1Harper Adams Agricultural College, Newport, Shropshire TF10 8NB, UK; 2Novartis Agriculture, Whittlesford, Cambridge CB2 4QT; 3HRI East Malling, West Malling, Kent ME19 6BJ, UK.

Background and objectives
Fusarium seedling blight of wheat in the UK is caused predominantly by Microdochium nivale and Fusarium culmorum. Contamination of UK wheat seed has increased in frequency over the last 30 years and in a 1992/93 survey of wheat seed, all samples tested were contaminated by M. nivale [1]. Assessment of samples is routinely undertaken by measuring the percentage of seed infected. This is a relatively simple test, but it usually only detects the fastest growing pathogen present within each seed and it takes no account of the inoculum concentration within individual seeds. Recent advances in PCR-based diagnosis of plant pathogens offer an hitherto unavailable opportunity to diagnose and quantify pathogens associated with Fusarium diseases of wheat whether symptoms are visible or not [2]. These techniques are being used to identify and quantify pathogen inoculum on seed to assist in the interpretation of fungicide seed treatment efficacy data. This will facilitate a fuller understanding of the relationship between inoculum density of each pathogen present, disease severity and yield loss in a range of environments.

Materials and methods
Plots of winter wheat cv. Hussar were inoculated with F. culmorum, M. nivale var nivale or M. nivale var majus at mid-anthesis and mist irrigated for two weeks. After harvest, seed was dried to 16 % moisture and cleaned. Four seed lots were chosen for a subsequent seed treatment trial. These were (a) commercially obtained Hussar, (b) seed from a misted/ M. nivale var majus treated plot, (c) seed from a misted/ F. culmorum inoculated plot and (d) seed from a misted/uninoculated plot. Seed lots were assessed for pathogens present by standard plate counts, quantitative PCR and diagnostic PCR from single seed extractions. For plate counts; 200 seeds were surface sterilised and placed on antibiotic amended Potato Dextrose Agar. After 7-14 days incubation at 13 C numbers of seed contaminated with M. nivale, F. culmorum and other Fusarium spp. were counted. For quantitative PCR; 5 g of each seed lot was freeze-dried, crushed and DNA extracted using CTAB. DNA of F. culmorum and the two subspecies of M. nivale was quantified using specific primers in the presence of competitive fragments. For diagnostic PCR; 60 single freeze-dried seeds from each batch were crushed and DNA extracted using CTAB. DNA of F. culmorum and the two subspecies of M. nivale was amplified using specific primers. DNA extraction and PCR was based on the method of Nicholson et al. [2].

Results and conclusions
Percentage of seeds infected with M. nivale and F. culmorum ranged from 17.5 to 54% and 1.5 to 67% respectively. The amount of pathogen DNA within the extracts ranged from 2.6 to 5.2 ng/Ál for M. nivale var nivale, 1.6 to 32.4 ng/Ál for M. nivale var majus and 0 to 90 ng/Ál for F. culmorum. Percentage seeds infected as determined by diagnostic PCR on individual seeds ranged from 2 to 55% for F.culmorum, 5 to 31% for M. nivale var majus and 7 to 14% for M. nivale var majus. Seventeen % of the F. culmorum inoculated seed contained both F. culmorum and M. nivale, of these, 3% contained both subspecies of M. nivale. Results showed that pathogen DNA was successfully amplified from both single seeds and bulk batches of wheat. The percentage infection determined by PCR of single seeds was lower than that by plate counts but PCR allowed the pathogens within individual seeds to be readily identified and quantified. Seed treatment efficacy data (percentage emergence, percentage establishment and seedling disease indices) will be evaluated in respect to the inoculum of Fusarium pathogens present on seed batches as determined by these three methods.

1. Reeves JC, Wray MW, 1994. BCPC Monograph No.57 - Seed Treatment: Progress and Prospects, pp.37-46.
2. Nicholson P, Lees AK, Maurin N, Parry DW, Rezanoor HN, 1996. Physiological and Molecular Plant Pathology 48, 257-71.