General enquiries on this form should be made to:randd.defra.gov.uk/Document.aspx?Document=BD1623... · Web viewAll analyses were undertaken using Minitab 14 statistical software

General enquiries on this form should be made to:Defra, Science Directorate, Management Support and Finance Team,Telephone No. 020 7238 1612E-mail: [email protected]

SID 5 Research Project Final Report

SID 5 (Rev. 3/06) Page 1 of 61

NoteIn line with the Freedom of Information Act 2000, Defra aims to place the results of its completed research projects in the public domain wherever possible. The SID 5 (Research Project Final Report) is designed to capture the information on the results and outputs of Defra-funded research in a format that is easily publishable through the Defra website. A SID 5 must be completed for all projects.

This form is in Word format and the boxes may be expanded or reduced, as appropriate.

ACCESS TO INFORMATIONThe information collected on this form will be stored electronically and may be sent to any part of Defra, or to individual researchers or organisations outside Defra for the purposes of reviewing the project. Defra may also disclose the information to any outside organisation acting as an agent authorised by Defra to process final research reports on its behalf. Defra intends to publish this form on its website, unless there are strong reasons not to, which fully comply with exemptions under the Environmental Information Regulations or the Freedom of Information Act 2000.Defra may be required to release information, including personal data and commercial information, on request under the Environmental Information Regulations or the Freedom of Information Act 2000. However, Defra will not permit any unwarranted breach of confidentiality or act in contravention of its obligations under the Data Protection Act 1998. Defra or its appointed agents may use the name, address or other details on your form to contact you in connection with occasional customer research aimed at improving the processes through which Defra works with its contractors.

Project identification

1. Defra Project code BD1623

2. Project title

Environmentally sustainable techniques to establish and manage wildlife seed mixtures (WM1), and pollen and nectar seed mixtures (WM2)

3. Contractororganisation(s)

NERC Centre for Ecology and Hydrology

54. Total Defra project costs £ 145,000(agreed fixed price)

5. Project: start date................ 01 April 2003

end date................. 31 March 2006


6. It is Defra’s intention to publish this form. Please confirm your agreement to do so...................................................................................YES NO (a) When preparing SID 5s contractors should bear in mind that Defra intends that they be made public. They

should be written in a clear and concise manner and represent a full account of the research project which someone not closely associated with the project can follow.Defra recognises that in a small minority of cases there may be information, such as intellectual property or commercially confidential data, used in or generated by the research project, which should not be disclosed. In these cases, such information should be detailed in a separate annex (not to be published) so that the SID 5 can be placed in the public domain. Where it is impossible to complete the Final Report without including references to any sensitive or confidential data, the information should be included and section (b) completed. NB: only in exceptional circumstances will Defra expect contractors to give a "No" answer.In all cases, reasons for withholding information must be fully in line with exemptions under the Environmental Information Regulations or the Freedom of Information Act 2000.

(b) If you have answered NO, please explain why the Final report should not be released into public domain

Executive Summary7. The executive summary must not exceed 2 sides in total of A4 and should be understandable to the

intelligent non-scientist. It should cover the main objectives, methods and findings of the research, together with any other significant events and options for new work.A. Management to enhance pollen and nectar resources for bumblebees and butterflies within

intensively farmed landscapes1. In recent years there have been serious declines in the diversity of bumblebees, butterflies and other

pollinating insects both in the UK and elsewhere Europe.2. Research has shown that the most effective means of increasing the abundance and diversity of

bumblebees on arable farmland is to sow simple, low cost mixtures of dicotyledons rich in pollen and nectar.

3. The potential benefits of this management prescription for butterflies are unknown. Also, there is increasing evidence that the effectiveness of this policy might be constrained by the short flowering season and longevity of the pollen and nectar species.

4. The aim of this study was to devise seed mixtures and cutting management regimes which are effective in overcoming these constraints.

5. The inclusion of a greater number of pollen and nectar species did not attract a correspondingly greater number and diversity of bumblebees or butterflies compared with the simple, low cost mixture. However, the complex mix was more effective in providing late season forage for the critically important reproductive stages of bumblebee colonies.

6. There were marked declines in the abundance of sown dicots in both mixtures between years 2 and 3 after sowing, reflecting the increasing dominance of grass species.

7. Timing of cutting rather than frequency had the most marked effects on the performance and regeneration of sown species, and the provision of pollen and nectar resources through the season.

8. The addition of an early June cut to the typical autumn cutting regime was effective in both enhancing and extending the provision of pollen and nectar resources available to bees and butterflies. However, timing of summer cutting was critical. Cutting in late June delayed flowering beyond the lifecycle of many bee and butterfly species, and had detrimental effects on the regeneration of the sown dicots. Also, any summer cutting will potentially destroy butterfly breeding habitat.

9. Removal of cut material, especially under infrequent autumn cutting, had a significant beneficial effect on the abundance of pollen and nectar species and their flowers.

10. It is recommended that pollen and nectar seed mixtures are refined by the inclusion of the best performing species to provide both mid- and late-season flowers (Trifoilum sp., Lotus corniculatus and Centaurea nigra), and the removal of competitive grass species. It may be more efficient to provide early flowering species in separate habitats scattered through the landscape. Summer cutting in May or early June, with the removal of herbage where possible, should be applied to half of the patch in order to extend the flowering season and longevity of the habitat, and minimise damage to butterfly breeding habitat.


B. Environmentally sustainable pest control strategies for Wild Bird Seed mixtures managed under the agri-environment schemes

1. Between 2003 and 2005 three integrated experiments were carried out on arable field margins with the overall aim of developing effective, but environment sustainable pest control strategies for Brassica-based Wild Bird Seed (WBS) mixtures managed under the agri-environment schemes.

2. A small plot experiment (1) tested the efficiency of three low cost post-emergence insecticide sprays (Toppel, Fastac and Mavrik) in reducing pest damage to annual seed-bearing crops. Establishment, survival and seed yield were measured, together with effects on non-target invertebrates.

3. Experiment 2 examined the effects of field-scale application of these insecticides on the seed yield of kale, a widely sown biennial seed-bearing crop.

4. Experiment 3 compared the effectiveness of a new generation of combined insecticide and fungicide seed treatments developed for oilseed rape (TMX and Chinook) with foliar spray (Mavrik) to control flea beetle in spring. These treatments were combined with additional applications of Mavrik in the summer to control pollen beetle.

5. Overall linseed was the most susceptible species to pest damage, followed by fodder radish and then kale. Quinoa and millet were not significantly damaged by pest species.

6. Insecticide application significantly reduced the amount of pest damage to fodder radish and linseed seedlings (Experiment 1), and increased the final cover of adult plants. Weight and abundance of fodder radish seed was significantly increased by insecticide application.

7. There were no significant effects of insecticide application to control pollen beetle in second year kale on either seed weight or abundance (Experiment 2).

8. In Experiment 3 severe flea beetle outbreaks in both 2004 and 2005 combined with dry spring weather resulted in significantly lower establishment of fodder radish and the elimination of linseed compared with 2003 (Experiment 1). This resulted in few consistent significant effects of pesticide application on cover, abundance or seed yield of sown species. Seed treatment with TMX significantly reduced seed abundance and yield of fodder radish in 2004. This may reflect incorrect dosing of the larger fodder radish during off label application of this product. In the following year this treatment resulted in the highest seed abundance and yield at one site.

9. All experiments showed that spring insecticide application as either foliar spray or seed treatment had very few consistent significant effects on soil surface active invertebrates recorded by pitfall trapping.

10. In contrast, summer pesticide application had a large number of significant effects on epigeal invertebrates, some of which are important food items for farmland birds, which were consistent across experiments and years. In all cases numbers of pest species (flea beetles and pollen beetles) were significantly reduced by pesticide application. Other groups showing a negative response included spiders (Araneae), flies (Diptera) bees and wasps, (Hymenoptera), bugs (Hemiptera) and Weevils (Coleoptera: Curculionide). Insecticide also had a significant negative effect on total invertebrate abundance, and the number of orders and families in several experiments. There was some evidence (Experiments 1 and 2) that Mavrik had fewer negative impacts on some invertebrates, such as spiders, compared with the other foliar insecticides (Toppel and Fastac).

11. It is recommended that Brassica-based WBS mixtures are managed as follows to minimise the risk of severe pest damage and maximise seed yield: i) avoid sowing too early (April – May recommended) into cold, dry or otherwise poor-quality seed beds; ii) always sow a wide range of pest resistant ‘insurance’ species with the Brassicas; iii) avoid establishment adjacent to oilseed rape; iv) rotate the location of patches every 1-2 years to avoid build up of pests; iv) dress the seeds of susceptible species (fodder radish, kale and linseed) with a combined insecticide and fungicide to reduce flea beetle damage and minimise risk to non-target invertebrates; vi) avoid summer pesticide application if possible to reduce the risk of significant damage to non-target invertebrate populations. Further research is required into the performance of mixtures based on species with greater pest-resistance, and those tolerant of herbicides used to control injurious weeds.

Project Report to Defra8. As a guide this report should be no longer than 20 sides of A4. This report is to provide Defra with

details of the outputs of the research project for internal purposes; to meet the terms of the contract; and to allow Defra to publish details of the outputs to meet Environmental Information Regulation or Freedom of Information obligations. This short report to Defra does not preclude contractors from also seeking to publish a full, formal scientific report/paper in an appropriate scientific or other journal/publication. Indeed, Defra actively encourages such publications as part of the contract terms. The report to Defra should include: the scientific objectives as set out in the contract; the extent to which the objectives set out in the contract have been met; details of methods used and the results obtained, including statistical analysis (if appropriate); a discussion of the results and their reliability;


the main implications of the findings; possible future work; and any action resulting from the research (e.g. IP, Knowledge Transfer).


A. Management to enhance pollen and nectar resources for bumblebees and butterflies within intensively farmed landscapes

1. Introduction

Over the last 25 years there have been significant declines in the diversity of butterflies, bumblebees and other pollinating insects in the UK and elsewhere in north west Europe (Thomas et al., 2004; Biesmeijer et al., 2006). Moreover, even once widely distributed species have suffered severe reductions in both geographic range and local abundance (Fox et al., 2001). Intensive agricultural management, loss of habitats and food plants, and increased pesticide use have been cited as the main contributing factors to these declines (e.g. Kevan, 1975; Smart et al., 2000; Steffan-Dewenter, Potts and Packer, 2005; Carvell et al., 2006). Butterflies are considered sensitive bio-indicators of environmental change for a wide range of less conspicuous co-occurring invertebrates (‘umbrella species’) (Fleishman, Murphy and Brussard, 2000). Similarly, bumblebees provide an essential pollination service for many agricultural crops and a wide range of wildflower species (Kevan, 1991; Free, 1993; Dicks, Corbet and Pywell, 2002), and are therefore considered ‘keystone species’ critical for ecosystem function.

The need to reverse these damaging impacts of modern agriculture on biodiversity has been recognised by reforms to the Common Agricultural Policy (Bignal, 1998). The UK agri-environment schemes (Defra, 2005a,b) aim to achieve this by a combination of less intensive management practices within the crop and the complete removal of land from agricultural production to create wildlife habitat. Recent research and monitoring has shown that one of the most effective means of increasing the abundance and diversity of foraging bumblebees on intensively managed arable land is to sow simple, low cost mixtures of dicotyledons rich in pollen and nectar on field margin strips (Pywell et al., 2005; 2006; Carvell et al., 2007). These typically include agricultural varieties of legumes (Fabaceae), such as Trifolium pratense and Lotus corniculatus. There is currently little information regarding the potential value of this habitat for other pollinating insects, such as butterflies and hoverflies. However, sowing field margins with more complex and costly mixtures of grasses and dicotyledons has been shown to enhance the number and diversity of butterflies compared with all grass seed mixture or vegetation regeneration naturally from the seed bank (Feber et al., 1996; Meek et al., 2002; Pywell et al., 2004).

