Introductions and Personnel Changes

- Brent Sanders has joined Terrall Seed Research.

- Ken Evans, Univ. of Kentucky, passed away recently.

- Grover Shannon is now soybean breeder with the University of Missouri at the Portageville station.

New Variety and Hybrid Releases

- Mississippi released a new group V-late soybean variety named 'Bolivar.'

- North Carolina released a new tobacco variety.

Stress vs Non-Stress Trials

Daryl Bowman, North Carolina State University - The value of including data from stress trials was examined. It was suggested that data from stress trials should be reported separately from that obtained from non-stress trials. When examining long term data, inclusion of data from stress trials improved predictability compared to when stress data was omitted. It was noted that varieties which yielded well in non-stress tests generally did not yield well in stress tests. Including data from stress trials appears to have merit.

A discussion of site-specific testing ensued. It was noted that many commercial breeders are breeding site-specific varieties for niche markets. These types of varieties will very likely not perform well in all test locations, but many university tests do not allow a choice of which locations a variety will be tested in. Locations outside those for which a variety is bred may behave like stress tests for that variety. Dr. Vance Watson noted that Mississippi currently has a statement to this effect in its variety test bulletins.

Growing MG III Soybeans in Mississippi

Lingxiao Zhang, Mississippi State University - Dr. Zhang reported on the current status of research at the Delta Branch Experiment Station on growing MG III soybean cultivars. Current studies have examined the effect of planting on MG III yields. Group III cultivars appear to have some potential in the Mid-South; however, there is a need for variety test data on MG III cultivars if they are to recommended for this area.

Use of Data Loggers as Tools in Yield Testing

Donny Glover, Terral Seed Inc. - Glover reviewed the use of data loggers to aid in data collection with specific emphasis on the Almaco and Harvestmaster products. Data loggers were more economical and efficient than recording data by hand. Each data logger model is different and you have to find one that works for you. Advantages of data loggers include:

1. reduced costs for bags, tags, & labor associated with weighing, dumping, and data input,

2. reduced turn-around time between harvest and reports, and

3. reduced input errors due to transcription errors and lost samples.

Disadvantages of using data loggers include:

1. They require higher levels of computer skills and management (e.g. uploading files, merging files, etc.)

2. Higher initial equipment costs.

Data loggers lend themselves to many collecting many types of data such as: combine data (weight, moisture, test weight, etc.); field book notes (height, maturity, disease, etc.); sensor-collected date (barcodes, temperature, humidity, etc); and GPS data)

Bill Pierce commented that they had some problems with the Grainguage system, particularly with large samples. The Almaco system was best when there were trash problems.

Public Sector Evaluation of GMO Soybeans

Harry Minor, Univ. of Missouri-Columbia - Questions remain as to whether GMO varieties perform differently under different management systems (e.g., Roundup-ready varieties tested with either Roundup or with conventional herbicides). Also, do GMO and conventional varieties produce comparable yields?

Currently the Univ. of Missouri tests Roundup-ready varieties with conventional checks randomized in the same field. The conventional checks are covered with plastic to prevent drift, and the entire field is sprayed with Roundup. However, covering the conventional checks even for a short time seemed to have a detrimental effect on the conventional check varieties. This system is very labor intensive, but it allows for direct comparisons between Roundup-ready and conventional varieties. Due to these problems, a new system of evaluating Roundup-ready varieties was tried this past year. Objectives were to compare 1) GMO varieties alone, 2) GMO + conventional varieties in a conventional test, and 3) contrast GMOs to a conventional check package of varieties. A split-block arrangement was used with conventional and Roundup-ready varieties separated by a buffer strip within each block. Performance of Round-up ready and conventional varieties was similar.

Three soybean production systems were also compared. These consisted of : 1) conventional varieties with conventional herbicides; 2) Roundup-ready varieties with Roundup only; and 3) Roundup-ready varieties with a conventional pre-emerged and a post application of Roundup. Yields of the three treatments (43.4, 41.4, and 42.6 bu/A, respectively) were similar.

The stand practice for evaluating Bt corn hybrids is to separate the Bt lines and embed them by replication into conventional corn trials. The check varieties are artificially infested with European corn borer (ECB). Yields have been similar in non-infested trials, probably due to low infestation levels of ECB. Yields of the best Bt lines dropped somewhat under heavy infestation compared to natural infestation.

A new method of measuring yield stability was discussed. The method uses a graphical presentation of mean yield and the associated variance for each variety in the form of a "whisker" plot. Variance is measured as deviations from average. Minor is currently working on a software package for this method.

Minor also discussed the use of a soil monitor to map fields that are to be used for yield trials. This technology is still in the test stage, but it appears to hold promise for identifying variations in soil type.

Measuring Yield Stability

Clarence Watson, Mississippi State Univ. - Yield stability of crop varieties is of interest to both producers and breeders. Producers are interested in stability of actual yield (in bu/A, lb/A, etc.), while breeders tend to be more interested in consistency of ranking. Numerous techniques are available for measuring both types of stability. Methods of measuring average performance or stability typically fall into four groups: 1) mean yield over environments, 2) regression analysis, 3) measures of variance, and 4) non-parametric rank-based methods. Selected yield stability measures from all four categories were calculated for the 1999 Mississippi Group IV soybean variety test data. The results varied considerably with choice of method. Reliable estimates of yield stability require large tests and numerous environments. Choice of method used to measure stability can greatly affect the conclusions drawn. Consequently, caution should be used in interpreting yield stability statistics . For comparing yield stability statistics from different tests, it is important to use the same method of measuring stability in both tests. Currently, most states are using some form of regression analysis. If yield stability is to be included variety test reports, some thought should be given to standardizing the technique. Another point to keep in mind is that others may be using your data to calculate yield stability, since most methods do not require raw data. If they are using different methods to measure stability, they may draw entirely different conclusions concerning the relative stability of a given set of varieties.

Deltasoy - an Internet-based Database System for the State Variety Trial

Lingxiao Zhang, Mississippi State University - Deltasoy is a new Internet-based variety selection program being developed by MAFES. Deltasoy is currently in the development and testing phase. It allows the user to search by variety, location, year, maturity group, irrigation status and Roundup-ready status. The user can limit the number of records to be displayed for each search and the data can be listed either in ascending or descending order. Advanced search options also allow the user to limit the search based on resistance to certain diseases. Deltasoy can be accessed at http://www.visit.to/deltasoy/.

Among the future goals for the web site are inclusion of variety test data from other states and interaction with the soybean model and GPS databases. For further details about Deltasoy or if you are interested in cooperating in the development of the Deltasoy database, please contact Dr. Lingxiao Zhang at lzhang@drec.msstate.edu .

Group Discussion

Don Dombeck- Asked for the group's opinion on the appropriate use of variety and hybrid checks. Should they be used to yield checks only or should they be used as maturity group checks in soybeans? Should we change the maturity group of an entry if we feel that it is misrepresented? Several states reported that they use maturity checks and may change the classification if the data warrants; however, the company is notified and offered the opportunity to withdraw the entry. It was felt that a minimum of two years of data would be required to make a final judgement.

Don Dombeck- stated that Patsy Sykes needs to be commended for her efforts in putting together and maintaining the SRIEG-33 website.

Walter Mayhew- made some comments relative to testing of Genetically-Modified Organisms (GMO). He pointed out that there are numerous groups, both pro and con, who are using our data to make their case for or against GMOs. We need to be very careful about how we generate and use GMO data.

Business Meeting

Minutes

The minutes of the 1999 meeting were approved. The minutes are listed on the SRIEG-33 web page.

http://www.mafes.msstate.edu/admin/ieg.html