Abstract:
Proc. Assoc. Advmt. Anim. Breed Genet. Voll2 FACTORS TO CONSIDER IN A MODEL FOR FERTILITY OF BEEF FEMALES IN NORTHERN AUSTRALIA C. J. O'Neill' , MXoates *and B.M. Burns' ' CSIRO Tropical Agriculture, Tropical Beef Centre, North Rockhampton, QLD, 4701 *Department of Biology, Faculty of Applied Science, University of Central Queensland, North Rockhampton, QLD, 470 1,and 3QDPI, Tropical Beef Centre, North Rockhampton, QLD, 470 1. SUMMARY A study of the fertility of beef cattle from a dry hopical environment was undertaken using data fkom the National Cattle Breeding Station, 'Belmont'. Data were collected from 1974 to 1990 from Hereford-Shorthorns @IS), interbred Brahman x HS (BX) and low (BL) and high (BH) grade B&mans. The dependent variable, pregnancy success and the fxed effects of environmental category, genotype and the combined effects of cow age and lactational status (ALS) were analysed using a least-squares analysis of variance. Environmental categories of good, average and poor years were based on pregnancy rate. Pregnancy rates of 3 yr old maidens and 5+ yr old non-lactating cows were consistently high and similar across all genotypes and environmental categories. Lowest pregnancy rates occurred in the lactating cows but in particular, lactating BH cows. Factors that need to be considered in constructing genetic prediction models of fertility in northern Australia are discussed. Keywords: Female fertility, Bos indicus, cow age, lactational status. INTRODUCTION Compared to southern Australia, the environment in northern Australia is characterised by relatively high and more variable levels of stresses. Brahman and Brahman derived genotypes predominate in this region because of their high levels of resistance to these stresses (Frisch 1987). Compared to Bos tuwws these genotypes possess relatively low levels of production potential, especially fertility (Chenoweth 1994). The combination of environment and genotype results in sub-optimal reproductive rates in this region. One option that may be available to northern beef producers is BREEDPLAN's estimated breeding values for the number of days to calving to improve a herd's fertility (Schneeberger et al. 1991). However, world wide, genetic evaluation techniques were generally derived in temperate environments and based on Bos taurus genotypes and do not consider either extreme environmental variability of the tropics, or the proportion of Bos indicus or the effects of lactation. Genetic prediction models for female fertility in the tropics should be modified in order to better reflect northern environment and genotypes. The present study aimed to highlight factors to consider in such models. 466 Proc. Assoc. Advmt. Anim. Breed. Genet. Voll2 MATERIALS AND METHODS Data were collected from 1974 to 1990 at the National Cattle Breeding Station, 'Belmont', located 26 km north of Rockhampton, Queensland, Australia. Harvey's PC version of Mixed Model Least Squares and Maximum Likelihood, (Harvey 1987) was used to estimate fixed effects and fust order interactions. Pregnancy success was chosen as the index for fertility. Environmental category based on pregnancy rate was used as a fmed effect. Years where pregnancy rates were 2 78% were defined as good years, rates between 78% and 68% as average years and rates 5 68% were defmed as poor years. The number of years in each category were approximately equal. Respective genotypes and their number of records were; Hereford-Shorthorns (HS), 1682; interbred Brahman x HS (BX), 1750; and low (BL = >5/8<15/16), 402 and high grade Brahman (BH = 215/16),766. Age of cow and lactational status, at the time of joining, were combined into a single fixed effect (ALS). For convenience the following classes of ALS are presented; 3 yr maiden, 4 yr lactating, 5 + yr lactating, 4 yr non-lactating and 5 + yr non-lactating. A detailed description of the environment, dataset and analysis can be found in 0'Neill(1995). RESULTS significantly (P<O.OOl) affected overall mean pregnancy rate (HS 74.5*1.3%, BX 69.0*1.2%, BL 74.&3.1%, BH 58.8*1.8%). ALS also had a highly significant effect on pregnancy rate (P<O.OOl). Highest values were found in 3 yr old maidens (81.5*1.6O) and 5+ yr old nonlactating cows (80.7i1.9%). Within lactating cows there was a significant (P<O.Ol) range in pregnancy rate corn 48&2.3% for 4 yr old cows to 69.