Selected articles and excerpts from research papers about Finnsheep...
Here is an example of study
Multi-line selection in a Finnsheep nucleus
Marja-Leena Puntila & Anne Nylander (Institute of Animal Production Agricultural Research Centre 31600 Jokioinen
Nucleus breeding in a 250 ewe Finnsheep flock was started in 1986, and was divided into three selection lines: meat, wool and fleece production. The meat line had almost twice the numbers of the other ones. The aim was to find appropriate measurements and evaluation methods required by the selection objectives of the three lines. The main objectives in the meat line were growth rate and scanned meat qualities; in the other lines, there were specific wool traits, with less emphasis on growth rate. Covariance components were estimated by an animal model, using the AI-REML algorithm in the DMU package. Heritabilities ranged from 0.32 to 0.55 for growth rates, and around 0.50 for muscle measurements. Estimates for wool and fur traits ranged from 0.13 to 0.52 and 0.06 to 0.38, respectively. Litter size remained constant through the years, averaging 2.72. Genetic trend for lamb birth weight was slightly upwards. The genetic change from 1986 to 1996 for 120-day live weight of all lambs in the flock was 1.2 kg, with a clear difference of 1.4 kg between the meat and the two other lines. Selection improved efficiency of growth, without changing the mature live weight of the Finnsheep ewe. A moderate positive genetic correlation of 0.29±0.04 was found between the weight of lambswool and the lamb live weight at shearing. A small genetic improvement of 0.2 kg in ewe wool weight was found. The results support the previous findings that Finnsheep has some special fleece characteristics, which can be improved by efficient selection.
Finnsheep are widely known for their high prolificacy. However, less is known about other breed characteristics. Meat production has been the main breeding objective in
Materials and methods
Description of the nucleus flock
A base population from the Pelso breeding flock, with a few outstanding ewes purchased at the beginning, was divided into three selection lines: meat, wool and fleece. The lines were kept separated. The number of white ewes over the years was about 250. The meat line was nearly twice the number of the others. The breeding work was concentrated on white Finnsheep. The same unit also functioned as a gene bank for coloured Finnsheep. Selection objectives for the meat line were growth rate, meat qualities and carcass traits; weight and quality of fleece for the wool line and fleece characteristics for the fleece line. In both the wool and fleece lines, a moderate growth capacity was presumed. More detailed information about the selection objectives has been reported by Puntila, Mäki-Tanila and Nylander (1990), although selection criteria have been somewhat revised since. Each year, some 20 to 30 per cent of the best ewe lambs were kept as replacement ewes. Only young rams with promising results were purchased from the national performance test; therefore, the nucleus flock was kept closed.Unfortunately, there were difficulties in finding outstanding individuals for wool and fleece lines. Six to eight rams were used yearly in the meat line and only three to four in other lines. Most of the rams were used for more than one year. Some rams were used later in other flocks. Inbreeding was avoided in the mating scheme.
Data recording and evaluation procedure
A total of 2778 matings was carried out and 2534 ewes were lambed. Lambing occurred in February and March. The overall number of lambs born was 6870, averaging 625 per year. Lambs were weaned at two months of age, except for a few during the first years, when the ewe lamb followed the dams until the 120 day-weighing. To increase profitability of the flock and growth rate of lambs from multiple births, a systematic, partially artificial rearing system of lambs was introduced in 1994. In this system, ewes had only one or two lambs (uniform in birth weight) to rear. All the excess ones were removed and transferred to a fully automated milk replacer feeding unit. The weaker lambs had been transferred earlier for artificial rearing. The lambs were weighed at birth (BW) (12-hours and 3-day weights), 6 weeks of age (42W) and at weaning (two months of age). Post-weaning live weights were recorded before turned out to pasture, at 120-days of age (120W), usually during the live animal evaluation and before slaughter. Ultrasonic measurements were taken at 120-days and, in some cases, before slaughter. This treatment was for the meat line lambs. Fleece characteristics were assessed on the lambs at about six months of age, soon after shearing and wool weighing. Fleece data included also fibre diameter measurements (n=101 lambs), based on the airflow method. Fleece traits, including neck skin thickness, were judged a month earlier. Skins after processing were also studied. Ewes were shorn twice; before lambing in the winter and in the autumn when they were housed. They had been weighed before going to pasture and at the beginning of indoor feeding.
