Skip to contents

Counts of yellow/white and sweet/starchy kernels on each of 4 maize ears by 15 observers.

Format

A data frame with 59 observations on the following 6 variables.

ear

ear, 8-11

obs

observer, 1-15

ys

number of yellow starchy kernels

yt

yellow sweet

ws

white starchy

wt

white sweet

Details

An ear of white sweet corn was crossed with an ear of yellow starchy corn. The F1 kernels of the cross were grown and a sample of four ears was harvested. The F2 kernels of these ears were classified by each of 15 observers into white/yellow and sweet/starchy.

By Mendelian genetics, the kernels should occur in the ratio 9 yellow starch, 3 white starch, 3 yellow sweet, 1 white sweet.

The observers had the following positions:

1Plant pathologist
2Asst plant pathologist
3Prof agronomy
4Asst prof agronomy
5Prof philosophy
6Biologist
7Biologist
8Asst biologist
9Computer
10Farmer
11Prof plant physiology
12Instructor plant physiology
13Asst plant physiology
14Asst plant physiology
15Prof biology

Source

Raymond Pearl, 1911. The Personal Equation In Breeding Experiments Involving Certain Characters of Maize, Biol. Bull., 21, 339-366. https://www.biolbull.org/cgi/reprint/21/6/339.pdf

Examples

if (FALSE) { # \dontrun{

library(agridat)

data(pearl.kernels)
dat <- pearl.kernels

libs(lattice)
xyplot(ys+yt+ws+wt~obs|ear, dat, type='l', as.table=TRUE,
       auto.key=list(columns=4),
       main="pearl.kernels", xlab="observer",ylab="kernels",
       layout=c(4,1), scales=list(x=list(rot=90)))

# Test hypothesis that distribution is 'Mendelian' 9:3:3:1
dat$pval <- apply(dat[, 3:6], 1, function(x)
                  chisq.test(x, p=c(9,3,3,1)/16)$p.val)
dotplot(pval~obs|ear, dat, layout=c(1,4), main="pearl.kernels",
        ylab="P-value for test of 9:3:3:1 distribution")

} # }