My soil experiment investigates soil-driven effects on P and Zn interaction in red cabbage and it is coming to an end in the coming week. This experiment was designed with the help of a statistician and thus it has a very much defined layout. It makes use of Graeco-Latin square – five P and five Zn treatments are positioned in a way that none of the five treatments is repeated in the same row or in the same column. This means every row and every column has all five P treatments and all five Zn treatments. It reminds me so much of my favourite pastime game, Sudoku. Here is an example of a Greaco-Latins square of 5 orders, where P treatments are labelled by Greek letters (P1, P2, P3, P4 and P5 as α, β, δ, γ and ε, respectively) and Zn treatments are labelled by Latin letters (Zn1, Zn2, Zn3, Zn4 and Zn5 as A, B, C, D and E, respectively). The labeling of treatments using a combination of Greek and Latin letters is actually where the name came from.

Graeco-Latin square of 5 orders (5 P treatments and 5 Zn treatments) with guard plants around (as brown squares).

Here is a real photo of the experiment:

Red cabbage in a 5×5 treatment experiment with guard plants around it.
Three Graeco-Latin squares, each on different soil in a glasshouse

The statistical worth of this experimental design still needs to be proven, but this is not  far away now. As always, I’m looking forward to new experience!