VIEW ARTICLE

Ecology and Epidemiology

Effect of Inocula Delivery Systems on Rhizobacterial Colonization of Underground Organs of Potato. J. B. Bahme, Graduate research assistant, Department of Plant Pathology, University of California, Berkeley 94720; Present address: AgBioChem, Inc., 10795 Bryne Avenue, Los Molinos, CA 96055; M. N. Schroth(2), S. D. Van Gundy(3), A. R. Weinhold(4), and D. M. Tolentino(5). (2)Professor, Department of Plant Pathology, University of California, Berkeley 94720; (3)Professor, Department of Nematology, University of California, Riverside 92521; (4)Professor, Department of Plant Pathology, University of California, Berkeley 94720; (5)Research plant pathologist, Instituto Guido Donegani, Novara, Italy. Phytopathology 78:534-542. Accepted for publication 19 October 1987. Copyright 1988 The American Phytopathological Society. DOI: 10.1094/Phyto-78-534.

The effect of delivering Pseudomonas fluorescens strain A1-B to soil by incorporation of bacteria-impregnated granules and injection through a low-pressure drip irrigation system on the resultant colonization patterns was determined at all stages of plant development on below-ground parts of field-grown potato in two different soil types. Strain A1-B was released from granules during irrigations and persisted at mean populations of approximately 10² cfu g^-1 in nonrhizosphere soil beneath the zone of granule incorporation. Populations of strain A1-B in the granules declined from an initial density of 10⁹ cfu g^-1 to 10⁵ by harvest time. Periodic application of strain A1-B through the drip irrigation system resulted in passive distribution of the bacteria throughout the top 30 cm of beds and established season-long mean population sizes of approximately 10⁵ cfu g^-1 in nonrhizosphere soil. Soil population densities of drip-introduced strain A1-B gradually increased during the season in silty clay loam but decreased in sandy loam. Mean rhizosphere population sizes (log cfu cm^-1) of strain A1-B on root segments 0–8, 8–16, and 24–32 cm distal to points of root origin in granule plots were 3.4, 4.4, and 0.7, respectively, in sandy loam, and 3.1, 2.2, and 1.4 for the same respective segments in silty clay loam. Rhizosphere populations on these respective root segments in drip plots were 4.1, 3.9, and 1.6 in sandy loam and 3.7, 2.8, and 2.2 in silty clay loam. The population densities and uniformity of colonization were much greater with granule and drip delivery systems than with seed-piece inoculation. Mean root tip population sizes (log cfu per tip) of strain A1-B on roots of 8 and 16 cm in total length were 3.3 and 0.7, respectively, in granule plots and 3.8 and 1.2 on roots of the same respective lengths in drip plots. Introduction of strain A1-B to soil by granule and drip delivery systems also resulted in large, relatively uniform populations of the bacterium on progeny tubers. Populations of strain A1-B in nonrhizosphere soil in drip plots accounted for 1 and 630% of the total detectable aerobic bacteria and native fluorescent pseudomonads, respectively. Rhizosphere populations of granule-introduced and drip-introduced strain A1-B on root segments adjacent to seed pieces constituted 0.22 and 0.17%, respectively, of the total detectable aerobic bacteria of these segments and 35 and 27% of the fluorescent pseudomonads for the same respective systems of delivery. Population densities of drip-applied strain A1-B were approximately 1 log unit greater in nonrhizosphere soil and on roots and progeny tubers in plots preplant-fumigated with metham sodium. Drip-irrigation injection of conidia of Trichoderma viride resulted in distribution of the fungus throughout beds and established season-long mean nonrhizosphere soil population sizes of log 3.5 cfu g^-1 in silty clay loam and log 1.0 cfu g^-1 in sandy loam.