ABSTRACT
Summary Sorghum and Digitaria cultivars grown in soil or nutrient solutions were inoculated with a strain of Azospirillum in short-term greenhouse experiments. Carbon compounds lost by roots in solution culture experiments were determined in order to estimate the quantity of energy substrates that might be obtained by rhizosphere organisms from growing plants. Other plants were treated with plant growth hormones, and responses were compared to those of plants treated with killed Azospirillum cells. The similarity of the responses suggests that growth-hormone production by the inoculant organism, rather than nitrogen (N) fixation, is responsible for part or all of the increase in plant growth rate. The estimated quantity of reduced carbon compounds (lysate) lost by roots to the nutrient medium would support substantial N-fixation activity if all were used for that purpose, but that clearly is not the case. The available evidence indicates instead that relatively little may be so used. Thus, the recognized low N-fixation efficiency of free-living heterotrophs and the demonstrated response of plants to bacterial growth hormones suggest that benefits from inoculation, even though they may be highly important in some circumstances, may not always involve N fixation.
