The effect of breeding hoggets on lifetime performance

Abstract

The reported study involved mature ewes that had either been presented for breeding and had lambed as a hogget (MP), had been presented for breeding as a hogget but did not become pregnant (MNP) or were not presented for breeding as a hogget (NM). Their subsequent performance as mature ewes was examined. Although breeding as a hogget reduced (P <0.05) two-tooth mating live weight and condition score in subsequent years, the differences were either smaller or not present at all (P>0.05). Breeding as a hogget increased (P<0.05) the number of foetuses per ewe lifetime (8.2±0.08, 7.0±0.13 and 7.1±0.12 for MP, MNP and NM respectively) but did not alter ewe (P>0.05) longevity. These results indicate that hogget breeding can be used as a management option to improve lifetime reproductive performance.

Keywords:

• ewe lamb
• hogget
• breeding
• reproduction
• lifetime performance

Introduction

In New Zealand, approximately 30% of ewe lambs (7–9 months of age; hoggets) are presented for breeding each year, but these ewe hoggets wean less than 4% of total lambs weaned in the national flock (MAF, 2010). These results indicate that increasing the percentage of ewe hoggets successfully bred and that rear a lamb to weaning is a means of increasing the national flock's productivity. A number of potential advantages of breeding ewe hoggets have been identified. These include higher net profits, improved utilisation of spring herbage, increased efficiency, early recognition of fertility potential, improved fertility level throughout the flock, increased rates of genetic gain and improved lifetime production of the ewe by breeding her at a younger age (Hight 1982; Kenyon et al. 2004b). However, Kenyon et al. (2004a) reported that the main reason farmers do not breed ewe hoggets is a perceived negative effect on breeding performance at 18–19 months of age (two-tooth).

A negative effect on two-tooth performance may negate any potential positive effect of successful breeding as a hogget on lifetime performance. However, Dyrmundsson (1973) stated that breeding at a young age in well-managed and nourished ewes has no detrimental effects on subsequent reproductive efficiency and, in some studies, has enhanced subsequent fertility leading to an increase in lifetime performance. More recently, Kenyon et al. (2008) managed pregnant and non-pregnant ewe hoggets as one group and reported that ewe hogget breeding reduced two-tooth live weight, body condition score and reproductive performance. Under these conditions, it may be hypothesised that breeding ewe hoggets would have a negative effect on ewe lifetime productive performance. Therefore, the aim of the present study was to record the performance of the two-tooth ewes in the study of Kenyon et al. (2008) during the subsequent 3 years to determine if breeding as a ewe hogget impacted on lifetime performance.

Method

Experimental design

The current experiment utilised 991 two-tooth (18–19 months at breeding) Romney ewes and was conducted from March 2005 until December 2008 at Massey University's Riverside farm, 10 km north of Masterton on the east coast of the North Island of New Zealand (40°50′S, 175°38′E).

Background

As previously described by Kenyon et al. (2008), to determine the effect of hogget breeding on two-tooth performance, ewe hoggets (n=1163; 8–9 months of age) were randomly divided into two groups in May 2004 and were either presented for breeding with rams for a 34-day period (n=873) or not bred (n=290). At the end of the breeding period, all ewe hoggets were merged into one group and managed under commercial grazing conditions until those that were diagnosed pregnant were grazed separately for lambing and lactation during the period from late September 2004 until mid-January 2005. A total of 568 lambs were weaned from the 873 ewe hoggets presented for breeding (mean birth and weaning weights were 4.16±0.14 and 21.50±0.19 kg respectively). The study design resulted in three groups of ewe hoggets:

1.	those that were presented for breeding and lambed (mated and lambed, MP)
2.	those that were presented for breeding but did not become pregnant (mated and not lambed, MNP)
3.	those that were not presented for breeding (not mated, NM).

