The effects of herbicide–adjuvant combinations and times of application on weed density and corn yield were determined in a 3-yr field study. At a given rate and time of application of bentazon/atrazine, Assist and Merge were equally effective adjuvants. Assist would be the adjuvant of choice for bentazon/atrazine rather than Merge because of its currently lower cost. Weed density depended on the rate and time of application of bentazon/atrazine. In general, bentazon/atrazine applied at early stages (7 d after emergence) of the development of corn seedlings or at high or intermediate rate (1.6 or 0.8 kg ai/ha) maintained low weed densities, with a relatively small range of variation over years. Delay in time (14 or 21 d after emergence) or reduction in herbicide rate (0.4 kg ai/ha) increased the risk of high weed pressure, although it was not always associated with yield loss. The best time for the application of postemergence herbicide appears to be a few days prior to the onset of the critical period of weed control.

A model has been developed to simulate the annual dormancy cycle of seeds of light-requiring species in the seed bank and the germination of exhumed seeds after irradiation. Simulation of dormancy and germination is based on a physiological model concerning the action of phytochrome in the seed. Dormancy is related to the amount of a hypothetical phytochrome receptor, which fluctuates in an annual pattern. Relief of dormancy is equivalent to an increase in the amount of receptor, and induction of dormancy is equivalent to a decrease in the amount of receptor. Annual changes in temperature are the driving force for annual changes in the amount of phytochrome receptor in seeds that are buried in the seed bank. From the average amount of phytochrome receptor in the seeds of a population, the model calculates the germination percentage that is reached when a seed sample from the population is exhumed, irradiated and incubated at a given temperature in darkness. In the model, relief of dormancy results in a widening of the range of temperatures over which germination can occur, and induction of dormancy results in a narrowing of this range. Model parameters were estimated by fitting the model to data from a burial experiment with seeds of Polygonum persicaria L., Chenopodium album L. and Spergula arvensis L. At regular time intervals during 3 years, subsamples of these seeds were exhumed and tested for germination in the laboratory. The simulation model gave a good description of the observed cyclic changes in germinability of exhumed seeds.