For Release June 25, 2006

Use Caution When Selecting Wheat Varieties

AGRI-VIEWS

by Chuck Otte, Geary County Extension Agent

As the 2006 wheat variety yield data starts to roll out in coming weeks, there are probably going to be some very unusual results. Proceed with caution when you start looking at those results. Be equally cautious when you start selecting the wheat varieties you are going to plant this fall. Beyond 2006 being a drought year, we have had some other very strange conditions that will probably taint the results this year!

Let’s get into the details. Perhaps the biggest anomaly of the 2006 wheat growing season was the virtual lack of rust diseases. In most years, leaf rust or stripe rust, or both, will cause the greatest reduction in yield. Rust resistance in wheat is something that can be a very fleeting benefit. If a popular variety is planted from Texas through Nebraska on a lot of acres, the rust races can adjust very quickly so that the race that is not affected becomes the dominant race.

Because of dry weather throughout much of the southern winter wheat belt, leaf rust and stripe rust were very late in developing in Texas and Oklahoma. Subsequently it was difficult to even find a rust pustule in Kansas until it was so late in the season it didn’t matter. While this was good for wheat production, it also means that older varieties, that we know are no longer resistant to leaf rust, may yield very well. Varieties, like Karl 92, still have good yield potential if there is very little, or no, rust disease pressure.

Naturally, the impact of the drought is going to be very evident in both overall yields, especially from the western Kansas experiment fields, and individual variety responses. So when you start seeing yield results be sure to take a long look at the data. Regardless of the source, make sure that the yield trial was replicated plots. In replicated plots, every variety was planted more than once, usually three or four times. By having this replicated plot design, the agronomists are able to perform statistical analyses to determine how much yield differential is needed for it to be statistically significant.  

If the results are all from unreplicated plots, you have no way to tell if the difference in yield between two varieties is because one is better than the other or if one was simply on better ground. When you are looking at results from replicated plots you should see a number called LSD, least significant difference. This is the statistically calculated figure that two yields must differ by for it to be significant. If this number is rather large, it means that there was a lot of variability in yields of all varieties. If it is small, it means that the varieties yielded very consistently.

The next thing I want you to look at is yield over a period of years. The K-State studies will usually have two, three and sometimes four year averages. These are very useful for showing how stable the yield of a variety is over a period of years. It a variety yielded well this year, but very poorly the past two or three years, it should tell you something. What I hope it is telling you is “don’t plant me!”

Finally I also want you to pick up a copy of the latest disease ratings of wheat varieties. This year, with no rust pressure, is unlikely to occur next year. Plant varieties that have good rust resistance. Look at all the results and then pick out three or four varieties and see how they do under your field conditions. Put the top public variety up against the best variety from a couple of different companies. Put the best head to head and see how they do. But don’t put a whole lot of faith in any results from 2006 wheat crop alone! 

-30-

Return to Agri-Views Home Page

Return to Ag Home Page