The measurement of linkage in heredity / by K. Mather.

  • Mather, Kenneth, 1911-1990.
Date:
[1938]
Catalogue details

Licence: Attribution-NonCommercial 4.0 International (CC BY-NC 4.0)

Credit: The measurement of linkage in heredity / by K. Mather. Source: Wellcome Collection.

    23/152 (page 9)
    0 Aa individuals will be, from the expansion of (è + è) 8 : No. of Aa individuals 876543210 1 8 28 56 70 56 28 8 1 Frequency . * 256 256 256 256 256 256 256 256 256 If we observe that a family actually obtained shows 7 Aa and ] aa members, we can apply a test of significance of the deviation from the expected 4 : 4 by the use of the above frequency distribution. The deviation is 3 from expectation in each class. Now deviations of 3 or more will occur when families of 8 : 0, 7 : 1, 1 : 7, or 0 : 8 are found. These families are expected with the frequencies rfe» and respectively. Then the total expectation of obtaining as large or larger deviation than the one 1 1 8- I 8 I 1 observed is -~r- i.e. or 0-070. z5o This is slightly greater than 7 per cent. It is generally considered that a deviation should not be taken as significant until the probability of obtaining it by chance is less than 5 per cent, and so the hypothesis can, in this case, be accepted as agreeing sufficiently well with the data. If we had been expecting a segregation of 3 : 1, or J : J, the correct binomial for a family of n would have been (J -f J) n . The general form for a ratio of X : y, where y = 1 — x, is (x -f y) n and the general term of the expansion giving the frequency of families with r individuals of one class and n — r of the other is While it is always possible to do a test of significance in this way, it is not always convenient to calculate the binomial expansion, particularly when n is large. A quicker technique is needed. Now the probability of a deviation being equalled or exceeded by chance is expressible as a function of the deviation divided