Bumblebee colonies require between 12 to 18 weeks of forage resources, depending on species, in order to complete their development and produce reproductives (queens and males) (Alford, 1975). Most rare and declining bumblebee species listed on the UKBAP found their colonies later in the season with peak foraging activity taking place between July and September. Recent research suggests that the peak flowering of pollen and nectar margins sown as part of the UK agri-environment schemes is on average 7.4 (±0.5) weeks, between late-June and mid-August (BD1624; R.Pywell unpublished data). There is also growing evidence that the agricultural varieties of legume sown in these mixtures are short-lived (<3 years) (Carvell et al.,2007). In the absence of alternative sources of mid- and late-season forage within intensively managed landscapes, it is therefore possible that such pollen and nectar margins may have no benefit or even function as sinks for bumblebee populations. Hence it is important to devise practical methods to prolong both the longevity of this habitat and its flowering season if benefits to bumblebee, butterflies and other pollinators are to be sustained in the longer term. Current management guidelines recommend cutting half this habitat in June to stimulate later flowering, with the remainder cut in the autumn. The effectiveness of this guidance remains untested. However, previous research suggests that frequent cutting increased flower abundance of some bumblebee and butterfly food plants, such as Hippocrepis comosa and Onobrychis viciifolia (Wells and Cox, 1993). Also, leaving the cut material in situ caused a significant increase in the proportion of competitive species, whereas removal had a significant beneficial effect on some legume species. Finally, mowing at the end of June had negative effects on butterfly abundance as this removed nectar sources (Feber et al., 1996). Summer mowing will also potentially destroy butterfly breeding habitat and larvae.

The aim of this experiment was to examine the effect of seed mixture composition and cutting regime on the provision of pollen and nectar resources for butterflies and bumblebees in intensively farmed landscapes. In order to achieve this aim we tested six experimental hypotheses:H1: Seed mixtures comprising a greater number of pollen- and nectar-rich plant species attract a greater number and diversity of foraging bumblebees and butterflies over a longer time period compared with more simple and lower cost mixtures;H2: Cutting more frequently will increase the abundance of dicot flowers, and therefore pollen and nectar resources for bumblebees and butterflies;H3: Cutting in early summer will extend the flowering period of pollen and nectar species, and therefore enhance their forage value for bumblebees and butterflies, compared with the typical management of autumn or spring cutting;H4: Removal of cut material, especially under infrequent autumn cutting, will be beneficial in maintaining the abundance of pollen and nectar species on fertile ex-arable soils in the long term compared with leaving the cut material in situ;H5: Autumn cutting and leaving the cut material in situ is essential for seed return and regeneration of pollen and nectar species in the long term;H6: There are synergistic interactions between seed mixture composition, cutting regime and herbage disposal method which enhance the abundance of pollen and nectar resources throughout the season, and also allow regeneration of the sown legume species.

The results are discussed in the context of future agri-environment scheme policies aimed at the enhancement and creation of habitat for pollinating insects on farmland.

2. Materials and methods

2.1 Experimental Design

In April 2003 a seed mixture comprising four pollen- and nectar-rich dicotyledons and a four fine-leaved grass species was sown at 20 kg ha-1 (Table 1; ‘simple mixture’) in 24 contiguous field margin plots each measuring 6 × 25 m at Hill Farm, Little Wittenham, Oxon (51°38'N 1°12'W). A combination of severe seed bed compaction and dry spring weather resulted in poor establishment. The experiment was re-sown in August 2003, but once again establishment success varied. It was therefore decided to re-locate the experiment to similar pollen and nectar margins also sown in April 2003 onto a more favourable seed bed at Manor Farm, Yorkshire (54º5′2″ N 0º49′11″ W). At this site two seed mixtures were sown along the margins of two arable fields (replicates). These main treatments were sown in two contiguous strips measuring 200 × 6 m in each field. Mixture 1. (‘complex mix’) conformed to that recommended under the arable options of the Countryside Stewardship Scheme (WM2) (Defra, 2002), comprising 6 dicotyledons (dicots) sown with 7 fine-leaved and tussocky grass species at 20 kg ha-1 at a cost of £140 ha-1 (€207 ha-1) (Table 1). Mixture 2. (‘simple mix’) conformed to that recommended under the new Entry Level Stewardship Scheme (EF4) (Defra, 2005a), comprising four dicots and four fine-leaved grasses at 20 kg ha -1 and a cost of £90 ha-1 (€133 ha-1). In the first year the vegetation was managed by cutting and removal of herbage on three occasions to control undesirable weeds and encourage establishment. In April 2004 each main treatment was sub-divided into eight contiguous 25 × 6 m sub-treatment plots and the following cutting regimes were applied at random both with and without removal of cut herbage: A. Cut April + June, B. Cut April + October, C. Cut June + October, and D. Cut October. In 2004 cutting was carried out on 21 April, 30 June and 3 October. In 2005 cutting was carried out on 20 April, 1 June and 4 October. The June cutting date was deliberately advanced 30 days in 2005 in order to investigate the effects of earlier summer cutting on the provision of flower resources. Cutting was carried out using a 1.6 m wide Ryetec 1600C rear-mounted flail collector mower (www.ryetec.co.uk). The rear collector box was left open to deposit cut and macerated herbage evenly across the sub-treatment plots as required.

2.2 Monitoring

In August each year the composition of the vegetation community was recorded from three 1 × 1 m quadrats placed at random within each sub-treatment plot. In each quadrat the

http://www.ryetec.co.uk/

percentage cover of all vascular plant species and bare ground were estimated as a vertical projection. In addition, in November 2004 and April 2005 seedlings of sown dicots were counted in 20 quadrats measuring 0.2 × 0.2 m arranged in a permanent diagonal transect across each sub-treatment plot. Percentage cover of bare ground, and that of chopped herbage were also recorded. Further counts were made from five 0.2 × 0.2 m quadrats in each plot in May 2006. Nomenclature follows that of Stace (1997).

The number of open flower units (congested raceme, e.g. Trifolium pratense, T. hybridum; erect raceme, e.g. Onobrychis viciifolia; auxiliary raceme, e.g. Lotus corniculatus; capitula, e.g. Centaurea nigra) of the sown pollen and nectar species was recorded from four 0.5 × 0.5 m quadrats placed at random within each sub-treatment plot. Counts were made on seven occasions between 20 May and 13 September in 2004, and on eight occasions between 26 May and 5 September in 2005. On each occasion the height of the vegetation in each sub-treatment was recorded from four drop disk measurements (diameter 300 mm, weight 200g) (Stewart et al., 2001).

On each visit the abundance and diversity of butterflies and foraging bumblebees was also recorded from a 25 × 6 m transect walked through the centre of each sub-treatment plot (Banaszak, 1980; Pywell et al., 2005; 2006). Walks were carried out between 10.00 am and 17.00 pm when weather conditions conformed to the Butterfly Monitoring Scheme (BMS) rules (temperature above 13°C with at least 60% clear sky, or 17°C in any sky conditions; no count at all if raining) (Pollard and Yates, 1993). The shade (ambient) temperature, percentage sunshine and wind speed were recorded at the end of each walk. Foraging bumblebees were recorded to species level and caste (following Prŷs-Jones and Corbet, 1991). Workers of Bombus terrestris and B. lucorum were collectively recorded as these cannot be reliably distinguished in the field. The cuckoo bumblebees (subgenus Psithyrus sp.), which are brood parasites of true Bombus species, were counted together as a group for analysis, but honeybees and solitary species were not noted. The flowering plant each bee was first seen to visit was also recorded to species level.

2.3 Statistical analysis

Mean percentage cover of all plant species was calculated per m2 for each sub-treatment plot together with the summary variables of species richness, and summed cover of sown dicots, sown grasses, unsown dicots and unsown grasses. Similarly, mean number of flowers per m2

for individual sown dicot species were calculated together with total sown dicot flower counts for each year. Counts of individual bumblebee and butterfly species from each visit were summed for each sub-treatment plot for each year. In addition, the summary groupings species richness and abundance of bumblebees and butterflies were calculated. Finally, the functional classification of ‘mobile’ or ‘immobile’ was applied to each butterfly species according to Warren (1992).

The effects of seed mixture and cutting regime on vegetation cover, seedling regeneration, dicot flowers, and bees and butterfly was investigated using a split-plot analysis of variance (ANOVA) model with seed mix as the main treatment (tested against the block × seed mix mean square), sub-treatments of cutting date in factorial combination with leave or removal of cut material, and various seed mix × cutting regime interactions (all tested against the error mean square). The two years were analysed separately. Vegetation percentage cover values were asin transformed and counts of seedlings, flowers, butterflies and bumblebees were log transformed as necessary following an assessment of the normality of residuals. All analyses were carried out using the SPLIT-PLOT ANOVA function of GenStat® 7.0 for Windows (Payne et al., 2002). Pairwise comparisons of each cutting sub-treatment were made using the ANOVA CONTRAST function of GenStat. Finally, the effects of seed mix and cutting regime on the distribution of dicot flowers, bees and butterflies through each season were examined by calculating median values for each sub-treatment plot and analysing them using the ANOVA model described above.

3. Results

3.1 Effects of seed mix and cutting regime on vegetation composition and seedling regeneration

In 2004 there were few significant differences in vegetation composition between the simple and complex seed mixtures (Table 2). Percentage cover of Trifolium hybridum and that of all sown dicots was significantly greater in the simple compared with the complex seed mix. Similarly, the cover of T. hybridum was significantly greater under the June + October cutting regime compared with cutting in April + June. Cover in both these sub-treatments was greater than cutting in either October or April + October. There were no significant differences in vegetation composition resulting from herbage removal or return.

In 2005 there were no significant differences in composition between the two seed mixtures. However, there were many more significant effects of cutting date on composition (Table 2). The cover of Centaurea nigra was significantly lower under the June + October cutting regime compared with all others. In contrast, both the cover of T. pratense and all sown dicots was significantly higher following April + June and June + October cutting compared with October and April + October cutting. Similarly, the richness of sown dicots was significantly greater under April + June cutting compared with all other cutting regimes. Herbage removal resulted in a significant increase in the cover of Lotus corniculatus, T. hybridum, T. pratense, and both the total cover and richness of sown dicots. There was a corresponding significant decrease in the cover of sown grasses. Finally, there were few significant interactions. Declines in the cover of C. nigra were accelerated under June + October cutting with herbage removal.

In November 2004 cutting the simple seed mixture resulted in a significantly greater cover of chopped herbage on the ground compared with the complex seed mix (Table 3). Both percentage cover of bare ground and cut herbage were significantly lower under April + June cutting compared with all other cutting regimes. In November the mean number of T. hybridum, T. pratense and total legume seedlings m-2 was significantly higher following cutting in April + October and October compared with either April + June and June + October cutting. Finally, herbage removal resulted in a significantly greater cover of bare ground, and a significantly lower (>50%) cover of chopped plant material. Cutting in April + June resulted in a lower cover of chopped material compared with the October cut sub-treatments regardless of whether it was removed or left in situ. There were no significant effects of herbage disposal method on seedling regeneration.

In April 2005 bare ground cover remained significantly higher in the June + October cutting treatment compared both with the April + October and June + October treatments. There was no significant difference in the cover of chopped herbage and plant litter between sub-treatments. The density of surviving T. hybridum seedlings remained significantly higher in the April + October cutting sub-treatment compared with the June cutting sub-treatments. Densities were also higher in the October cutting sub-treatment compared with the June + October regime. The number of surviving and T. pratense and total legume seedlings remained significantly higher in the April + October and October cut sub-treatments compared with the June cutting sub-treatments.

In May 2006 there were no significant effects of seed mix on bare ground or seedling regeneration. Similarly, there were very few effects of cutting date. There was some evidence that cutting in April + October continued to result in significantly less bare ground cover. Herbage disposal regime resulted in no significant effects on seedling regeneration. The only significant interaction was that removal of the cut material from the complex seed mix resulted in considerably more bare ground in May 2006 compared with the simple seed mix.

