3.7.2 An important method for studying mutation-causing substances involves killing female mice 17 days after mat- ing and examining their uteri for living and dead embryos. The classical method of analysis of such data assumes that the survival or death of each embryo constitutes an inde- pendent binomial trial. The accompanying table, which is extracted from a larger study, gives data for 310 females, all of whose uteri contained 9 embryos; all of the animals were treated alike (as controls).25

(a) Fit a binomial distribution to the observed data. (Round the expected frequencies to one decimal place.)

(b) Interpret the relationship between the observed and expected frequencies. Do the data cast suspicion on the classical assumption?

Number of embryos Number of Dead Living female mice 10 9 136 8 103 234567 68 7 50 6 13 5 6 4 1 3 1 2 0 1 0 9 0 0 Total 310