Grasses can be very good hosts as well as onion family composites and legumes. If you use mycorrhizae free perlite, vermiculite, cocopeat media you can plant single seeds of interest and from recovered sporangiospores about 100 will conclusively inoculate your strain. In your growout in isolated bags the strain identity can be kept and in a growth cycle millions of sporangia will result. Look to David Douds and collaborators for articles on simple on farm propagation systems. Once you have your increase the strain can be mixed with mycorrhizal free material and the plants can be tested with and without. Check out David Douds on Google scholar and the methods are well described.

When propagating mycorrhizal fungi it is important not to have a high Phosphorus level as this will make the plants less susceptible. Natural areas of vegetation in the area of your interest may have a variety of mycorrhizae to start of with. While mycorrhizal fungi are not specific in their relationship on different hosts the distribution of fungi shift and not all strains are as useful for the growth promotion. In inoculants many researchers have used mixtures of strains in order that some will adapt to soil and host combination.

To propagate a wide variety of mycorrhizae I would base the media on perlite vermiculture and use 20% good compost. I would use sorghum or maize, onion, clover and sunflower as my host mixture. I would use natural grasslands of the soils you are interested in a relatively untouched perennial area and use soil samples to mix in your mixture. This should capture a diverse mixture of beneficial species and strains. Make sure not to have overly rich Phosphorus so you do not turn off the plant susceptibility.

Mycorrhizal inocula for commercial scale can be produced using following steps

1. Select a marginal area, not highly fertile, as the inocula consists of soil and root debries of host plant, which needs to be lifted, dried and packed. Soils with moderate to low phosphorus content yield better spore count.

2. Use a fast growing, highly fibrous rooted host plant, we used Eleusine coracana or finger millet for many Glomus species.

3. Mixing of some porous soil additives such as cocopeat (obtained from the powdery coir pith of coconut husk, a waste product of coconut husk fiber industry), which is way cheaper than other porous additives such as perlite.

4. Host plant can be grown till inflorescence opening or about 8-10 weeks, spore count can be checked by wet sieving of the dry soil and by trypan blue staining of host roots. Sufficient spore count 400-700 per 10 g soil could be obtained easily.

5. After the growth period, the host plant is cropped close to the soil and the soil along with host roots is air dried and packed and used as VAM/AM fungal inocula. Large quantities of bulky inocula can be obtained from as little as 4-5 square meters of area.