Two lines of Large White pigs were selected for increased rate of growth and decreased backfat thickness respectively over five generations. Each line was bred from four boars, each mated to four gilts per generation, these being selected on a within-family basis to minimize inbreeding. Response to selection was measured by direct comparison with contemporary control stock. The control line, originating from the same foundation stock as the selection lines, consisted of 15 groups of one boar and two gilts, mating between groups being designed to reduce the rate of inbreeding.

Over three generations the control stock exhibited random variation in the traits measured except rate of daily live-weight gain and backfat thickness ‘K’. The evidence suggests that the time trend in these was mainly, if not entirely, non-genetic in nature.

The low rate of progress achieved in both lines is ascribed to the selection method adopted and to losses of potential breeding stock. It is concluded that the success of such a small-scale breeding operation is dependent on achieving optimum husbandry conditions, and even then it cannot achieve the same rate of progress as in larger herds. The necessity of using a control strain to assess genetic progress in experiments of this nature is emphasized.