tiial plan . .. .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 56 Figure 9.2 Arena con figuration for arena behaviour test. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 58 Figure 9.3 The effect of pasture allowance on M B S ( least squares mean ± standard error) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 66 Figure 9.4 The effect of pasture al lowance on proportion of ewes that stayed at the taggi ng site after tagging ( least squares mean ± standard error). .......................... 1 66 Figure 9.5 The effect of li tter size at bi1th and pasture al lowance l ate pregnancy on ewe litter survi val from birth to tagging (least squares mean ± standard error). .......... 1 69 Figure 9.6 The effect of pasture allowance on proportion of lambs to bleat before ewe contact after taggi ng (standard errors i ncl uded ) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 72 Figure 9.7 The effect of pasture al lowance on proportion of ewes to make lamb contact after taggi ng (standard errors i ncluded ) . .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 73 Figure 9.8 The effect of pasture al lowance on proportion of lambs to stand i n the fi ve­ mi nute observation period i m mediately after tagging (standard errors i ncluded). l 74 Figure 9.9 The effect of pasture al lowance on the proportion of lambs to locate their dam ' s udder i n the fi ve-mi nute observation peiiod i m mediately after taggi ng (standard errors i ncluded) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 74 Figure 9. 1 0 The effect of pasture al lowance on the prop01tion of lambs to fol low their moving dam after taggi ng (standard errors included ) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 75 Figure 9.7a and 9.7b. The effect of pasture allowance on the proportion of lambs to have ewe contact within fi ve minutes from taggi ng. Graphs of esti mated 'behaviour' functions for twi n lambs ( Figure 9.7a) and for triplets ( Figure 9.7b) . . . . . . . . . . . . . . . . . . . 1 76 xvi.

(20) Figure 9.8a and 9.8b. The effect of pasture allowance on the proportion of lambs to stand within five minutes from taggi ng. Graphs of esti mated ' behaviour' functions for twi ns (Figure 9.8a) and for tri plets ( Fi gure 9.8b) . . ... ...................... ................. l 77 Figure 9.9a and 9.9b. The effect of pasture allowance on the proportion of lambs to locate thei r dam ' s udder within five minutes from taggi ng. Graphs of estimated 'behaviour' functions for twi n lambs ( Figure 9.9a) and for triplets ( Figure 9.9b) . 1 77. .................. . ......................................... . . ................................................ . ........ . ........ Figure 9. 1 Oa and 9. 1 Ob. The effect of pasture allowance on the proportion of lambs to fol low their movi ng dam withi n five mi nutes from tagging. Graphs of estimated 'behaviour' functions for twi n lambs (Figure 9. 1 Oa) and for triplets ( Figure 9. 1 Ob) . . . . . . . . . . ............................. . . . . . . . . . . . . . . . . . ................................................ ........................ 1 77 Figure 9. 1 1 a and 9. 1 1 b. The effect of pasture allowance on the proportion of lambs to suckle within fi ve mi nutes from taggi ng. Graphs of esti mated behaviour functions for twin l ambs (Figure 9. 1 1 a) and for tri plets (Figure 9. 1 1 b) . .............................. 1 77 Figure 9. 1 2 The effect of pasture allowance and l i tter size at birth on lamb survi val at taggi ng (standard errors i ncluded) . ........... ..................... ....................................... 1 79 Figure 9. 1 3 Behaviour confl icts for the dam when her litter is bei ng ear-tagged (where D= Dam, L l , L2 and L3 =Dam ' s Litter, L= Flock Lamb, E=Fiock Ewe) . .......... 1 83. xvii.

(21) Introduction In New Zealand, many sheep breeders have selected and bred ewes for i ncreased fecundity over the last two decades. Selection within existing breeds has been used to establish highly prol ific flocks ( Davis et al., 1 987). Highly fecund breeds such as the Fi nnish Landrace and East Friesian were introduced i n the 1 990' s with the objective of i ncreasing flock fert i l i ty (Meadows, 1 997).. I mproved lambing percentage is the biggest contri butor to higher p�ofits on New Zealand sheep farms (Geenty, 1 997). The national mean lambi ng percentage in 2002 was 1 24%, which was significant ly greater than the 200 1 season' s record of 1 1 9% (Anon, 2002). Davis et al. ( 1 983) reported that triplets replace si ngles a l itter ize increases above 1 .7 lambs per ewe. The i ncreased proportion of ewes havi ng triplets i s of concern to farmers and to i ndustry as Iamb mortality i s assumed to be highest i n tri plets before 24 hours post-partum (Amer et al., 1 999).. Lamb survivability is an i ncreasing problem . Lamb losses of 30% have been recorded between pregnancy scan n i ng and tai l ing (Anon, 2002; Aspi n, 1 997) and above a l i tter size of 2.3 lambs born per ewe lambi ng, farmers can expect a reduction i n farm profitabi lity (Amer et al., 1 999). Twi n and tri plet born l ambs have higher mortali ty rates than si ngles ( Hal l et al., 1 988; Johnson et al., 1 982; H i nch et al., 1 983; Scales et al., 1 986). Many studies emphasise the importance of lamb birth weight reporting lower survi val to weani ng in lambs weighi ng less than 3 kg at bi rth ( Dalton et al., 1 980; Johnson et al., 1 982; Hight and Jury, 1 970; Nowak and Li ndsay, 1 992). But overa l l , the relationship between lambing rate and lamb survi val i poorly understood i n highly fecund ewes.. Most lamb deaths occur i n the fi rst two days from birth and adequate maternal care is requi red to minimise losses and minimise the effects of detri mental environmental factors (Poindron et al. , 1 984). However the proper expression of adequate maternal behaviour is made more difficult by modern l i vestock systems which i ncrease fecundity and thus i ncreased demands upon the motheri ng abi l ity of ewes (Chenoweth and Landaeta-Hernandez, 1 998). 1.

