Wheat variety performance trials are conducted by Michigan State University (MSU) each year at several locations throughout Michigan's winter wheat production area. Entries to the trials include MSU experimental lines, promising lines from neighboring states and commercial varieties from other universities and private seed companies. The primary objective of this testing program is to provide the agronomic data needed to determine which lines to release as commercial varieties. A second objective is to show Michigan wheat growers which varieties perform best in Michigan. This year�s results are summarized in the accompanying tables.

Although wheat producers are always interested in how varieties perform in a given year and location, performance in single year and location should never be used in selecting a variety to plant. It is best to select a variety on the basis of data from at least three years of testing. Varieties selected with such comparisons are more likely to perform well under a wide range of conditions.

Multi-Year Performance Summary (Table 1)

Each line in the table has data for a single variety. The column bordered by double lines has this year�s average yield. The table is arranged so that the varieties appear in order of �97 average yield with the highest yielding variety first and the lowest yielding variety last. Not all varieties have been tested in all years so the table has several blank cells. To the right of the �97 yield column are multi-year yield averages. Only data for varieties included in the relevant year�s tests are included here. See the section titled �Experimental� for details on how the trials were conducted and more detail on what the data in each column�s data represent.

At the bottom of each table are the means, L.S.D.s, and C.V.s for the 1997 data columns. The L.S.D. (least significant difference) is the statistical measure of how big a difference needs to be to be considered real. If the difference between two means is greater than or equal to the L.S.D., then the varieties are probably really different for that trait. Otherwise, there is insufficient evidence to claim that the varieties are actually different. The C.V. (coefficient of variation) is indicative of the trial�s precision for a given trait. Lower C.V. values indicate more precise trials.

In any given year or at any given site, several varieties will usually fall into the group of 'highest yielding' varieties. The composition of that group and the identity of the absolute "winner" can and do change from location to location and year to year. This means that the single best variety cannot be determined in advance for a specific site. What you can do is identify a group of varieties whose past performance and agronomic characteristics indicate that they are most likely to be winners in the upcoming season. It is a good idea to plant two or more varieties. That increases the chance of having the best adapted variety for the particular conditions that are likely to prevail during the ensuing season. Selecting two varieties can reduce losses from diseases and insects that occur when a given variety's pest resistance is overcome by a change in the pest population.

Single Site Yield Performance Summary (Table 2)

Columns in this table represent yield (bushels/acre) at each of the eight sites. The last column is the average of all sites. Each row in the Table represents a single variety in the test.

Experimental

The 1997 State Wheat Variety Trial was harvested at eight county sites: Lenawee, Ionia, Saginaw, Kalamazoo, Huron, Tuscola, Sanilac, and Ingham. Plots were 11 feet long and had 7 rows at 6" row spacing. Individual sites were implemented as three replication alpha lattices (15 blocks of 5 plots each). Seeding rates were standardized to 1.8 million seeds per acre. Planting dates were all within normal ranges. Fall fertility varied with cooperator practice. Spring nitrogen was applied as urea (80 lbs/acre) at greenup. No fungicides were applied. All plots at a site are harvested on a single day. Yield was calculated using the entire area of the plot including the wheel tracks between plots. Yield, test weight, and moisture data were acquired electronically on the plot combine at the time of harvest. Yield comparisons are only valid within a column. All scores are based on a 0-9 scale, where 0 is the best possible score. "Septoria leaf blotch" scores are for foliar symptoms only and probably reflect both S. nordorum and S. tritici infections. Sprouting score data are based on greenhouse evaluation of 5 heads from all three replications at both the Ingham and Ionia county sites. Heads were picked immediately before those sites were harvested. After two days of drying, the heads were subjected to continuous misting for 5 days. Data for 50% pollen shed indicates the number of days past January 1st before that variety reached the point where 1/2 of its heads where flowering. Plant height was measured at the tip of average heads in a plot.

MSU makes no endorsement of any wheat variety or brand. Cooperator support is gratefully acknowledged.

Site summary information:

Table 1a. Multi-year data agronomic data. MSU wheat trials.

Table 1b. Multi-year data agronomic and disease data. MSU wheat trials.

Table 1c. Multi-year data agronomic and disease data. MSU wheat trials.