This research presents a modified floatation method to extract weed seed from soil samples and an automated imaging technique to enumerate seed following extraction. Rapid floatation of weed seed from soil in a K2CO3 solution was facilitated by centrifugation. High recovery of seed from soil samples was accomplished in a relatively short time. Recovery of seed of giant foxtail and velvetleaf from a silt loam soil was near 100% and recovery of redroot pigweed was 86%. Germination of common lambsquarters, giant foxtail, and velvetleaf seed was reduced by exposure to K2CO3 under certain conditions. However, these reductions do not appear to reduce the utility of this extraction technique. The automated image capture technique for counting seed showed similar limitations as with manual counting. Seed count accuracy generally decreased with increased number of seed and smaller seed sizes. In general, the image capture technique compared favorably with manual seed counting and has the potential to further improve seedbank assessment methods.

Laboratory, greenhouse, and field studies were conducted to determine the allelopathic potential of lantana and Siam weed on the germination and growth of five crops. Emergence and dry weight of bioassay species were affected when lantana or Siam weed debris was present on the soil surface or incorporated into the soil. However, presence of Siam weed on the soil surface did not affect dry weight of rape. Emergence and dry weight of crops (except emergence of spinach) were not affected when crops were grown in soil samples collected from beneath field-grown lantana. Dry weight of Chinese cabbage and chilli was reduced when these crops were grown in Siam weed-contaminated soil. Germination of Chinese cabbage, chilli, and rape decreased progressively when exposed to increasing concentrations of aqueous lantana extract. However, the lantana extract at full strength (66.7 g L−1) did not reduce germination of spinach and cucumber seed. Siam weed extract, when applied at full strength to seed of spinach, Chinese cabbage, rape, and chilli, reduced germination by 10, 12, 21, and 19% of control, respectively. Full-strength extract of lantana and Siam weed decreased seedling length and fresh weight of all crops.

Knowledge of timing and extent of weed emergence before and immediately after crop seedbed preparation is needed to decrease need for preplant herbicides and increase efficacy of postemergence weed control in crops with either mechanical or chemical methods. Such knowledge is important for weeds that infest most crops over a wide area. For these reasons a mechanistic seedling emergence model based solely on soil temperature was developed for common lambsquarters. The model was validated using four sets of field data collected in 1988, 1990, and 1991 near Morris, MN. Agreement of predicted and observed emergence values across all site-years was 0.95 and the coefficient of determination (R2) was 0.98 (P < 0.001). Agreement for individual site-years was 0.96, 1.08, 1.08, and 0.98 and associated R2 values were 0.99, 0.99, 0.99, and 0.98 (P < 0.001 for each site-year), indicating close agreement between predicted and actual emergence values.