(22) Farmers and scientists are searching for ways to improve lamb-reari ng abi l i ty i n highly fecund sheep, as cun·ent lamb losses are not acceptable from a production and animal wel fare perspecti ve.. The objecti ve of this thesis i s : Can we improve lamb rearing petformance in highly fecund sheep?. This thesis has taken a hol istic and appl ied approach to investigating the genetics, behaviour and management of the ewe-lamb relationship and investigated the integration of these factors under commercial pastoral fann i ng conditions. The relevant li terature has been reviewed and trials have been designed to i nvestigate the fol lowi ng questions : •. Can we identify the Lamb and ewe behaviours that encourage ewe- Lamb attachment in highly prolific flocks?. •. Can we identify the management practices on fatm that enhance ewe-lamb attachment ?. •. Can we select for improved mothering ability and Lamb swvival under these conditions?. The research areas investigated to answer these questions are presented in Figure I . The format of the thesi s fol lows: •. The first stage of the research required knowledge of current lamb reari ng performance on high performing sheep farms i n New Zealand. Chapter 2 explores current reproducti ve rates, lamb losses and management practices through a survey of sheep farmers with high produci ng flocks.. •. Genetic parameters for ewe maternal behaviour score, li tter survival and lamb survi val are esti mated in Chapter 3. The magnitude of environmental and management effects on these ewe and lamb traits has been esti mated.. •. Environmental effects are i nvestigated fu1ther i n Chapter 4 and Chapter 9. Chapter 4 i n vestigates the effect of ewe physiology i n late pregnancy on maternal behaviour and lamb reari ng performance. Maternal nutri tion i explored further in Chapter 9 and i nvesti gates the effect of pasture allowance i n late pregnancy o n ewe-lamb attachment behaviours and subsequent lamb rearing performance.. 2.

(23) •. Breed and l i ne di fferences i n some traits are often regarded a i nd icative of genetic variation. The objective of Chapters 5 through to 7 i s to i nve tigate Fi nn and Texel s which have been previously recognised by New Zealand farmers to have marked ly di fferent behaviour. The e recently i ntroduced breeds were u ed to explore the relationship between ewe and lamb behaviours and lamb reari ng petformance. Chapter 8 i n vestigates the behaviour of a Romney flock di vergently elected for maternal IGF- 1 . A relationship was sought with lamb reari ng performance.. •. Arena tests were used i n Chapter 8 and 9 to determi ne i f maternal behaviour and lamb reari ng performance can be predicted before parturi tion.. Figure I Some. Photograph taken by Julie Everett-Hincks 3.

(24) 4.

(25) CHAPTER 1. Literature Review. Photograph taken by Julie Everett-Hincks.

(26) Chapter I. 1 . 1 Introduction The abi lity of the ewe to rear her litter to weani ng is determi ned by the successful execution of a number of processes. These are dli ven by genetics, behaviour and the environment (summarised in Figure I ). A better u nderstanding of perinatal ewe and lamb behaviour wi l l benefit ani mal production and welfare (Chenoweth and Landaeta­ Hernandez, 1 998) . The li terature review concentrates o n the study of ewes and lambs managed outdoors. It reviews studies on ewe and lamb behaviour, genetics and farm management as they relate to lamb survi val and lamb reari ng performance to weani ng.. 1 .2 Background The evolutionary success of sheep has been due largely to i ts behaviour. The highly developed flocking i nstinct and doci li ty of sheep have fac i l i tated thei r domestication and enabled more effective coexi stence with man ( Ryder, 1 983; Zohary et al., 1 998).. Domestication is the process by which wild ani mals adapt to man and a captive environment. It can be viewed as both an evolutionary process and a developmental phenomenon ( Price, 1 998). According to Price ( 1 998) artificial selection is the only genetic mechani sm unique to domestication and may be applied i ntentional ly (consciously) or i nadve1tently (unconsciousl y). Zohary et al. ( 1 998) proposed that many of the morphological, physiological and behavioural traits that characterise domestic sheep and disti nguish them from their wi ld ancestors were shaped to a l arge extent by unconscious selection. Grandin and Deesi ng ( 1 998) suggested that domestication i nfluenced quantitati ve rather than quali tati ve behaviour responses, hei ghten i ng the response threshold above normal levels of sti mulation. They too believed that unconscious selection played a major role and is largely due to mans' role as a buffer between the ani mal and the environment.. 6.

