You need to determine a few things before starting to isolate rhizosphere bacteria: 1- Why do you want to isolate rhizobacteria? to select PGPR, or biocontrol agents? Remember only 1 to 5% of rhizobacteria are PGPR so you need to test a very large number of isolates to find a few intereting ones. 2- To study their diversity? For this aim you can use molecular methods, pyrosequencing combined with bioinformatics can give you very interesting results. Let me remind you a few definitions: a) the surface of roots is what we call therhizoplane (some authors include very very strongly attached particle of soils. b) soil sticking to the roots when you shake them vigourously is what we refer to as the rhizosphere. But an important point frequently neglected is to check if your plant is colonized with mycorhizae or not. If your plant is mycorrhizal then you should talk about the mycorrhizosphere, so most reports refering to the rhizosphere are indeed studying the mycorrhizosphere. You also need to determine which crop you want to study and the area or region targeted. Now the protocole for isolation is relatively simple and unfortunately their is no standard method: For PGPR or biocontrol isolates you will have to visit many fields growing the same crops and to talk with the farmers to obtain permission and to know the cropping history (previous crops, crop rotation etc...) and fertilization history, and you need to sample 3 plants with their roots, and you should choose those that are vigourous. Any information you have on the sampled field and the cultivars planted will be very helpful. Note that for ecological study your sampling should be representative to the field sample and should be done according to an experimental design. The plant roots with attached soils (you can take many subsamples, and don't forget to take roots to color and observe if there is a mycorrhizal colonization) are transferred to the lab on ice in sterile bags. All tools used should be desinfected. Once in the lab transfer to a fridge and process as soon as possible (1-3 days max). Some of my students separate bulk soil from the roots by vigourously shaking the roots in the field, you need to collect the soil for chemical analyses (pH, org. matter, N,P,K, Ca, Mg K). The rest is like any plate count of microbes in soil, but for the first dilution you can put the roots with the attached soil in a fisrt bottle containg 90 ml saline, and then you shake mechanically on a reciprocal or orbital shaker for 15-20 min) In this bottle you will have the rhizospheric or mycorhizospheric microorganisms. If you transfer the roots to another dilution bottle and shake for a longer period of time you will have the rhizoplane bacteria. If you want to enumerate bacteria you need to collect the soil in the first dilution bottle and determine if dry weight (after drying at 1050C). If you want the endophytes you need to surface desinfect root surface before homogenizing with a mortar and pestle. The dilution used should give you single colonies on agar media, for a first isolation you can use 10% Tryptic soy agar with any fungicide to inhibit fungal growth. Now for identification and characterization purpose you need first to indicate what is exactly your aim? Hope this is helpful, wishing you all the best. http://femsec.oxfordjournals.org/content/48/1/1.full http://aem.asm.org/content/64/12/5004.short