From September 1999 to July 2000, N and P concentrations of fine roots were measured with the method of sequential soil core at bimonthly intervals in a mixed forest of Tsoong's tree (Tsoongiodendron odorum Chun) and Chinese fir (Cunninghamia lanceolata (Lamb.) Hook.) in Sanming, Fujian. The results showed that N, P concentration of Chinese fir and Tsoong's tree in fine roots were negatively related to root diameter size. The concentrations of N and P in living roots and dead roots were compared. The order of N concentration in fine roots in different samples was Tsoong's tree>undergrowth>Chinese fir, while that of P was undergrowth>Tsoong's tree>Chinese fir. For Chinese fir, the seasonal change of N, P concentrations in fine roots with various diameter classes showed a single-apex curve with a maximum in September. For Tsoong's tree, maximized concentration of N in fine roots appeared in July or September and maximized P concentration in May.
Based on the measurement of monthly litterfall and their gross calor ic values, the seasonal dynamics of energy return through litterfall were determ ined in a pure and a mixed T. odorum (Tsoongiodendron odorum Chun) forests with Ch inese fir (Cunninghamia lanceolata (Lamb.) Hook.) in Sanming, Fujian Provinc e. Annual ene rgy return through litterfall was estimated as 12.648×10 6J·m -2 for the mixed fo rest, being 4 2% higher than that of the pure forest, and a large proportion of the energy return comprised leaf litter. The conversion efficiency of solar rad i ation energy into litterfall was 0 56% for mixed forest and 0 54% for pure for es t, respectively. The monthly energy flux in litterfall of Chinese fir showed a t hree-apex curve, peaked in March, August and December, respectively, which was s imilar to that in various fractions of leaf, twig, flower and fruit litter. The consistency in monthly patterns among different litter fractions of Chinese fir was attributed to their solid connections all the while. The monthly energy flux in litterfall of T. odorum culminated in January, May and August, the same was true for its leaf and twig litter. However, energy flux in flower litter only oc curred during March to May and that in fruit litter appeared in January and Marc h. The monthly dynamics of energy flux through litterfall of the two forests wer e both determined by their respective litterfall pattern of Chinese fir. Seasona l energy flux in litterfall for both mixed and pure forests followed the sequenc e of spring>winter>summer>autumn, but fluctuations in the former were less disti nct than those in the latter.
