G DIJST¹, PJ OYARZUN², SN NEATE³, M DE BRUIJNE¹ and TD VATCHEV⁴

¹DLO-Research Institute for Plant Protection (IPO-DLO), PO Box 9060, 6700 GW Wageningen, the Netherlands; ²CIPP, Quito, Ecuador; ³CSIRO Division of Soils, Adelaide, South Australia; ⁴Plant Protection Institute, Kostinbrod Sofia, Bulgaria

Background and objectives
In Tulip and Iris, Rhizocton1a solani AG 2-t causes severe sprout and bulb rot in the early cold season. At the warmer end of the growth season R. ;solani AG 4 attacks [1]. Disease spreads to patches that differ in size. Similar to some other soil borne fungal diseases [2], we expected that differences in soil suppressiveness account for the differences in disease spread and should be taken into account for site-specific soil management and IPM. Objectives of this study, carried out at IPO-DLO, were to reduce inoculum production by cultural and biological means, to measure and improve the natural soil suppressiveness against disease spread and to predict risks.

Results and conclusions
To detect the target for control, production and survival of AG 2-t was monitored in the field and in storage during and between growth seasons. Reisolation and identification and pathogenspecific IP-testing for infectivity showed that colonized material was not always infective and that the pathogen could not always be recovered from infective material. AG 2-t inoculum or infectivity was recovered from inside bare patches to -45 ;cm deep, and from 15 ;cm (max.) beyond the patches. Biocontrol with Verticillium biguttatum reduced production by 15 ;cm, but did not have a detectable effect on the sporadic survival. The few surviving sources hold a risk, as the fresh new crop stimulates the soil inoculum potential. For commercial predictions these effects need to be modelled and based on better detection.

In a search for prevention of patch formation we studied disease spread under regulated soil water and temperature conditions [1]. Soil suppressiveness was determined in a row-test as spread of bulb rot in Iris by R. ;solani AG 2-t at 12°C and by R. ;solani AG 4 at 18°C. For both pathosystems, more or less the same 30 commercial field soil samples performed as extremely conducive or suppressive. Suppressiveness was more pronounced at 18°C. In contrast to ANOVA, PCA (principal component analysis) yielded a useful differentiation between the natural soil receptivity, taking both disease severity and the type of response into account. Soil suppressiveness increased in time after amending with wheat straw, whereas matured GFT compost was ineffective. Soil disinfestation destroyed the soil suppressiveness and antagonist V. ;biguttatum but not the pathogens. Further addition of straw brought the soil suppression back to its original level. The effects are currently being tested in field plots.

For routine use in risk prediction, a simple indicator is sought to replace the laborious row-test. Disease severity in a one-bulb-test was inadequate, as well as colony extension on different carriers on or in soil. The differences in natural suppressiveness between 30 field soils correlated with combinations of their chemical soil factors. Culturally induced suppressiveness in the bioassay correlated with enhanced microbial activity and biomass. The reliability of such indicators for the field should be determined.

References
1. Dijst G, Schneider JHM, 1996. In: Sneh B, et al. eds. Rhizoctonia Species. Dordrecht: Kluwer, pp. 279-289.
2. Oyarzun PJ, Dijst G, Zoon FC, Maas PVVT, 1997. Phytopathology 87, 534-541.