(27) Chapter I. Duri ng domestication a number of morphological changes i ncluding a reduction i n sexual di morphism, a reduction i n male horn size, reduced camoufl age coloration and shortening of l i mbs relative to body size occurred in sheep (Zohary et al., 1 998).. Humans protect domesticated sheep and with a lowered predation pressure comes rel axed selection pressure for behavioural traits that are effecti ve again t predator . Thus there is a reduced fl ight di stance and aggression (Zohary et al., 1 998). Domestication reduced the ri sks for females canying twi ns therefore rel ax i ng selection agai nst twi nning and i ncreasing the frequency of twins (Zohary et al. , 1 998).. Human control over feeding and mating have reduced i ntraspecies competition (Grandi n and Deesi ng, 1 998). Zohery e t al. ( 1 998) proposed that this reduced i ntra-population competition, arising from a richer anthropogenic environment compared to the ani mals wild environment, wou ld lead to automatic selection for accelerated sex ual maturation, i ncreased twinning and i ncreased fat storage.. As a result of genetic selection, humans have changed the morphology and physiology of sheep and produced many breeds differi ng in a range of traits. Key and Maclver ( 1 980) questioned the extent that selection for various physical characteristic resulted i n genetical ly con-elated changes i n behaviour. Dwyer and Lawrence ( 1 998a) suggested that the i ntensification of sheep farming and the selection of sheep for i ncreased production have led to a decrease i n the expression of behaviours associated with survivabil ity under extensive conditions. Therefore feral breeds of sheep were considered to be superior mothers than the more selected breeds ( Dwyer et al., 1 998b).. 1 .3 Behaviour Behaviour is an adaptive mechanism, which responds to and alters the i nternal and external environments of an animal . It can be considered a dynamic i nterface between the i nternal and external envi ronment (Grandi n and Deesing, 1 998). Maternal behaviour i ncludes those behaviours of the dam which occur around the time of birth and which are associated with the del i very and survival of the young to 7.

(28) Chapter I. weaning. The behaviours i ncl uded becoming i solated, seeking shelter, nest buildi ng, parturition, clean i ng and sti mu lati ng the neonate to suckle and establi shment of the mother-young bond (Chenoweth and Landaeta-Hernandez, 1 998 ).. The onset of maternal behaviour i n sheep i s control led by levels of oestrogen and progesterone in gestation and the release of oxytocin in delivery ( Dwyer et al., 1 999b). Ewes, especial ly i nexperienced ewes, rel y pri mari ly on olfactory cues from amniotic fl uid on the lambs ' coat for the normal development and mai ntenance of maternal behaviour at parturition ( Levy and Poi ndron, 1 987; Alexander et al., 1 986; V i nce et al., 1 985, Poi ndron et al., 1 980b). The post parturient ewe becomes attached to any newly born lamb, which she has licked for 20-30 mi nutes (Smith et al., 1 966). The behaviour of the ewe before, duri ng and after parturi tion has a major i n fluence on lamb survi val (Nowak et al., 2000). The first hour post-pa1tum is the most i m portant i n the establ i shment o f a n exclusive mother-young bond (Alexander e t al., 1 986). The development of a strong maternal bond is an important prerequisite for lamb survi val and it is i n fluenced strongly by continuous association between the ewe and her lamb i n the time after birth ( Murphy et al., 1 994a). The formation of the bond is the result o f a rapid learning process infl uenced by both ewe and lamb behaviour (Shi l l ito-Walser, 1 978). Shi l ito-Wal ser et al. ( 1 983) observed that as lambs got older ewes moved away more quickly when disturbed. Lindsay et al. ( 1 990) observed that i ncreased vocal behaviour of lambs was strongly associated with thei r earlier recognition of their mother and better bondi ng. Nowak ( 1 996) suggested that vocal communication between the ewe and the l amb is central to adequate bond formation and has demonstrated that lamb bleati ng behaviour is i n volved i n the attachment process. Pol lard ' s New Zealand study showed that when lambs were between 6 and 1 2 hours old, ewes reari ng tri plets bleated more when thei r litters were i ntact and less when one of their lambs was removed compared with ewes with a single lamb ( Pollard, 1 992). Pollard ( 1 992 ) found that a 61 2 hour old lamb, which was separated from its dam, was more l i kely to approach a ewe that was bleati ng than a si lent ewe and concl uded that ewe bleats help in long di tance recognition of the dam. Lamb vocalisations are modified by thei r reari ng experience and appear to represent a signal of need (Dwyer et al., 1 998b).. 8.

(29) Chapter I. Ewe-lamb bonding and lamb survi val are maximised by management practices that i ncrease the time spent at the birth site by the dam after parturition ( Nowak, 1 996). Ewes that remai n longer on their birth site tend to have hi gher lamb survi val rates (Cioete et al., 1 998b ). Bondi ng and consequently survival of twi ns can be considerably i mproved i f dams remain on the bi rth site for a m i n i mum of 6 hours (Alexander et al., 1 986; Nowak, 1 996).. Separation and desertion are l ikely when the ewe is disturbed (Smith, 1 965 ) and may result from the conflict between the flocki ng behaviour of sheep and the isolation associated with maternal behaviour ( Ki l gour and Szantar-Coddi ngton, 1 995). Alexander et al. ( 1 986) showed that the i ncidence of separation from twi ns by i nexperienced ewe. was high with animals that moved from their bi rth site soon after gi vi ng birth and suggested that i ntensi ve management by shepherds cou ld aggravate the loss of lambs. However, Murphy et al. ( 1 994a) suggested that it is not so much that the dam is on the birth si te for 6 hours after parturi tion that improves lamb survi val but si mply the fact that the dam is together with all of her lambs.. The frequency of rejection behaviours decl i ned with maternal experience ( Dwyer and Lawrence, 2000) and more frequent abnormal post lambing behaviour occur among 2year-old primi parous ewes than older ewes (Sharafeldi n and Kandeel , 1 97 1 ). Poor maternal behaviour i n pri miparous ewes is often considered a cause of lamb death (Arnold and Morgan, 1 975) and Dwyer and Lawrence (2000) suggest that the behaviour of a pri m iparous ewe is reasonably predictive of her behaviour in subsequent pregnancies as i ndi vidual ewes were consistent i n thei r expression of maternal care across parities. Lindsay et al. ( 1 990) suggested trai n i ng primiparous ewes to the presence of man to reduce emotivity and subsequently i mprove mothering abi l ity.. Under extensive pastoral conditions Meri nos are slow to recognise the size of their li tters and inherently poor at mai ntai ning contact with more than one lamb in the neonatal period (Alexander et al., 1 983b; Stevens et al., 1 982). In one Australian study 46% of twin beari ng Meri nos became permanently separated from one lamb, mostly on the day after giving birth and in at least 54% of the cases there was no obvious precipitati ng factor and 34% of twi n beari ng ewes were separated temporari l y from at. 9.

