Below-ground nitrogen transfer between different grassland species: Directquantification by N-15 leaf feeding compared with indirect dilution of soil N-15

Nitrogen (N) transfer from one species to another is important for the N cy cling in low-input grassland. In the present work, estimates obtained by an indirect N-15 dilution technique were compared with estimates obtained by a direct N-15 leaf feeding technique over two complete growing seasons in r ed clover-ryegrass and white clover-ryegrass mixtures under field condition s. The direct technique confirmed that N transfer between clovers and ryegrass is a bi-directional process. The transfer of N from both clovers to ryegra ss occurred within 25 days upon the first labelling event. A very high N tr ansfer occurred from white clover to the associated ryegrass, 4.5 and 7.5 g m(-2) in the 1st and 2nd production year, respectively. The corresponding values for transfer from red clover to the associated ryegrass were 1.7 and 3.6 g m(-)2. Quantified relatively to the total above-ground N content of white clover- ryegrass and red clover-ryegrass mixtures, the N transfer exc eeded 50% and 10%, respectively, in three out of seven harvests. The N tran sfer from N-15 labelled grass to associated clovers constituted a relativel y constant proportion of approx. 8% of the above-ground N content of the mi xtures. Estimates based on the soil N-15 dilution technique generally underestimate d the net N transfer by more than 50% compared to the direct N-15 labelling technique. Furthermore, the indirect N-15 dilution technique estimated onl y marginal differences between red and white clover in the quantities of N transferred, whereas the direct N-15 labelling technique showed the N trans fer from white clover to the associated ryegrass to be significantly higher than that involving red clover. It is concluded that N transfer is a much more dynamic and quantitatively important process in grassland than previou sly recognised.