The trait that I selected for was milk production because it would be a trait that could fluctuate more and give a better indication of the success of my selections. My goal for milk production was to increase it by 5000 pounds in the 10 or so years that we ran our herds through the game.
I started my selection by culling all cows that were standing open, bar none. Next, I culled all cows producing below 13,000 pounds until year 15, when my herd had improved enough to increase my selection minimum to 14,000 pounds. Heifers and heifer calves that had negative values for estimated milk production were culled next. In the years when my selective breeding had eliminated any negative estimated milk production values, I culled first from the milking cows (below 13 or 14 thousand pounds) and then the heifers from the bottom up after ranking their milk production estimates. I culled these heifers until my herd was close to 60 bred cows (the upper herd tolerance number). Heifer calves were not culled until they reached breeding age except in the years when they could be culled because they had negative milk production estimates.
I chose sires by one criterion only: the highest milk production estimate. It did not matter whether he had 2 or 2000 daughters in production as long as his estimated production was the highest. I bred all my heifers and cows to this one bull. Of course, the highest ranking bull changed from year to year and only toward the end did I have to use the bull twice, so, although I wasn’t concerned with inbreeding, I happened to be able to avoid it anyway. There were two years (14 and 17) when I decided to try using a bull from my herd. In both cases, I selected the highest bull calf and kept him until he was of breeding age and then randomly selected five lower ranking heifers to breed to him. In both cases the results were not outstanding and the bull was culled that next year. (In retrospect, I didn’t give him a chance to prove himself by selected the lowest ranking heifers and should not have culled him before his daughters were in production.)
All of the matings were done with one bull so it is not really “best to best”, it was more like “best to best; mediocre to best; not-so-good to best”. Nor could you call it random mating except in the cases where I used bulls from my own herd.
A summary of the herd’s results is attached. I had to execute one additional year to get the number of lactating cows to comply with the given acceptable parameters. By looking at the summary, we can see that I didn’t reach my goal of increasing milk production by 5,000 pounds. However, looking at the herd average estimated transmitting ability, the herd average BV, and the population average BV, estimated milk production was increased by a great deal. To me these are the numbers that are important because these are the genetic indicators of the milk producing ability. The herd average phenotype milk production may not have increased by 5,000 pounds (although it came very close) but that could be due to any number of environmental factors. This becomes especially clear when looking at years 13 through 16, where the actual pounds of milk production fluctuates up and down and yet the estimated milk production numbers continue to climb.
This project was a fun, interesting, and valuable tool in helping us to see what it is like to put the selection and mating systems that we learn about in class into use. The practical application makes it clearer what it is that farmers and animal breeders do with real animals.