Re-evaluating thresholds for pollen beetle in oilseed rape

Project Report No. 495
Re-evaluating thresholds for pollen beetle in oilseed rape
Stephen Ellis and Peter Berry
ADAS, High Mowthorpe, Duggleby, Malton, North Yorkshire YO17 8BP


The aim of this project was to produce up-to-date thresholds for pollen beetle control by relating the potential for pest damage to the inherent tolerance of the crop to pest damage. The arrival of insecticide resistant pollen beetle in the UK makes it imperative that treatments are not applied unless really necessary to protect yield. The project hypothesises that oilseed rape crops produce significantly more flowers than required to produce the optimum pod number to maximise yield so there are excess flowers, which could be sacrificed to pollen beetle attack before yield is lost.

The research has resulted in a number of conclusions which have a significant impact on our understanding of the pest/host relationship between pollen beetle and oilseed rape:

  • A single pollen beetle is capable of destroying an average of nine buds. This information was previously unknown and is pivotal in determining thresholds for the pest.
  • Winter and spring rape crops produce a similar number of excess flowers suggesting no inherent difference in their tolerance to pollen beetle attack. This contradicts current understanding.
  • There is no difference in the number of excess flowers between hybrid and conventional varieties contradicting the perceived wisdom that hybrid varieties are potentially more susceptible to pollen beetle damage
  • Sparse crops have a greater tolerance to pollen beetle attack than more dense plantings. This result appears counter-intuitive but is supported by the fact that crops with fewer plants/m2 had more excess flowers per plant
  • Plant number and GAI (green area index) are potentially good indicators of excess flower number
  • A dynamic threshold scheme is suggested in which the treatment threshold is no longer a single value for all crops. Instead it varies in relation to the number of excess flowers produced by different varieties in different seasons. This is an important change in the developmental approach to thresholds which has potential for application to other pest/crop interactions.

An experiment in which a range of beetle populations (0, 5, 10, 15, 20 and 50/plant) were confined on rape plants suggested that a single beetle on average is capable of destroying nine buds. Excess flower numbers (flower numbers minus pod numbers at harvest) were assessed in a range of hybrid and conventional spring and winter rape varieties sown at a range of seed rates (10 to 200 seeds/m2).

Spring oilseed rape crops produced a similar number of excess flowers to winter oilseed rape crops, which indicates that they are potentially equally tolerant of pollen beetle attack. This is a significant change from current advice that suggests that spring crops are inherently more susceptible to pollen beetles than winter crops.

Hybrid, open pollinated and semi-dwarf varieties produced a similar number of excess flowers suggesting they are also potentially equally tolerant of pollen beetle attack, although there were significant differences between specific varieties e.g. Castille had relatively few excess flowers.

Again this result contradicts the perceived wisdom which suggests that hybrid crops are more susceptible to pollen beetle because they are sown at lower seed rates than conventional varieties. Crops with fewer plants/m2 had more excess flowers per plant than more dense crops, suggesting that thin or 'backward' crops may not be as susceptible to pollen beetle attack as initially thought.

The project has demonstrated that it is possible to predict variation in the number of excess flowers per plant within a season from measurements of plants/m2 or GAI (green area index) at green bud. Both parameters showed strong negative relationships with excess flowers per plant.

However, there were large seasonal differences in excess flower number and further work is required to predict seasonal variation over a number of years. A conceptual pollen beetle threshold scheme has been proposed in which the pollen beetle control threshold is negatively related to plants/m2.

This implies that the threshold will change dependent upon the number of excess flowers which may be influenced by both variety and season. To move away from the concept of the threshold being a single value applicable to all cropping situations represents an important development in the evolution of pest control strategies.

Further work is required to validate the prediction scheme, particularly whether crops are less tolerant to losing buds from the main raceme compared with later formed buds.

Only small differences were detected between pollen beetle numbers measured in the field margins compared with the field centre. There was a weak trend for more pollen beetles along the southern side of a field, but the effect was not consistent.

Related Publications

Document download

View a printer friendly version of this publication

Download this publication PDF


File size: 594KB