(30) Chapter I. least one lamb (Alexander et al., 1 983b). I n Meri nos the prop01tion of separations decreased as the time ewes remai ned near the birth site i ncreased.. Ewe-lamb separation causes emotional stress which i n fluences the behaviour and cortisol response of young ani mals more marked ly than older animals (Napolitano et al., 1 995). These authors fou nd that the average dai ly weight gai n was lower in lambs that had been separated earl ier from the ewe and suggest that the lowered performance is the consequence of a reduced abi l ity of young animals to cope with emotional and nutritional stresses (Napolitano et al. , 1 995).. Thus the temporary separation of a lamb from its dam and its l i ttermates is detri mental to its survi val (Nowak and Li ndsay, 1 992; Alexander et al., 1 983b; Stevens et al., 1 982) and later performance (Napolitano et al., 1 995 ). Orgeur et al. ( 1 998) reported that when ewes and lambs temporari ly separated from each other they express their di stress by an i ncrease in bleati ng and locomotor activity. H i gh-pitched bleating is affected by ewe breed and may reflect the l i kelihood of separation of the ewe from the lamb ( Dwyer et al., 1 998b) but Li ndsay et al. ( 1 990) suggested that the best mothers were those that. were less active in stressfu l si tuations and those that bleated more often.. Ani mals that are used to humans are more easily managed and less timid than those that were not (Markowitz et al., 1 998). Lambs that have 40 minutes of positive human contact at 1 -3 days of age are less afraid of people (Markowitz et al., 1 998). Murphy et al. ( 1 994a) reported that ani mals of quiet temperament grow faster than animals that are. restless, nervous or aggressi ve ani mals.. Attempts have been made to fi nd associations with 'emoti vity' and reaction of animals to an external sti mulus (Alexander et al., 1 984) and i ndications are that lamb survi val is lower i n ewes that are more emotive ( Putu et al., 1 988a). The Arena Behaviour Test has been used i n Australia to predict the lamb-reari ng abi l i ty of Meri no ewes ( Ki lgour, 1 998). Ki lgour ( 1 998) showed that arena behaviours were useful predictors of lamb survi val . Ki lgour and Szantar-Coddington ( 1 995, 1 997) and Ki lgour ( 1 998) reported di fferences i n the total di stance travel led and the number of bleats i n ewes selected for greater fertility when compared to a random control flock i n the arena behaviour test. Murphy et al. ( 1 994a) showed that resu lts from their mod i fied arena test were highly 10.

(31) Chapter I. repeatable when assessed at weani ng, hogget age and 2.5 years. Shi l l i to-Walser et al. ( 1 983) tested three breeds of sheep in a T maze to observe the behaviour of ewes when gi ven a choice between goi ng near to their own lambs or goi ng to a group of ewes and lambs from their flock. Maternal behaviour was strongest in Dalesbred, selected for h i l l farmi ng, with young lambs and the flock tendency strongest i n Soays, least domesticated sheep, with older lambs ( Shi l l i to-Walser et al., 1 983).. The quanti fication for maternal behaviour is difficult and presents logi tical problems especial l y as the expres ion of this trai t may be i nfluenced by: pari ty, number of lambs, experience, nutrition and the weather ( Poindron and. Le. Nei ndre, 1 980a).. O'Connor et al. ( 1 985) quanti fied the potential infl uence of maternal behaviour on lamb survi val and growth across a number of tradi tional and highly fecu nd ewe breeds and across ewe ages. They established a positi ve relati onship between a M aternal Behaviour Score ( M B S ) and lamb survi val and growth when con·ected for litter size, ewe breed and age (O' Con nor et al., 1 985). The M B S was recorded on a 5-poi nt scale on the response of the ewe to a shepherd handl i ng and taggi ng her lambs within 24 hours from birth. Lamb survi val to wean ing i ncreased 1 0% with an i ncrease i n M B S from 2 to 5 and there was an i ncrease i n weight weaned per ewe mated with i ncreasing M B S (O' Connor, 1 990). M B S is a good i ndicator not only for lamb growth but also for the strength of the pre-weaning ewe-lamb relati onship (O'Connor, 1 996). Further studies on M B S by Parker and Nicol ( 1 993 ) showed that lambs of M B S 4 and 5 ewes located thei r dam's udder i n 30% less time than lambs of M B S 3 ewes. Rapid location of the udder i ncreased i mmunoglobu lin concentration in lamb plasma 24 hours after birth and may contribute to the i mproved lamb survi val and subsequent ewe producti vity associated with high M B S ewes (Parker and Nicol, 1 992) .. Pol lard ( 1 989) highlighted the impo1tance o f neonatal lamb behaviour i n ewe-lamb attachment, especially as litter size i ncreases. Arnold and Morgan ( 1 97 5 ) reported that poor maternal behaviour was the direct cause of 1 6% of lamb deaths, however fai l ure of the lamb to dri nk after standing caused 23% of lamb deaths. Poi ndron et al. ( 1 980b) empha i sed the i mportance of lamb behaviour and reported that the behaviour of the neonatal lamb i n fl uences the maternal responsi veness of the ewe. However many bel ieve that maternal behaviours i n fluence lamb behaviour providing evidence that 11.