3.2 Effects of seed mix and cutting regime on flower resources

In 2004 the cumulative number of sown dicot flowers through the season was always higher in the simple compared with the complex seed mix (Fig. 1a). However, there was no significant difference between the median values (ANOVA F1,1 = 0.94ns). Similarly, there were few significant effects of seed mixture on the total number of sown dicot flowers summed for all seven visits (Table 4). Only the total number of T. pratense flowers was significantly higher in the simple compared with the complex seed mix. In contrast, cutting regime had very

marked effects on the cumulative number of dicot flowers through the season (Fig. 1b). The April + October and October cutting regimes resulted in significantly greater median flower densities than the April + June and June + October cutting regimes (F3,14 = 42.80***). The sum total density of T. hybridum flowers was significantly greater following October cutting compared with all other treatments (Table 4). Flowers of this species were also significantly more abundant following April + October and June + October cutting compared with April + June cutting. The total abundance of T. pratense flowers and all sown dicots were significantly more abundant in both the April + October and October cut sub-treatments compared with April + June and June + October cutting. Also, total flower abundance was significantly higher following June + October cutting compared with April + June cutting. Cutting regime also affected vegetation height. October cutting resulted in a significantly taller sward than all other sub-treatments. Cutting in April + October resulted in the next tallest vegetation, followed by cutting in June + October. Finally, cutting in April + June resulted in significantly shorter vegetation. Herbage disposal method had no significant effect on the cumulative number of dicot flowers (Fig. 1c) (F1,14 = 0.04ns). Only Onobrychis viciifolia flowers were significantly more abundant following herbage removal (Table 4). There were no significant interactions between seed mix, cutting date or herbage disposal method.

Between 2004 and 2005 there was a marked decline in total flower abundance of sown dicots (mean per visit 2004 = 75.7±7.0 m-2; 2005 = 32.0±4.6 m-2). In 2005 the cumulative number of sown dicot flowers through the season remained higher in the simple compared with the complex seed mix (Fig. 1a), but there was no significant difference in medians (F1,1 = 30.46ns). However, the total number of sown dicot flowers summed over all eight visits was significantly greater in the simple compared to the complex seed mix (Table 4). Also, the average height of the vegetation was significantly taller in the complex compared with the simple seed mixture.

Date and frequency of cutting continued to affect significantly the cumulative flower distribution through the season (Fig. 1b). Median values for the April + June and June + October cutting regimes were significantly greater than those of the April + October and October regimes (F3,14 = 8.04**). Sum total C. nigra flowers were significantly more abundant in the April + October and October cut plots compared with April + June and June + October cutting (Table 4). Flowers of this species were also more abundant in the April + June sub-treatment compared with the June + October sub-treatment. The number of L. corniculatus flowers responded positively to April cutting with significantly higher abundance in the April + October and April + June cut plots. The abundance of T. hybridum flowers was significantly higher following April + June and June + October cutting compared with October cutting. The abundance of T. pratense flowers and all sown dicots was significantly enhanced by cutting in April + June and June + October. Once again, the vegetation was significantly taller in the October cut sub-treatments compared with all others. Vegetation cut in April + October was also significantly taller than in either the April + June and June + October cut sub-treatments. Removal of the cut herbage resulted in a consistently higher cumulative distribution of dicot flowers over the season and a significantly higher median value (Fig. 1c) (F1,14 = 12.64**). Herbage removal also resulted in a significant increase in the total abundance of T. hybridum, T. pratense flowers and all sown dicots (Table 4). There was a significant seed mix × cut date interaction for vegetation height. This reflected the disproportionately greater reduction in height of the complex seed mix resulting from June cutting compared with the simple mix. The significant seed mix × cut date interaction for T. pratense flower density reflected the greater reduction the abundance of flowers in the simple mix compared with the complex mix as a result of April + October and October. The seed mix × cut date interaction for C. nigra was due to the absence of this species in the simple mix. There was also a synergistic effect of cutting in April + October and October in combination with herbage removal on the abundance of C. nigra flowers. Finally, leaving the cut herbage in situ had a significant detrimental effect on the abundance of L. corniculatus flowers in the complex seed mix, but not in the simple mix.

A similar experiment was established at Manor Farm in September 2003 to examine the effects of cutting regime on the flowering of different T. pratense varieties (R. Pywell, unpublished data; Appendix 1). The results from 2004 showed a similar pattern of extended flowering period in response to early June cutting. There was also a marked decline in flower

abundance between year 2 (2004) and 3 (2005). In 2005 the June cutting regimes both increased overall flower abundance and extended the flowering period compared with cutting in April or October.

3.3 Effects of seed mix and cutting regime on bumblebee and butterfly abundance

In 2004 the cumulative number of bumblebees and butterflies recorded through the season was always higher in the simple compared with the complex seed mix (Figs. 2a & 3a). However, there was no significant difference between the median values (bees ANOVA F1,1 = 2.71ns; butterflies F1,1 = 1.00ns). Only the mean richness of bumblebee species per visit was significantly higher on the simple seed mix compared with the complex mix (Table 5). Timing and frequency of cutting had very marked effects on the cumulative number of bumblebees (Fig. 2b). April + October and October cutting resulted in significantly greater median bee densities than April + June and June + October cutting (F3,14 = 22.62***). The abundance of Bombus terrestris / B. lucorum, B. pascuorum and B. hortorum summed for all seven visits was significantly higher following April + October and October cutting compared with April + June and June + October cutting (Table 5). B. lapidarius was significantly more abundant in the October cut sub-treatments compared with all others. Abundance of this species was also significantly higher in the April + October cut plots compared with both sub-treatments cut in June. Both total bee abundance and species richness were also significantly higher following April + October and October cutting compared with April + June and June + October cutting. Finally, the abundance of male and queen bumblebees was significantly higher following April + October and October cutting. There were relatively few significant interactions. The seed mix × cut date interaction for B. terrestris / B. lucorum and B. lapidarius reflected the comparatively greater reduction in numbers resulting from April + June and June + October cutting of the simple seed mix compared with the complex mix. There was no effect of herbage disposal method on cumulative number of bees (Fig. 2c) (F1,14 = 0.00ns).

In 2004 effects of cutting regime were less marked on butterflies and there was no significant difference between median values (Fig. 3b) (F3,14 = 1.00ns). The abundance of the butterfly species Maniola jurtina, Aphantopus hyperantus, Aglais urticae and Cynthia cardui was all significantly higher in the April + October and October cut sub-treatments compared with the June cut sub-treatments (Table 6). Pieris rapae was significantly more abundant in the October cut plots compared with all others. The abundance of mobile butterfly species, total butterflies and butterfly species richness was also significantly higher in the April + October and October sub-treatments. Abundance of immobile butterfly species was significantly higher following April + October and October cutting compared with June + October cutting. Immobile species were also significantly more abundant following October cutting compared with April + June cutting. Herbage disposal technique did not have a significant effect on cumulative butterfly numbers over the season (Fig. 3c) (F3,14 = 1.00ns). Finally, the significant seed mix × cut date interaction for Aphantopus hyperantus reflected the very low numbers of this species recorded in the simple seed mix.

Between 2004 and 2005 there were a marked declines in total abundance of both bumblebees (mean per visit 2004 = 10.3±0.1 per 125 m2; 2005 = 4.2±0.1 per 125 m2) and butterflies (2004 = 1.3±0.1 per 125 m2; 2005 = 0.8±0.1 per 125 m2) recorded on each plot. In 2005 there were no significant seed mix effects on bumblebees (Fig. 2a; F1,1 = 20.25ns) (Table 5). April + June cutting resulted in a significantly higher median value of bumblebees compared with April + October cutting (F1,14 = 5.01*) and October cutting (F1,14 = 5.29*) (Fig. 2b). The abundance of B. pascuorum was significantly higher in the April + June cut plots compared with those cut in April + October and October (Table 5). This species was also significantly more abundant in the June + October cut plots compared with those cut in October alone. B. hortorum was significantly more abundant following April + June cutting compared with April + October and October cutting. In contrast, abundance of male bumblebees was significantly higher on the October cut plots compared with those cut in April + June and June + October. Removal of herbage resulted in a consistently higher cumulative number of bees (Fig. 2c), but there were no significant differences between the medians (F1,14

= 2.23ns). However, removal significantly increased the sum abundance of B. lapidarius, total bumblebees and bee species richness. There was a synergistic beneficial effect of June cutting on total bumblebee abundance of the simple seed mix, but not the complex mix. In

contrast, male bee numbers only responded positively to October and April + October cutting in the complex mix. In 2005 there was no significant difference in butterfly median values between the simple and complex seed mixes (Fig. 3a) (F1,1 = 1.00ns). However, the total abundance of mobile butterflies and the richness of butterfly species were significantly higher in the simple seed compared to the complex mix (Table 6). Over the season there was no effect of cutting date on butterfly distributions (Fig. 3b) (F3,14 = 1.58ns). Maniola jurtina was significantly more abundant following April + June cutting compared with the April + October cut. Mobile butterfly species were also significantly more abundant following April + June cutting compared with the April + October cutting. Total butterfly abundance was significantly higher in the April + June and October cut sub-treatments compared with April + October cutting. Finally, there was no effect of herbage disposal on cumulative butterfly distribution (Fig. 3c) (F1,14 = 2.01ns). In terms of interaction effects, April + June cutting benefited M. jurtina and total immobile butterflies more in the complex mix compared with the simple mix. Also, cutting the complex mix in April + June and cutting the simple mix in October resulted in a larger than expected increase in mobile butterflies, total butterflies and butterfly species richness.

4. Discussion

4.1 Pollen and nectar seed mixtures

There were marked differences in the composition and floral resources of the two prescribed pollen and nectar seed mixtures. Cover of sown dicots, especially Trifolium sp. and Lotus corniculatus, was higher in the simple, less costly seed mixture compared with the complex mixture. Consequently the simple mixture produced a consistently higher density of dicot flowers (35-45%), particularly those of Trifolium sp., in both years. These effects can be explained in terms of differences in seed mixture composition and sowing density, and the performance of individual species (BD1433; Pywell et al., 2002; 2003). The seed mixtures differed in several important aspects, namely the complex mix contained two tall and competitive grass species (Phleum pratense and Festuca pratensis) which were both sown at relatively high rates. This resulted in a significantly taller, and presumably more competitive, sward. In contrast, only fine-leaved, low-growing and therefore less competitive grasses were sown in the simple mixture. Also, seed sowing density of several dicot species was significantly higher in the simple seed mixture, for example individual seed rates of T. hybridum and Lotus corniculatus were 33% and 50% higher respectively. Finally, two of the seven dicots sown in the complex mixture either failed to establish (Vicia sativa ssp. sativa), or established poorly and did not persist (Medicago lupulina). Establishment of Vicia species is known to be adversely affected by hard-coat dormancy and seedling herbivory by molluscs (BD1425; Pywell et al., 2007). Similarly, M. lupulina is a short-lived, early successional species of calcareous grasslands (Gibson and Brown, 1991) which is unlikely to be ecologically adapted for survival in productive, neutral grassland established on fertile, heavy soil. For these reasons, the complex and more costly seed mixture containing a greater number of pollen- and nectar-rich plant species did not attract a greater number and diversity of foraging bumblebees and butterflies compared with more simple, low cost mixture. However, the long-term sustainability of pollination services within intensively managed landscapes requires the provision of pollen and resources for both the foraging workers and the reproductive components of bumblebee colonies. The complex mix included Centaurea nigra, a dicot which flowers in late summer. This species attracted a significantly greater number of male bumblebees in the sub-treatments cut in April + October and October in 2005 compared with the simple mix.

4.2 Cutting regime

The results suggest that timing rather than the frequency of cutting had the most marked effects on the performance and regeneration of the sown species, and the provision of pollen and nectar resources through the season. Cutting has direct effects on the vegetation and associated invertebrates through the removal of the vegetation structure and flowers (Morris, 1990). It will also have indirect effects on the plant community by encouraging vegetative

growth, enhancing flower production, creating gaps for germination and by altering the competitive balance between species (Bullock, 1996). Persistence of sown T. pratense in productive grasslands has been shown to diminish rapidly after 3-4 years under cutting management (Mela, 2003) and cutting and grazing (Prigge, Bryan and Goldman-Innis, 1999). In this study we found that adding a summer cut to the typical autumn cut appeared to enhance the cover and persistence of T. pratense and T. hybridum at the expense of competitive grasses. Further research is required to determine the precise mechanism of this observed effect, and if it is an effective means of maintaining sown pollen and nectar species in the longer term. However, tall, late-flowering dicots, such as C. nigra, were reduced by this June cutting. Timing of summer cutting also appeared to be the critical factor controlling the provision of pollen and nectar resources for bumblebees and butterflies. Cutting in late June delayed the re-flowering of T. hybridum by c.50 days and that of T. pratense by c.70 days. This effectively removed the dicot flower resource available to bumblebee colonies during the July and August period which is considered critical for reproductive success (Goulson, 2003). Similarly, cutting at this time would have also removed both nectar and potential oviposition sites for farmland butterfly species, such as Polyommatus icarus and Maniola jurtina (Feber et al., 1996). Furthermore, the very low densities of Trifolium seedlings emerging in the autumn and following spring suggested that this desynchronisation of flower resources and pollinators may have resulted in potentially catastrophic consequences for seed production and regeneration of these species (Hawkins, 1958).

