Both of these show the distribution of phenotypes. In directional selection, the distribution of phenotypes forms a "bell curve." Selection against one of the extreme phenotypes causes the distribution to "move" in one direction or the other. An example might be plants whose flower color is determined by incomplete dominance: white, pink, or red. Pink flowers may be the average phenotype, but if we start to remove red flowers from the population, the "mean" phenotype will be shifted toward white flowers.
In disruptive selection, the average phenotype is selected against. This produces a "two-humped" bell-type curve, and the greater distribution is split between the two phenotype extremes. If we have the same type of incomplete dominance as mentioned in the previous paragraph, assume that the pink flowers are selected against. This means that the two "humps" shown in the distribution will be centered around the red and white phenotypes.
