Breeding Better Cows

by M. H. FOHRMAN

IT WOULD be possible to raise our national level of dairy production from four thousand five hundred-odd pounds of milk a year to 7,500 or more and from 185 pounds of butterfat to at least 300. The ways and means of doing it are at hand, but there is reason to believe that little measurable progress in breeding better dairy cattle has occurred in the past decade.

Let us first consider some background details. Cows in Dairy Herd Improvement Association herds showed an average increase of 319 pounds of milk and 14 pounds of butterfat between 1935 and 1945; in 1944, the 561,587 cows on test gave an average of 8,296 pounds of milk and 336 pounds of butterfat. Such figures, however, do not truly measure progress in breeding, because the averages are subject to the influence of markets, prices of milk and feeds, and labor conditions. Besides, the admirable responses to incentives offered to enlarge wartime production must be considered in the light of what would have been produced in a normal situation. We cannot assume that the increase is to be attributed entirely to genetic improvement through better breeding practices.

Much more significant are records that compare the production of daughters with that of their dams. We analyzed such records of 29,598 cows in 708 good herds and found that the dams averaged 452 pounds, of butterfat, against 451 pounds for the daughters. These were above-average animals : Their owners were employing their best judgment in selecting herd sires and were culling out the poorer cows, yet the final results were disappointing as far as improvement through breeding was Concerned.

It is true that the benefits of any advance in teaching better breeding Would come only after a long time because of the slow growth of dairy cattle and the vast number of animals that comprise the national dairy herd. Any improvement in the upper levels, however, will gradually filter down through the cattle population as better bred stock becomes available from these top herds. That, of course, throws a large responsibility for better breeding on the owners of the herds that supply bulls to be used in farm breeders' herds, so that the breeders, in turn, will be passing on better breeding stock from their own herds.

Which procedures offer the greatest assurance of this continuous progress? We believe the experience of the Bureau of Dairy Industry gives the answer. Since 1919, our various experimental breeding herds have been following a program of breeding that calls for the continuous use of sires that have already proved their transmitting ability, as determined by comparing the production records of their daughters with those of the dams of these daughters.

Our logic is sound. We know that the ability to produce milk and butterfat is inherited from both parents. When a sire's daughters demonstrate that they are better producers than their own dams, it is a safe conclusion that what they inherit from their sire is better than what comes to them from their mothers. When the increased ability is great enough to be indicative, and a large proportion of the daughters outproduce their dams, then we have the proved sire with which to work. Used in sequence, each of these sires makes a contribution toward the betterment of the germ plasm of the females in the herd. The system replaces the older idea that the only helpful guide to selection of breeding stock was the production record of the individual cow, and that, in the case of young sires, most of the emphasis was to be put on the record of the dam.

Many practical breeders have experienced the success that follows the use of an outstanding sire, but they have been slow to realize that such sires can be properly evaluated only after they have daughters in production, and that means proved-sire breeding. Objections are voiced to the use of older bulls in the herd; good proved sires are hard to find and difficult to handle, and perhaps not so fertile as younger bulls. But none of these objections is tenable if one is willing to put forth the earnest effort needed to breed better cattle.

This method of breeding may have been classed as theoretical a generation ago, but we have used it long enough at Beltsville to offer it now as a successful way to build a high level of production. Our herds have all been handled under uniform conditions, and the results can be interpreted in terms of genetic improvement. A series of 7 unrelated sires have been used in the Holstein herd, and 15 in the Jersey herd. The figures illustrate the effective way that the good inheritance has influenced the average production level of the females in the herd; they do not express the average production in the year indicated, but are a production inventory of the animals in the herd on July 1 of each of these years. No culling has been practiced to influence the averages; all females have been kept in the herd as long as they were useful. In all cases, a total has been made of the highest milk and butterfat record on a mature-equivalent basis of each cow in the herd; this total is divided by the number of cows to give the average shown.

The figures tell their own story: In both breeds, the herd cows in 1945 averaged almost 200 pounds more butterfat than those of 1926. The decline in numbers in 1936 was due to an outbreak of tuberculosis that forced the disposal of many of the older animals. All records were made on three milkings daily for a 365-day lactation period, and the feeding and handling were always uniform. Therefore, it was the good germ plasm contributed by the proved sires that effected the increase in producing ability. The figures may be compared with records of the Dairy Herd Improvement Association, given in the second table.

Outbreeding and linebreeding have been used in both Jersey and Holstein herds at Beltsville, and with bulls of satisfactory transmitting ability there has been no measurable difference in the results. It is not the system of mating that is important, but the quality of sires. Inbreeding may be helpful in concentrating the superior qualities found in good males or females, but its use may be accompanied by a loss in size, and the animal may not be able to express a good inheritance. Experience has shown that inbreeding to certain animals has worked successfully, but it must be tried in the case of any particular individual, and may result in loss of animals and time if the results are unsatisfactory. Hybridizing from inbred strains within a breed takes a long time and many animals in order to establish anything approaching purity; besides, the combination of such inbred strains offers nothing more than a return to the level of production from which the original lines were derived.

One recent development has added much to the value of the proved-sire system. It is the policy of most well organized artificial-breeding associations to use proved sires as much as possible. If the procedure is followed closely, many members of such associations ultimately will have progeny in their herds from a series of good sires.