In contrast, early June cutting in the following year delayed re-flowering of Trifolium species for a similar period of time (this effect was reproduced in an additional experiment: see Appendix 1). However, in this case peak re-flowering was coincident with peak abundance of foraging bumblebees in late July and early August. This was therefore a highly effective means of extending the flowering period of pollen and nectar species and enhancing their abundance by two- to seven-fold. This significantly increased the abundance of long-tongued bumblebee species (Bombus pascuorum and B. hortorum) which specialise on foraging on T. pratense (Pywell et al., 2005). Seed counts from 5 random T. pratense flower heads collected from each sub-treatment plot in September 2005 also found that early June cutting had significant, beneficial effects on seed production (April + June cut = 77.3±6.7 viable seeds head-1; April + October cut = 50.4±6.7 seeds head -1; ANOVA F1,14 13.26; P=0.003**). Importantly, there was no detrimental effect of early June cutting on seedling regeneration in the following year.

4.3 Herbage disposal

The method of herbage disposal also had important ecological effects on vegetation composition, seedling regeneration and the provision of pollen and nectar resources. The removal of cut material once or twice a year resulted in a significant increase in the cover (45%) and richness (25%) of sown dicot species at the expense of grasses. Cutting and removal of herbage resulted in an instant reduction in competition for space and light compared with the more gradual reduction resulting from leaving the cut material in situ. It is also likely that removal resulted in nutrient off-take and reduction in soil fertility (Tallowin et al., 2002). Similarly, removal also resulted in significant increases in the density of sown dicot flowers, particularly Onobrychis viciifolia and Trioflium sp. (Wells and Cox, 1993), and therefore increased the abundance and richness of bumblebees. In contrast, cut and macerated herbage left in situ will act as a physical barrier to light reaching the underlying plant species. The intensity and duration of this effect will be dependent on the depth and composition of the cut material, and the time taken for it to decompose or be eroded by wind or water. Plant species will vary greatly in their ability to tolerate the stress induced by this type of shading. This will have indirect effects on plant community composition by altering the competitive balance between species.

Removal of the cut material did not appear to affect the ability of the sown dicots to regenerate from seed (Smith, Pullan and Shiel, 1996; Pywell et al., 2007), suggesting that sufficient seed is returned to the soil surface either before or during the cutting operation. In montane meadows in Japan the mechanical operation of late-August hay-making was found to deposit large quantities of viable seed on the soil surface which could contribute to the persistence of T. pratense (Sakanoue, 2005). However, it is likely that the thick mat of cut

and macerated vegetation rich in viable seed resulting from cutting and leaving in situ in this study is a poor medium for germination and survival. Further research is required into the factors constraining the regeneration of Trifolium species in productive habitats managed for the provision of pollen and nectar resources.

4.4 Management to enhance pollen and nectar resources for bumblebees and butterflies

The results of both this and the additional study (Appendix 1) demonstrated that pollen and nectar margins sown under the current UK agri-environment scheme guidance are only effective for 3-4 years despite intensive cutting management. The most practical means of guaranteeing a continuity of pollen and nectar resources is to re-establish this habitat either in situ or preferably elsewhere on the farm after 3 years. It is recommended that the composition of future pollen and nectar seed mixtures is revised and simplified to include a combination of good performing dicot species which flower in mid-season (e.g. T. pratense, T. hybridum, L. corniculatus) and late-season (e.g. C. nigra, Malva moschata). The latter is critically important to provide foraging resources for the reproductive stages of bumblebee colonies. Further research is required to determine if the exclusion of grass species will reduce competition, increase the regeneration of sown dicots by seed and therefore improve the longevity of these habitats. However, the potential value of the sown grass species as larval food plants for butterflies species should also be considered. It may be more pracrical to sow separate dicot and grass strips or patches to provide both foraging and breeding habitats. Neither seed mixture provided early season forage for nest-searching queen bumblebees or early emerging butterfly species. It may also be more efficient to provide these resources in separate habitats scattered throughout the landscape, for example by either planting or conserving early flowering hedgerow species (Salix cinerea, Malus sp., Lamium album, Glechoma hederacea) and patches of novel crop species (e.g. Lunaria annua).

The current management guidelines of cutting pollen and nectar margins in summer is effective in increasing the cover and extending the flowering period of sown dicots. Cutting only half of the sown margin also reduces the potential damage to breeding populations of butterflies, and will retain some tall vegetation habitat for nesting bumblebees and other invertebrates. However, we recommend that the date of cutting is advanced to May or early June at the latest in order to ensure the provision of pollen and nectar resources are synchronised with peak foraging bumblebee and butterfly numbers. The precise date of cutting will require adjustment for different latitudes and seasonal variation in growing conditions. Ideally cut material from the pollen and nectar margins should be removed. However, this increases the time taken, cost and complexity of the cutting operation for land managers. To overcome this problem the practical and ecological effects of traditional hay-making and baling require further investigation.

5. Acknowledgements

This study was funded by a commission (BD1623) from the Department for Environment, Food and Rural Affairs. The authors are grateful to Richard Brown for managing and hosting the experiment. We also thank Carole Freeland for data input, and James Bullock and Tim Sparks for advice on statistical analysis.

6. Tables and figures

Table 1. Sowing rate and seed mixture composition.

Species1. Complex mix 2. Simple mix% kg ha-1 % kg ha-1

Agrostis capillaris 5 1.0Cynosurus cristatus 20 4.0 20 4.0Festuca ovina 10 2.0Festuca pratense 20 4.0Festuca rubra ssp commutata 10 2.0 28 5.6Festuca rubra ssp juncea 5 1.0 20 4.0Phleum pratense 8 1.6Poa pratensis 12 2.4

Centaurea nigra 2 0.4Lotus corniculatus 2 0.4 4 0.8Medicago lupulina 2 0.4Onobrychis viciifolia 5 1.0 8 1.6Trifolium hybridum 2 0.4 3 0.6Trifolium pratense (var. Altaswede) 5 1.0 5 1.0Vicia sativa ssp sativa 4 0.8

TOTAL 100 20 100 20

Table 2. Effects of seed mixture, cutting date and herbage disposal method on mean percentage cover and richness of sown plant species in 2004 and 2005.

Centaurea nigra

Lotus corniculatus

Trifolium hybridum

Trifolium pratense

Sown dicots Sown grasses Richness sown dicots

Seed mix df 2004 2005 2004 2005 2004 2005 2004 2005 2004 2005 2004 2005 2004 2005Complex 1.02 3.74 4.94 1.17 30.83 10.29 26.13 11.77 63.16 27.10 45.55 68.37 3.25 2.44

Simple 0.00 0.00 11.23 0.42 42.52 9.71 31.71 18.75 85.86 29.61 20.83 64.24 3.06 1.94F-values 1,1 11.31 11.66 0.38 0.53 4944** 0.00 2.39 114 3170* 0.32 61.48 1.44 0.22 0.74Cut date

Apr + Jun 0.63 1.67a 7.54 1.79 40.63b 13.29 30.13 24.50a 79.22 42.09a 27.71 52.40 3.29 2.96aApr + Oct 0.46 2.88a 8.38 0.92 25.17c 8.42 34.58 7.58b 68.80 20.34b 36.59 70.43 3.04 2.04bJun + Oct 0.13 0.22b 4.80 0.04 53.29a 10.05 16.25 23.67a 75.09 34.06a 32.53 57.02 3.12 2.04b

Oct 0.83 2.71a 11.63 0.42 27.63c 8.25 34.71 5.30b 74.92 16.93b 32.60 70.53 3.17 1.71bF-values 3,14 2.96 4.31* 1.56 3.09 11.36*** 2.32 2.87 9.23*** 2.18 7.69** 1.68 2.16 0.48 4.67*

Herbage disposalLeave 0.31 1.45 6.90 0.23 35.86 8.25 29.35 10.11 72.71 20.20 32.59 68.91 3.08 1.90

Remove 0.71 2.29 9.27 1.35 37.50 11.75 28.48 20.42 76.30 36.51 32.13 56.27 3.23 2.48F-values 1,14 2.98 0.80 1.28 4.65* 0.17 7.70* 0.01 7.55* 1.12 17.53*** 0.02 7.77* 0.91 5.51*

InteractionsCut date × Herbage disposal 3,14 2.98 4.31* 0.35 0.86 0.09 1.65 0.29 2.83 0.96 1.82 0.61 0.76 0.56 0.37

Seed mix × Cut date 3,14 2.98 0.80 1.80 0.17 2.16 0.24 0.89 0.29 2.94 0.32 2.11 0.26 1.96 1.39Seed mix × Herbage disposal 1,14 2.96 1.30 0.06 3.88 0.14 0.94 0.14 0.29 0.66 0.24 0.11 0.17 0.46 0.26

* P<0.05, ** P<0.01; *** P<0.001, blank – no significant difference

Table 3. Effects of seed mixture, cutting date and herbage disposal method on mean percentage cover of bare ground and cut herbage, and the number of legume seedlings per m2 between 2004 and 2006.

% bare ground % cut herbage & litter Trifolium hybridum m2 Trifolium pratense m2 Total legumes m2

Nov2004

Apr 2005

May 2006

Nov2004

Apr 2005

May 2006

Nov2004

Apr 2005

May 2006

Nov2004

Apr 2005

May 2006

Nov2004

Apr 2005

May 2006

Seed mixComplex 12.18 14.62 14.82 18.32 10.15 - 21.02 7.03 19.06 14.06 19.77 7.19 56.02 36.80 34.06

Simple 11.96 13.20 10.39 32.02 11.26 - 2.89 2.97 1.25 46.88 28.67 3.75 116.80 48.20 7.81F-values 1,1 0.00 0.06 1.61 499.82* 1.20 - 0.42 0.50 1.07 1.16 1.81 1.00 1.00 2.52 1.05Cut date

Apr + Jun 2.65b 13.04b 10.03 0.58b 6.36 - 0.00b 1.09b 0.63 0.00b 0.31b 1.25 1.41b 3.75b 4.38Apr + Oct 14.29a 8.72b 7.49 31.07a 13.11 - 17.03b 14.69a 33.13 99.22a 76.56a 9.38 228.44a 129.22a 50.63Jun + Oct 14.85a 18.92a 18.29 29.69a 6.26 - 0.47b 0.31c 3.75 0.31b 0.47b 5.63 5.78b 1.25b 17.50

Oct 16.49a 14.95ab 14.60 39.35a 17.08 - 30.31a 3.91ab 3.13 22.34a 19.53ab 5.63 110.00a 35.78a 11.25F-values 3,14 5.11* 3.78* 3.35* 13.37*** 2.45 - 5.03* 7.92** 0.92 10.75*** 4.87* 1.02 23.60*** 10.23*** 1.26

Herbage disposalLeave 6.71 14.28 11.32 36.26 13.47 - 8.98 5.78 17.50 37.73 39.22 6.88 112.66 67.50 27.19

Remove 17.43 13.54 13.88 14.08 7.93 - 14.92 4.22 2.81 23.20 9.22 4.06 60.16 17.50 14.69F-values 1,14 14.57** 0.12 0.95 22.91*** 2.65 - 0.03 0.64 0.84 0.02 0.57 0.73 0.03 0.05 0.47

InteractionsCut date × Herbage disposal 3,14 1.94 2.16 1.15 4.09* 1.45 - 2.75 1.38 1.16 0.10 0.50 2.15 0.52 1.24 1.44

Seed mix × Cut date 3,14 0.77 1.44 4.12* 8.51** 0.04 - 0.48 0.79 0.92 1.86 0.43 0.59 1.98 0.40 1.00Seed mix × Herbage

disposal1,14 0.62 0.50 1.28 1.30 2.18 - 0.37 0.16 0.99 0.18 1.44 0.08 0.62 0.78 0.09

* P<0.05, ** P<0.01, *** P<0.001, blank – no significant difference

Table 4. Effects of seed mixture, cutting date and herbage disposal method on mean vegetation height (cm) and the number of sown dicot flowers per m 2 summed for all visits in 2004 and 2005.

