This experiment was conducted to determine whether light conditions during cultivation affect the number or species composition of emerging weeds. Plots were cultivated on 31 May 1996 under the following light conditions: 1) Ambient nighttime light (full moon) plus tractor headlights, 2) ambient nighttime light plus tractor headlights covered by transparent green film (green tractor lights), 3) ambient nighttime light with no tractor headlights, 4) ambient mid-day light. A fifth treatment was not cultivated at all, but was treated with glyphosate on 31 May to kill emerged plants. Results discussed below are based on weed densities determined 3 weeks after cultivation. Cultivated plots had more broadleaf weeds than uncultivated plots (119/m2 vs. 40/m2). Annual grass populations were the same in cultivated and uncultivated plots (20/m2). Common lambsquarters (Chenopodium album) and carpetweed (Mollugo verticillata) were the most common annual broadleaf weeds, representing more than 80% of the annual broadleaf population in cultivated plots. Light conditions during cultivation did not influence the total number of broadleaves, number of annual grasses, or total number of weeds emerging. Lambsquarters emerged in greater numbers after daytime or night cultivation using green tractor lights (48/m2) than after night cultivation using standard tractor headlights (32/m2). These results suggest no practical benefit to night cultivation for reducing overall weed density. A similar study in 1995 led to a similar conclusion. Although measurable effects of light conditions on weed emergence were observed in both years, the magnitude and consistency of the effects were not enough to suggest changes in cultivation practices without further research.