Commercial cane sugar yield of 18 sugarcane (Saccharum spp. hybrids) genotypes evaluated in a randomised block design at two locations in peninsular India over three crop cycles, giving six environments.
sugarcaneA data frame with 216 rows and 6 variables:
Factor with 6 levels, the environment (location-by-crop-cycle).
Factor with 2 levels, SGA and TCSM.
Factor with 3 levels, IP, IIP and R.
Factor with 2 levels, the replication (block) within environment.
Factor with 18 levels, the genotype.
Numeric, commercial cane sugar yield in t/ha.
The trials were conducted by the ICAR-Sugarcane Breeding Institute at
Sakthi Sugars, Sivaganga (SGA) and Tiruttani Cooperative Sugar Mills,
Thiruvalangadu (TCSM), both in Tamil Nadu, India. The three crop cycles are
plant I (IP), plant II (IIP) and ratoon (R). The six environments are the
location-by-cycle combinations, labelled SGA-IP, SGA-IIP,
SGA-R, TCSM-IP, TCSM-IIP and TCSM-R.
Two replications are distributed. A third column of values present in an
earlier working version of these data was found, on audit, to be identically
the arithmetic mean of the two observed replicates in the SGA-R,
TCSM-IIP and TCSM-R trials, for every one of the 18 genotypes.
Such a value is a derived quantity, not an independently randomised
experimental unit: including it leaves the error sum of squares of those
environments unchanged while doubling the error degrees of freedom, which
halves the error mean square; because the genotype means are unchanged, the
genotype sum of squares is at the same time multiplied by three halves, so
the genotype F-ratio is inflated by exactly three. Only the two genuine
replicates are therefore included, giving a fully balanced series of six
two-replicate randomised block experiments.
data(sugarcane)
str(sugarcane)
with(sugarcane, table(env, rep))
round(tapply(sugarcane$ccsy, sugarcane$env, mean), 3)
Run the code above in your browser using DataLab