Submitted to: International Turfgrass Science Journal
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: June 19, 1997
Publication Date: N/A
Interpretive Summary: It is important for turf users as well as researchers to know that one variety is different from another. Historically, performance was measured in a field-grown turf plot were morphological characteristics were measured and used to separate one variety from another. Statistical models and efficient field plot designs with appropriate mean separation techniques were necessary to ensure differences among varieties were detected. Statistical differences, or lack thereof, could be declared by a researcher, but economic differences were less well defined and an end-user may have been willing to risk planting a variety that was nonsignificant but slightly different. With the large number of varieties becoming available, judgmental descriptions may be more acceptable than statistical rejection zone tactics. Transgenic plants and other modern molecular technologies also require final performance testing to "prove" a new variety. Protection of intellectual property rights will make it necessary to distinguish turfgrass varieties as molecular techniques increase in popularity. Statistical methods used to separate cultivars are based on allelic and genotypic frequencies within the cultivars. Variability can be detected and characterized using recently developed analysis of molecular variance (AMOVA) techniques and relatedness among cultivars can be visualized with cluster and principle component analyses. These procedures make it easier for an end-user of turf varieties to be more fully informed before making decisions.
Technical Abstract: Distinguishing turfgrass cultivars from each other has been an objective of public and private research programs. Historically, performance of a characteristic measured in a field-grown turf plot or observation of morphological descriptors were used to separate one cultivar from another. Application of traditional statistical models and efficient field plot designs with appropriate mean separation techniques is necessary to ensure differences among cultivars are detected with a probability of alpha Type I error. Economic differences, however, may be realized if an end-user is willing to risk a Type II error at the beta level. With the large number of cultivars becoming available, rejection zone tactics could well be replaced with publication of confidence intervals and t- tests for comparisons of interest with calculated probability levels (p-values). Introduction of transgenic plants into outcrossed cultivars and other modern molecular technologies will still require final performance testing to "prove" a new cultivar. Protection of intellectual property rights will make it more necessary to distinguish turfgrass cultivars and molecular techniques will increase in popularity as identification tools. Statistical methods used to separate cultivars are based on allelic and genotypic frequencies within the cultivars. Variability can be detected and characterized using analysis of molecular variance (AMOVA) and relatedness among cultivars can be visualized with cluster and principle component analyses.