MNP and NM ewe hoggets were managed as one group during the period from September 2004 until mid-January 2005. Subsequently, all three groups were merged and managed under commercial conditions for the remainder of the present study. Kenyon et al. (2008) previously reported the effects of hogget breeding on two-tooth (2-year-old) ewe performance during 2005. Briefly, in summary, MP two-tooth ewes were lighter (44.8±0.23, 50.8±0.37 and 51.6±0.34 kg for MP, MNP and NM respectively) and of lower body condition (2.8±0.02, 3.3±0.03 and 3.3±0.03 for MP, MNP and NM respectively) at breeding than both MNP and NM ewes, and remained lighter to the weaning of their lambs. These differences resulted in slightly lower reproductive performance although there was no difference in the weight of their lambs at birth or at weaning.

Present study

To evaluate the effect of breeding ewe hoggets on their adult performance, 991 of the previously mentioned ewe hoggets were studied in the present work. The three groups comprised MP ewes (n=532), MNP ewes (n=211) and NM ewes (n=248) that were 17 months of age in January 2005.

Ewe live weight and body condition

In 2006, 2007 and 2008, ewes were weighed un-fasted and body condition scored (Jefferies 1961; scale 1–5 with 1 = emaciated and 5 = obese) at breeding, pre-lambing (approximately 2 weeks pre-lambing) and weaning of the lambs (approximately 100 days after the mid-point of lambing). The 2005 data (two-tooth, 2-year-old) have been reported by Kenyon et al. (2008).

Reproductive performance

In April 2005, 2006, 2007 and 2008, when the ewes were aged approximately 18, 30, 42 and 54 months, ewes were joined with crayon-harnessed Romney rams (ram:ewe ratio of 1:100) for a period of 34 days (two 17-day reproductive cycles). Crayon-harness marks on the rumps of the ewes were used to determine the proportion of ewes mated in the first 17 days of the breeding period and ewes pregnant to the first 17 days of breeding.

Pregnancy status was then diagnosed by ultrasound scanning approximately 50 days after the end of breeding each year. Ewes were diagnosed as being either non-pregnant, single-, twin- or triplet-bearing (the latter two groups pooled as multiple-bearing for some analyses).

Lamb birth and weaning weight

In 2005 and 2006, within 24 h after birth, lambs were weighed, identified to their dam and birth rank recorded. Lambs were weighed again at weaning. Lamb live weight data were not recorded in 2007 or 2008. The 2005 data were reported by Kenyon et al. (2008).

Data analysis

All data were analysed using SAS 2006. Ewe live weight and body condition score were analysed using the Mixed procedure, with a mixed linear model for each year. The model included the fixed effects of ewe group (MP, MNP, NP), pregnancy diagnosis category (non-pregnant, single- or multiple-bearing (due to relatively low numbers of triplet-bearing ewes, twin- and triplet-bearing ewes were characterised as multiple-bearing) and the interaction between ewe group and pregnancy diagnosis category. Non-significant (P>0.10) interactions remained in the model.

Ewe breeding performance (mated in the first 17 days of breeding, pregnant to the first 17 day of breeding) and pregnancy status (non-pregnant, single- or multiple-bearing) each year were analysed as a binomial trait after logit transformation using the Genmod procedure with ewe group as a fixed effect. Data were back-transformed and presented as least-square means with a 95% confidence interval (CI). The umber of foetuses each year per ewe presented for breeding was analysed using the Mixed procedure with ewe group as a fixed effect for each year.

To determine a proxy for ‘lifetime’ (2004–2008, age approximately 8 months to 5 years) reproductive performance of each ewe present at the beginning of the study (2004), each ewe's pregnancy scanning data were collated (non-pregnant, single-, twin- and triplet-bearing) for each year. Ewes not bred in 2004 (NM) were given a nominal value of 0 for that year's scanning data. Ewes that were not present at later pregnancy scanning (i.e. those that had been removed from the flock or had died during 2006 to 2008) were given a nominal value of 0 for the remaining year's pregnancy scanning data (referred to as ‘Lifetime data 2005–2008’). The data were then analysed using the Mixed procedure with ewe group as a fixed effect. In addition, the lifetime reproductive performance of only those ewes that were present for the entire study (i.e. those that had complete scanning data for 2005–2008, plus 2004 if bred) was also analysed using the Mixed procedure with ewe group as a fixed effect (referred to as ‘Lifetime ewes present 2008’).

