Phenotypic and genetic parameters for fertility in sheep, in terms of number of lambs born and number weaned, have been estimated in a flock of medium Peppin Merinos. Repeatability of fertility traits over all ages is low, but there is an age effect, the record at 3 years of age having a higher value than those at 2 or 4 years. The regression of subsequent performance on a difference of 1 lamb at the initial lambing was higher for the difference between 1 and 2 lambs than between 0 and 1 lamb, which indicated that selection for twins is likely to raise fertility in the current flock more rapidly than selection against barrenness. The estimate of heritability for 2 years of age was negligible for each fertility trait, but the estimates for the 3-year-old record were higher, the value for lambs born being over 0.3. Mass selection for number of lambs born at 3 years of age would thus be expected to lead to appreciable genetic progress, while there would be at least some progress in number of lambs weaned. Phenotypic and genetic correlations were also estimated between fertility at different ages and 10 sheep and wool traits measured at 15–16 months of age. Phenotypically, fertility is positively correlated with body weight and negatively with skin wrinkle score. Genetically, it is suggested from the estimates that fertility is positively correlated with body weight and staple length and negatively correlated with fibre diameter, clean scoured yield, and wrinkle score. No phenotypic or genetic correlation was found between greasy or clean wool weight and either measure of fertility. The application of the findings to breeding for higher fertility is discussed. In an appendix, a maximum likelihood method of estimating heritability for all-or-none traits in half-sib data is presented.

Medium-wool Merino ewes, from groups that, as a result of selection since 1954 for this character, have a low (O group) and high (T group) incidence of multiple births, were injected with 0,375,750, or 1500 i.u. pregnant mare serum gonadotrophin (PMSG) on day 13 or 15 of the oestrous cycle (day 1 is the day of oestrus). The two groups were compared on the basis of the occurrence of oestrus, number of ovulations, ovarian follicle score, plasma progesterone level, and weight of luteal tissue. Conception rate in the two groups was compared by examining the reproductive tracts on day 20 of pregnancy, when the number and size of embryos were assessed. Injection of PMSG on days 13 and 15 gave similar results, so these data were pooled for all subsequent comparisons. Dose response lines for PMSG were constructed, the above ovarian characters being used as response types, and the curves were compared by relative potency analysis. On the basis of number of ovulations, plasma progesterone, and luteal tissue weight, PMSG was approximately three times as potent in T ewes as in O ewes. That is, for a given dose of PMSG, the T ewes produced an ovarian response three times as great as that of O ewes. At low doses of PMSG (4 1500 i.u.) the T ewes also produced more follicles in the ovary than 0 ewes. With 1500 i.u. PMSG this was reversed, since the O ewes produced many follicles that did not ovulate. The relative loss of potential embryos by day 20 was similar in the O and T ewes examined, and increased with increasing dose of PMSG. In ewes given 1500 i.u. PMSG only 20-35% of potential embryos were recovered. Embryo size on day 20 was not significantly influenced by either group or dose of PMSG. In untreated ewes the normal ovulation rate differed significantly between the O group (1.00) and T group (1.50). Ovarian data collected from untreated adult ewes of the O and T groups substantiate the differences observed in the PMSG study. It is not yet known whether the between-group differences in fecundity have arisen through changes in pituitary gonadotrophin secretion, changes in ovarian sensitivity, or both.

Ratios are frequently advocated as criteria for selection, in either animal or plant breeding; for example, production per unit of food eaten is currently receiving attention, particularly with Merino sheep. Clean wool weight per head (W) is already widely used as a selection criterion, but suggestions have been made that wool weight per pound of body weight (B) or per unit skin area (estimated as B0.6, or measured) should be used instead, as estimates of the efficiency of conversion of food into wool. This paper develops formulae for writing the heritability of any ratio, as well as its genetic correlations with its numerator and its denominator, in terms of the relationships between the numerator and the denominator. From these formulae the likely result of selecting for the ratio can be predicted, for a given range of values of the genetic and phenotypic parameters of the numerator and denominator. The formulae can be used for any ratio, provided gene x environment interactions and epistasis can be neglected. In particular, it is shown that the genetic correlation between W and W/B or W/B0.6 is consistently positive for a wide range of values of the parameters of W and B, so that selection for high wool weight will also increase efficiency, as measured by either ratio. The genetic correlation between B and W/B or W/B0.6, on the other hand, is nearly always negative, sometimes very strongly so, and selection for high values of W/B or W/B0.6 will decrease body weight. The theoretical predictions are confirmed by results from experimental selection groups, in which the observed correlated responses are all in the predicted direction, and in most cases agree well in magnitude. It is suggested that more information is required on the importance of body size itself, in relation to factors such as fertility and survival, before the value of selection on W/B or W/B0.6 can be accurately assessed.

