AMMONIFICATION OF COMPLEXES PREPARED FROM GELATIN AND AQUEOUS EXTRACTS OF LEAVES AND FRESHLY-FALLEN LITTER OF TREES ON DIFFERENT SOIL TYPES

The rates of decomposition of polyphenol-gelatin complexes were studie d in two sets of experiments. In the first set, ammonification rates o f complexes formed from aqueous extracts of green, pre-dehiscent leave s and from aqueous extracts of freshly-fallen litter of a range of tre e species and one shrub were determined using a standard soil inoculum and the unspecialized microflora of distilled water. The second set o f experiments investigated the rates of ammonification of complexes pr epared from green, pre-dehiscent leaves of a smaller range of species, using inocula prepared from the soils beneath the trees from which th e leaves had been collected and with the medium buffered to the approp riate soil pH. The quantity of polyphenol-gelatin complex formed diffe red between species, within species between green pre-dehiscent leaves and freshly-fallen litter, and in the case of birch, sycamore, oak, a nd hybrid larch, within species between sites. Aqueous extracts of bot h green, pre-dehiscent leaves and fresh litter of elm, ash, and beech did not form precipitates with gelatin. Six of the 14 complexes formed from aqueous extracts of green, pre-dehiscent leaves were ammonified to some extent by the unspecialized microflora in unsterilized distill ed H2O, which suggests that the complexes have a relatively low resist ance to ammonification. Complexes formed from leaves of hybrid larch a t Grizedale forest, Sitka spruce, western hemlock, and grand fir, rele ased no more mineral N with the soil inoculum than with microbial inhi bitors, which suggests that those complexes are very resistant to ammo nification. Complexes formed from aqueous extracts of freshly-fallen l itter released more N in the presence of the soil inoculum than did co mplexes formed from aqueous extracts of green, pre-dehiscent leaves of the same species. In general, the release of mineral N from the compl exes was fitted well by double exponential regressions, with R2 values between 68 and 94%. From these regressions, the time taken for half o f the N in a complex to be released ranged from 2 days (complex prepar ed from leaves of sycamore on a brown earth of high base status) to 56 weeks (complex prepared from leaves of birch growing on a brown podzo lic soil). The regressions showed that complexes prepared from green, pre-dehiscent needles of western hemlock, Norway spruce, and Sitka spr uce would never release 50% of their N in the experimental conditions.