Swedish University of Agricultural Sciences, Box 7044, 750 07 Uppsala, Sweden

Background and objectives
Wheat dwarf virus (WDV) is a geminivirus with a leafhopper vector (Psammotettix lineatus) [1]. The vector has two generations a year and overwinters as an egg. To serve as a virus vector, the nymph must feed on an infected plant, since the virus is not transmitted to the eggs. Lolium ssp. and Poa ssp. are among the perennial grasses that can serve as virus sources [2]. WDV causes disease in winter wheat. Typical for the disease is empty headings that never get out of the sheath. Seriously affected plants will become chlorotic and never leave the hummock-stage. Dramatic yield reductions have been recorded [3].

In Sweden, serious outbreaks of the disease occurred during the beginning of the century. Prior to 1997, the last epidemic of wheat dwarf was in 1942. Numerous hoppers were observed in 1955, but there were no reports of WDV from this year. WDV was found at seven locations 1996. During 1997 it developed into an epidemic with more than 100 diseased fields. Studies of the epidemiology of WDV are therefore needed. The objective of this study is to investigate the factors promoting outbreaks of WDV.

Materials and methods
The incidence and activity of the vector, as well as the frequency of WDV-carrying vectors during the growing season, were monitored. The hoppers were caught by water and suction traps and tested individually by ELISA. The proportion of plants infected during autumn was monitored likewise. 1500 hoppers and 2500 plants have been tested during 1997. Preliminary data on cultivation practices, such as seed rates, date of sowing and crop rotation with fallow, have been collected from WDV affected farms.

Results and conclutions
Water trap catches, recording the activity of the leafhopper, show that at temperatures under 15C, the vector seems inactive. Nonetheless, from sowing date experiments we know that the vectors are still there and can infect the plants that emerged in the beginning of October, when temperatures around 10C were recorded. Still, data from farmers' fields shows that an early sowing date increased the risk for WDV infection. Under Swedish conditions, an early sowing date is in the last week of August and a late one in the second half of September.

The frequency of WDV carrying hoppers was measured from the end of August to the middle of October. In the first record, from fields sown in 1996, 30% of the hoppers were carrying WDV. Thereafter, from catches in fields sown in 1997, about 5% of the hoppers carried virus throughout the autumn. Records of the proportion infected plants at farms where the disease occurred in 1997, started from the beginning of October when the plants were about ten to fourteen days old. At this time the virus was restricted to the roots. Two to four weeks later, virus could be detected in the upper parts of the plants. By the end of the autumn, 15-30% infected plants were recorded in the field experiments. During spring 1998, outwintering of plants will be monitored, as well as secondary spread of WDV by nymphs during the summer. Information on cultivar responses to WDV and effects of chemical control of the vector will be collected.

Difficulties with field experiments in plots might occur with a disease like wheat dwarf. The vector has great mobility and often the virus has a patchy distribution in field. Complementary studies in farmers fields may therefore be important.

1. Lindsten K, Vacke J, Gerhardson B, 1970. National Swedish Institute of Plant Protection Contribution 14, 281-297.
2. Vacke J, 1972. Vyzkumnych Ustavu Rostlinne Vyroby Praha-Ruzyne 17, 151-162.
3. Moreau JP, Lapierre H, Navarro D, Debray P, Fohrer F, Lebrun I, 1992. Phytoma 443, 21-25.