The performances of lambs, in different lines, were first analysed by the least squares analysis (SAS GLM procedure), to evaluate non-genetic factors. The fixed effects included birth year, sex, type of birth and rearing, age of the dam and age of the lamb. For wool and fleece traits, age at shearing, evaluation and live weight of lamb were used as covariates. For maternal genetic influence, variance components for direct and maternal effects, and covariance components between effects were estimated by REML procedures with the single trait animal model. The DMU package, with AI-REML algorithm (Jensen & Madsen, 1994), was used. Estimates of genetic trends for different traits were based on the mean predicted breeding values.
Results and discussion
The overall fertility in the flock was estimated at 90%. Litter size at birth for yearling ewes was 1.60 (s.e. 0.06) and 2.68 (s.e. 0.07) for older ewes. The least squares means for prolificacy and litter weight at birth and 42-day of age, at the first lambing and later lambings, are presented by lines in Table 1.
Means within a column and within a class not followed by the same letter differ significantly at the 0.05 level. At first lambing, there was no difference between lines for litter size at birth and 42-days of age. The older ewes of the fleece line seemed to be slightly more prolific. Total litter weights of older ewes were quite uniform between lines, whilst non-selected gene bank coloured ewes produced lighter litters, as was expected. Least-square means for lamb weights atbirth, 42-day and 120-day of age, by lines, are shown in Table 2.
Birth weights were not significantly different between the meat, wool and fleece lines; only coloured animals, as a reference group, differed significantly, not only for birth weights but also for later live weights. At six week of age, the lambs in the meat and wool lines had kept the same growth rate but after weaning, the lambs in the meat line were slightly significantly heavier than the lambs in the wool line. It should be pointed out that the considerable improvement in the live weights at 6 weeks and 120-days of age was achieved after introducing the artificial rearing system for lambs from litters of two and more lambs per ewe. Good growth performance in the wool line resulted largely from the high genetic merit of a few sires.
Heritabilities estimates for growth rate from birth to 120-days of age, ranged from 0.55 to 0.32; for ultrasonic muscle area and depth measurements, they were around 0.50. For maternal genetic influence, variance components, for direct and maternal effects, were also estimated. Heritabilities for direct (h2) and the direct-maternal genetic correlation (rdm) for lamb live weights from birth to 120-days of age are given in Table 3.
Direct and total heritabilities of lamb live weight increased with the age until slaughtering, while maternal heritability declined. These findings are in agreement with Swedish studies (Näsholm and Danell, 1996), where it was postulated that, if maternal effects exist but are not considered, estimated additive genetic variance will include, at least, part of the maternal variance. It is then expected that estimates of direct heritability are lower when maternal effects are included, if the direct-maternal genetic correlation is positive. In this study, maternal heritability estimates for birth weight were lower than the estimates presented by Maria at al (1993) for Romanov sheep, Burfening and Kress (1993) and Näsholm and Danell (1996), but consistent with the estimates of Tosh and Kemp (1994), in prolific Romanov sheep. Direct-maternal correlations were positive and seemed to increase with age of lamb. Näsholm (1994) also assumed these to be positive. Contrary assumptions have been presented by Robison (1981), Maria et al (1993) and Tosh and Kemp (1994). Heritabilities from bivariate analyses were quite similar to those from single trait analyses. The positive, rather high, maternal correlations between BW and 42W indicate a high maternal effect on lamb growth. Therefore, maternal genetic effect should be included in the genetic model, which has not yet
been taken into account in the official Finnish Sheep Recording Scheme.
Coefficient of variation was largest in BW (24%), decreasing with age (at 120-day, 15%). Variation for meat qualities in
Finnsheep was remarkably low, the coefficient of variation for ultrasonic measurement ranging from 0.10 to 0.17. Genetic
correlation between 120W and the eye muscle area was 0.33 (s.e. 0.10), indicating that there is a certain potential to
simultaneously improve growth rate and meat qualities.
