Alfalfa {nlme}R Documentation

Split-Plot Experiment on Varieties of Alfalfa


The Alfalfa data frame has 72 rows and 4 columns.


This data frame contains the following columns:

a factor with levels Cossack, Ladak, and Ranger
a factor with levels None S1 S20 O7
a factor with levels 1 2 3 4 5 6
a numeric vector


These data are described in Snedecor and Cochran (1980) as an example of a split-plot design. The treatment structure used in the experiment was a 3times4 full factorial, with three varieties of alfalfa and four dates of third cutting in 1943. The experimental units were arranged into six blocks, each subdivided into four plots. The varieties of alfalfa (Cossac, Ladak, and Ranger) were assigned randomly to the blocks and the dates of third cutting (None, S1—September 1, S20—September 20, and O7—October 7) were randomly assigned to the plots. All four dates were used on each block.


Pinheiro, J. C. and Bates, D. M. (2000), Mixed-Effects Models in S and S-PLUS, Springer, New York. (Appendix A.1)

Snedecor, G. W. and Cochran, W. G. (1980), Statistical Methods (7th ed), Iowa State University Press, Ames, IA

[Package nlme version 3.1-57 Index]