The percentage of ewes present at breeding each year was used as a proxy for longevity within the flock, with the number present at breeding in 2005 being the base value. Ewe survival was analysed as a binomial trait using the Genmod procedure. If a ewe was not present at weaning she was characterised as no longer being part of the flock. Each year, non-pregnant ewes were removed from the flock at pregnancy scanning (2005, n=33; 2006, n=21; 2007, n=7; 2008, n=45). Ewes were also removed from the study if they were culled due to poor health or if they died. However, no individual records were kept for this.

Lamb birth and weaning weight were analysed using the Mixed procedure with a mixed linear model. The model included the fixed effects of ewe group, birth rank (or rearing rank), sex of the lamb and two- and three-way interactions between these effects. All interactions were non-significant (P>0.10) and removed from the model. Birth weight was used as a covariate in the model for the weaning weight analysis. Date of birth was used as a covariate in the models for lamb birth and weaning live weight. Lamb survival to weaning in 2005 and 2006 was analysed as a binomial trait after logit transformation using the Genmod procedure with a linear model that included ewe group.

Results

Ewe live weight

In 2006, MP ewes were lighter (P <0.05) than NM ewes at mating (52.8±0.72 kg MP ewes and 55.2±0.96 kg for NM ewes) and were lighter (P <0.05) than MNP and NM ewes at pre-lambing (68.4±0.62 kg, 71.4±1.39 kg and 71.1±1.09 kg for MP MNP and NM ewes respectively; see Fig. 1). However, no differences (P>0.10) between ewe groups were observed at weaning of their lambs in 2006.

Figure 1  Effect of ewe group (mated and lambed [MP], mated and not lambed [MNP] or not mated [NM] as a hogget in 2004) on ewe live weight A and body condition score B at mating (MT), pre-lambing (PL) and weaning (WW) in 2005 (05), 2006 (06), 2007 (07) and 2008 (08). *MP is different (P <0.05) from MNP; #MP is different (P ≪0.05) from NM; †MNP is different (P <0.05) from NM. Mating live weight and body condition score in 2005 have previously been reported by Kenyon et al. (2008).

In 2007, MP and NM ewes were heavier (P <0.05) than MNP ewes at mating (63.2±1.24 kg, 62.8±1.09 kg and 59.2±1.29 kg for MP, NM and MNP ewes respectively). However, no differences (P>0.10) between ewe groups were found at pre-lambing or at weaning of the lambs.

In 2008, no differences (P>0.10) between ewe groups were found on ewe live weight at mating and pre-lambing. At weaning of the lambs, MP ewes (59.3±0.57 kg) were slightly heavier (P <0.05) than NM ewes (56.8±0.84 kg).

Body condition score

Body condition score of the ewes in 2006 was not different (P>0.10) between ewe groups (Fig. 1). In 2007, NM ewes had higher (P <0.05) body condition scores than MP ewes at mating (3.0±0.09 and 2.8±0.10 for NM and MP ewes respectively) and higher (P <0.05) body condition scores than MP and MNP ewes 2 weeks pre-lambing (2.8±0.07, 2.6±0.08 and 2.6±0.08 for NM, MP and MNP ewes respectively).

In 2008, no differences (P>0.10) between ewe groups were found on body condition scores at mating or pre-lambing. At weaning of the lambs, MP ewes had higher (P <0.05) body condition scores than NM ewes (2.4±0.04 and 2.2±0.06 for MP and NM ewes respectively).

Reproductive performance

During the period 2006 to 2008, ewe group had no effect (P>0.10) on the proportion of ewes bred either during the first 17 days of breeding or ewes that became pregnant during the first 17 days of breeding (data not shown). The proportion of non-pregnant ewes at scanning was also not affected by ewe group (P>0.10) in all years (Table 1). In 2006, the proportion of MP ewes being single-bearing was less (P <0.05) than that of MNP ewes and the proportion of MP ewes being multiple-bearing was greater (P <0.05) than MNP ewes. No effect (P>0.10) of ewe group was observed for the proportion of single- and multiple-bearing ewes in either 2007 or 2008. The average number of foetuses per ewe at pregnancy scanning did not differ between the groups in each year.