Mean vegetation height (cm)

Centaureanigra

Lotuscorniculatus

Onobrychis viciifolia

Trifoliumhybridum

Trifoliumpratense

Total sown dicots

Seed mix df 2004 2005 2004 2005 2004 2005 2004 2005 2004 2005 2004 2005 2004 2005Complex 36.85 41.05 0.94 13.31 26.13 5.13 0.50 0.88 291.13 130.94 102.56 32.00 423.31 182.25

Simple 32.78 26.11 0.00 0.00 86.94 9.25 1.13 1.13 402.00 214.00 145.38 105.38 635.88 329.75F-values 1,1 3.21 379* 1.86 48.89 0.62 31.39 1.00 4.00 0.92 11.67 2776* 9.35 1.20 1204*Cut date

Apr + Jun 25.24d 24.89c 0.00 3.13b 22.50 17.13a 1.50 1.88 213.50c 235.63a 22.38b 135.88a 261.38c 393.63a

Apr + Oct 37.26b 39.18b 1.00 11.13a 64.00 9.50a 0.50 0.50 373.88b145.13a

b 261.25a 24.50b 700.63a 190.75bJun + Oct 31.77c 24.88c 0.00 0.75c 38.88 0.75b 0.88 1.50 324.38b 217.00a 18.50b 100.50a 383.50b 320.50a

Oct 45.00a 45.40a 0.88 11.63a 100.75 1.38b 0.38 0.13 474.50a 92.13b 193.75a 13.88b 772.88a 119.13bF-values 3,14 33.77*** 49.98*** 1.75 13.90*** 2.71 11.74*** 0.96 1.25 32.83*** 4.23* 24.60*** 11.65*** 58.27*** 9.10***

Herbage disposalLeave 34.51 34.15 0.63 4.44 48.56 4.75 0.25 0.25 342.25 122.44 134.56 43.81 526.75 175.69

Remove 35.12 33.02 0.31 8.88 64.50 9.63 1.38 1.75 350.88 222.50 113.38 93.56 532.44 336.31F-values 1,14 0.18 0.59 0.58 1.70 0.59 4.07 4.79* 4.15 0.21 9.64** 0.73 8.25* 0.03 15.28**

InteractionsCut date × Herbage disposal 3,14 0.69 1.51 0.33 3.84* 0.36 0.93 0.49 0.56 0.75 1.35 0.25 3.11 0.26 2.39

Seed mix × Cut date 3,14 0.13 19.83*** 1.75 13.90*** 1.06 1.18 1.04 0.17 2.06 0.34 0.59 4.18* 3.20 1.80Seed mix × Herbage disposal 1,14 0.28 0.39 0.58 1.70 0.27 9.41** 0.53 0.12 0.50 0.02 0.07 2.66 0.91 0.34


Table 5. Effects of seed mixture, cutting date and herbage disposal method on species richness and mean total number of bumblebees per 150 m 2 plot in 2004 and 2005. Only species with mean total abundance of >1 per sub-treatment plot included.

B. terrestris/B. lucorum

B. lapidarius B. pascuorum B. hortorum Totalbumblebees

Richness bumblebees

Males Queens

Seed mix df 2004 2005 2004 2005 2004 2005 2004 2005 2004 2005 2004 2005 2004 2005 2004 2005Complex 1.75 0.75 15.06 17.94 33.75 8.63 3.25 0.44 53.81 27.75 3.06 2.63 2.38 8.38 0.63 0.31

Simple 3.19 1.06 33.94 15.56 50.25 20.63 2.94 1.38 90.31 38.63 3.38 3.13 1.38 1.44 0.56 0.31F-values 1,1 10.80 25.00 16.66 0.00 48.27 8.46 0.06 4.59 16.97 6.40 444* 4.00 2.56 27.94 0.02 0.00Cut date

Apr + Jun 0.50b 0.88 7.75c 14.63 6.88b 24.50a 0.25b 1.75a 15.38b 41.75 2.50b 3.38 0.00b 1.25b 0.00b 0.50Apr + Oct 3.25a 0.50 35.25b 16.88 84.25a 7.13bc 6.50a 0.38b 129.25a 24.88 4.00a 2.25 3.75a 6.13ab 1.38a 0.13Jun + Oct 1.13b 1.00 12.13c 11.00 2.88b 21.25ab 0.25b 1.25ab 16.38b 34.50 2.63b 3.13 0.38b 1.50b 0.00b 0.50

Oct 5.00a 1.25 42.88a 24.50 74.00a 5.63c 5.38a 0.25b 127.25a 31.63 3.75a 2.75 3.38a 10.75a 1.00a 0.13F-values 3,14 8.67** 0.78 77.93*** 1.90 31.94*** 4.19* 10.62*** 3.79* 47.29*** 1.12 8.92*** 1.79 6.35** 6.45** 9.31*** 1.08Herbage disposal

Leave 2.00 0.75 23.81 12.81 44.25 9.81 3.44 0.75 73.50 24.13 3.25 2.50 1.13 3.94 0.88 0.13Remove 2.94 1.06 25.19 20.69 39.75 19.44 2.75 1.06 70.63 42.25 3.19 3.25 2.63 5.88 0.31 0.50F-values 1,14 1.80 0.78 0.50 7.35* 0.35 4.18 0.46 0.72 0.09 8.01* 0.01 5.54* 3.72 1.20 5.97* 3.23

InteractionsCut date × Herbage disposal

3,14 0.52 2.29 1.56 1.03 0.12 1.54 0.03 1.49 0.39 1.89 0.01 1.23 1.39 0.22 0.66 0.12

Seed mix × Cut date

3,14 3.38* 1.45 13.78*** 1.36 0.98 3.16 0.54 1.26 2.88 4.53* 0.09 1.03 0.71 5.86** 0.47 0.24

Seed mix × Herbage disposal

1,14 0.07 1.54 0.15 0.47 0.73 0.20 0.00 0.72 0.17 0.14 1.25 3.85 0.41 1.36 0.66 0.00


Table 6. Effects of seed mixture, cutting date and herbage disposal method on species richness and mean total number of butterflies per 150 m 2 plot in 2004 and 2005. Only species with mean total abundance of >1 per sub-treatment plot included.

Maniola jurtina

Aphantopus hyperantus

Pieris rapae Pieris brassicae

Aglais urticae Cynthia cardui

Mobile butterflies

Immobilebutterflies

Totalbutterflies

Richness butterflies

Seed mix df 2004 2005 2004 2005 2004 2005 2004 2005 2004 2005 2004 2005 2004 2005 2004 2005 2004 2005 2004 2005Complex 0.75 1.31 0.56 1.25 0.50 1.19 0.50 1.00 4.00 0.00 0.63 0.00 5.81 2.94 1.69 3.06 7.50 6.00 2.81 3.06

Simple 0.75 0.81 0.13 0.94 0.25 1.25 0.63 1.13 7.25 0.81 1.38 0.19 9.69 4.00 1.06 2.38 10.75 6.38 2.56 4.00F-values 1,1 0.00 0.64 49.00 0.51 0.44 1.00 1.00 1.00 27.04 6.76 36.00 9.00 8.56 289* 6.25 0.54 13.80 0.14 0.08 225*Cut date

Apr + Jun 0.50b 2.00a 0.13b 0.75 0.38 2.25 0.00b 1.25 0.13b 0.50 0.00b 0.13 0.63b 5.38a 0.63bc 3.50 1.25b 8.88a 1.13b 4.25Apr + Oct 0.75b 0.38b 0.75a 1.50 0.13 0.25 0.75b 0.50 11.88a 0.13 1.88a 0.00 14.88a 1.13b 2.13ab 2.13 17.00a 3.25b 3.75a 2.25Jun + Oct 0.00b 0.88ab 0.13b 0.75 0.38 1.00 0.25b 1.13 0.13b 0.13 0.25b 0.13 1.13b 2.75ab 0.38c 2.13 1.50b 4.88ab 1.25b 3.00

Oct 1.75a 1.00ab 0.38ab 1.38 0.63 1.38 1.25a 1.38 10.38a 0.88 1.88a 0.13 14.38a 4.63ab 2.38a 3.13 16.75a 7.75a 4.63a 4.63F-values 3,14 6.93** 3.88* 4.47* 1.44 0.98 2.00 9.18*** 0.90 18.86*** 1.62 4.16* 0.30 32.76*** 3.70* 6.35** 1.83 23.38*** 3.35* 19.90*** 3.22Herbage

disposalLeave 0.56 1.06 0.25 1.13 0.44 1.25 0.69 0.75 4.69 0.56 0.75 0.06 6.75 3.44 1.06 2.94 7.81 6.38 2.50 3.94

Remove 0.94 1.06 0.44 1.06 0.31 1.19 0.44 1.38 6.56 0.25 1.25 0.13 8.75 3.50 1.69 2.50 10.44 6.00 2.88 3.13F-values 1,14 1.80 0.00 1.80 0.04 0.37 0.01 1.87 2.33 1.63 1.22 1.01 0.30 0.84 0.00 2.38 0.71 2.01 0.07 0.65 1.75

InteractionsCut date × Herbage disposal

3,14 0.73 1.31 1.27 1.35 1.35 2.30 5.60** 1.40 0.64 1.62 0.80 1.12 0.68 4.31* 1.11 3.07 0.70 4.36* 1.98 3.25*

Seed mix × Cut date

3,14 0.53 3.75* 7.13** 1.82 0.98 1.04 0.47 1.90 0.13 1.62 1.05 0.30 0.42 0.98 3.14 3.65* 1.72 1.95 1.96 1.23

Seed mix × Herbage disposal

1,14 1.80 1.18 1.80 0.32 0.37 0.55 1.87 2.33 0.88 1.22 1.01 0.30 0.52 1.74 0.00 0.36 1.32 1.33 0.77 0.51


Fig. 1. Effects of a) seed mixture, b) cut date and c) herbage disposal on mean (±SE) cumulative number of sown dicot flowers per m2.

0

200

400

600

800

1000

20/05/04 09/06/04 29/06/04 19/07/04 08/08/04 28/08/04 17/09/04Cum

ulat

ive

sow

n di

cot f

low

ers

m-2

Complex SimpleComplex Simple

Leave RemoveLeave Remove

2004 2005

20042005

2004 2005

a) Seed mixture

b) Cut date

c) Herbage disposal

0

50

100

150

200

250

300

350

400

450

26/05/05 15/06/05 05/07/05 25/07/05 14/08/05 03/09/05Cum

ulat

ive

sow

n di

cot f

low

ers

m-2

0

100

200

300

400

500

600

700

800

20/05/04 09/06/04 29/06/04 19/07/04 08/08/04 28/08/04 17/09/04Cum

ulat

ive

sow

n di

cot f

low

ers

m-2

0

100

200

300

400

500

26/05/05 15/06/05 05/07/05 25/07/05 14/08/05 03/09/05Cum

ulat

ive

sow

n di

cot f

low

ers

m-2

2005

0

100

200

300

400

500

600

700

20/05/04 09/06/04 29/06/04 19/07/04 08/08/04 28/08/04 17/09/04Cum

ulat

ive

sow

n di

cot f

low

ers

m-2

0

50100

150

200250

300

350400

450

26/05/05 15/06/05 05/07/05 25/07/05 14/08/05 03/09/05Cum

ulat

ive

sow

n di

cot f

low

ers

m-2

OctApr + Oct Jun + OctApr + Jun OctApr + Oct Jun + OctApr + Jun

Fig. 2. Effects of a) seed mixture, b) cut date and c) herbage disposal on mean (±SE) cumulative number of bumblebees per 125 m2 plot.



a) Seed mixture

b) Cut date

c) Herbage disposal

0

20

40

60

80

100

120

20/05/04 09/06/04 29/06/04 19/07/04 08/08/04 28/08/04 17/09/04

Cum

ulat

ive

bum

bleb

ees

125

m2

0

10

20

30

40

50

26/05/05 15/06/05 05/07/05 25/07/05 14/08/05 03/09/05

Cum

ulat

ive

bum

bleb

ees

125

m2

0

20

40

60

80

100

120

140

160

20/05/04 09/06/04 29/06/04 19/07/04 08/08/04 28/08/04 17/09/04

Cum

ulat

ive

bum

bleb

ees

125

m2

0

10

20

30

40

50

60

26/05/05 15/06/05 05/07/05 25/07/05 14/08/05 03/09/05

Cum

ulat

ive

bum

bleb

ees

125

m2

010

2030

4050

6070

8090

100

20/05/04 09/06/04 29/06/04 19/07/04 08/08/04 28/08/04 17/09/04

Cum

ulat

ive

bum

bleb

ees

125

m2

0

10

20

30

40

50

60

26/05/05 15/06/05 05/07/05 25/07/05 14/08/05 03/09/05

Cum

ulat

ive

bum

bleb

ees

125

m22004 2005

2004 2005

2004 2005


Fig. 3. Effects of a) seed mixture, b) cut date and c) herbage disposal on mean (±SE) cumulative number of butterflies per 125 m2 plot.

