Abstract Glasshouse experiments were conducted to study the response of non-host Brassica rapa and host Sorghum bicolor to inoculation with the arbuscular my- corrhizal fungus (AMF) Glomus etunicatum when given different levels of N (0.9 mmol kg –1 sand, 2.7 mmol kg –1 sand, 8.1 mmol kg –1 sand) and P (3.6 μmol kg –1 sand, 10.7 μmol kg –1 sand, 32.0 μmol kg –1 sand) fertiliser. On both plant species, the presence of G. etunicatum inocu- lum (+AMF) was associated with significant changes of shoot δ 15 N values, with +AMF plants having larger aver- age δ 15 N values than uninoculated plants (–AMF). These values are the largest average differences in shoot δ 15 N yet recorded for AMF and nutrient effects. B. rapa shoot δ 15 N average differences ranged from 1.67‰ to 2.70‰, while for S. bicolor they range between 2.07‰ and 4.40‰. For shoot δ 13 C only the non-host B. rapa re- sponded to ±AMF and added N. Although the harvested dry weight biomass (–35.2% B. rapa; +39.8% S. bicolor) of both plant species responded to AMF inoculation, no direct relationship was observed between isotopic dis- crimination and growth inhibition for the non-host B. rapa. In this paper we discuss some implications regard- ing AMF inocula on the basis of our findings and current literature. Keywords Arbuscular mycorrhiza · Non-host · Nitrogen-15 natural abundance · Carbon-13 natural abundance · Brassica rapa Introduction The plant genus, Brassica, belongs to one of the few ma- jor plants groups, Cruciferae (Capparales), that generally do not form mycorrhizal symbioses with Glomales fungi (Gerdemann 1968). Most work addressing the interac- tion between arbuscular mycorrhizal fungi (AMF) and Brassica species focuses on either the non-host impact on co-cultured arbuscular mycorrhizal host plants and/or on AMF development in host-fungal symbioses. Infor- mation about the direct impact of AMF on non-hosts has not been sufficiently explored to generalise about the consequences of this. Koide and Li (1989) reported an initial depression of growth when B. hirta was given a soil inoculum containing mycorrhizal propagules. Manjunath and Habte (1992) observed a reduction in biomass production when B. nigra was cultured with low amounts of added P and inoculated with AMF. AM hosts normally show an improved nutritional state when inoculated with AMF, and this is thought to occur via an improved capacity to take up soil nutri- ents with a low mobility, especially P (Rhodes and Gerdemann 1980). AMF are also involved in the acquisi- tion of the highly water-soluble N forms, such as NO 3 – and NH 4 + (e.g. Smith and Smith 1990). The use of stable isotope natural abundances to ex- plain plant-AMF interactions is in the empirical phase of gathering information for the design of hypotheses (Handley et al. 1993, 1999; Azcón-G.-Aguilar et al. 1998), but has shown already that the colonisation of non-diazotrophic plants by AMF changes the δ 15 N val- ues of host plants, when compared with non-mycorrhizal treatments, and that there are interactions with abiotic factors such as drought and N supply. The problems in- herent in designing such experiments, however, preclude easy mechanistic explanations. Microbial contamination is one such problem. The implementation of large experiments requires the pro- duction and isolation of large quantities of homogene- ous and viable single-species AMF inocula, making it technically impractical to use axenic inocula. To our H.M.A.C. Fonseca ( ✉ ) Centro de Biologia Celular, Departamento de Biologia, Universidade de Aveiro, 3810-193, Aveiro, Portugal e-mail: hfonseca@bio.ua.pt Tel.: +351-23-4370772, Fax: +351-23-4426408 R.L.L. Berbara Universidade Federal Rural do Rio de Janeiro, Departamento de Solos, Seropédica, Itaguaí, RJ, CEP 23851-970, Brazil M.J. Daft School of Life Sciences, University of Dundee, Dundee, DD1 4HN, Scotland Mycorrhiza (2001) 11:151–158 DOI 10.1007/s005720100125 ORIGINAL PAPER Henrique M.A.C. Fonseca · Ricardo L.L. Berbara Melvin J. Daft Shoot δ 15 N and δ 13 C values of non-host Brassica rapa change when exposed to ± Glomus etunicatum inoculum and three levels of phosphorus and nitrogen Accepted: 16 May 2001 / Published online: 10 July 2001 © Springer-Verlag 2001