Glasshouse experiments with Ricinus communis showed that the presence/absence of a VA mycorrhizal fungus (Glomus clarum) changed the δ15N value of the host by as much as 2‰ when the plants were given urea (released as NH4+) as their only N‐source. This small change in Δ15N would create a large error in calculating sources of plant N. In particular, these results throw into doubt any models of N‐cycling which assume that soil N can be treated as a single source. The correct N‐source value for VAM‐infected NH4− ‐using plants may be the δ15N of soil NH4++ 2‰. Treatment effects were also found in the distribution of δ15N and % N among plant organs. Plants with VAM had a lower N:P atom ratio and were larger in total biomass. Carbon discrimination (δ13C) was greater in the VA‐infected plants. The measured effects of VAM infection suggest that for some plants the fungus may be the primary site of N assimilation. A parallel experiment with Eucalyptus globulus and the ectomycorrhizal fungus Hydnangium carneum resulted in no significant differences in any of the variables measured for this host‐fungus pair when the sole N‐sources were inorganic (NO3− and NH4+ released from urea). Ectomycorrhizal fungi are diverse in their physiological behaviour, and these data should not be taken as being representative of the whole group. More work is required with other types of mycorrhiza and more complex sources of N. Future work will include a water balance to partition the effects of water use and nutrient supply in determining δ13C. An on‐line combustion‐ANCA‐MS method is described for fully automated measurement of natural abundance levels of 15/14N and 13/12C for plant materials. This method achieves the required precision while dramatically increasing sample throughout.

No abstract

17Both inoculants greatly improved the growth of the trees and colonization spread to the crops once 18 the trees were 6 months old. However, benefits of inoculation to crop growth were not observed due 19 to increased competition from the larger inoculated trees growing in a restricted soil volume. Of the 20 two inoculant fungi, Glomus etunicatum appeared to be more mobile as it spread more rapidly, 21 formed higher levels of colonization at increasing distances from the tree and was responsible for 22 most of the mycorrhizal cross-contamination. In contrast, colonization of tree and crop roots by 23 Gigaspora albida was higher nearest the tree. This work demonstrated the benefits of mycorrhizal 24 fungus inoculation for tree growth and confirmed that trees and crops share the same AM fungi. 25Trees may therefore act as reservoirs of mycorrhizal fungi, either inoculant or indigenous, for 26 surrounding crops or other annual vegetation. It was also shown that tree pruning, the normal 27 2 practice in agroforestry systems, did not reduce mycorrhizal infection or prevent spread to crops.

No abstract

No abstract

No abstract

No abstract

No abstract