Johannsen (1905) Bean Experiment:
Limits to Directional Selection

    (1) Johannsen (1905) sampled a large number of beans (Phaseolus vulgaris) from an outbred line. (2) He created two lines, one drawn from the largest beans and one from the smallest, from the two tails of the distribution. (3) Over several generations, the two lines diverged over a range of seed sizes, including some that were larger or smaller than the largest and smallest of the original outbred line: Directional Selection on size was thus possible.

    (4) Past a certain point, no further divergence in size was observed, despite continued sampling of beans at the two extreme. Seed size in either line remained variable, which Johannsen attributed to environmental variation. (5) Deliberate selection of the smallest seeds from the Large line, and of the largest from the Small line, produced the same range of seed size that had been observed in the two pure lines. Though genetically uniform, the two lines continued to be subject to environmental variation.