Table 1  Effect of ewe group (mated and lambed [MP], mated and not lambed [MNP] or not mated [NM] as a hogget in 2004) on the proportion (%) of non-pregnant, single- and multiple-bearing ewes and overall number of foetuses per ewe presented for breeding in 2006 (3 years old), 2007 (4 years old) and 2008 (5 years old). The table shows least-square means (CI and ±SE). Different superscripts between rows within columns are significantly different (P<0.05).

		Proportion of ewes
Ewe hogget treatment	n	Non-pregnant	Single-bearing	Multiple-bearing	Foetuses per ewe^1
2006
MP	512	2.4 (1.36–4.36)	34.7 (30.4–39.2)^a	62.9 (58.3–67.2)^b	163±2.6
MNP	176	3.6 (1.64–7.86)	44.8 (37.4–52.5)^b	51.5 (43.9–59.0)^a	155±4.4
NM	231	4.4 (2.29–8.18)	38.8 (32.4–45.7)^ab	56.8 (49.9–63.4)^ab	159±3.8
2007
MP	453	0.51 (0.13–2.03)	16.2 (12.8–20.1)	83.3 (79.3–86.7)	192±2.6
MNP	166	1.35 (0.34–5.24)	18.9 (13.4–26.0)	79.7 (72.5–85.4)	192±4.3
NM	198	1.69 (0.55–5.12)	18.6 (13.6–25.1)	79.7 (73.1–85.0)	190±3.9
2008
MP	378	5.33 (3.34–8.40)	17.9 (14.0–22.5)	76.8 (71.9–81.1)	177±3.5
MNP	144	6.45 (3.26–12.37)	19.4 (13.3–27.3)	74.2 (65.8–81.1)	173±5.7
NM	170	8.11 (4.66–13.73)	15.5 (10.6–22.3)	76.4 (68.8–82.5)	177±5.3
^1Number of foetuses identified per ewe presented for breeding×100/1.

Lifetime reproductive performance 2005 to 2008

The total number of foetuses per ewe present at the beginning of the study in 2004, including ewes that were not present in all 5 years (Lifetime data 2004–2008), was greater (P <0.05) in MP than either MNP or NM ewes (Table 2). When the analysis included only those ewes still in the flock at breeding in 2008, this relationship was still present (P <0.05).

Table 2  Effect of ewe group (mated and lambed [MP], mated and not lambed [MNP] or not mated [NM] as a ewe hogget in 2004) on the lifetime reproductive performance (number of foetuses identified per ewe at pregnancy scanning) for all ewes (Lifetime 2004 to 2008) or for only those ewes that had pregnancy scanning data for each year (Lifetime ewes present 2008). The table shows least square means (±SE).

Ewe hogget treatment	n	Lifetime 2004 to 2008^1	n	Lifetime ewes present 2008^2
MP	532	6.5^b±0.11	378	8.2^b±0.08
MNP	211	5.0^a±0.18	144	7.0^a±0.13
NM	248	5.3^a±0.16	170	7.1^a±0.12
^1Lifetime 2004 to 2008: this included all ewes present at mating 2005. Those that were not present at all pregnancy scannings post-2005 were given a nominal value of 0 for the subsequent years they were not present. Hoggets not presented for breeding in 2004 (NM) or that did not become pregnant (MNP) were given a nominal value of 0 for that year.
^2Lifetime ewes present 2008: this only included ewes still present in 2008.

Lamb live weights in 2006

MP ewes gave birth to lighter (P <0.05) lambs than MNP ewes (Table 3). Singleton-born lambs were heavier (P <0.05) than twin- and triplet-born, and twin lambs were heavier than triplet-born lambs at birth. Ewe group had no effect (P>0.10) on lamb weaning weight in 2006. However, sex of the lamb and rearing rank affected weaning weight: male lambs were heavier (P <0.05) than female lambs and singleton-born and reared lambs were heavier (P <0.05) than multiple-born and reared lambs. In addition, triplet-born but twin-reared lambs were the lightest (P <0.05) at weaning compared with the other rearing ranks.

