Abstract A method is described in which 1 year-old chestnut coppice was fed in situ with air highly enriched in 13CO2 (23%). After 3 days, 13C concentration increases in shoots were measured by mass spectrometry. Respiratory losses between 13C feeding and harvest were estimated using two different methods: (i) a model involving the temperature response of respiration and (ii) direct measurement of 13C content of the CO2 respired by the shoots during the night.
Carbon allocation to roots was deduced by subtracting from the given amount of 13C, the amount remaining in shoots and the 13C respired by the shoots.
The method was tested twice during the growing season. Very little carbon was allocated to roots in late July, but over 80% of assimilated 13C went to roots at the end of September. Despite some approximations in the 13C respiratory losses estimations, the method allowed evaluation of carbon allocation to roots with an error of about 5%.