†Early series treatments denote weed presence at sugar beet emergence;

^‡Late series treatments weed addition after sugar beet emergence;

^§ Weed presence from emergence to harvest (season-long weed presence)

^¶Season-long weed-free treatment

Table 2. Model selection p-values for sugar beet leaf and root measurements from the field study. Null hypothesis is that removal of the fixed effect term (weedy vs non, duration) does not reduce model fit; significant P-values suggest the term should remain in the model. P-values associated with final selected model terms are printed in bold.