Table 3  Effect of ewe group (mated and lambed [MP], mated and not lambed [MNP] or not mated [NM] as a hogget in 2004) on sex of the lamb and birth and rearing rank on birth weight and weaning weight of lambs born to ewes in 2006. The table shows least-square means (±SE). Different superscripts between rows within columns are significantly different (P<0.05).

	n	Birth weight (kg)	n	Weaning weight (kg)
Ewe hogget treatment
MP	782	4.7±0.04^a	656	27.2±0.47
MNP	254	4.9±0.05^b	213	27.1±0.51
NM	327	4.8±0.05^ab	278	27.3±0.50
Sex of the lamb
Male	699	5.0 ±0.04^b	597	28.3±0.47^b
Female	649	4.7 ±0.04^a	547	26.1±0.48^a
Birth rank/rearing rank^1
Single/single	339	5.7±0.04^c	281	30.7±0.25^d
Twin/twin	949	4.7±0.03^b	734	26.7±0.15^b
Triplet/triplet	75	4.0±0.09^a	24	26.3±0.74^b
Twin–single	—	—	80	29.5±0.41^c
Triplet–twin	—	—	26	24.6±0.71^a
Triplet–single	—	—	2	25.3±2.52^abc
^1Twin–single, born as a twin but reared as a singleton; triplet–twin, born as a triplet but reared as a twin; triplet–single, born as a triplet but reared as a singleton.

Lamb survival to weaning in 2005 and 2006

Ewe group had no effect (P>0.10) on lamb survival in 2005 (91% [95% CI = 88–93%], 91% [86–94%] and 90% [86–93%]) or 2006 (84% [81–86%], 84% [79–88%] and 85% [81–88%]) for lambs born to MP, MNP and NM ewes respectively.

Longevity of ewes

Ewe group had no effect (P>0.10) on the proportion of ewes still present in the flock in 2006 or 2008. However, in 2007, a greater (P <0.05) proportion of the MP ewes were present compared with MNP ewes (Fig. 2).

Figure 2  Effect of ewe group (mated and lambed [MP], mated and not lambed [MNP] or not mated [NM] as a hogget in 2004) on the proportion (%) of ewes present in 2005, 2006, 2007 and 2008. *MP is different (P <0.05) from MNP.

Discussion

The aim of this study was to investigate the effect of hogget lambing on subsequent performance of ewes to the weaning of their lambs as a 5-year-old ewe. In comparison with many earlier studies, it is important to note that, in this work, ewes that had lambed as a hogget were not given preferential feeding (Kenyon et al. 2008). Preferential feeding could have limited the ability of the study to identify potential negative consequences of ewe hogget breeding. Kenyon et al. (2008) previously reported that two-tooth (2-year-old) reproductive performance was slightly poorer in MP ewe hoggets. While these data are not reported in the present paper, they were used to calculate indices of lifetime reproductive performance. The two-tooth live weight and body condition score data have also been utilised in Fig. 1.

Kenyon et al. (2008) reported that in this cohort of ewes, hogget breeding reduced two-tooth live weight and condition score. The subsequent live weight and condition scores of ewes during 2006 to 2008 did differ on some occasions between the ewe groups. However, the relative size of the differences, when apparent, was relatively small (maximum of 3 kg weight and 0.2 condition score). The results of previous studies support the negative impact of hogget breeding on two-tooth live weight (Keane 1974; Baker et al. 1981; McMillan & McDonald 1983). However, other studies found that the live weight effect was no longer apparent in later years (Cannon & Bath 1969; Dyrmundsson 1973; Baker et al. 1978, 1981). Combined, these studies suggest that hogget breeding does not have a permanent adverse effect on mature ewe live weight.

The percentage of ewes still present in the flock in 2008 as a proportion of those presented for breeding at 18 months of age (an indicator of ewe longevity) did not differ between those bred as a hogget or not. The lack of a negative effect of hogget breeding on longevity has been also previously reported to at least 6 years of age (Cannon & Bath 1969; Baker et al. 1978, 1981; Ponzoni et al. 1979). These findings do not support the commonly heard concern among farmers that breeding at a young age reduces longevity.