(32) Chapter I. maternal behaviour at parturi tion is cri tical to lamb survi val (O'Connor et al., 1 992b; Astroshi and Osterberg, 1 979, Dwyer and Lawrence, 1 999a). Shi l l i to-Walser ( 1 978) also appreciated the importance of effective lamb behaviour concluding that the formation of the maternal-offspri ng bond in sheep is the result of a rapid learn i ng process i n fl uenced by the behaviour of both the ewe and the lamb.. According to Alexander ( 1 988), mother-young acceptance is more l i kely if the newborn lamb stands soon after bitth, sucks soon after standing, fol lows the mother closely and moves to the mother if separated. A New Zealand study at Ruakura on mi xed age Romney ewes concluded that the time taken by a lamb to stand up after bi rth was the best avai lable i ndication of vigour (Wal lace, 1 949). Wal lace ( 1 949) observed little or no relationship between vigour and sex, type of birth, age of ewe, birth weight, type of presentation, or time between onset of l abour and birth. However, Astroshi and Osterberg ( 1 979) observed that an i ncrease in birth weight shottened the time from birth to standi ng. Dwyer et al. ( 1 999b) showed that the greatest vatiation i n lamb activity was related to maternal progesterone. It is bel ieved that in prolific sheep, birth weight influences behaviour trai ts, but litter size does not, except i ndirectl y through i ts effect on bi tth weight, so the smal ler the lamb the less active it is (Atroshi and Osterberg 1 979).. Many studies have demonstrated breed di fferences i n peri natal ewe behaviour ( Dwyer et al. , 1 998b, 1 999b; Dwyer and Lawrence, 1 999a; Fahmy et al., 1 996; Shi l l i to-Walser et al., 1 983 ; Key and Mac l ver, 1 980; Lindsay et al., 1 990; Cloete et al., 2002 ) and have. attempted to defi ne what patticular characteristics make a good mother. O' Connor et al. ( 1 985, 1 992b ) have i ncreased our understandi ng of maternal behaviour in tradi tional and high ly fecund sheep, however little i s known of the effect of lamb behaviour i n large l i tters o n formation of the ewe-lamb bond and little is known of thei r behaviour under extensi ve New Zealand farming condi tions. Research on farm under commercial pastoral conditions is l i mited with the majoti ty of outdoor studies performed on Merinos, which are not natural ly fecund (Alexander et al., 1 979; Putu et al., 1 988b ).. 12.

(33) Chapter I. 1 .4 Selection The variation of motheri ng abi lity within and between breeds shows promise for genetic improvement of this i m portant characteri stic ( Manktelow, 1 996 ). However i mprovement in motheri ng abi lity is dependent upon our success in effecti vely integrating evolutionary behaviours with modern production systems and then selecting those behaviours that are advantageous for production ( Chenoweth and Landaeta­ Hernandez, 1 998 ). This statement assumes that we can adequately identify evolutionary behaviours and apply selection pressure to them, without comprom ising other valuable trai ts. Therefore a better understandi ng of the relationship between lamb survi val and ewe behav iour are critical for the development of sustai nable breeding programmes ( S i mm et al. , 1 996 ). Breed and line di fferences i n some trai ts are often regarded as indicati ve of genetic variation ( H inch, 1 997 ) . Breed di fferences were reported for early neonatal progress i n lambs ( S iee and Spri ngbett, 1 986; Alexander et al., 1 990a,b; Cloete et al., 1 998b; Cloete et al., 2002 ). supporting the contention that it is paitial ly under genetic contro l . Fahmy e r al. ( 1 996 ) reported n o breed di fference i n peri natal behaviour i n Fi nn and Camb1idge ewes ( 20% Fi n n ), but Holmes ( 1 976 ) found poorer maternal abi lity in Fi n n ewes compared to Forest C l u n ewes and suggested that the poorer maternal abi l i ty cou ld have ari sen through lack of selection pressure associated with selection for prolificacy from i ntensive husbandry, as Fi nn sheep in Europe are predom inantly lambed i ndoors. Whately et a l . ( 1 97- 0 observed that Meri nos and Rom neys in New Zealand exhibited poorer motheri ng abi lity than the Perendale, Cheviot and Border Leicester Romney cross sheep, crossbred sheep were best. Meri no dams were most l i kely to abandon lambs and Romneys had the poorest abi lity to seek shelter when lambing in poor weather. Alexander et al. ( 1 983a) reported poor maternal behaviour in twin-beari ng Merino, Dorset Horn and Border Leicester ewes in Austral ia.. Dalton. et. al. ( 1 980 ) found large di fferences between breeds in lamb survi val under New. Zealand conditions. Studies of lamb survi val in prolific breeds have shown that lamb losses duri ng the first few days after birth consti tute between 50 and 75% of al l lamb losses ( Petersson and Danel l, 1 985; Dalton. et. al., 1 980; Hinch. er. al., 1 986; Maund et. al., 1 980). These studies showed a positi ve re lationsh ip between birth weight and lamb. 13.