0

2

4

6

8

10

12

14

20/05/04 09/06/04 29/06/04 19/07/04 08/08/04 28/08/04 17/09/04

Cum

ulat

ive

butt

erfli

es 1

25 m

2

0

2

4

6

8

10

26/05/05 15/06/05 05/07/05 25/07/05 14/08/05 03/09/05

Cum

ulat

ive

butt

erfli

es 1

25 m

2

02

46

81012

1416

1820

20/05/04 09/06/04 29/06/04 19/07/04 08/08/04 28/08/04 17/09/04

Cum

ulat

ive

butt

erlie

s 12

5 m

2

0

2

4

6

8

10

12

26/05/05 15/06/05 05/07/05 25/07/05 14/08/05 03/09/05

Cum

ulat

ive

butt

erlie

s 12

5 m

2

0

2

4

6

8

10

12

14

20/05/04 09/06/04 29/06/04 19/07/04 08/08/04 28/08/04 17/09/04

Cum

ulat

ive

butt

erlie

s 12

5 m

2

0

2

4

6

8

10

26/05/05 15/06/05 05/07/05 25/07/05 14/08/05 03/09/05

Cum

ulat

ive

butt

erlie

s 12

5 m

2



a) Seed mixture

b) Cut date

c) Herbage disposal

2004 2005

2004 2005

2004 2005


Appendix 1. Effects of a) Red clover variety, b) cut date and c) herbage disposal on mean (±SE) cumulative number of Trifolium pratense flowers per m2 (R. Pywell, unpublished data).

0

100

200

300

400

500

600

27/05/05 16/06/05 06/07/05 26/07/05 15/08/05 04/09/05 24/09/05

Cum

ulat

ive

T. p

rate

nse

flow

ers

m-2

Britta

Milvus

Wild SomersetWild Berkshire

a) Clover variety

b) Cut date

c) Herbage disposal

2005

0

20

40

60

80

100

120

140

160

30/05/06 19/06/06 09/07/06 29/07/06 18/08/06 07/09/06 27/09/06

Cum

ulat

ive

T. p

rate

nse

flow

ers

m-2

Britta

Milvus

Wild Somerset

Wild Berkshire

2006

Britta Milvus Wild Somerset Wild BerkshireBritta Milvus Wild Somerset Wild Berkshire

0

100

200

300

400

500

600

27/05/05 16/06/05 06/07/05 26/07/05 15/08/05 04/09/05 24/09/05

Cum

ulat

ive

T. p

rate

nse

flow

ers

m-2

AprJunJun and OctOct

010

2030

405060

7080

90100

30/05/06 19/06/06 09/07/06 29/07/06 18/08/06 07/09/06 27/09/06

Cum

ulat

ive

T. p

rate

nse

flow

ers

m-2

AprJunJun and OctOct

2005 2006

Apr Jun Jun and Oct OctApr Jun Jun and Oct Oct

0

100

200

300

400

500

27/05/05 16/06/05 06/07/05 26/07/05 15/08/05 04/09/05 24/09/05

Cum

ulat

ive

T. p

rate

nse

flow

ers

m-2

leaveremove

0

10

20

30

40

50

60

70

80

30/05/06 19/06/06 09/07/06 29/07/06 18/08/06 07/09/06 27/09/06

Cum

ulat

ive

T. p

rate

nse

flow

ers

m-2

leaveremove

2005 2006

leave removeleave remove

B. Environmentally sustainable pest control strategies for Wild Bird Seed mixtures managed under the agri-environment schemes

7. Introduction

There is good evidence that increased winter mortality is a key factor causing serious declines in the breeding populations of at least seven farmland bird species, including the BAP priority species Tree Sparrow (Passer montanus) and Reed bunting (Emberiza schoeniclus) (e.g. Siriwardena et al., 2000; Peach et al., 1999). This has been linked to the loss of winter food resources caused by highly efficient and intensive modern agricultural practices, and in particular the decline of winter stubbles and mixed farming (Stoate et al., 2004). Field experiments and monitoring have demonstrated that sowing mixtures of seed-bearing crops is an effective means of supplementing the over-winter food supply for farmland birds (Stoate et al., 2004; Bradbury et al., 2004). Since 2002 this management prescription has been widely adopted by the UK agri-environment schemes (AES) and it is considered central to reversing the decline of farmland birds in the short term (Vickery et al., 2004). So-called ‘wild bird seed mixtures’ (WBS) (option EF2; Anon., 2005) have proved to be a popular option under the AES, with over 7,000 ha sown by 2005 and a draft target of over 23,000 ha by 2015 under the Cereal Field Margins Habitat Action Plan (Anon., 2006; Stevenson, 2007).

However, many WBS mixtures contain species which are highly susceptible to attack by common agricultural invertebrate pests. Moreover, the severity of pest attack is likely to be increased because these mixtures are invariably spring-sown. Field observation suggest that widely sown Brassica species, including kale (Brassica oleracea L. Acephala group) and fodder radish (Raphanus sativus) are particularly susceptible to damage by a complex of pest species, including yellow-striped flea beetles (Phyllotreta spp.) (Coleoptera: Chrysomelidae) and pollen beetles (Meligethes aeneus) (Coleoptera: Nitidulidae) (Meek et al., 2003). Adult yellow-striped flea beetles overwinter in non-crop areas, hibernating in tussocky grass, in debris under hedges and similar situations. In spring they migrate into fields and feed on the cotyledons, leaves and stems of young seedlings. Feeding damage results in seedling mortality, reduced seedling growth, delayed crop development, uneven maturity and lower seed yield (Putnam 1977; Lamb 1984). Small circular gouges taken mainly from bottom leaf surfaces cause plants to take on a peppered or ‘shot holed’ appearance. Female beetles lay their eggs in the soil near suitable plants in May/June. Larvae feeding on underground portions of plants may also result in decreased plant vigour.

Pollen beetle is an important pest of oilseed rape and other Brassica crops throughout Europe (Winfield, 1992). Furthermore, there is anecdotal evidence that pollen beetle damage can cause a reduction in the final seed yield of the brassica component of WBS mixtures (Meek et al., 2003). Adult pollen beetles overwinter in the soil in sheltered, well-drained sites along hedgerows and similar shelter barriers. The beetles become active during late March to early April in the UK, but do not fly until about a month later, when mean daily temperatures rise above 15ºC. The pollen beetle feeds on pollen from a large number of plant families, but has specialised in only ovipositing in buds of the Brassicaceae. Damage is done by adults and larvae feeding on buds and flowers of the plants (Williams and Free, 1978). Females chew holes in the bases of unopened flower buds and lay 1-3 eggs in each hole. The larvae feed initially on pollen but later move to unopened flowers and finally onto the newly formed seedpods. These activities can significantly reduce final seed yield. Furthermore, both adults and larvae eat pollen from buds and flowers, and it is this feeding which poses an additional threat to pollination and therefore seed production of these crops (Cook, et al., 1999).

The high mobility of flea beetles and pollen beetles mean that crop rotation provides only limited control, so farmers are dependant on several methods of chemical control including seed treatments, granular insecticides and field sprays (Lamb 1984, 1989). The effectiveness of these insecticides in controlling pest species in WBS mixtures remains untested. However, many commercial insecticides are classed as broad-spectrum, so there is considerable potential for damage to non-target invertebrates. Short-term reductions in the availability of invertebrate prey in the critical spring/summer period may have deleterious effects on the breeding success of some farmland bird species.

This study comprised three integrated experiments with the overall aim of developing effective, environment sustainable pest control strategies for WBS mixtures managed under the agri-environment schemes. The experiments were undertaken at both the small-plot and field-plot scales, and tested the following non-exclusive hypotheses:

H1: Common agricultural pests (flea beetle and pollen beetle) cause severe damage to component species of WBS mixtures, and this may result in poor establishment and reduced seed yield for farmland birds in winter (Experiments 1-3); H2: Flea beetle damage at the seedling stage causes a greater reduction in plant density and seed yield than pollen beetle damage to the flowers at the adult stage (Experiment 3);H3: The effects of flea and pollen beetle damage on seed yield are synergistic (Experiment 3).H4: It is possible to control these pest species and reduce their damaging effects through the application of widely available insecticides (Experiments 1-3);H5: Recent advances in pesticide seed treatment technology provide a more effective means of controlling pest species for a longer period, and have a lesser detrimental effect on non-target species compared with conventional foliar pesticide (Experiment 3);H6: Some pesticide regimes have serious detrimental effects on non-target invertebrate species which form an important component of the diet of many farmland birds (Experiments 1-3);H7: It is possible to recommend cost-effective and environmentally sustainable pest management regimes for WBS mixtures managed under the agri-environment schemes.

8. Materials and methods

8.1 Experiment 1: Effects of pest control measures on annual wild bird seed mixtures

8.1.1 Experimental design

In April 2003 a 3 × 200 m arable field margin on sandy loam soil at Hill Farm, Little Wittenham, Oxon (51°38'N 1°12'W) was sown with an annual WBS mixture comprising four seed-bearing crops at 7.5 kg ha -1

(Anon., 2005; Table 7a; cost approx. £40 ha-1). The margin was divided into twenty contiguous plots of 3 × 10 m each and one of four pest control treatments was applied in a randomised block design with five replicates: These comprised: 1. an unsprayed control; and the application of three widely available synthetic pyrethroid insecticides, 2. Toppel 10® (EC 100 g l-1 cypermethrin A.I. applied at 250 ml ha-1, United Phosphorus Limited); 3. Fastac® (EC 100 g l-1 alpha-cypermethrin A.I. applied at 150 ml ha-1, Cyanamid Limited) and 4. Mavrik® (EC 240 g l-1 tau-fluvalinate A.I. applied at 200 ml ha-1, Makhteshim-Agan UK Limited). Mavrik was the only pesticide of those selected which does not have a detrimental effect on honey bees. This suggests that it may have less damaging effects on other non-target invertebrates. The pesticides were each applied according to the label recommendations to each plot using a AZO pedestrian sprayer with a 3 m boom fitted with six Lurmark orange, 110º fan nozzles. Pesticide application was at the 2-4 leaf seedling stage of fodder radish (12 May) and again at the 6-8 leaf stage on 30 May to control flea beetles. A further application was applied at the green to yellow bud stage (21 June) to control pollen beetles.

8.1.2 Monitoring

Three permanently marked 50 × 50 cm quadrats were placed at random within each plot avoiding the outer 1 m. Seedling counts of fodder radish and linseed were made in each quadrat at the seedling stage on 16 May and again on 4 June 2003. On both occasions the percentage leaf area damaged by flea beetles was scored (nil, minor (<5 holes per leaf), moderate (6-20) and severe (>20)) for a sub-sample of fodder radish seedlings within each quadrat. On 14 August all vegetation in each quadrat was clipped off at ground level and placed into a labelled polythene bag. Each bag was sorted and the material oven dried at 80ºC for 24 hours to determine the number of stems, seed pods / heads and seed yield of the four sown species. Prior to harvesting the percentage cover of all vascular plant species and bare ground was estimated from a 1 × 1 m quadrat centred over each 50 × 50 cm quadrat.

