High variation in the percentage of root length colonised by arbuscular mycorrhizal fungi among 139 lines representing the species subterranean clover (Trifolium subterraneum)

• We examined colonisation of subterranean clover by arbuscular mycorrhizal fungi.
• Colonisation was 12–68% of root length across 137 lines representing the species.
• Differences in colonisation between two cultivars were consistent across 11 soils.
• Key root traits for phosphorus-uptake changed little when the fungi were present.
• Potential may exist to develop cultivars with differing mycorrhizal colonisation.

Development of more phosphorus (P)-efficient pasture legumes could improve the P balance of pasture systems. Trifolium subterraneum (subterranean clover) is the most widely-grown annual pasture legume in southern Australia and is host to arbuscular mycorrhizal fungi (AMF). To examine the potential to alter colonisation level through breeding, 97 lines from the core collection of subterranean clover (core lines) and 42 cultivars (cultivar lines) were grown in a glasshouse in a low-P sandy acid field soil with indigenous AMF (Experiment 1). To examine the robustness of rankings, two cultivars (Riverina and Woogenellup) were also grown in 11 field soils (Experiment 2). To examine whether the presence of AMF changed other root traits important for P uptake, cultivars Riverina and Woogenellup were grown in two low-P field soils (sand, sandy loam), either pasteurised to remove indigenous AMF or unpasteurised, and with and without addition of inoculum of AMF (Experiment 3). The range of colonisation for core lines and cultivar lines was similar (∼12–68% of root length), however, the core lines had higher mean colonisation (37% vs 33%) and a greater proportion of lines with >40% of root length colonised (41% vs 24% of lines). Woogenellup was more heavily colonised than Riverina in eight soils and had a similar level of colonisation in three soils. In the sand, pasteurisation resulted in shoot DM increasing by up to 100% (possibly due to reduction in root diseases). Inoculation, in pasteurised and unpasteurised soil, resulted in an increase in shoot DM, shoot P concentration and shoot P content of ∼100%, increased average root diameter by up to 13%, decreased specific root length by up 30% and decreased root mass ratio; effects on average root hair length were variable. In the unpasteurised sandy loam, shoot P concentrations were higher than in the sand, and inoculation slightly increased shoot P concentration, but not shoot DM; the effects of inoculation on root traits were mostly small. We conclude that potential may exist to select subterranean clover for higher or lower colonisation by AMF using cultivar lines. Also, indigenous AMF may be less effective at enhancing plant P uptake and growth than inoculants. Finally, key root traits for enhancing plant P-uptake may change in the presence of AMF, but not sufficiently so to warrant deliberate inclusion or exclusion of AMF when screening germplasm for such traits.