(34) Chapter I. survi val as litter size i ncreased and i denti fied that dams older than 3 years were more l i kely to support the survi val of large li tters. Dawson and Carson (2002) reported simi l ar lamb mortality rates across a range of ewe genotypes however weaned Jamb output was greatest for the B l uefaced Leicester x B l ackface ewe genotype. Wiener et al. ( 1 983) showed that crossbred Jambs had better survi val rates than purebreds but this advantage was onl y evident between 3 days and 6 weeks of age. Selecti ve breedi ng has been advocated as a means of i mproving ewe Jamb rearing abi lity and lamb survi val under pastoral conditions (Haughey, 1 993). However recent estimates of the genetic components of Jamb survi val i ndicate l i mi ted genetic variation (Petersson and Danell, 1 985; Hal l et al., 1 995; Lopez-Y i l lalobos and Garrick, 1 999; Morris et al., 2000; Fadi l l i and Leroy, 200 I ; Cloete et al., 2002) and that lamb survi val i s control led mainly by non-genetic factors. However, Petersson and Danel l ( 1 985) suggest that the maternal genetic influence of lamb survi val, that is the dam ' s provision of an environment i n which the offspri ng performs, is more important than the directly transmitted genetic component of the offspri ng.. Selecti ve breedi ng for ewe Jamb rearing abi lity, defined as the ratio of Jambs weaned to lambs born, has resulted in a reduction in Jamb m01tal i ty (Haughey, 1 983; Cloete and Schol tz, 1 998a) and a reduction in the i nterval from stand i ng to suckl i ng (Cioete and Scholtz, 1 998a). The inference that reari ng abi l ity is heritable has been confi rmed i n few studies w i t h Meri nos ( Haughey, 1 984; Haughey et al., 1 985) and has been reviewed by Piper et al. ( 1 982). The repeatabi l ity of reari ng performance has also been esti mated at I 0% by Haughey ( 1 984 ). Purser and Young ( 1 983) reported that Jamb reari ng abi l i ty was a repeatable trait and that the more Jambs reared the better the subsequent performance appeared to be. If the Jamb was not reared from the ewe' s fi rst parity lamb mortality at age three and parity two was 26.8% but if the previous Jamb was reared the mortal ity was only 1 3 .5% (Purser and Young, 1 983). Dal ton et al. ( 1 980) suggested cul li ng 2 year old ewes that were not pregnant would have a greater effect on i mprovi ng flock Iamb survi val than cul l ing ewes that had lost a lamb. Morris et al. (2000) found that environmental variances due to permanent maternal effects (for example uteri ne capacity, pel vic width, mi lking and maternal abi l ity) were found to contribute mostly to the repeatabi l i ty of ewe reari ng performance providing evidence that improved lamb. 14.

(35) Chapter I. surv i val has to be seen mainly as a successful partnershi p between mother and offspring throughout pregnancy, parturition and lactation. There has been selection for behaviour traits thought to i mprove lamb survival, however these studies are scarce ( Lam be et al., 200 I ; Cloete et al., 2002). Lam be' s team reported a heritabili ty of 0. 1 3 and a repeatabi l ity of 0.32 ( Lam be et al., 200 I ). Cloete et al., (2002) showed that the time to stand from birth and time to suck from standing have a. genetic component and that it di ffers between breeds. In the study of Owens and Armstrong ( 1 985) an i ncrease in lamb birth weight was associated with neonatal lamb vigour in Booroola lambs, although only regression coefficients were reported. New Zealand sheep farmers have a range of options to influence the genetic capabi lity of their flock. New sheep breeds and the genes they can contribute to the national sheep flock can provide an opportunity to i mprove lamb-rearing abi l i ty. However published estimates of genetic parameters for lamb rearing abi lity are limited, particularly in highly fecund sheep and selection i s l i mi ted as lamb rearing abi lity is a sex-limited trait. Further research is needed to determine the heritabi lity of lamb and ewe behaviours that relate to lamb survi val and lamb rearing abi lity.. 1 .5 Environment and Management Bei lharz and Luxford ( 1 984) discussed the limiti ng effect of the environment on reproducti ve progress and accepted that extreme variants like Booroola genes and the appl ication of fertility treatments l i ke Fecundin used for i ncreasing litter size, can only be uti lised successfully i f the environment i s i mproved. They maintain that even relati vel y smal l increases in litter size wi l l need improvements in the environment if they are not to have deleterious side effects ( Bei lharz and Lux ford, 1 984). Amer e t al. ( 1 999) reported that above a litter size of 2.3 lambs born per ewe lambi ng, farmers can. expect a reduction in farm profitability. Triplets replace single as litter size increases above 1 .7 lambs per ewe and Iamb mor1ality is assumed to be highest in triplets before 24 hours post-partum and di ffers with type of environment and farm (Amer et al., 1 999).. 15.