The effects of pesticide application on soil surface active invertebrates was determined using pitfall traps (plastic cups of diameter 8 cm, depth 11 cm) sunk into the ground with the top level with the soil surface (modified after Luff, 1996). Each trap was one-third filled with a preservative solution of 50% propylene glycol and 50% water. Four traps were equally spaced along the centreline of each plot. The traps were opened immediately after pesticide application and remained open for 5 days. The effects of pesticide application on epigeal arthropods was investigated using a Dietrick insect suction sampler with a 35-cm diameter intake (D-Vac Ltd, Ventura, CA, USA) (Dietrick, 1961). Each plot was sampled on 22 June, one day after the final pesticide application. Each sample comprised four 30 s samples taken from random locations avoiding the outer 1 m of each plot, giving a total sample area of 0.38m2. For both pitfall traps and D-vac samples all invertebrates were identified to Class or Order with the exception of Coleoptera which were identified to Family. Pest species (flea beetles, pollen beetles) were counted as separate categories. Finally, summary variables of total invertebrates and number of groups (Classes, Orders and Families) were calculated.

8.1.3 Statistical analysis

The effects of pest control treatments on the seedling abundance, percentage pest damage, cover of adult plants and final seed yield of sown and unsown species were examined using two-way analysis of variance (ANOVA) with block and treatment in the model. The effects of pest control measures on soil surface active and epigeal invertebrates were investigated in the same way. Appropriate data transformations (logarithmic and arcsine) were undertaken prior to analysis to meet assumptions of normality of residuals. Pairwise comparisons of the four pest control treatments were made using Tukey’s multiple comparison tests. All analyses were undertaken using Minitab 14 statistical software (Ryan, Joiner and Ryan, 2000).

8.2 Experiment 2: Effects of pest control measures on biennial wild bird seed mixtures


In April 2004 two field corners of 0.6 ha and 0.5 ha at Lower Farm, Radley, Oxon (51°41'N 1°13'W) were sown with a biennial WBS mixture comprising three seed-bearing crops at 7.5 kg ha -1 (Anon., 2005; Table 7b; cost approx. £120 ha-1). Each corner was sub-divided into eight contiguous plots of 20 × 20 m each and one of four pest control treatments was applied in a randomised block design with four replicates: These comprised: 1. an unsprayed control; 2. Toppel 10®; 3. Fastac®; and 4. Mavrik® (see section 8.1.1 for details). The pesticides were each applied according to the label recommendations using a tractor-mounted boom sprayer at the green to yellow bud stage of kale (15 April 2005) to control Pollen beetles.

8.2.2 Monitoring

On 21 August 2005 percentage cover of vascular plant species and bare ground was estimated from four 1 × 1 m quadrat placed at random within each plot, avoiding the outer 5 m. At the same time all seed-bearing stems of Kale were counted and harvested from each quadrat and placed into a labelled polythene bag. Each bag was sorted and the material oven dried at 80ºC for 24 hours to determine the seed yield.

The height and density of the kale canopy (up to 2 m) meant that the only practical way of sampling the effects of pesticide application on epigeal arthropods was by sweep netting (Standen, 2000). A standard sweep net of 0.5 m diameter with a 0.7 m handle was vigorously swept through the canopy of flowering kale 2 days after pesticide application (17 April). One sample unit comprised 10 strokes of approximately 1.0 m width at 1.0 m intervals along three 10 m sample zones situated in the centre of each plot. All invertebrates were identified to Class or Order with the exception of Coleoptera which were identified to Family. Pest species (flea beetles, pollen beetles) were counted as separate categories. Finally, summary variables of total invertebrates and number of groups (Classes, Orders and Families) were calculated.


The effects of pest control treatments on the percentage cover and final seed yield of kale were examined using two-way analysis of variance (ANOVA) with block and treatment in the model. The effects of pest control measures on epigeal arthropods were investigated in the same way. Appropriate data transformations (logarithmic and arcsine) were undertaken prior to analysis to meet assumptions of normality of residuals. Pairwise comparisons of the four pest control treatments were made using Tukey’s multiple comparison tests (Ryan, Joiner and Ryan, 2000).

8.3 Experiment 3: Developing environmentally sustainable pest control strategies for wild bird seed mixtures


In April 2004 and 2005 identical experiments were established at Hill Farm, Little Wittenham, Oxfordshire (51°38'N 1°11'W) and Widmere Farm, Marlow, Buckinghamshire (51°35'N 0° 47'W). At Little Wittenham twenty four plots measuring 24 × 30 m with 30 m discard were established along the margin of a single field growing winter barley. At Marlow sixteen plots (2 replicates) each measuring 20 × 30 m with a 30 m discard were established round the margin of a field growing spring oilseed rape and a further eight plots (1 replicate) were established in a nearby field growing the same crop. At each site in both years the plots were sown with an annual WBS mixture comprising four seed-bearing crops at 10.0 kg ha-1 (Anon., 2005; Table 7c; cost approx. £50 ha-1). One of eight pest control treatment regimes was applied to each plot in a randomised block design with three replicates. These comprised four treatments to control flea beetle in spring: 1. No treatment (control), 2. TMX® seed treatment (FS 100 g A.I. thiamethoxam systemic insecticide + fludioxonil and mefenoxam 100 g A.I. fungicide, applied at 15 ml product kg -1 seed, Syngenta UK Limited),

3. Chinook ® seed treatment (FS imidacloprid insecticide 100 g A.I. + beta-cyfluthrin fungicide 100 g A.I. applied at 20 ml product kg-1 seed, Bayer CropScience UK Limited), and 4. Mavrik® foliar spray (EC 240 g l -1

tau-fluvalinate A.I. applied at 200 ml ha-1, Makhteshim-Agan UK Limited). These spring pest control treatment were combined with two summer pest control treatments in factorial combination: 1. No treatment (control) and 2. Mavrik foliar spray (details as above). The pesticides were each applied according to the label recommendations. Mavrik spray was applied using a tractor-mounted boom sprayer at the 2-4 leaf seedling stage and again at the 6-8 leaf stage of fodder radish to control flea beetles, and finally at the green to yellow bud stage to control pollen beetles.

8.3.2 Monitoring

Permanently marked 50 × 50 cm quadrats were placed in two equally spaced lines each comprising four quadrats situated 4 m from the plot edge. Counts of fodder radish and linseed plants were made in each quadrat on three occasions in each year. On the first count the amount of flea beetle damage was also scored (nil, minor (<5 holes per leaf), moderate (6-20) and severe (>20)) for fodder radish and linseed seedlings within each quadrat. At Little Wittenham counts were made in June, July and October of 2004, and again in June, August and October 2005. At Marlow counts were made in May, June and September 2004 and again in May, July and August 2005. Variation in recording dates reflected different sowing dates and rates of crop development between sites and in different years. Immediately following the final count all sown species in each quadrat were clipped off at ground level and placed into a labelled polythene bag. Each bag was sorted and the material oven dried at 80ºC for 24 hours to determine the number of stems, seed pods / heads and seed yield of the four sown species. Prior to harvesting the percentage cover of all vascular plant species and bare ground was estimated from a 1 × 1 m quadrat centred over each 50 × 50 cm quadrat.

The effects of pest control treatments on soil surface active invertebrates was determined using pitfall traps (plastic cups of diameter 8 cm, depth 11 cm) sunk into the ground with the top level with the soil surface (Luff, 1996). Each trap was one-third filled with a preservative solution of 50% propylene glycol and 50% water. Six traps were placed in a 3 × 2 grid with c. 6 m between traps and the plot edge. The traps were opened immediately following spray application at the 2-3 leaf stage of fodder radish remained open for 5 days. The effects of pesticide application on epigeal arthropods was investigated by sweep netting 1 day after spray application for pollen beetle. A standard sweep net of 0.5 m diameter with a 0.7 m handle was vigorously swept through the canopy of flowering fodder radish. One sample unit comprised 20 strokes of approximately 1.0 m width at 1.0 m intervals along a 20 m transect situated in the centre of each plot. All invertebrates were identified to Class or Order with the exception of Coleoptera which were identified to Family. Pest species (flea beetles, pollen beetles) were counted as separate categories. Finally, summary variables of total invertebrates and number of groups (Classes, Orders and Families) were calculated. Additional measures of the effects of pest management regimes on flea and pollen beetle activity were made by placing four yellow sticky traps (Andersen et al., 2006) (Aeroxon®, Agralan UK Limited) with 5 m spacing along the centreline of each plot (Andersen et al., 2006). Each trap measured 25× 10 cm and was suspended c.20 cm above the soil surface. Traps were exposed for 24 hours immediately following Mavrik foliar pesticide application in the spring and again in early summer. Counts of flea and pollen beetles were made from each trap.


The effects of pest control treatments on the seedling abundance, percentage pest damage, cover of adult plants and final seed yield of sown species were examined using two-way analysis of variance (ANOVA). The factorial design enabled individual effects of spring and summer pest control to be examined, together with any interactions. The effects of pest control measures on soil surface active and epigeal invertebrates were investigated in the same way. Appropriate data transformations (logarithmic and arcsine) were undertaken prior to analysis to meet assumptions of normality of residuals. Pairwise comparisons of the four spring pest control treatments were made using Tukey’s multiple comparison tests. All analyses were undertaken using Minitab 14 statistical software (Ryan, Joiner and Ryan, 2000).

9. Results

9.1 Experiment 1: Effects of pest control measures on annual wild bird seed mixtures

The May census at the 2-4 leaf stage showed there were no significant differences in fodder radish or linseed seedling densities between pest management treatments (Table 8a). A high proportion of fodder radish (63-85%) and linseed (43-57%) seedlings were assessed as having moderate pest damage in all treatments. At the June census (6-8 leaf stage) there remained no significant difference in the density of

fodder radish plants between treatments (Table 8b). However, the density of surviving linseed plants was significantly lower in the unsprayed control compared with those plots treated with the insecticides Toppel and Fastac. A significantly higher proportion of fodder radish plants were assessed as having severe pest damage in the control (65%) compared with all of the plots sprayed with insecticide (0-12%). Similarly, the proportion of pest-damaged linseed was significantly lower in the plots treated with Fastac compared with the other treatments. In August the percentage cover of fodder radish was significantly higher (50%) in the plots sprayed with Toppel compared with the unsprayed control (Table 9). The cover of linseed was significantly higher (30-50%) in the plots sprayed with any insecticide compared with the control. In contrast, the cover of quinoa was significantly higher in the control compared with the plots sprayed with Toppel. The cover of millet and unsown species was higher in the plots sprayed with Mavrik compared with those spayed with Toppel.

At the August harvest there remained no significant difference in the final density of adult fodder radish plants between treatments (Table 10a; Fig. 4a). However, plots sprayed with Toppel yielded a significantly higher number of radish seed pods, total seeds and a higher seed weight compared with the control. Seed yield was increased from 56 (±17) kg ha-1 to 115 (±13) kg ha-1 following application of Toppel, representing an average increase in yield of 51% (Fig. 4b). Seed yield was also increased to 79 (±11) and 82 (±15) kg ha -1

following application of Fastac and Mavrik respectively. There was a significantly higher final density of linseed plants in the plots sprayed with Fastac compared with the control (Table 10b). This resulted in a higher seed yield of 255 (±21) kg ha-1 compared with 192 (±38) kg ha-1 in the control. However, there were no significant differences in any measure of linseed seed yield between treatments. There were significantly fewer quinoa plants in the plots sprayed with Toppel compared with all other treatments (Table 11). Similarly, total seed yield of quinoa was significantly higher in the unsprayed control compared with plots sprayed with Toppel and Fastac. Seed yield of this species was also significantly higher in the plots sprayed with Mavrik compared with those sparyed with Toppel. There were no significant differences in plant density or seed yield of millet between treatments.

There were very few significant effects of foliar insecticide application on the abundance of soil surface active invertebrates as measured by pitfall trapping (Table 12). Application of the insecticides Toppel and Fastac significantly reduced the activity and density of flea beetles compared with in the unsprayed control. The abundance of Cantharid beetles was significantly higher in the plots sprayed with Toppel compared with the control. Finally, there were no overall significant effects of pesticide on total abundance or richness of invertebrates in pitfall traps. There were more significant effects of pesticide application on the abundance of epigeal invertebrates (Table 13). Numbers of pest species (flea beetles, pollen beetles) were all significantly higher in the unsprayed control compared with all of the insecticide treatments. Aphids were significantly more abundant on the plots sprayed with pesticide compared with the unsprayed control in the summer. This may reflect a reduction in the abundance of predatory species. The abundance of Staphylinid beetles was significantly higher in the plots sprayed with insecticide compared with the control. This mostly predatory group may have been attached by the greater abundance of aphid prey. Also, abundance of this family was significantly higher in plots sprayed with Fastac compared with those sprayed with Mavrik. There were no significant effects of pesticide on total abundance or richness of invertebrate groups.

