Dairy farmers around the world are faced with an increasing amount of information and number of choices regarding their breeding programs. Semen is now available from hundreds of bulls from a variety of countries with breeding values presented on a variety of traits. However, many non-production traits are not evaluated and presented by all countries. The rate of stillbirths and dystocia (difficult calving) are examples of traits that are not always evaluated.

Although the majority of selection emphasis in dairy breeding is placed on production, many non-production traits are important in maximizing longevity and reducing losses due to mortality and illness. Unfortunately, many non-production traits are influenced more by environment and less by genetic ability thereby reducing heritabilities of these traits. As well, there is often a maternal genetic effect that complicates evaluation and selection on these traits. Furthermore, genetic variability of health and reproductive traits may contain more non-additive (non-transmittable) genetic components which further reduce effectiveness of traditional selection for these traits. Regardless, traits such as stillbirths and dystocia do constitute financial losses to the dairy farmer and should be considered in breeding and management programs.

Table 1. Normal direct heritability ranges for heifers (diagonal; cows in brackets) and genetic correlations (off-diagonal) with associated traits (Philipsson et al., 1979).

		Stillbirth	Dystocia     Birth weight    Gestation

							       length


Stillbirth	.00 - .05	0.6 - 0.8        0.4	     0.2 - 0.3

		(.00 - .02)

Dystocia			.03 - .20        0.9	        0.3

				(.00 - .08)

Birth weigth				     .10 - .40	     0.4 - 0.5

					    (.10 - .40)

Gestation length					        .50

							       (.50)

The normal range of heritabilities for stillbirths and associated traits were given by Philipsson et al. (1979) in Table 1. As shown in the table, heritability of stillbirths is very low, particularly for cows (.00 to .02), and therefore, direct selection against stillbirths would be relatively ineffective. Fortunately, genetic correlations with other traits do exist. A particularly strong correlation (0.6 to 0.8) exists between stillbirths and dystocia and a moderate correlation (0.4) exists between stillbirths and birthweight. These associated traits do present other alternatives to selecting directly against stillbirths. For example, selection for calving ease (selecting against dystocia) instead of rate of stillbirths, provides many advantages. Heritability of direct and maternal calving ease is 0.11 and 0.12, respectively (Dwyer, 1984), which is several times higher than heritability for stillbirth rate. Since rate of stillbirths and calving ease are highly correlated, it is more efficient to select indirectly for rate of stillbirths by selecting for calving ease. According to Meijering (1985):

"... it is questionable whether sire evaluation for stillbirths is worth the effort at all in the present situation, considering the extremely small sire variance. For, even if heritability was three times as high as presently estimated, indirect selection through dystocia would be as efficient as direct selection for stillbirth given the present effective progeny group sizes and genetic correlations."

In addition to more effective protection against stillbirths, selection for calving ease also reduces the associated costs of assisted calving, longer days open, and reduced milk production of the dam.

The potential also exists for stillbirths to be caused by bulls that are carriers for undesirable recessive genes, such as mulefoot or BLAD (bovine leucocyte adhesion deficiency). Fortunately, A.I. centres regularly test bulls for these diseases to ensure that farmers are aware of the potential problems associated with recessive carriers.

Several non-genetic factors have been examined for their impact on stillbirth rate and calving performance, the most important being the dam's parity, sex of calf, nutritional status of dam, and season of calving. Heifers have been found to have stillbirth rates 2 to 4 time higher than cows (Van Dieten, 1963; Grommers et al., 1965; Laster and Gregory, 1973). Sex of calf is thought to have a significant effect on calving ease, but has little effect on stillbirths (Cloppenburg, 1966; Philipsson, 1976; Hassig and Scholte, 1979). Also, heifers or cows with improper nutritional regimes may experience reduced calving performance and consequently higher stillbirth rates (Arnett et al., 1971; Lowman, 1979). Slightly higher rates of stillbirths have been reported in the summer months (Lindstrom and Vilva (1977), presumably due to lower surveillance rate when animals are in pasture.

In light of the very low heritability of stillbirths and the significant effect of many non-genetic and environmental factors, raw mean stillbirth rates associated with individual sires could be extremely misleading. In most cases, the reason some sires will appear to have more stillbirths are due to non-genetic factors or by chance. When progeny groups sizes become large (500 daughters) the probability that high numbers of stillbirths are caused by chance are reduced. Simple Chi-squared tests could be used to illustrate the effect of small sample sizes on significance of average number of stillbirths. However, systematic non- genetic factors including parity, season of calving, or breed interactions cannot be ruled out. The proper procedure to evaluate sire differences for stillbirth rate involves careful BLUP sire evaluation techniques, but with a low heritability, as in this case, very large progeny group sizes would be needed to determine any differences between sires. Unfortunately, large progeny group sizes reduce the number of bulls that can be tested and, therefore, reduces the rate of genetic progress for production traits. Furthermore, standard BLUP evaluations assume multivariate normality of data (Henderson, 1973), and with categorical variates such as dystocia and stillbirth, this assumption is not met, thereby reducing the probability that animals are correctly ranked (Gianola, 1980; Portnoy, 1982).