(36) Chapter I. Ewes must uti l ise reserves to support the growth of mu lti ple lambs i n utero (Jel bart and Dawe, 1 984). The energy cost during pregnancy in the ewe is largely met by i ncreased feed i ntake, except i n very late pregnancy when i ntake may decline. The ewe may respond to i ncreased nutrient demand in late pregnancy and lactation by eating more, by mobi lising maternal tissues, by improving the overal l effecti veness of digestion or by i ncreasing the efficiency of uti l i sation of nutrients by tissues (Jel bart and Dawe, 1 984). Mel ior ( 1 983) recognises that foetal growth rate is determi ned by maternal nutri tion and placental growth. Placental development is especial ly important in ewes lambi ng 1 50% or more to ensure l ambs are sufficiently heavy at birth and to gi ve them the best chance of surv i val (Mel ior, 1 983; Bel l , 1 984).. In late pregnancy and early lactation the nutritional requirements of ewes are very high, particularly for ewes with twins (Hol mes, 1 975). The last 4-6 weeks before lambing are cri tical ( Scales et al., 1 986). When multi ple bearing ewes were offered addi tional feed in late pregnancy to i ncrease l i veweight, lamb mortality was reduced, suggesting meri t i n i mproved pre-lamb feeding for ewes carrying more than one lamb (Scales e t al. , 1 986). Scales and eo-workers reported that a I Okg i ncrease in ewe l iveweight duri ng the last 6 weeks of pregnancy resulted in an i ncrease of 0.46kg in singles and 0.52kg i n twi ns at birth ( Scales et al., 1 986). Ideal ly ewes with multiples shou ld be identi fied as early as possi ble, separated and preferential ly fed (Geenty, 1 997).. Hight and J ury ( 1 970) reported that lambs born to ewes fed wel l i n late pregnancy are better equipped to survive conditions predisposing to exposure and starvation as they have more energy stored as brown fat reserves and mai ntain their suckl i ng drive for longer than lambs born to poorly fed ewes. Moore et al. ( 1 986) reported that lambs born from ewes fed higher pasture al lowances had higher rectal temperatures at birth than those born from low al lowance ewes i ndicati ng the lambs poorly fed i n utero tend to achieve high metabolic rate more slowly than those which recei ve a good nutrient supply (McCutcheon et al., 1 98 1 ). Moore and eo-workers ( 1 996) also observed that increased lamb vigour was associated with i ncreasing bitth weight for all lambs, but i ncreasi ng rectal temperature was associated with i ncreasing birth weight only for male lambs. However, Robi nson et al. ( 1 977) reported that the level of feedi ng in late pregnancy did not affect foetal weight i n highly fecund Finn crosses.. 16.

(37) Chapter I. Inadequate feeding of ewes can result in poor maternal behaviour and poor ewe behaviour is an important cause of lamb death (Nowak, 1 996). Dwyer et al. (2003) demonstrated that a moderate reduction in maternal nutrition i n late pregnancy resulted in a quantitati ve reduction i n the expression of maternal behaviours, i n particular MBS, at parturition under intensive conditions i ndoors. The ame author did not ob erve a change in neonatal lamb behaviour from the effect of maternal under nutrition and concluded that neonatal progress was affected by birth weight ( Dwyer et al., 2003). Nowak ( 1 996) observed a considerable improvement in twin lamb bondi ng with their dam and twin lamb survival when the mother remained on the bit1h site for a mini mum of 6 hours. Lindsay et al. ( 1 990) observed that the time spent on the birth ite i ncreased when ewes with twi ns were on a higher plane of nutri tion 6 weeks before lambing. Putu et al. ( 1 988a) observed that pasture conditions duri ng late pregnancy had no effect on. ti me spent by the ewes on the birth site. However twin-beari ng ewes remained on the birth site for sign ificantly longer than ewes with a single lamb. In twin bearing ewe low level of nutrition resulted in a higher propm1ion of permanent de ertions of at lea t one of thei r twin lambs ( 1 9.2%) compared with ewes on a high level of nutrition (4.3%) ( Putu et al., 1 988a). If paddock pasture quality and feedi ng levels are good and the paddock has adequate shelter then the ewe wi l l stay at the bi rth site with her lambs for a longer period ( Pollard and Littlejohn, 1 999; Putu et al., 1 988b ). Studies have attempted to define suitable husbandry practices for the i ncreased lambing rate and to encourage ewe-lamb attachment and subsequently survival immediately after birth. In a national survey of farmer practices at lambing, Aspin ( 1 997) identified that when selecting paddocks for lambing priority considerations were: pasture cover, shelter, stocking rate and topography. However these factors were not defi ned. Pollard and Littlejohn ( 1 999) i n vestigated shelteri ng behaviour of lambing sheep under extensive conditions and types of shelter did not affect lamb productivity in the one­ year of sntdy, where weather was moderate. They did however observe that least used artificial shelter were near roadways and human acti vity rei nforcing the ewe's need to seek isolation. Ducker and Fraser ( 1 973) investigated husbandry practices indoor and also compared indoor with outdoor practices at lambing. They concluded from their indoor trials that poorer husbandry resulted in increased lamb mortali ty, particularly amongst twin lambs. The ewe-lamb relationship benefits from the absence of 17.

(38) Chapter I. interference from flock mates with lambs i ngesting more colostrum with outdoor lambing (Ducker and Fraser, 1 973). However for New Zealand conditions, the effect of season and weather on survi val is the most important source of variabi l ity ( John on et al., 1 982; Pollard and Littlejohn, 1 999; McCutcheon et al., 1 98 1 ).. Studies investigati ng the effect of nutri tion on the behaviour of the ewe and lamb are scarce and have predominantly been perfo1med on Merinos, which are not naturally fecund ( Putu e t al., 1 988a), under artificial conditions ( Dwyer e t al., 2003 ), or have not adequately defi ned the behaviours that are affected (Moore et al., 1 986).. Many studies have investigated ewe and lamb behaviour and Pol lard ( 1 989) and O'Connor et al. ( 1 985) have made a significant contri bution i n vestigating ewe and lamb behaviour in litter sizes larger than two under extensive New Zealand conditions. However, this l iterature review highlights the research deficits that exist and therefore this thesis attempts to add to the current literature by investigating the management and genetic factors which affect maternal and neonatal lamb attachment behaviours for larger l itters under pastoral grazing conditions and explores thei r relationship with lamb survi val to weaning.. 18.

