An Alternative Possibility for Seed Coat Color Determinaton in Mendel's Experiment

Corresponding author: James R. Myers

IT is possible that in his test of the hypothesis of independence among three genes influencing seed characters of pea (cotyledon color and shape and flower or seed color) Mendel relied on partial dominance of seed color to classify heterozygotes as well as homozygotes in his cross. It would make sense for him to do this if possible, rather than growing progeny to classify flower color. He indicated that hybrid seed coats "are often more spotted and the spots sometimes coalesce into small bluish-purple patches." Correns also described partial dominance of seed coat color, but it is not clear to me whether he used the information in classifying his populations. Apart from these descriptions, I do not know of any subsequent researchers who used partial dominance of seed coat color to classify heterozygotes. While on average the testa of the heterozygote is more intensely pigmented, variations among seeds from different plants as well as among seeds of the same plant are observed. In some crosses, differences between the pigmented homozygote and heterozygote may be sufficiently discrete for accurate classification. In other crosses, the lines between heterozygotes and homozygotes may be blurred. If it were possible to recreate the original crosses, one might be able to determine whether Mendel would have had the opportunity to classify his material on the basis of seed characteristics alone.

Classification is complicated by additional seed coat color genes of which neither Mendel nor Correns appeared to be aware. Correns's description of an "orange-red seed coat, which turns brown on aging" may describe the M gene. Mendel describes the seed coat as being "grey, grey-brown, or leather-brown," the latter of which might be the M gene. This gene does not appear to be involved in partial dominance. Rather, it is the intensity of anthocyanin spotting that gives a clue as to the genotype. Anthocyanin spotting is controlled by two loci (F and Fs), each independent of the other. We do not know whether one or both were present in the parental material of Mendel. Mendel does not report independent segregation of anthocyanin spotting. If one of these genes were segregating, it would complicate the classification of partial dominance of seed color. Both Mendel and Correns reported parents that did not show the purple spotting, but when crossed, a new character (purple spotting) was observed in the F₁ and subsequent generations. This result could be explained if the parents had the genotype aa FF x Aa ff [white flowers/colorless seed coat x purple flowers/gray (unspotted) seed coat] (A, purple flowers and gray seed coat; a, white flowers and colorless seed coat; F, anthocyanin spotting on seed coat; f, absence of anthocyanin spotting; aa prevents expression of F). Anthocyanin spotting would appear in the F₁: Aa Ff.

But the following would be expected to segregate in the F₂: A-F- (purple flowers/spotted seed); A-ff (purple flowers/gray but nonspotted seed); aaF- (white flowers/colorless seed coat); aaff (white flowers/colorless seed coat). Such segregation would produce an expectation of 3:1 for the spotted to nonspotted seeds among the purple flower class instead of the 2:1 ratio expected if the spotting in the heterozygote could be unambiguously identified. In this issue of GENETICS, C. E. Novitski has performed the customary ² tests and has shown that both alternatives are statistically in agreement with Mendel's data, although Mendel's 2:1 ratio gives a somewhat better fit (Myers' ² = 29.2, P = 0.30; Mendel's ² = 15.3, P = 0.95).

This article has been cited by other articles:

				D. J. Fairbanks and G. B. Schaalje The Tetrad-Pollen Model Fails to Explain the Bias in Mendel's Pea (Pisum sativum) Experiments Genetics, December 1, 2007; 177(4): 2531 - 2534. [Abstract] [Full Text] [PDF]