Winter annual cover crops used as green manures and N level effects on cotton (Gossypium hirsutum L.) were evaluated in two identical field experiments during 1990. Treatments in the experiments were cover crops of crimson clover (Trifolium incarnatum L.), Austrian winter pea [Pisum sativum spp. avense (L.) Poir], rye (Secale cereale L.), or winter fallow and four N levels (0, 50, 100, 150 lb N/acre applied in a split application). The experimental design was randomized complete block in split-plot arrangement with cover crops as the main plots. The experiments had four replicates.

Study 1 was conducted on a Norfolk sandy loam and Study 2 was on a Goldsboro loamy sand. All plots were disked twice to incorporate the winter covers on April 26. 'Coker 315' was planted on May 7 and soil moisture was measured at 6 inch intervals to 36 inches two days later. Nitrates in the top 8 inches of the soil were monitored from the end of May through early August and petiole nitrate levels were measured in the youngest fully mature leaves collected weekly beginning at the end of June through August. Cotton population was determined at harvest.

There was no difference between the cover crop treatments in soil moisture at any depth in either soil type. On the Norfolk soil, soil nitrates were greater in the clover and pea winter cover plots (without added N) than the unfertilized winter fallow plots until early August. However, petiole nitrate levels were greater in these plots only in June and early July. Soil nitrate status of the Goldsboro soil was high, with mid-season levels of over 20 ppm occurring in the unfertilized winter fallow plots. Petiole nitrate status of the unfertilized cotton in the clover and pea plots were slightly greater than the unfertilized winter fallow plots. In both soil types, cotton grown on plots that had rye as a cover crop and 0 added N had lower petiole nitrate levels than the unfertilized winter fallow plots, but the petiole nitrate levels of the rye cover crop plots with 50 lbs. added N were similar to the levels of the winter fallow with 50 Cotton stands were slightly reduced in the legume winter cover plots compared to the winter fallow and rye cover crop plots in both soil types. On the Norfolk soil, population was greatest in plots that had a rye cover crop. cotton lint yield plateaued at 50 lbs N/acre in the fallow and rye plots on the Norfolk soil. In that soil, yield from the clover and pea plots without added N were intermediate between the 0 and 50 lb levels in the winter fallow treatment. On the Goldsboro soil, there was no N response in the winter fallow treatment due to the high residual soil N.