© 1999 Kluwer Academic Publishers. Printed in the Netherlands. 127 Contribution of Lolium perenne rhizodeposition to carbon turnover of pasture soil

Carbon rhizodeposition and root respiration during eight development stages of Lolium perenne were studied on a loamy Gleyic Cambisol by 14CO2 pulse labelling of shoots in a two compartment chamber under controlled laboratory conditions. Total 14CO2 efflux from the soil (root respiration, microbial respiration of exudates and dead roots) in the first 8 days after 14C pulse labelling decreased during plant development from 14 to 6.5 % of the total 14C input. Root respiration accounted for was between 1.5 and 6.5 % while microbial respiration of easily available rhizodeposits and dead root remains were between 2 and 8 % of the 14C input. Both respiration processes were found to decline during plant development, but only the decrease in root respiration was significant. The average contribution of root respiration to total 14CO2 efflux from the soil was approximately 41%. Close correlation was found between cumulative 14CO2 efflux from the soil and the time when maximum 14CO2 efflux occurred (r=0.97). The average total of CO2 efflux from the soil with Lolium perenne was approximately 21 g C-CO2 d1 g1. It increased slightly during plant development. The contribution of plant roots to total CO2 efflux from the soil, calculated as the remainder from respiration of bare soil, was about 51%. The total 14C content after 8 days in the soil with roots ranged from 8.2 to 27.7 % of assimilated carbon. This corresponds to an underground carbon transfer by Lolium perenne of 6–10 g C m2 at the beginning of the growth period and 50–65 g C m2 towards the end of the growth period. The conventional root washing procedure was found to be inadequate for the determination of total carbon input in the soil because 90 % of the young fine roots can be lost