Accurately representing plant development is essential for applying phenology knowledgement to investigate the effects of climate on weed management. Development in wild melon (Cucumis melo L.) is driven by temperature; thus, it could be simulated by thermal-time (TT) accumulation using limited accumulation when a lower optimum temperature (Topt) is exceeded. Experiments were conducted to investigate wild melon phenology (development rate) and seed production in soybean (Glycine max L.) at seven different sowing dates (April to August) in a completely randomised design (CRD) at Research Farm of Gorgan University of Agricultural Sciences and Natural Resources, Iran, during 2012. Results indicated that a slight shift in developmental rates occurs among plantings dates, except for those plants sown in August. The estimated TT for April–August planting dates were ~411 Celcius degree days, 448 Celcius degree days, 733 Celcius degree days, 672 Celcius degree days, 604 Celcius degree days, 558 Celcius degree days and 251 Celcius degree days respectively. Depending on planting date, weed emergence occurred at 5–20 days after planting. During the 79, 75, 92, 81, 71, 67 and 61 days of wild-melon growth, the mean number of fruits per plant and seeds per fruit were significantly different at each sowing date. Wild melon could produce a lot of fruits and seeds (up to 5000) within a growth cycle (average in 75 days) and also weed management is needed during the May and June because of the highest seed production of wild melons that emerged during May. The results attained here suggest that temperature alone could not reflect the effect of environment on C. melo development at each given growth stage. Thus, other environmental factors, such as daylength, maybe needed to better estimate weed development. Future research may use multiplicative models to clarify this claim. These results highlighted the value of testing a model over a wide range of environments.
