50-year German wheat trial: old Kanzler variety yields 15% less than modern ones
In trials run by an agricultural cooperative in Lower Saxony, the old winter wheat variety Kanzler yielded about 15% less than modern varieties. It was also far more prone to disease, while extra nitrogen fertiliser did not raise yields.
At a trial field in Ubbendorf, in the Nienburg district of Lower Saxony, Germany, the agricultural cooperative Raiffeisen-Warengenossenschaft Niedersachsen Mitte eG has been running field trials for half a century. To mark the anniversary, old and modern winter wheat varieties were compared, showing what plant breeding has delivered in yield and disease resistance. The findings are relevant for farmers choosing varieties and planning fertilisation.
The old variety falls behind
The most telling comparison was between the old variety Kanzler, modern varieties and a hybrid wheat. Kanzler, approved in 1980, was once a well-known winter wheat. In this trial its yield was roughly 15% lower than that of the modern comparison varieties. The publication gives an example calculation: at a yield of 90 dt/ha, the gap would be 13.5 dt/ha, or €270 per hectare at an assumed wheat price of €20 per dt. This is illustrative only, as the actual trial yields were not stated.
Kanzler was also considerably more susceptible to disease, especially yellow rust. Healthier varieties can help limit pesticide use, but how much fungicide can be saved depends on the variety, disease pressure and the conditions of the year.
The fertilisation trial
In the fertilisation trial, Kanzler and the modern variety Informer were compared across eight fertilisation variants based on earlier practice, current recommendations and artificial intelligence calculations. In fertilised variants, Kanzler yielded about 10–15% less than Informer. Within each variety, there were no statistically significant yield differences between the fertilisation variants, so more fertiliser brought no demonstrable extra yield.
One possible reason is that the soil released considerably more nitrogen than expected. An additional nitrogen dose planned in an advisory variant for straw decomposition was, in the trial managers' assessment, not necessary. The more heavily fertilised stand looked more even and vigorous, but N-tester readings and crude protein levels indicated the extra nitrogen was not needed.
ChatGPT and Google Gemini were also asked for fertiliser recommendations; both AI-based variants applied 140 kg of nitrogen per hectare. Such tools can support planning, but whether their advice is useful depends on the available data and expert review.
The results come from a single site and year and cannot be generalised.