(39) CHAPTER 2. Lamb rearing success on high performing sheep farms. Photograph taken by Julie Everett-Hincks. Related publications: Everett- H i ncks et al. ( 200 1 ) . S urvey of high performi ng sheep farmers: lamb rearing experience. The 6'11 Congress for the Association of Advanced. A n imal Breeding and Genetics..

(40) Chapter 2. 2 . 1 Summary Data col lected from 1 8 1 New Zealand sheep farmers were segregated into two groups; a high farm group with at least 1 50% lambs weaned and a low farm group with less than 1 50% at weaning. The percentage lamb losses per ewe were simi lar between groups, therefore by maintaining losses simi lar to the Low group, the High group had greater lamb production to weaning. High lamb rearing success to weaning was observed in both groups ( 1 63%, 1 34%) compared to the national average of 1 1 5% to tai l i ng for the 1 999/2000 season (Anon, 2002), which was not surprising as high performing sheep farmers were surveyed. Both groups identi fied motheli ng abi l i ty as the most i mportant factor affecting lamb survival. Management practices were investigated to explain how the High farm had greater lamb production to weaning. Ewe feeding levels plior to lambing were simi lar for the two groups, however shepherding frequency was signi ficantly higher in the High farm group compared with the Low farm group and higher compared to shepherding frequencies reported in earlier studies (Aspi n, 1 997). Increased shepherding frequency ha probably led to greater lamb production by maintai ning low lamb losses i n the High farm group.. 2.2 Introduction A report by McKinnsey and Company recom mended that farmers should target an on­ farm producti vity i ncrease of 5% per annum to offset a long-term decrease i n farm product plices (Anon, 2000). Thi s increase wi l l depend on the adoption of technologies and farm management disciplines used by leading farmers. The lamb meat market remai ns strong and has increased from an average 1 999 season lamb price of 40.30 to $7 1 .25 for the 200 I season (Anon, 2002). Improved lambing percentage is the most important factor used to generate higher profits on sheep farms (Geenty, 1 997). Farmers wanti ng to produce more lambs per ewe mated are usi ng highly fecund breeds such as the Fi nnish Landrace and East Friesian, hogget mating or 20.

(41) Chapter 2. i ncreasing ewe ovulation rate through the use of ferti lity treatments. However, an increase i n the number of lambs born per ewe is only valuable i f lambs survive and comes at a cost if pregnancy and lambing are not managed appropriately. In 1 98 1 Rohloff and eo-workers surveyed 1 80 Otago and Southland high performance flocks ( Rohloff et al., 1 982). The aim of that survey was to assess what management changes had accompanied the tise in lambing percentage. Feed supply duri ng lambing and early lactation and lamb mortality were the main concerns of farmers with flocks exceedi ng 1 50% lambi ng. Lamb mortality ranged from 9.4% for lower lambing percentage flocks ( 1 30- 1 49%) to 1 3.8% for the highest performi ng flocks ( + 1 70 % ). This paper explores current reproductive rates, lamb losses and management practices through a survey of sheep farmers with high producing flocks.. 2.3 Materials and Methods One hundred and eighty one sheep farmers identified by farm consultants and breed groups took patt in a survey. The farmers were achievi ng a high lambi ng percentage for their farming conditions. The questionnaire was designed and compi led by Julie Everett-Hi ncks, Kevin Stafford and Hugh Blair from Massey University. The survey form contained twenty four questions and requested that answers be based on 1 999 production figures. Information was requested on farm descri ption, sheep selection and cul l i ng policies, size and age structure of the flock and management policy. Lamb production records were requested with particular emphasis on percentage lamb losses, lamb reari ng performance and cause of ewe and lamb deaths. To objectively separate the respondents i nto two groups a cut-off lambing percentage of 1 50% was used. Data were segregated based ori the Monitor Farm mean for the 1 999 season plus two standard deviations (i .e. 1 25% + (2. x. 1 3%)), which was 1 50% (Anon,. 1 999). 69% of the respondents had greater than 1 50% lambs weaned per ewe mated, whereas 3 1 % had less than 1 50% at weaning. The groups w i l l be referred to as High 21.

Figure

Table 2. 1 Summa1y description of farms surveyed.
Table 2.2 The importance of factors affecting lamb survival according to respondents from High and Low farm groups
Figure 2.3 Mean feeding levels for different ewe groups at difI erent stages of pregnancy f
Figure 2.4 Mean shepherding frequency for sheep raising singles, twins and triplets on High and Low farms (standard error included)
+7

References

Related documents

Junior Positive mental health inventory prepared by Agashe and Helode (2009) was used as psychological instrument in the present study.Results reveal significant impact of yoga

The use of sodium polyacrylate in concrete as a super absorbent polymer has promising potential to increase numerous concrete properties, including concrete

All correlation coefficients were negative, indicat- ing that a higher burden of adherence barriers measured by ABQ-HIV or its subscales was associated with a lower MMAS score and

Here we present a comparison between the bacterial community of faecal samples from the rabbit and the community of another caecotrophagic species, the guinea pig, to com- pare

The structure and magnitude of the oceanic heat fluxes throughout the N-ICE2015 campaign are sketched and quantified in Figure 4, summarizing our main findings: storms

Nitri fi cation and sedimentary denitri fi cation occurred near the river mouth, nitri fi cation prevailed further offshore under the plume, and fi nally, phytoplankton

Effects of separate and combined stresses of salinity and waterlogging on shoot and root K + content of selected 8 barley varieties under hydroponic conditions.. 6 day old

In 2008, the first pilot experiments were conducted in seven organic winter oilseed rape fields (total area: 4.2 ha) in North-western Switzerland.. One half of each field was