1) Permission is granted to reproduce the tables in their entirety provided the source is referenced and the data are not manipulated or reinterpreted. A conspicuous disclaimer which states "endorsement or recommendation by Purdue University is not implied" must accompany any information reproduced.

2) Advertising statements by an individual company about the performance of its entries can be made as long as they are accurate statements about the data as published, with no reference to other companies’ hybrids. A statement similar to "See the official Purdue University bulletin ‘Performance of Commercial Dent Corn Hybrids in Indiana, 2002, B- 816 for details." must be included in the ad. Additional copies of this publication are available to Indiana residents from their local Purdue Cooperative Extension Offices.

This document can be accessed electronically by the following method.

1. World Wide Web (Netscape, Mosaic, etc.): by pointing to the URL address:
http://www.agry.purdue.edu/ext/variety.htm

_______________________________________________

Predominant soil types or production levels commonly found in northern, central, and southern Indiana were represented by four locations in each region. At each location, plots consisted of four rows and were, replicated three times within a randomized complete block design. All data was collected on only the middle two rows of the plot .Participating companies supplied relative maturity data reported in the index as days from emergence to kernel black layer. The plots were planted from pre-counted packets of seed using a plot research air planter, and harvested with a self-propelled harvester without gleaning. Stand counts for each individual plot were recorded during the month of June. The percent stand is determined based on the number of kernels planted and the number of plants emerged, and therefore, indicates only germination and emergence losses.

Average harvest moisture is presented for each hybrid, and yield data are calculated in bushels of shelled grain per acre, adjusted to 15.5% moisture. No yield adjustment was made for stand except for continuous skips, which are determined to be of mechanical origin. Lodged plants (stalks broken below the ear) were counted just prior to harvest. Test weight data were adjusted to 15.5% moisture.

Each farm cooperator prepared the seedbed using conventional tillage practices and applied herbicides at labeled rates. Fertilizer application was based upon the management program of each cooperator and the desired plant population of each test. The planting populations were 32,000 ppa, for the north and central tests 30,000 ppa for the south and 24,800 ppa for the sand test.

The performance data for each test within a region is reported in a table. Hybrids in each table are listed by yield in descending order. Comparisons can be made only between hybrids within a table. Comparing two hybrids from different tables (i.e. different environments and/or production levels) would likely lead to an erroneous conclusion.

With the use of statistical analysis, it is possible to determine whether a difference in performance between two hybrids is real or due to chance, at a chosen level of confidence. The least significant difference (LSD) listed at the bottom of each table for each column of data should be used to distinguish such differences. This bulletin presents data with an LSD of 10%. If the difference between two hybrids were equal to or greater than the LSD, the difference would be attributable to hybrid differences and not random chance in 18 out of 20 (90%) instances when the two hybrids are evaluated under conditions like those of the test. A difference which is less than the LSD is likely due to chance, rather than true hybrid performance. When no significant difference for a given parameter is found among hybrids, "ns" (nonsignificant) replaces an LSD value.

While much care is exercised to keep experimental error to a minimum, variation in data among any given hybrid’s replications can and does occur. For example, very rarely is a hybrid’s yield exactly the same for all replicates, although relatively small amounts of variation are certainly considered "normal". A relatively large amount of variation for many of the hybrids in a test can render the data useless for determining real, instead of chance, differences in performance. Plot replication is carried out in order to sample the uncontrollable variation within a test site, which has been chosen for its uniformity of soil type and slope. Other factors which may introduce error, but which can be reasonably controlled, include uniformity of tillage, herbicide and fertilizer application, timeliness of planting and harvesting, and the condition of equipment used to perform the research. Ideally, all hybrids within a test should have an equal chance to perform.