Here's the backstory: S had an existing graphics system that does your standard plotting (histograms, scatter, box'n'whisker, etc.), and then the Bell labs people came by and made a new, incompatible system called "Trellis". (In the clone mirror world where S is called R, it's called "Lattice".) The only non-reference docs for Lattice are the Trellis ones, but those are the docs you ought to read.
Your reward for reading this post are these cool infovis stories, which motivate Trellis.
The barley experiment was run in the 1930s. The data first appeared in a 1934 report published by the experimenters. Since then, the data have been analyzed and re-analyzed. [...] Then in the early 1990s, the data were visualized by Trellis Graphics. The result was a big surprise. Through 60 years and many analyses, an important happening in the data had gone undetected.I get a weird thrill out of the thought that a different visualization of data can reveal new aspects to it.
(I also discovered today that a Cairo backend for R exists, and is actually in Debian, but it hasn't made its way to Dapper yet. I guess I have to live with these hideous fonts a while longer...)