&1.9% for 7 to 9 yr old cows. Genotype a Cza a 4 Yr 4 yr Lactating EZl 5+ 5+ yr 3 yr Maiden yr Lactating Non-lactating m Non-lactating 6 c F a' w -w 80 70 60 50 40 30 20 Environmental category category_ Figure 1. interaction of ALS by environmental The response of the different classes of ALS to the different environmental categories (Figure 1) and the response of the different genotypes to the different classes of ALS (Figure 2) were highly significant (P<O.OOl). Three yr old maidens and 5+ yr old non-lactating cows achieved relatively hia and similar pregnancy rates in all three environmental categories (Figure 1). Lactating Classes 467 Proc. Assoc. Advmt. Anim. Breed. Genet, Voll2 experienced the greatest decline in pregnancy in the poor years. The 4 yr old lactating cows had significantly lower pregnancy rates than both maidens and 5+ yr old non-lactating cows in good (P<O.O5), average (P10.01) and poor (P<O.OOl)years. Pregnancy rates of 3 yr old maidens were consistently high and similar across all genotypes (Figure 2). Pregnancy rates of the two nonlactating classes of ALS were also similar to each other and consistently high for all genotypes except BH. The pregnancy rates of 4 and 5+ yr old lactating HS cows were similar, unlike the corresponding pregnancy rates in the other three genotypes. For lactating BH cows, the pregnancy rate was the lowest and the difference in pregnancy rate between 4 and 5+ yr old cows was greatest. E [za 4 yr Lactating m EZSj 5+ 5+ yr 3 yr yr Maiden Lactating 100 ^^ r m 4 Yr Nan-lactating Non-lactating BL Genotype Figure 2. Interaction of ALS by genotype. DISCUSSION A number of researchers are concerned that aspects of the new animal breeding technologies do not adequately reflect the genetics of beef production in the tropics (eg. Notter and Hohenboken 1990; Davis 1993). Therefore, the following discussion deals with components of models that the authors felt should be considered for female fertility in the tropics. Dependent variable. Large yearly variations in pregnancy rate on 'Belmont', and other locations in northern Australia ( Holroyd et al. 1990), may render days to calving, as the target trait, inappropriate for genetic prediction models for tropical environments. ln the analysis of days to calving, nonpregnant cows are assigned a value for days to calving (Meyer et al. 1990). The very low rates of pregnancy in some years, especially for lactating cows (Figure l), ensure that for any estimation of breeding value for days to calving a large proportion of cows (i.e. the non-pregnant cows) would be allocated this value. Such an allocation would be expected to reduce the normality of the distribution of the variable. The same conclusion has been reached by Ponzoni and Gifford (1994). 468 Proc. Assoc. Advmt. Anim. Breed. Genet. Vol12 category. Grouping years into good and poor environmental categories for a semiarid environment, DeNise and Torabi (1989) found that genetic parameters for growth change in response to the level of environmental stress. The differential response of ALS in the different categories (Figure 1) may mean that the genetic parameters for fertility in the good years are different from the genetic parameters in the poor years. Environmental Of the numerous studies of the fertility of `B&mans only a few studies (eg. McCarter et 1991) considered the influence of Brahman content on fertility. In studies of liveweight data of other breeds the proportion of purebreeding was found to be a significant effect on growth (eg. Graser and Hammond 1985). Evidence from these studies and from Figure 2 suggests that not accounting for various proportions of Brahman breeding would be ignoring important sources of variation. Genotype. al. Age and lactational status of cow. In the dry tropical environment of northern Australia, both age and lactational status of the cow are factors that affect fertility (Figures 1 and 2). The significantly higher reproductive rates of both 3 yr old maidens and non-lactating cows, over lactating cows, confirm previous reports from tropical environments (Frisch et al. 1987). Meyer (1992) has shown that ignoring maternal effects can significantly inflate the estimate of h' for preweaning growth traits. Ignoring significant ALS effects on reproductive traits could also distort the estimate of h2 for these traits. This study indicates that for a of target trait needs careti category, proportion of Bos factors on genetic parameters model of female fertility to be suitable for northern Austalia the choice consideration and that fixed effects should include environmental indicus and both age and lactational effects. The influence of these warrants breeders to instigate kther investigations into this issue. REFERENCES Chenoweth, P.J. (1994) Aust. VetJ 71:422. Davis, G.P. (1993) AustJAgricRes. 44:179. DeNise, R.S.K. and Torabi, M. (1989) XAnimSci. 67:2619. Frisch, J.E. (1987) J.Agric.Sci.Camb. 109:213. Frisch, J.E., Munro, R.K. and OWeill, C.J. (1987) Anim.Reprod.Sci. 15:l. Graser, H-U. and Hammond, K. (1985) Aust.J.Agric.Res. 35527. Harvey, W.R. (1987) Unpublished statistical manuscript. Holroyd, R.G., James, T.A., Doogan, V.J., Fordyce, G., Tyler, R. and O'Rourke, P.K. (1990) Aust.JExp.Agric. 30:727. McCarter, M.N., Buchanan, D.S. and Frahm, R.R. (1991) JAnim.Sci. 69:2754. Meyer, K. (1992) Livest.Prod.Sci. 31:179. Meyer, K., Hammond, K., Pamell, P.F., Mackinnon, M.J. and Sivarajasigam (1990) Livest. Prod. Sci. 25: 15. Notter, D.R. and Hohenboken, W.D. (1990) Proc. 4th. Wld.Cong. Genet.App. Livest. Prod. XV:347. O'Neill , C.J. (1995) M.App.Sc. Thesis, Central Queensland University. Ponzoni, R.W. and Gifford, D.R. (1994) J.Anim.Breed.Genet. 111:52. Schneeberger, M., Tier, B. and Hammond, K. (1991) Proc.Aust.Assoc.Anim.Breed. Genet. 9:194. 469