NEARLY all soils contain fungi that can penetrate the plant roots and cause vesicular-arbuscular (v.a.) mycorrhiza; most plants become infected with them. Because of the ubiquity and wide host range of the fungi, inoculation as practised for some crops, for example, legumes with nodule bacteria and forest trees with ecto-trophic mycorrhizal fungi, has not been seriously considered for plants with vesicular-arbuscular mycorrhiza. Two recent developments suggest that it deserves consideration. At least six different Endogone species are now known to cause these mycorrhiza1,2 and there is growing evidence that their effects on the host can differ. It has also been established that in some steamed and irradiated soils low in available phosphorus, inoculation with Endogone spores leads to marked increases in growth of up to 300 per cent (see reviews by Nicolson3 and Gerdemann4), In unsteril-ized soils the response to inoculation is usually thought to be much smaller because (a) the added inoculum has to compete with the normal soil microflora, and therefore infection develops more slowly, and (b) even uninoculated plants gradually become infected from the indigenous Endogone population. It nevertheless seemed to be interesting to determine how far growth could be improved in unsterilized soils if seedlings were already mycorrhizal when planted. © 1969 Nature Publishing Group.