Andrew N Thompson

Improving ewe reproductive performance remains a priority for the Australian sheep industry. Here we present a protocol for veterinary practitioners to investigate cases of poor reproductive performance. The proposed methodology can be readily adapted to account for resources available and risk factors specific to the farm in question. This protocol will allow practitioners to determine the (a) timing and (b) likely causes of foetal/lamb loss to inform strategies targeted at improving reproductive performance.

Split paternity rates in multiple born lambs of syndicate-mated Merino flocks have previously not been reported, primarily due to the cost of genotyping. Pedigree data from litters born to genetically diverse syndicate mated ewes in three Merino Lifetime Productivity flocks across Australia were analysed to examine rates of split paternity, or heteropaternal superfecundation. Over three joinings at three sites, 1082 twin or triplet litters were marked, of which 577 were heteropaternal (53.3%). There was no effect of age of dam, year of birth, sire or maternal grandsire on heteropaternity rates. These high rates of heteropaternity confirm the need to genotype all progeny from multiple births in syndicate mated flocks to ensure accurate genetic evaluation. INTRODUCTION Split paternity, or heteropaternal superfecundation, occurs when females are successfully mated by multiple males during the same oestrus cycle. Decreased costs of determining paternity using DNA means that it is now feasible to obtain pedigrees from syndicate mating which may have some practical advantages over single-sire mating. Being able to syndicate mate sheep and determine parentage using DNA requires less paddocks than single sire joining and reduced labour in terms of feeding, labour for lambing rounds (preparation for lambing through identification of dams and mothering up) together with less disturbance during the lambing period as well less risk of failed matings with infertile single-sire mated rams. Very little data about heteropaternal superfecundation has been published for either sheep or cattle. In cattle, where twinning rates are traditionally very low (1-5% depending on breed), McClure et al. (2017) examined rates of heteropaternal superfecundation in Irish herds, citing data from the Irish Cattle Breeding Federation database. They reported that, with an average of 1.7% twin rates, 0.98% of these were heteropaternal. By contrast, in syndicate-mated sheep, surprisingly high levels of heteropaternity, detected by DNA technology, have recently been reported for flocks in Ireland under relatively intensive conditions (Berry et al. 2020). The only information on the occurrence or frequency of heteropaternity in multiple-born lambs in naturally mated Merinos showed even higher levels (46-59%) in lambs born in 2012-2015 as part of the Merinolink Genomic Validation Project (Martin 2016). We predicted that heteropaternity would also be high in extensively run Merino flocks in different parts of Australia with varying genotypes. We tested this in three naturally mated Merino flocks in Western Australia and New South Wales over 2-3 years. We also examined the effect of heteropaternity on total weaning weight of litters to see if there was any difference in the weaning weights of single-sired twin lambs compared to heteropaternal twins.

Selection of Merino sheep for use in breeding programs includes the combination of visual assessment and measurement of production traits. Genetic evaluation of these assessments and traits takes into account non-genetic effects to improve the accuracy of breeding value predictions. The hypothesis tested in this paper was that visual classing assessment of sheep on a traditional three-point and a novel five-point visual scoring system are heritable and both are also affected by non-genetic effects such as birth and rearing type. Using data from the first two years of classing ewes in the Merino Lifetime Productivity project at Pingelly, WA, moderate heritability estimates were observed for both scoring systems (0.24±0.08 for three-point and 0.17±0.07 for five-point). Both traits were moderately repeatable (0.31-0.36+0.03). Birth and rear type impacted visual scoring grades significantly, indicating that they should be accounted for when visually classing Merino sheep. INTRODUCTION Merino sheep breeding routinely combines objectively measured production selection and visual assessment to improve the quality and quantity of wool produced as well as improve structure and conformation. Merino sheep classing by visual selection is based on a number of subjectively assessed traits such as wool quality, quantity and conformation, assessed by professional sheep classers using different scoring systems to class animals into different categories (Brown et al. 2002; Robinson et al. 2007). When used in conjunction with estimated breeding values for production traits, visual classing can add value to selecting superior animals for replacement, with greater accuracy and efficiency than using a single method alone (Mortimer et al. 2010). Genetic evaluation of Merino production data includes accounting for any fixed effects or non-genetic factors, such as birth type (whether the animal was born as a single or multiple), rear type (whether the animal was raised as a single or multiple), the age of the dam, the date of birth of the animal and whether that animal was born to a maiden or experienced mature ewe (Hadfield and Kruuk 2007; Brown et al. 2016). These non-genetic factors influence the phenotype of the animal and can often influence how it is classed visually. For example, twin born and reared lambs are typically smaller and produce broader and less wool than single counterparts (Swan et al. 2008, Thompson et al. 2011a,b). Accurate estimates of these fixed effects need to be included when estimating breeding values, to ensure accurate estimates of genetic merit. Research has shown that visually assessed classer grades have a heritability between 0.12 and 0.2 and have favourable genetic and phenotypic correlations with liveweight, wool quality and structural traits (Mortimer et al. 2009). These subjectively measured traits are also significantly influenced by birth type, rear type and other non-genetic factors. In addition, Clarke and Thompson (2021) found that classers were influenced by subjective assessments of liveweight, clean fleece weight and fibre diameter when grading of animals. In this study non-genetic factors had a significant effect on classing outcomes such that at the first seven month old professional classing 69% of the culls were twins and only 31% were singles. Conversely 70% of the tops were singles and only 30% were twins. The current study uses an expanded data set from the Merino Lifetime

The Merino Lifetime Productivity project is evaluating the lifetime productivity of the ewe progeny of industry sires at five sites around Australia. Additional work at the Pingelly site is evaluating the productivity of the wether progeny. Liveweight, carcase traits and wool production data for wether progeny born in 2016 and 2017 have been collected. This paper presents a preliminary analysis of the value of production of the 2016 and 2017 drop wethers at yearling and adult age. The difference in value of production to 18-22 months of age between the top and bottom ranking sire groups was approximately $50 per head, excluding differences in mortality between sire groups. In future analyses the wether data will be compared to the productivity of the ewes as well as combined with ewe data to calculate flock profitability so comparisons can be drawn between sires in feed intake and whole-body energy.