Improving risk assessment and control of saddle gall midge (Haplodiplosis marginata)


Cereals & Oilseeds
Project code:
01 February 2013 - 31 March 2016
AHDB Cereals & Oilseeds.
AHDB sector cost:
Total project value:
Project leader:
Steve Ellis ADAS, High Mowthorpe, Duggleby, Malton, North Yorkshire, YO17 8BP


pr568-final-project-summary pr568-final-project-report

About this project


Saddle gall midge is a sporadic, but a periodically and locally-important, pest of cereals in the UK. Severe, widespread outbreaks occurred in continuous cereals on heavy land in 2010 and 2011 with yield losses in the most severe cases reaching 70%. The occasional nature of the pest in the UK means that experience of the problem is minimal. The aims of this project were:

1. To monitor midge development in relation to meteorological data to improve understanding of the pest’s life-cycle and to facilitate improved forecasting of outbreaks.
2. To measure the impact of midge infestation on yield of wheat and barley. 
3. To evaluate the efficacy of a range of timings and products for midge control.
4. To use data from objectives 1–3 to propose provisional thresholds for saddle gall midge.

Soil sampling was an effective method of assessing levels of saddle gall midge and monitoring its development. Detection of midge pupation provides an early warning of adult emergence and is an important component of risk assessment for the pest. A fungal parasite Lecanicillium spp of midge larvae was identified and had a dramatic effect on larval development, such that few became pupae. Yellow water traps were more effective than yellow sticky traps or emergence traps at catching saddle gall midge adults. In collaboration with Harper Adams University, developmental models using soil and air temperature were evaluated and predicted the timing of pest emergence to within eight days. Also prototype pheromone traps were tested and were very effective at trapping adult male saddle gall midge.

There was no clear relationship between larval infestation and crop yield, probably due to the low level of pest infestation throughout the project. Chemical control of saddle gall midge reduced tiller infestation by up to 92% but had no impact on crop yield, indicating that damage does not always equate to loss of yield. Sprays targeted at the first appearance of saddle gall midge adults were generally most effective at reducing pest infestation but those targeted at larvae were ineffective. Lamda-cyhalothrin and thiacloprid both gave good control of the pest. A single spray of lamda-cyhalothrin generally resulted in the lowest levels of pest infestation.

Levels of saddle gall midge were generally too low to make any progress on the development of threshold and there was no evidence to suggest that larval threshold of greater than 500 larvae/m2 should continue to be used until experimental evidence suggests otherwise.

A basic IPM strategy is proposed for the pest which takes into account individual field risk, pest numbers, monitoring of pest development and the timing of adult emergence to determine the need for insecticide treatment. The project should help to reduce the unpredictability of saddle gall midge attack and help to minimise unnecessary insecticide treatment against the pest.