Because there is a positive correlation of 0.4 between calf size and stillbirth, the possibility exists that if the female being bred is smaller than average or the calf size is larger than average, the rate of stillbirths could increase. This could be the situation if the dam is a heifer as mentioned above. This could also occur if a significant difference for size exists between breeds. In many populations around the world, North-American Holsteins have been used on smaller varieties of the Friesian breed or other breeds. Because of the size difference between breeds and the effect of heterosis on calf size, initial crosses between breeds should be made carefully. For example, only cows or larger heifers should be considered during the first few generations of introducing the larger Holstein strain. Also, North-American sires are routinely evaluated for calving ease making selection against dystocia possible. Nutrition programs emphasizing rapid growth of heifers may also reduce stillbirths due to calving difficulties. Size has increased in the Canadian Holstein breed over the last 15 years, but better nutrition program for heifers were also emphasized during the same period. As a result, the stillbirth rate for heifers has fluctuated but remained on average below 7%, as shown in Figure 1. The stillbirth rate for cows has remained much more constant with an average of 2.9%.

Figure 1.

A logical course of action for countries importing the North-American Holstein would be:

• Avoid using sires rated "difficult" for calving ease.
• Only use North-American sires on cows or well-grown heifers.
• Encourage nutritional programs for heifers that emphasize growth.
• Include genetic evaluations for calving ease and stillbirth rate (if available) in the total merit index, with economic weights appropriate for the importing country, that include the cost of dystocia and stillbirth.

Table 2. The percentage of stillbirths in Canadian Holsteins by age of dam * (Miglior, 1993).

		No. Observations    No. Stillbirths      % Stillbirths

Heifers		   518,085	        35,963	            6.9%

Cows		 1,751,626	        51,184	            2.9%

All		 2,269,711	        87,147	            3.8%

References

Arnett, D.W., Holland, G.L., and Totusek, R., 1971. Some effects of obesity in beef females. J. Anim. Sci., 33: 1129-1136.
Cloppenburg, R., 1966. Geburtsverlauf bei nachkommen von Schwarzbunten bullen einer westfalischen Besamungsstation. Thesis, Georg-August University, Gottingen.
Dwyer, D.J., 1984. The effect of selection bias on calving ease evaluation of Holstein sires. Ph.D. dissertation. University of Guelph, Guelph, Ontario, Canada.
Gianola, D., 1980. A method of sire evaluation for dichotomies. J. Anim. Sci. 51: 1266-1271.
Grommers, F.J., Brands, A.F.A., and Schoenmakers, A., 1965. Mortaliteit van kalveren bij de partus van Nederlandse runderen. Tijdschr. Diergeneeskd., 90: 231-244.
Hassig, H. and Schlote, W., 1979. Untersuchungen zum Geburtsverlauf bei Fleckvieh-farsen. Zuchtungskunde, 51: 275-288.
Henderson, C.R., 1973. Sire evaluation and genetic trends. In: Proc. Anim. Breeding Genetics Symp. in Honour of J.L. Lush. Am. Soc. Anim. Sci. and Am. Dairy Sci. Assoc., Champaign, IL, pp. 10-41.
Laster, D.B. and Gregory, K.E., 1973. Factors influencing peri- and early postnatal calf mortality. J. Anim. Sci., 37: 1092-1097.
Lindstrom, U.B. and Vilva, V., 1977. Frequency of stillborn Calves and its association with production traits in Finnish cattle breeds. Z. Tierz. Zuchtgsbiol., 94: 27-43.
Lowman, B.G., 1979. Pre-calving management and feeding of the beef cow in relation to calving problems and viability of the calf. In: B. Hoffmann, I.L. Mason and J. Schmidt (Editors). Calving Problems and Early Viability of the Calf. Martinus Nijhoff, The Hague, pp. 392-407.
Meijering, A. 1984. Dystocia and stillbirth in cattle - A review of causes, relations and implications. Livest. Prod. Sci., 11: 143-177.
Meijering, A. 1985. Sire evaluation for calving traits by Best Linear Unbiased Prediction and nonlinear methodology. Z. Tierzuchtg. Zuchtgsbiol. 102: 95- 105.
Philipsson, J., 1976. Studies on calving difficulty, stillbirth and associated factors in Swedish cattle breeds. II. Effects of non-genetic factors. Acta Agric. Scand., 26: 165-174.
Philipsson, J., Foulley, J.L., Lederer, J., Liboriussen, T. and Osinga, A., 1979. Sire evaluation standards and breeding strategies for limiting dystocia and stillbirth. Report of an EEC/E.A.A.P. working group. Livest. Prod. Sci., 6: 111-127.
Portnoy, S. 1982. Maximizing the probability of correctly ordering random variables using linear predictors. J. Multivar. Anal. 12: 256-269.
Van Dieten, S.W.J., 1963. Mortaliteit van kalveren bij de partus à terme van MRIJ-runderen. Thesis, University of Utrecht. The Netherlands.