1948 Ecology, 86(7), 2005, pp. 1948–1957  2005 by the Ecological Society of America DAMPING-OFF EPIDEMICS, CONTACT STRUCTURE, AND DISEASE TRANSMISSION IN MIXED-SPECIES POPULATIONS W. OTTEN, 1 J. A. N. FILIPE, 2 AND C. A. GILLIGAN Epidemiology and Modelling Group, Department of Plant Sciences, University of Cambridge, Downing Street, Cambridge CB2 3EA, UK Abstract. In this paper we introduce a method to analyze the inter- and intraspecies transmission rates and contact structure between susceptible and infected plants in epidemics in mixed plant populations. Using spatiotemporal maps of damping-off epidemics caused by Rhizoctonia solani in replicated microcosms of 400 randomly mixed radish and mustard seedlings, we first show that as an epidemic progresses through a mixed plant population it changes the contact structure between infected (I ) and susceptible (S ) plants, so that resistant plants are challenged progressively more frequently, thereby slowing the epidemic. We estimate the transmission rates for each of the four S–I pair combinations. All trans- mission rates rose initially with time and then declined, with overall lower rates for the more resistant mustard species. However, the interspecies transmission rates were unex- pectedly high. We show how these rates relate to relative changes in infectivity and sus- ceptibility of each host species, and how a species that is weakly susceptible in a pure stand can still act as a strong donor of disease in a mixture, thereby lowering the efficacy of the mixture to reduce disease. We discuss the broader consequences for the dynamics of disease in natural and managed populations. Key words: disease control; mixed host populations; primary and secondary infection; Rhizoc- tonia solani; soil-borne diseases; transmission of infection. INTRODUCTION There is wide interest in the spread of diseases through mixed-species populations in order to under- stand invasion in natural communities and the control of disease in mixed agricultural crops. Infections caused by soil-borne pathogens play an important role in regulating dynamic patterns in natural plant com- munities, mediating the rate of growth and the ability to withstand competition (van der Putten 2000). In ag- ricultural environments, crop mixtures are widely ap- plied at various scales of genetic heterogeneity ranging from isogenic lines that differ in a single resistant gene, to mixtures of cultivars or species, and in a variety of spatial patterns (e.g., in rows, or randomly mixed) (Finckh et al. 2000, Mundt 2002). The suppressive ef- fects of mixed-species populations on invasion have been demonstrated in numerous mathematical (Brachet et al. 1999, Garrett and Mundt 1999, Jeger 2000) and empirical studies (Burdon and Chilvers 1977, Garrett and Mundt 2000, Zhu et al. 2000, Mundt 2002). Where- as mathematical studies are ordinarily based on generic epidemiological concepts such as dispersal kernels and transmission rates, the analysis of empirical studies is Manuscript received 15 July 2004; revised 15 November 2004; accepted 24 November 2004. Corresponding Editor (ad hoc): D. M. Rizzo. 1 E-mail: wo200@cam.ac.uk 2 Present address: Department of Infectious Disease, Im- perial College London, St Mary’s Campus, Norfolk Place, London W2 1PG, UK. mostly aimed at various measures of efficacy of a mix- ture. These measures include the disease severity in a mixture relative to single-genotype populations, the ra- tio of areas under the disease progress curve, the delay in the appearance of infection, or the ratio in the daily progress rates (Lannou et al. 1994, Garrett and Mundt 1999, Andrivon et al. 2003). Such measures, however, do not explain the underlying mechanisms driving an epidemic. In this paper we combine both approaches. We derive simple regression relationships from generic epidemiological models, and apply those to epidemics in mixed plant populations. We show how spatial maps of disease progress can be used to obtain empirical estimates of epidemiological processes such as the rates of disease transmission between individual plants. The analysis is an extension of a method previously tested for homogeneous plant populations (Otten et al. 2003, 2004b, Filipe et al. 2004). Here we demonstrate its application to damping-off epidemics in mixed plant populations. As the method is based on generic epi- demiological principles, it can be applied to a range of pathogens and significantly improve our understanding of epidemics in mixed-species populations. We demonstrate the method for the analysis of damp- ing-off epidemics caused by the soil-borne fungal plant pathogen Rhizoctonia solani, a ubiquitous and variable soil-inhabiting fungus (Sneh et al. 1996; see Plate 1). Many isolates of R. solani cause worldwide econom- ically important diseases on many plants including ce- reals, field crops, ornamentals, potato, and forest trees.