The heritability estimate of 0.44 for the mature ewe live weight (autumn weight) is consistent with the literature : estimates varying between 0.30 and 0.50 (Stobart at al, 1986; Woolaston, 1986; Näsholm, 1994).
Means, standard deviations, heritability estimates, genetic and phenotypic correlations for wool weight and quality traits are shown in Table 4.
Heritability estimates for lamb fleece weight, staple length and wool grade, were moderately high (0.35-0.50), but generally lower for evenness, lustre and density. The moderately high heritability estimate for wool grade resulted from the relatively objective method of assessing fineness, based on crimp frequency. The heritability estimate of 0.41 (s.e. 0.06) for fleece weight also in good agreement with the estimates of Fogarty (1995), for wool and dual purpose breeds and with those of Notter and Hough (1997) in Targhee sheep. The mean heritability for the annual fleece weight of adult ewes was 0.37. The genetic correlation of lamb fleece weight with live weight at shearing was 0.29 (s.e. 0.04) (phenotypic correlation 0.37), which means that there is no antagonism in selecting simultaneously for both fleece and live weight. This estimate in Finnsheep is somewhat smaller than in Targhee sheep (Notter and Hough, 1997), and that stated in the review of Fogarty (1995). Fleece weight had quite a favourable genetic evaluation (Mortimer and Atkins, 1994). The mean fibre diameter for lamb fleece was 23.4 microns (CV 7.4%). This finding is in good agreement with a Finnish study from progeny testing of 11 selected Finnsheep sires by Haykal (1981). In the current study, fleece of ewes (n=53) had a fibre diameter ranging from 25.4 to 34.0 microns (CV 6.6%). The age of ewe tended to have an effect on fibre diameter.
Subjectively assessed fleece characteristics showed low heritability estimates, except for the size of the curl (Table 5). The genetic correlations between traits varied from 0.23 to 0.66. Phenotypic correlations were somewhat less, between 0.21 and 0.35. The objective measurement, thickness of skin (neck) with a h2 of 29±0.05, is a valuable tool for selection of light skins.
A ten-year breeding study allowed the estimation of genetic trends. BW remained unchanged between the lines. Higher genetic improvement was associated with increased lamb live weight at 42W and 120W. The rate of genetic change in the meat line was not much larger than in the wool line. This can be explained by different factors. Firstly, the number of ewes in the wool line was rather small and the same sires were used over several years. Secondly, in the meat line, the main selection criteria, within the last few years, had been meat qualities, based on BLUP-values for the ultrasonic measurements, instead of growth rate. Estimated genetic responses for 120W, within the past seven years, have been 0.9 kg in the nucleus flock, compared with 0.6 kg in the National Sheep Recording Scheme. Annual genetic response for 120W in the meat line was approximately 0.14 kg. Shrestha et al (1996) in Canada, reported, over 20 years of selection for lamb live weights, the following genetic improvements in Finnsheep, at birth, 21, 70 and 90 day of age; 0.02, 0.02, 0.03 and 0.05 respectively. In
This study provided evidence that there is genetic variability in Finnsheep for the main traits used in the multi-line selection. The magnitude of the heritabilities and genetic correlations demonstrate that the applied selection criteria in the lines have resulted in an improvement for most of the traits. Genetic parameters indicated that simultaneous improvement of growth and fleece characteristics can be achieved. Prolificacy remained stable across the years. Genetic improvement for lamb live weight increased with age, without changing birth weight and mature ewe live weight. The trends for lamb live weight showed a slightly larger response in the meat line, as expected. The results from live animals and processed skins support previous findings that Finnsheep has some desirable fleece characteristics, which can be improved by intensive selection. The main selection objective for Finnsheep will be meat qualities in the near future. Annual response for meat qualities has been estimated to be 3 to 5%, supposing the possible high selection intensity in Finnsheep. A 3-year field study has already started for development of a recording scheme and a breeding programme for meat traits.
| & Della Jones,
Gippfinn Finnsheep Stud
RMB 4518, Morwell 3840