Weed fecundity: Current status and future needs

Seed rain is the major input to populations of annual weeds. During the last decade data on seed production have been published for several weed species that accurately reflect the reproductive potential under varying conditions. Many weed species produce thousands of seed per plant, and several species can produce in excess of a million seeds per plant. Knowledge of seed production by a weed over a range of densities, and in proportional mixtures with competing crops or other weed species, allows for predicting seed rain under different cropping conditions. Seed production by weeds at and below the single season economic threshold, and/or for weeds remaining in the system after weed control activities have ceased, provide information useful to making long-term management decisions. Estimating enormous seed rain from high densities of uncontrolled weeds is of limited value in relation to decision making. Serious problems exist with some protocols for assessing weed seed production, limiting the utility of the data generated. A serious problem exists with regionality of data, and with acceptance of incomplete data. Variations in weed seed production suggest that the utility of population models developed in one region is limited to the conditions under which the work was conducted. Long-term experiments are needed to evaluate how changes in management impact seed rain and population dynamics. Development of accurate and reliable data on seed production is essential if weed science is to improve the ecological basis of weed control decision making. Inclusion of the weed seedrain component of weed population dynamics requires a paradigm change.