Although work has been published showing relationships between milk production of the ewe and growth of the lamb, little has been done to investigate the effects of physical impairment or deficiency in mammary function in the ewe on the survival and growth of the lamb. This point has been studied by examining data for survival, birth weight, daily gain in weight to weaning age, and weaning weight from 621 Merino lambs born in the F. D. McMaster Field Station, C.S.I.R.O., flock from 1946 to 1949. When the performance of lambs from ewes having impaired or imperfect udder function was compared with that of lambs whose dams had normal function it was found that the lambs from ewes having defective function had a lower survival rate, a lower daily gain in weight to weaning age, and lower weight at weaning. The differences were statistically significant. Examination of the dry ewe flock in 1946 revealed that approximately 6 per cent. of all ewes aged from 1 to 7 years had damaged or imperfect udders. There was an association between age and incidence of defective udders, the incidence being highest among the oldest ewes. Culling of affected ewes from the flock was found to have very little effect in increasing the overall percentage of lambs raised to weaning age, or increasing their weaning weight. It is doubtful if culling for this fault would be beneficial in the average commercial Merino flock.

Two groups of Australian Merinos have been under selection for low (O) and high (T) reproduction rate since 1954, and a third (B) for high rate since 1959. The O and T groups were founded on Peppin ewes from one flock which had borne singles (or twins) in two successive years, mated to single- (or twin-) born rams from the same flock. The B group was founded on two quintuplet rams and ewes born in multiple births (13 in one intake, 91 in a second), from a medium-wool non-Peppin flock belonging to the Seears Brothers, of 'Booroola', Cooma, which had been under selection for multiple births for 10–15 years. In 1959, the lambing percentage of the flock (lambs born to ewes joined) was stated to be 170-180. Selection for both sexes in the early years of the experiment was based on the presence (or absence) of multiple births in the record of the dam's first three lambings, or in that of the ewe herself, again during her first three lambings. In later years this was combined with a dam's ranking coefficient based on all available records, the coefficient being based on deviations from average performance at the ages of record, and the heritability of repeated records. The sex of her litter mates was found to have no effect on a ewe's lambing performance. The litter size in which she was born had no effect on her lambing performance when she belonged to the O or B group, but twin-born ewes in the T group were sometimes at a disadvantage compared with singles-for their lambings at 2–4 years if their dams were adult, and for those at 5–7 years if their dams were 2-year-olds. Selection response was analysed by examining T – O and B – O differences in performance for ewes born in successive calendar years, the two criteria of performance being the percentage of ewes with at least one multiple birth in their first three lambings, and ranking coefficients based on the same lambings (at 2–4 years). A regression of each difference on time is an estimate of the annual rate of genetic change due to selection; no attempt has been made to relate such changes to what might be predicted from parameters. Adjustments were made for the handicap suffered by twin-born T group ewes by adding 10.0 to the percentage showing at least one multiple birth for those whose dams were adult, 3.4 to the 2-4-year-old ranking coefficient of those ex adults, and 2.6 to the 5-7-year-old ranking coefficient of those ex 2-year-old dams. No unbiased correction for age of dam was possible; instead, T – O and B – O differences were calculated separately for progeny of adults and progeny of 2-year-olds, and a weighted mean difference computed. The groups ran at Deniliquin until 1964, and at Armidale from 1965. In spite of the change in environment, and a bad drought in 1965, regression lines could be fitted to all differences except the 2-4-year-old ranking coefficient in group T. Initial responses in percentage of ewes with at least one multiple birth in their fist three lambings were 35.2 and 40.1 for T – O and B – O respectively, and the linear regression coefficients, measuring continuing response, were 2.39 ± 0.58 and 5.52 ± 1.33. The initial responses account respectively for 68 and 62% of the average superiority of the high lines during the experiment. Initial responses for ranking coefficients at 2-4 years of age were 8.2 and 21.7 for T – O and B – O respectively, corresponding approximately to 0.26 and 0 + 72 lambs born/ewe/lambing. Continuing responses fluctuated more than for percentage of ewes with multiple births; regression coefficients of 0.67 ± 0.17 were fitted for T – O from 1954–56 to 1964, and 3.6 ± 1.2 for B – O from 1961 to 1968. The former is equivalent to 0.02, and the latter to 0.11 lambs born per ewe per year. Initial responses contributed 57–76% of the average superiority of the high line (T or B) over O, depending on the regression coefficient used. For the 1968 drop, the difference in ranking coefficient at 2–4 years was equivalent to 0.38 lambs born/ewe/lambing for T – O, and 1.04 lambs born/ewe/lambing for B – O. Ranking coefficients based on lambings at 5–7 years are the result of selection on the ewes themselves, in addition to genetic gains. The T – O and B – O differences in these coefficients showed approximately the same rate of annual change as those at 2–4 years. Corrections were made to some data from T ewes in estimating response, but in practice twin ewes might give a lower performance at their early lambings. However, the actual performance of the flocks shows that any such disadvantage was not serious. In 1972, ewes aged 2–7 years in the three groups bore respectively 1.11, 1.36 and 2.10 lambs per ewe joined. Clearly, the reproduction rate has responded to selection, the rate of response being far higher in the B group than the T. Clearly, also, initial selection of the base animals made a major contribution to the superiority of the high lines, though there has also been a continuing response.

