Improvement of food animal species through genetic selection began long ago, but little genetic progress was made in dairy cattle until four major developments occurred in the 20th century. First was the development of a national milk recording system, known as the Dairy Herd Improvement (DHI) program, that provided data on which to base selection decisions. Second was the introduction of artificial insemination (AI) via fresh and, later, frozen semen, because this allowed the widespread use of superior sires across great geographical distances. Third was the development of a genetic evaluation system based on herdmate comparisons, because this provided accurate predicted transmitting abilities (PTA) that could be used to identify superior animals within or across herds. Fourth was embryo transfer (ET), which greatly increased the impact of superior females by enhancing the female reproductive rate. As a result of these developments, the Holstein cow changed dramatically over the past 100 years, as shown below.
Today’s dairy cow is not only more productive, but she is also much taller, with more angularity or dairy character (at the expense of some strength) and greatly improved udder conformation and foot and leg structure.
Genetic selection initially focused on high milk volume, but over time the emphasis shifted toward fat yield and, more recently, protein yield. The relative values of milk, fat, and protein differ between countries, depending on the proportion of milk that is used for cheese production and the presence or absence of quotas. The trend in economic weights for the production components of the United States Department of Agriculture (USDA) selection indices is shown below.
|Traits included||USDA economic index (and year introduced)|
|Somatic cell score||...||...||...||-6%||-9%||-9%||-8%|
|Daughter pregnancy rate||...||...||...||...||...||7%||7%|
As shown above, milk and fat yield were given equal emphasis 35 years ago, but in the mid-1970s genetic evaluations for protein yield became available. Protein yield quickly replaced milk as the primary trait for genetic selection. However, it is important to note that a very small weight on milk yield (or even a zero or slightly negative weight) does not imply that milk yield will remain constant because milk volume is highly correlated with fat and protein yield. Protein yield remained the primary selection goal until recently when fat yield regained some importance and when productive life and other functional traits began to receive significant attention. At present, 48% of the selection goal focuses on increasing income through improvement of milk, fat, and protein yield.
As shown in the table above, 52% of the emphasis in the USDA Net Merit index is now focused on reducing expenses through improvement of functional traits, such as measures of longevity, udder health, mobility, fertility, and calving ability. The following photo of “Granny,” a famous Wisconsin cow, indicates the functions she must perform each day to make a positive economic contribution to the dairy farm.
If a cow can perform all of these functions effectively, she can produce amazing results. For example, the cow shown above produced more than 195,000 kg of milk in roughly 4,500 days in milk during her lifetime.
Selection for improved health and longevity began in the early 1980s with the advent of linear type appraisal programs. Over time, the focus of selection for improved conformation shifted from overall classification score, which is indicative of a cow’s ability to succeed in the show ring, to improvement of udder traits and foot and leg traits, which are indicative of a cow’s ability to survive in a commercial dairy herd. The primary means of improving these traits is through selection for udder composite (UDC) and feet and legs composite (FLC), which represent weighted averages of the PTAs for udder traits and foot and leg traits, respectively. In the USDA Net Merit index, UDC and FLC receive 6% and 3% of the total economic weight, respectively. Research shows that a one point increase in UDC provides 20 additional days of longevity, whereas a one point increase in FLC provides 10 additional days of longevity. Selection for linear type traits, particularly udder traits, is practiced in nearly every leading dairy country, and it has led to dramatic changes in the appearance of the Holstein cow. Body size composite (BDC) was added to the USDA Net Merit index in 2000, but with a negative weight, such that feed costs for maintenance could be reduced and such that the trend toward increasing frame size over time could be reversed. Currently, BDC receives 3% of the total economic weight in the USDA Net Merit index.
In the early 1990s, scientists and practical breeders recognized that a cow’s physical appearance did not fully explain her ability to survive and prosper in a commercial dairy herd environment. In fact, many of the leading causes of involuntary culling (culling for reasons other than low milk production) were only weakly correlated with the physical appearance of the cow. The following graph shows the main reasons for culling in U.S. dairy herds, according to the National Animal Health Monitoring Service.
As shown above, reproduction, mastitis, and lameness account for about 70% of all culling in U.S. dairy herds. Only 20% of the cows are removed due to low milk production, and this percentage has decreased over time. For these reasons, scientists began to develop and implement genetic selection programs for direct measures of animal health and fertility in the early to mid-1990s.
The first trait considered was length of productive life (PL), which was evaluated by USDA in 1994 and is a measure of the number of months from first calving until death or culling.
As shown above, tremendous genetic variation exists in productive life. Under equivalent management, an average daughter of 11HO8046 Ramos will remain in the herd 13.6 months longer than an average daughter of 6HO921 Token. Some challenges exist in selection for improved productive life, mainly that reliability of genetic evaluations is low until a bull’s daughters have had a chance to finish multiple lactations, but it is relatively easy for a dairy farmer to improve the longevity of his or her herd by two or three months per generation through genetic selection. Productive life currently receives 20% of the total economic weight in the USDA Net Merit index.
Selection against clinical mastitis occurs in the Nordic countries, but the rest of the world uses information on somatic cell count (SCC) or somatic cell score (SCS) instead. Participation in SCC recording programs is excellent because milk processors pay incentives for high-quality milk. In the United States, genetic evaluations for somatic cell score also became available in 1994. Significant genetic variation exists between sire families in somatic cell score as well. Although not shown, the best Holstein bull at present is 29HO9330 Blue Ribbon, whose first lactation daughters have average SCS of 2.45, whereas first lactation daughters of the worst Holstein bull, 36HO248 Win Tex, have average SCS of 3.49. Because SCC doubles with every one point increase in SCS, there is a two-fold difference in SCC between daughters of these two sires under similar management conditions. Furthermore, a one point increase in SCS corresponds to an increase of approximately 13% in the incidence rate of clinical mastitis.
