Site preparation is a required silvicultural practice for establishing loblolly pine plantations in the Piedmont physiographic region of the southeastern U.S.; however, relatively little is known about its effect on soil and site factors that influence growth and yield. In this study, the effect of seven different site preparation prescriptions on competing vegetation, tree nutrition, and the spatial distribution of planted seedlings was examined. Three treatments had a soil tillage component and differed with respect to the number of operations employed in removing residual trees and logging slash prior to discing. One treatment involved roller chopping and broadcast burning, one an application of the herbicide glyphosate followed by a broadcast-burn, one involved shearing all raking fogging-debris into windows, and one was no site preparation prior to planting. All treatments were applied to 12 sites in the South Carolina and Georgia Piedmont. After four years in the field, 756 trees within the site preparation treatment areas were randomly selected for analysis of foliar nutrients, soil nutrients, and competing vegetation. Herbaceous, woody shrub, and hardwood competition levels were not significantly different among site preparation treatment areas after four years. However, hardwood competition levels had increased at a faster rate during the last two growing seasons on chopped and disced areas than on the other areas. Hardwood competition became the predominant factor limiting pine basal diameter when 83% of the total basal area was in hardwoods or hardwood basal area levels exceeded 3.5 m²ha⁻¹. Potassium was identified as the most limiting nutrient 38% of the time, compared to 28% for phosphorus, 14 % for nitrogen, 7% for calcium, aiid 13% for magnesium. A significant linear relationship between soil and foliar nutrients confirmed these results. No treatment effect on nutrient deficiencies was evident. Foliar nutrient critical levels were derived using the Diagnosis and Recommendation Integrated System (DRIS) and were the same as those reported in the literature for nitrogen, phosphorous, and magnesium while potassium and calcium critical levels were determined to be twice as high as those reported (0.52% and 0.19%, respectively). The spatial distribution of seedlings at planting and after 2 years was determined. Spatial patterns varied from uniform to random as the degree of logging slash removal decreased. After two years, the spatial distribution shifted toward random and clustered. The degree of change was similar in all but the herbicide burn and untreated areas, which exhibited the greatest change and whose mortality tended to be clustered. / Ph. D.