Doctoral Dissertations

Date of Award

5-1992

Degree Type

Dissertation

Degree Name

Doctor of Philosophy

Major

Plant, Soil and Environmental Sciences

Major Professor

Henry A. Fribourg

Committee Members

John H. Reynolds, William L. Sanders, John C. Waller, J.B. McLaren

Abstract

Most of the information concerning the effects of the tall fescue (Festuca arundinacea<\em> Shreb.) endophytic fungus (Acremonium coenophialum<\em> Morgan-Jones Gams) on beef (Bos taurus<\em> L.) steer performance has been obtained from grazing trials conducted as independent endeavors. These trials may be related over space and time. Datasets from 12 trials conducted during the last 13 years at nine locations in seven eastern US states were pooled to provide combined estimates of steer daily gains on tall fescue pastures free (E-) of or infested (E+) with A. coenophialum<\em>. Treatments included E- fescue (≤ 5% E + ); moderately infested fescue (≥ 20% to ≤ 35% E+); highly infested (≥ 50% to ≤ 97% E+); and in tall fescue-clover (Trifolium<\em> spp. L.) mixtures, endophyte-free (E- CD, moderately infested (MECL), and highly infested (HECL) at the same E + levels with about 25% and 10% clover in spring and summer stands, respectively. Spring, summer, spring plus summer together, and fall plus winter together datasets were analyzed separately using Henderson's mixed model procedure (MMP). In addition to incorporating the variance components of the random effects into the mixed model equations, mean daily gain estimates were adjusted for the initial weights and steer grazing days ha⁻¹ covariates. Seasonal steer performance generally reflected pasture E+ level and clover incidence. Mean daily gains were variable for the treatment X location combinations, but most of the variation occurred within highly infested treatments. Mean daily gains were comparable for models analyzed with and without the steer grazing days ha⁻¹ covariate, but the standard errors of the means were smaller for those models which included this covariate. The MMP permitted the estimation of the fixed effects of treatments and treatments X locations over a broad inference space of future years and different pastures. Since the combined analysis was able to estimate treatment effects which were not obtainable in each discrete study, combining datasets may be a feasible way to circumvent some of the financial and logistical constraints that force undesirable comprises in the conduct of grazing and other expensive or time-consuming research. The establishment of cooperative projects, using common treatments and identical protocols, would further increase the sensitivity of combined analyses.

Download
Files over 3MB may be slow to open. For best results, right-click and select "save as..."

Share

COinS