Although breeding as a hogget reduced two-tooth reproductive performance in 2005 (Kenyon et al. 2008), it had a positive effect on number of foetuses per ewe lifetime. This clearly indicates that any reduction in two-tooth reproductive performance is more than outweighed by the positive effect of an additional breeding as a hogget. Others have previously reported that hogget breeding did not negatively affect ewe pregnancy rates and number of lambs born per ewe as the ewe aged from 2 to 6 years (Cannon & Bath 1969; Baker et al. 1978, 1981, Ponzoni et al. 1979). It should be remembered that the extra potential benefit from a greater number of foetuses per ewe lifetime, by breeding as a hogget, will only result in a greater weight of lambs weaned per ewe if farmers ensure that hoggets are fed to reach suitable target live weights pre-breeding and in pregnancy (Schreurs et al. 2010).

Kenyon et al. (2008) found that hogget breeding had no effect on the weight of lambs weaned per ewe as a two-tooth in 2005. This lack of difference was also found in 2006. Furthermore, in both years, hogget breeding had no effect on lamb survival to weaning. Similarly, no negative effect on lamb weaning weight and apparent lamb survival has been reported previously (Cannon & Bath 1968; Baker et al. 1978; Ponzoni et al. 1979) and, indeed, Baker et al. (1981) reported that those ewes that which were bred as hoggets weaned a heavier lamb in later years, with greater apparent survival.

A second comparison that can be undertaken based on this study is between ewes that lambed as a hogget and those that were presented for breeding but did not become pregnant as a hogget. Kenyon et al. (2008) reported that, as two-tooths in 2005, those ewes that were presented for breeding but did not lamb as a hogget were heavier than those that did lamb, but their reproductive performance did not differ. Therefore, Kenyon et al. (2008) suggested they were less productive. This is supported by the work of Baker et al. (1981) who found that ewes that did not lamb as a hogget tended to wean fewer lambs as a two-tooth. Moore et al. (1983) also reported that ewes that did not become pregnant as hoggets were also less likely to get pregnant and wean a lamb as two-tooths, compared with those that had lambed as a hogget. These results might indicate that performance as a 3–5-year-old ewe might also be lower in those that were not pregnant as hoggets. However, in the present study there was no difference in mating pattern, pregnancy scanning performance per year or over the combined 4-year period (2005–2008) or for lamb weaning weight or survival between ewes that had or had not lambed as a hogget. Baker et al. (1978) also reported no difference in reproductive performance or weight of lambs weaned between ewes that did not lamb as a hogget and those that successfully lambed. However, in the present study those that did not lamb as a hogget had fewer foetuses per ewe lifetime than those that successfully bred as a hogget. Fogarty et al. (2007) reported that ewes that had successfully reared a lamb as a hogget subsequently reared more lambs to weaning in their second and third lambings than those that did not lamb as a hogget; this was considered due to higher fertility and lamb survival rates. Similarly, Baker et al. (1981) reported that ewes that did not lamb as a hogget weaned fewer lambs in the subsequent 3 years than those that did lamb as a hogget. Combined, these data suggest that farmers may wish to cull ewe hoggets that do not breed successfully. A possible management option for farmers is to present more hoggets for breeding than they require as replacements and cull those that do not achieve pregnancy.

Conclusion

When the results of the present study are combined with those of previous studies, it is apparent that hogget breeding results in greater lifetime reproductive performance without negatively affecting the longevity of the ewe. These results were observed even when two-tooth live weight and performance had been negatively affected. Therefore, hogget breeding should be used by New Zealand farmers as a way of improving the reproductive performance of their flocks.

Acknowledgements

The authors wish to acknowledge the funding provided by Massey University, Beef + Lamb NZ (formally Meat & Wool New Zealand) and the C. Alma Baker Trust and the technical help provided by the Riverside farm staff and Massey University technical staff. PR Kenyon is partly funded by the National Research Centre for Growth and Development.