Estimates of heritabilities and of phenotypic and genetic correlations are given, based on extensive measurements on medium Peppin Merino ewes at 15–16 months of age. In general these substantiate results obtained by other workers and, in particular, confirm the high heritabilities of the traits measured. An effort has been made to try to detect possible changes in additive genetic variance for the trait under selection (clean wool weight). Estimates are obtained for data from animals at different stages of selection: (A) either unselected, or with little selection history, and (B and C) with varying amounts of selection. For stage A data the average estimated additive genetic variance was 0.31. There are problems involved in estimating from stage (B+C) data but an upper limit average value of 0.22 was obtained. Thus, although a decrease in additive genetic variance has occurred, its statistical significance is unknown and conclusions about the decrease must necessarily be tentative. In practically all cases the estimates of phenotypic and genetic correlations are of the same order of magnitude, and for the genetic correlations may be summarized as: See PDFAll other combinations of traits have negligible genetic correlations (in the range –0.20 to + 0.2).

Four numerical characteristics-follicle depth, follicle curvature, number of follicles per unit area of skin, and ratio of number of secondary to number of primary follicles-describing the size, shape, and arrangement of wool follicles have been measured in Peppin Merino sheep at 4-5 months (weaning), 15-16 months (two-tooth shearing) and at later ages ranging from 2 1/2 to 7 1/2 years. Estimates of their repeatability, heritability and phenotypic, genetic, and environmental correlations with 10 wool and body characteristics are reported. All four follicle characteristics were found to be highly inherited and sufficiently correlated with wool characteristics to be of interest to both the wool biologist and the sheep breeder. Fixed environmental effects influenced the expression of some follicle characteristics, while others, notably follicle curvature at any age, were unaffected, and therefore potentially more useful as practical selection aids. Groups of sheep selected for clean wool weight with control of (i) fibre diameter and wrinkle score, and (li) crimp frequency and wrinkle score, exhibited changes in the four follicle characteristics, which agreed with what the genetic correlation estimates would predict. The role of follicle characteristics in the biology of genetic control of wool growth is portrayed by fitting causal models invoking follicle characteristics as intermediates between the gene and the wool character. The analysis separates three independent genetic control systems, the identity of which corresponds closely to factors postulated in previous theoretical studies.

Estimates of death rates are given for four mating groups in a flock of approximately 1000 Merino ewes, for each year of age from 1½ to 10½. Three of the groups are under selection for clean wool weight and other characters, and the fourth is an unselected control. The flock is run under extensive grazing conditions, and death rates are based on ewes dead or missing between one pen-mating and the next. Between 1951 and 1957, the average death rate for ewes 1½ to 7½ years old was 2.2 per cent. per annum, while at the older ages the average was no higher than 7.3 per cent. During the drought year 1957-58 losses averaged 3.8 per cent. for ewes up to 6½ years old, then rose steeply with age to 45.6 per cent. for ewes 9½ years old. In neither period did the selected groups differ from the unselected control. Starting with the 1953 drop, all rams in the unselected control group have been retained for the study of age effects on them. Average losses to date in each age group have been less than 5 per cent.