The next major development in U.S. genetic improvement programs was the implementation of USDA sire evaluations for daughter pregnancy rate (DPR) in 2003. Daughter pregnancy rate is a measure of female fertility, namely a cow’s ability to begin cycling in a timely manner after calving and to conceive and maintain a pregnancy when inseminated. This trait is measured as days open (DO), but it is converted to 21-day pregnancy rate (i.e., the proportion of non-pregnant cows that become pregnant during a 21-day period) for the purpose of genetic evaluation. A 1% increase in DPR corresponds to a decrease of 4 DO.
Just as for PL and SCS, huge genetic variation between sire families exists for daughter pregnancy rate as well. Under similar management, daughters of 11H5086 Blastoff will have a 7.1% higher 21-day pregnancy rate than daughters of 7H6965 Eagle, and Blastoff daughters will have 28 fewer days open per lactation. Although management issues should be considered first when addressing reproductive problems on a commercial dairy farm, it is relatively easy to achieve an improvement of 1% in 21-day pregnancy rate per generation through genetic selection. Daughter pregnancy rate now receives 7% of the emphasis in the USDA Net Merit index.
Calving time remains a source of frustration and economic loss for many owners of Holstein cattle. Problems with dystocia (calving difficulty) and stillbirths due to large calf size and poor maternal calving ability lead to increased labor and veterinary costs, as well as increased calf and cow losses and impaired production and fertility in the subsequent lactation. Genetic evaluations for direct calving ease (the service sire’s propensity to sire large calves) have been available for more than two decades, but little genetic progress has been made for two reasons. The first reason is that calving ease PTAs have been used largely as a corrective mating tool – bulls that sire small calves are mated to yearling heifers, whereas bulls that sire large calves are mated to lactating cows. Although this is a reasonable management practice by the farmer, it is problematic from a breed improvement point of view because the genes that cause large calf size remain at high frequency in the breed. A better strategy would be to simply avoid using service sires that cause dystocia, as many of the beef breeds have done. The second reason is that large calves tend to become large heifers that win prizes at fairs and cattle shows, so many Holstein breeders have intentionally used service sires with high dystocia rates as a means of creating desirable show calves.
Through developments by scientists over the past four years, dairy farmers in the United States now have access to genetic information for four calving traits: service sire calving ease (SCE), daughter calving ease (DCE), service sire stillbirth rate (SSB), and daughter stillbirth rate (DSB). The traits DCE and DSB are considered as “direct,” which means that they measure the effect of the service sire (sire of the calf), whereas SCE and SSB are considered as “maternal,” which means that they measure the effect of the sire of the cow. Of these four traits, SCE and DSB should receive the most emphasis, as they have the highest heritability and the greatest economic value.
As shown above, large genetic differences exist in service sire calving ease. When yearling heifers are mated to 7HO6417 O-Man, only 3% will have significant calving problems. On the other hand, 18% of heifers mated to 28HO583 Leduc will have severe calving problems. Although not shown, the best bull for daughter calving ease is 14HO2736 Ito; only 4% of his daughters have dystocia when they grow up and have calves of their own. On the other hand, the worst is 6HO2002 Flip (not shown), as 12% of his daughters have calving problems.
As opposed to SCE and DCE, which are expressed as the percentage of difficult births in first-calf heifers, SSB and DSB reflect the percentage of stillborn calves from both first-calf heifers and older cows. The top and bottom bulls for DSB are shown above. When daughters of 7H5687 Forbidden grow up, only 4.5% of their calves will be stillborn. On the other hand, 15.4% of calves from daughters of 7H5710 Dane will be stillborn. Although results for SSB are not shown, only 5.4% of calves sired by 1HO5306 Bond will be stillborn, when he is used as a service sire, as opposed to 11.1% of calves sired by 1H6149 Garter. Collectively, SCE, DCE, SSB, and DSB receive 5% of the total economic weight in the USDA Net Merit index.
Genetic selection programs have become extremely efficient due to implementation of national data collection systems, development of herdmate comparison methods for genetic evaluation, widespread use of AI for producing replacement heifers, and heavy use of ET for producing young bulls. Rapid progress has been made in milk, fat, and protein yield, as well as udder conformation and overall type. Progress has been negligible for calving traits, and female fertility has declined. During the past 10 to 15 years, genetic evaluations have been developed for productive life, somatic cell score, daughter pregnancy rate, direct and maternal calving ease, and direct and maternal stillbirth rate. Tremendous variation exists between sire families for these “functional” traits. At present, the USDA Net Merit index places 48% of the emphasis on increasing income, whereas 52% of the emphasis is placed on reducing expenses. Changes in the national selection index quickly lead to changes in the sires used for AI contract matings and ET matings of elite females, which in turn lead to changes in the type of AI bulls available to dairy farmers for producing the next generation of replacement heifers. As such, dairy cattle breeding programs can change direction very rapidly, and producers in the United States and other countries are already starting to reap the benefits of the selection programs for improved health, fertility, longevity, and calving ability that are discussed in this paper.
Kent A. Weigel
Department of Dairy Science, University of Wisconsin
Genetic Programs Administrator
National Association of Animal Breeders