9.2 Experiment 2: Effects of pest control measures on biennial wild bird seed mixtures

The application of insecticide to control pollen beetle attack resulted in no significant differences in percentage cover or abundance of kale (Table 14). Similarly, there were no significant differences in the cover of unsown species or bare ground. There were no significant effects of insecticide application on total seed yield of kale (Table 15). However, there was some evidence that the application of insecticides resulted in an increase in the mean seed weight per kale plant (Toppel 16.3 (±5.4) g and Fastac 12.9 (±2.3 g per plant) compared with the unsprayed control (8.5 (±2.9) g per plant), but these differences were not significant.

However, there were a large number of significant effects of insecticide application on epigeal invertebrates living in the kale canopy (Table 16). Insecticide significantly reduced the abundance of both pollen beetles and flea beetles in the sweep net samples compared to the unsprayed control. In the case of pollen beetles the reduction was dramatic (93-96%). Similarly, insecticide application resulted in a significant reduction in the abundance of spiders compared with the control. Toppel also significantly reduced spider abundance compared with Mavrik. Fastac resulted in a significant reduction in Diptera compared with the control. The abundance of non-ant Hymenoptera and Curculionide beetles was significantly reduced by all insecticides. Finally, total abundance of invertebrates was significantly lower following insecticide application compared to the control. However, pest species accounted for much of this abundance (78%). The number of invertebrate families and orders present in the sweep samples was significantly higher in the control compared to the plots sprayed with Toppel and Fastac, but not Mavrik.

9.3 Experiment 3: Developing environmentally sustainable pest control strategies for wild bird seed mixtures

There were very few significant effects of spring and summer pest control treatments on the density, percentage cover and levels of pest damage in the wild bird seed crop species at Marlow (adjacent spring oilseed rape) or Little Wittenham (adjacent winter barley) in either 2004 or 2005 (Tables 17-19). Seed treatment with TMX significantly reduced the density of fodder radish seedlings at the June and September counts compared with the other treatments at Marlow in 2004 (Table 17). However, there was no effect of this treatment at Little Wittenham in the same year, or at either site in 2005. The establishment and survival of linseed was severely affected by flea beetle damage at both sites in both years (Table 17). Linseed was eliminated from the seed mixture by the final harvest census at Marlow in both years, and at Little Wittenham in 2005. There were no significant effects of pest control treatments on survival.

A significantly higher proportion of fodder radish seedlings in the untreated control plots showed severe pest damage at Marlow in May 2004 compared with plots treated with Chinook or TMX (Table 18). However, this pattern of damage was not repeated at Little Wittenham. Finally, seed treatment with TMX resulted in a significant reduction in the percentage cover of fodder radish compared with the other spring pest control treatments at Marlow in 2004 (Table 19). However, there was no effect of this treatment at Little Wittenham in 2004, or at either site in 2005. There were no significant pest treatment effects on any other sown or unsown species.

There were a greater number of significant effects of pest treatment on final seed yield of fodder radish in 2004 (Table 20). At Marlow seed treatment with TMX significantly reduced the number of mature pod m -2, seeds m-2 and total seed weight g m-2 compared with all other treatments (Table 20a). Similarly, at Little Wittenham TMX treatment resulted in a significant reduction in seed abundance and yield compared with the control and Chinook seed treatment (Table 20b). Summer pesticide application to control pollen beetle resulted in a significant reduction in the number of mature pods and the total seed weight compared with the control at Marlow. However, this effect was not repeated at Little Wittenham. Finally, there was a significant negative interaction between spring seed treatment (TMX and Chinook) and summer Mavrik application on seeds m-2 and total seed weight g m-2 at Little Wittenham. In 2005 there were significantly more fodder radish seeds per pod and seeds m-2 in the TMX treated plots compared with the Mavrik foliar spray (Table 20b). There were no significant effects of either spring or summer pest control treatments on seed yield at Little Wittenham in this year.

Spring pest control treatments resulted in large significant differences in the number of flea beetles caught on yellow sticky traps at both sites in 2004 (Tables 21a,b). Significantly lower numbers of flea beetles were caught in the Mavrik treated plots compared with all other treatments at Marlow in May (Table 21a). Also, significantly greater numbers of this pest species were trapped in the Chinook plots compared with the untreated control. Similarly, significantly fewer flea beetles were trapped in the Mavrik plots at Little Wittenham compared with the control and TMX plots in early June 2004 (Table 21b). There was no significant carry over of spring pest control treatment effects at Marlow in late June (Table 21a). However, the summer application of Mavrik resulted in a significant reduction in flea beetles. There was a small, but significant carry over of spring pest control effects at Little Wittenham (Table 21b) with the Mavrik treated plots having lower densities of flea beetles compared with the TMX plots. However, summer application of Mavrik resulted in a large and significant reduction in the density of flea beetles at this site in late June. There were no significant effects of pest control measures on pollen beetle numbers and activity as estimated by this method. Similarly, there were no significant interactions between spring and summer pest control measures on the abundance and activity of pest species.

Overall there were very few significant effects of spring or summer pest control treatments on soil surface active invertebrate species as measured by pitfall traps at either site in both years (Tables 22 and 23). The effect which were detected were not consistent between sites or years. Spring pest control had no significant effects on the abundance and activity of any invertebrate group recorded at Marlow in 2004 (Table 22a). Millipedes (Order Diplopoda) were significantly more abundant in the unsprayed control compared with those plots treated with Mavrik in the summer. Similarly, Diptera were more abundant in the control compared with the summer Mavrik treatments. At Little Wittenham in 2004 there were significantly fewer spiders (Order Araneae) caught in pitfall traps in the plots treated with Marvik compared with the control and Chinook treatments (Table 22b). In contrast, there were significantly more isopods recorded in the Mavrik plots compared with Chinook. Numbers of Anthicid beetles were significantly lower in the TMX plots compared with the other treatments. Finally, there were significantly more Cantharid beetles caught in the Chinook treated plots compared with the control.

At Marlow in 2005 there were significantly more spiders caught in the untreated control compared with the plots treated with Mavrik and TMX (Table 23a). Also, there were significantly more Hemiptera in the control compared with the Chinook treated plots. At Little Wittenham there were significantly more Hemiptera in the Chinook plots compared with Mavrik and TMX (Table 23b). There were significantly fewer Coleoptera larvae and Anthicid beetles in the TMX plots compared with all others. Finally, there were significantly more flea beetles and weevils (Coleoptera: Curculionidae) in the unsprayed control plots compared with those treated with Mavrik in the summer.

There were a comparatively greater number of significant effects detected on epigeal invertebrates as estimated by sweep netting, and these were much more consistent between sites and years (Tables 24 and 25). At Marlow in 2004 the abundance of the following groups was significantly greater in the unsprayed control compared with the plots spayed with Mavrik in the summer: spiders, insect larvae, non-aphid Hemiptera, flea beetles, and weevils (Table 24a). Moreover, the mean number of invertebrate orders and families was significantly higher in the control compared with the Mavrik treated plots. Similarly, at Little Wittenham in 2004 abundance of the following groups was significantly greater in the control compared with the summer Mavrik treatment: non-aphid Hemiptera, non-ant Hymenoptera, flea beetles, and weevils (Table 24b). In this case both total invertebrate number, and richness of orders and families were both significantly higher in the control compared with the Mavrik treatmeant. At Marlow in 2005 similar effects were recorded, with the abundance of the following groups significantly higher in the control compared with the summer Mavrik treatment: insect larvae, non-aphid Hemiptera, thrips (Order Thysanoptera), non-ant Hymenoptera, flea beetles, pollen beetles and weevils (Table 25). Both total number of invertebrates, and number of orders and families were significantly higher in the control compared with the Mavrik treatment. Similarly, at Little Wittenham in 2005 the following groups were significantly more abundant in the control compared with the Mavrik treated plots: thrips, Diptera, and total invertebrates.

10. Discussion

10.1 What is the potential for common agricultural pests to reduce the establishment and seed yield of Wild Bird Seed mixtures?

The results confirmed there is considerable potential for pest species to reduce the establishment and yield of widely sown seed bearing crops in Wild Bird Seed (WBS) mixtures (Meek et al., 2003). There were large differences in the susceptibility of the different species to pest damage. Linseed was the most susceptible to flea beetle attack in spring. This may partly reflect the small seed size of linseed which produces a relatively small seedling which is more vulnerable to pest attack than larger-seed Brassicaceae species, such as fodder radish and kale. Other important factors include the presence secondary plant compounds which may act as either pest feeding stimulants or deterrents (Barlet and Williams, 1991). Fodder radish was also susceptible to damage by a complex of flea beetle attack in spring and pollen beetle damage in summer resulting in reduced establishment and seed yield. However, this species was never completely eliminated from the WBS mixtures. There was little evidence of pest damage to the other sown species, millet and quinoa. Further research is required into the performance of mixtures based other brassica and non-brassica species with greater natural pest-resistance, and those tolerant of herbicides used to control injurious weeds.

There were large differences in the pattern and intensity of pest attack in all three years and between sites. For example linseed had an average abundance of 87.9 adult plants m -2 in 2003 (Experiment 1), but was almost eliminated from Experiment 3 in 2004 (4.4 m -2) and 2005 (0.0 m-2). Similarly, fodder radish established well in 2003 (48.8 m-2) compared with 2004 (6.9 m-2) and 2005 (4.2 m-2). This is unlikely to reflect the difference in plot size between Experiments 1 (30 m2) and 3 (480 m2) because many previous studies have demonstrated that populations of specialist herbivores tend to decrease on individual plants with increasing area of host plants (e.g. Horton and Capinera, 1987; Al-Doghairi, 1999). This almost certainly reflected the severity of flea beetle outbreaks in 2004 and 2005, combined with dry spring weather which slowed seedling growth during this critical phase. Similarly, there was some evidence that the close proximity of crops likely to harbour flea and pollen beetles, particularly oilseed rape, may have increased the severity of pest attack at Marlow.

10.2 Does seedling damage by flea beetles result in a greater reduction in seed yield than pollen beetle damage to the flowers at the adult stage? Are these effects synergistic?

The severity of pest attack in both 2004 and 2005 meant that the results of the separate pest management regimes examined in Experiment 3 were inconsistent and inconclusive. Also, there were very significant

interactions between spring and summer pest control treatments. Nevertheless, there was generally more evidence of significant beneficial effects of flea beetle control measures on the abundance, cover and seed yield of sown species (11 out of 32 tests) compared with pollen beetle control (2 out of 32 tests). However, it is possible that the single pollen beetle control treatment (Mavrik spray) was not the most effective means of controlling this pest species.

10.3 Is it possible to control these pest species and reduce their damaging effects through the application of widely available insecticides?

Experiments 1 and 2 confirmed that the application of low cost, widely available foliar pesticide sprays, either at the small plot- or field-scale, was equally as effective in reducing the abundance of flea beetles and pollen beetles in WBS mixtures (Al-Doghairi, 1999). Casual observations suggest that the effects of the insecticides lasted for 3-5 days after application. This suggests that frequent application may be necessary in some years and in situations where pest pressure is particularly high. This may pose unacceptable risks to non-target invertebrates and adjacent non-crop habitats. Prolonged insecticide application may also increase the resistance capacity of the pest species. Further research is required into the effectiveness of naturally-occurring repellent and deterrent compounds (Palaniswamy and Wise, 1994) which are also permissible on organic farms.

10.4 Do new generation pesticide seed treatments provide a more effective means of controlling pest species than conventional foliar insecticides?

Research on other brassica crops suggests that seed treatment with systemic insecticide gives longer-lasting protection to both the foliage and root system of seedling plants compared with frequent application of post-emergence foliar sprays (Ester, de Putter and van Bilsen, 2003). The single, targeted application of the seed treatment is also considered an environmentally friendly alternative for protecting brassica crops against pests that occur frequently compared with frequent spraying of the whole crop areas with broad-spectrum insecticide. Unfortunately pest pressure was so severe in both years of Experiment 3 that both these new seed treatments and conventional foliar spray failed to reduce the abundance of flea beetles. Consequently, the effects on cover, abundance and seed yield of sown species were inconclusive. