. Journal of Agricultural Research. tes with respect to allmeasured characters is showTi in figure 8. From the character of these distributions it would seem that in thecase of the dwarf variation a relatively few hereditary elements willaccount for the differences between this and tlie normal form. Thebehavior of the leaf lengths and shapes is very different from that en-countered where short, broad leaves of nondwarf stock are crossed withrelatively long, slender leaves of some other strain. In such cases thefrequency distributions are unimodal, the indication being that severalhereditary fa
. Journal of Agricultural Research. tes with respect to allmeasured characters is showTi in figure 8. From the character of these distributions it would seem that in thecase of the dwarf variation a relatively few hereditary elements willaccount for the differences between this and tlie normal form. Thebehavior of the leaf lengths and shapes is very different from that en-countered where short, broad leaves of nondwarf stock are crossed withrelatively long, slender leaves of some other strain. In such cases thefrequency distributions are unimodal, the indication being that severalhereditary factors are concerned in the differences between the parents. Aug. i8, 1933 Inheritance of Dwarfing in Maize 305 CORRELATIONS OF MEASURED CHARACTERS The character of tlie distributions in most cases precludes the use ofthe correlation coefficient, since the measured character so often is bimo-dal. Recourse may be had to fourfold groupings, dividing the populationarbitrarily into two groups of the measured character, but when such /^O. <^ 3 6 7 5 9/0 // /2 /3/^ /5 /6 /7 Fig. 4.—Frequency distribution for width of leaf. Shaded portion, dwarf plants; solid lines, brachytic plants; broken lines, normal plants. divisions are made with due regard to the character of the distributionsone class is often zero or Yory small. Under such conditions a correlationcoefficient is practically without meaning, and such coefficients have notbeen calculated. In those cases where the data justified the use of thebiserial correlation, the coefficients have been calculated and are givenin Table V. For the most part, however, the frequency polygons willgive a clear conception of the nature of the inheritance. L 3o6 Journal of Agricultural Research Vol. XXV. No. 7 /oo 90 80 70 30 20 /O o s • /o 20 SO
Size: 1204px × 2074px
Photo credit: © Reading Room 2020 / Alamy / Afripics
License: Licensed
Model Released: No
Keywords: ., bookcentury1900, bookdecade1910, bookidjournalofagricul25assouof
