In an experiment on permanent grass at Rothamsted during 1975–9 fertilizer-N was applied each year either by injecting an aqueous solution of urea (supplying 250, 375 or 500 kg N/ha) in spring, or by broadcasting ‘Nitro-Chalk’ granules (supplying 100, 200, 300, 400 or 500 kg N/ha) in six equal dressings for each of six cuts.
Dry-matter production was largest on plots injected with urea through knives 30 cm apart, and more N was recovered from the injected than from the broadcast applications. Aqueous urea injected at the 60 cm knife spacing nitrified more slowly and persisted in the soil longer than urea injected at 30 cm spacing; this persistence caused grass to grow more uniformly throughout the season, but yields were less. Injecting the nitrification inhibitor sodium trithiocarbonate (STC) with the aqueous urea postponed N uptake much less than doubling knife spacing, but the inhibitor substantially diminished percentage N03-N in harvested grass.
In spring 1977 individual plots were split to measure N residues. Half-plots thus received N at the specified rates, either in 4 successive years (1975–8) or in two pairs of successive years (1975 and 1976; 1978 and 1979). In 1977 urea injected in the 2 previous years gave large residual effects, which were increased by STC and also by injecting in bands 60 instead of 30 cm apart. Broadcast ‘Nitro-Chalk’ had much smaller residual effects. In 1979 residual effects of N applied in the 4 previous years were apparently small, regardless of the method of application, because clover became abundant on plots not given N.
In 1978 dry-matter production was smaller where N had been given each year during 1975–8 than where N was withheld in 1977. Analysis showed this was caused by a shortage of potassium. This effect was most pronounced where 375 or 500 kg N/ha had been injected in bands 60 cm apart.
The results showed that a single, injected application of aqueous urea increased yields of dry grass as effectively as equivalent repeated dressings of ‘Nitro-Chalk’.