91. Selection for Low Residual Feed Intake Is Only Justified in Long-Fed Cattle.

Abstract Over 30 years ago I modelled crossbred sheep and cattle growth using a modified Gompertz curve. The cattle trial evaluated GxE by raising the cattle on Low, Medium or High quality pasture. On High quality pasture there was large heterosis for mature weight (21%) and not for rate of maturity, the same result as for sheep. However, on Medium quality pasture there was less heterosis (13%) and negligible (1%) on Low quality pasture. I hypothesised that the effect of heterosis must be due to increased appetite rather than improved efficiency per se as heterosis is expressed only when that greater appetite was satisfied. I returned to the sheep data set and, as expected, heterosis effects were greater in lambs raised as singles than twins. At the same time Hutton Oddy was working on lines selected for either High or Low weight. He found very few differences in partial efficiencies of metabolism but came to the grand conclusion that “Big lambs suck”, i.e. the differences were driven by differences in appetite. During the 1990’s Peter Parnell and Robert Herd developed residual feed intake selection lines following the work of Robert Koch. The cattle RFI lines demonstrated that post-weaning RFI is heritable and differences exist in mature cows. However, most of the difference between the lines can be attributed to differences in fat so productivity and profitability differences very much depended on stocking rate and season conditions. My review of multiple projects also demonstrated that variation in RFI is associated with variation in appetite rather than maintenance efficiency. This has led to a complete paradigm shift from being a believer in selection for improved efficiency to now focus on output. Indeed, our work on maternal productivity has influenced us to now believe that selection for greater appetite is actually better as when feed is plentiful, it is also cheap and so having animals that can utilise cheap feed by gaining condition and setting themselves up to be more resilient for times of feed shortage is actually preferable to selecting those with low RFI. In feedlot animals, feed is expensive, cost of gain is a primary driver and it is disruptive to bring animals into yards to weigh and take a relatively inaccurate measure of feed intake. Recently I modelled genetic variation in a long-fed Wagyu program and demonstrated that variation in feed intake that could not be accounted for by variation in carcass weight, fat depth, muscle or marbling, could be accounted for by variation in visceral fat depots. Thus, selection for low RFI in long-fed in addition to productivity cattle is warranted.