There is a relatively common planting spacing experimental design called a Nelder trial. It is really cool looking from the air, generating a typical wheels-and-spokes pattern.
There was a also a John Archibald Nelder, very famous British statistician (rock-star level IMHO), who did extremely important work on generalized linear models (GLM) among other things. Literally, he wrote the paper† and the book‡ and many of us have used his work one way or another.
I don’t know why, but I always thought that the first Nelder—let’s call him Forestry Nelder—was different from the elegant-looking Stats Nelder. I was asking a colleague Why would have Stats Nelder been working in forestry problems? The answer has two parts: 1. Forestry Nelder and Stats Nelder are the same person, and 2. Nelder didn’t come up with the design thinking of forestry.
It happens that Nelder worked at the National Vegetable Research Station, Wellesbourne where he wrote “New Kinds of Systematic Designs for Spacing Experiments”, published in Biometrics in 1962. The subject of the design was Brassica, more specifically Brussels sprouts and cabbage. It appears that in 1965 Gene Namkoong, a forest quantitative geneticist and tree breeder working in the US, was the first to adapt the family of Nelder’s designs in forestry⸸. Interestingly, the design is rarely used today in tree breeding, but it is much more common in silviculture.
And that is how we got these amazing trials in forestry.
† Nelder and Wedderburn. 1972. Generalized Linear Models. J. R. Statist. Soc. A, 135: 370–384. ‡ McCullagh and Nelder. 1983. Generalized Linear Models. Chapman & Hall. ⸸ Namkoong. 1965. Application of Nelder’s designs in tree improvement research.
