Spatial and temporal population dynamics of a phyllosphere colonizing Bacillus subtilis biological control agent of sugar beet cercospora leaf spot Douglas P. Collins, a Barry J. Jacobsen, b, * and Bruce Maxwell c a Seattle Tilth Association, Seattle, WA, USA b Department of Plant Sciences and Plant Pathology, Montana State University, P.O. 3150, Bozeman, MT 59717-3150, USA c Department of Land Resources and Environmental Sciences, Montana State University, P.O. Box 3150, Bozeman, MT 59717-3150, USA Received 7 January 2002; accepted 3 October 2002 Abstract This study examines the spatial and temporal variation of populations of the biological control agent (BCA) BacB, a Bacillus subtilis isolate, in the field and growth chamber in the presence of the fungus, Cercospora beticola, the causal agent of Cercospora leafspotofsugarbeet.TheuseoftheselectiveBCAsupportsubstrate b-glucan,appliedat0,0.5,and1.0%ofthespraysolution,did notinfluencedifferencesintotalpopulationnumbers(spores+vegetativecells)ofaspontaneousrifampicinresistantisolateofBacB (Rif+)overa14daysprayperiod.BacBRif+,appliedasasporeformulation,declinedfrom10,000CFU=cm 2 onday0.5–100CFU/ cm 2 on day 14 at the three levels of b-glucan tested. Distribution of BacB Rif+ populations was modeled on a leaf scale, with and without b-glucan.Higherpopulationsofvegetativecellsweremorelikelyat14dayswith1% b-glucanthanwith0% b-glucan. BacB populations were more aggregated without b-glucan than with the nutrient substrate. There was no correlation between BacB density and Cercospora leaf spot disease severity, indicating that neither antibiosis nor parasitism is likely an important mechanism of disease control. Ó 2002 Elsevier Science (USA). All rights reserved. Keywords: Spatial distribution; Population dynamics; Bacillus subtilis; Cercospora leaf spot; Sugar beet phyllosphere ecology; Geostatistics 1. Introduction Bacterial biological control agents have been used in the phyllosphere to control frost injury (Wilson and Lindow, 1994), fire blight of pear (Wilson and Lindow, 1993), postbloom fruit drop of citrus (Sonoda et al., 1996), early leafspot of peanut (Kokalis-Burelle et al., 1992), Cercospora leaf spot of sugarbeet (Kiewnick and Jacobsen, 1998), and other diseases. However, the number of commercialized biopesticides for use in the phyllosphere remains low. Developing a biological control agent (BCA) for the phyllosphere is challenging. The phyllosphere is a relatively harsh environment with a paucity of nutrients and large fluxes of radiation and moisture (Kinkel, 1997). Employing a BCA requires demonstrating consistent disease control, understanding theecologicalmechanismsinvolvedwithdiseasecontrol, and understanding the population dynamics of the BCA, all in the environment in which it is expected to perform. Microbial ecology studies in general, and phyllo- spheremicrobialecologystudiesinparticular,mustdeal with the tremendous variation common on spatial and temporal scales with microbial populations (Beattie and Lindow,1999;Kinkeletal.,1995,1996).Largenumbers of relatively small sample units are necessary for a re- alistic model or understanding of the spatial dynamics of a phyllosphere microbial population. High spatial variation in abundance or aggregation of epiphytic bacterial populations is often, though not always, better described by a lognormal rather than a normal distri- bution of density (Hirano et al., 1982). Spatial distri- bution of BCAs, in relation to their capacities to effect Biological Control 26 (2003) 224–232 www.elsevier.com/locate/ybcon * Corresponding author. Fax: 1-406-994-7600. E-mail address: uplbj@montana.edu (B.J. Jacobsen). 1049-9644/02/$ - see front matter Ó 2002 Elsevier Science (USA). All rights reserved. doi:10.1016/S1049-9644(02)00146-9