2.8.23
TAKE-ALL PROGRESS CURVES HAVE DIFFERENTIAL EFFECTS ON THE YIELD

A SCHOENY1 MH JEUFFROY2 and P LUCASl

1INRA, Station de Pathologie végétale, BP 29, 35653 Le Rheu Cedex, France; 2INRA, Unité d'Agronomie, 78850 Thiverval-Grignon, France

Background and objectives
Take-all, caused by Gaeumannomyces graminis var. tritici, is a damaging disease in all major wheat-production areas and in seasons with favourable climatic conditions, yield losses can be dramatic. There is still a lack of information precisely characterizing the effect of take-all on yield components throughout the cropping season.

This study was undertaken in order to link the epidemiological development of take-all with alterations in the elaboration of yield components of winter wheat.

Materials and methods
Field trials were conducted at Le Rheu (Brittany, west France) in 1995, 1996 and 1997. The five sites were naturally infected second or third winter wheat crops (cv. Soissons). An experimental seed treatment fungicide (nil and high rates) was combined with a nitrogen fertilization (low and high levels) in a block design with three to six replications. Disease assessments were made at 3-week intervals from tillering until grain filling. At each date, a representative sample of 40 plants was collected in each plot. The presence or absence of symptoms (incidence) and the percentage of roots with symptoms (severity) were recorded for each root system. A disease progress model was fitted to the observed data [1]. Final harvest samples were examined for yield components: ear and grain numbers were counted and grain weight was assessed.

Results and conclusions
The crop history-seed treatment combination permitted us to achieve various disease progress curves in terms of incidence and severity. Yield components were compared within the same year for high and low disease levels, respectively, achieved with the nil and high rates of fungicide. Comparisons between different years were made indirectly using a model of yield component elaboration.

Early development of take-all affected both grain number per m2 and 1000-grain weight whereas late development affected only 1000-grain weight. The importance of the losses depended on the severity of the attacks. The benefits of nitrogen fertilization to reduce yield losses due to take-all is also discussed.

References
1. Colbach N, Lucas P, Meynard JM, 1997. Phytopathology 87, 26-32.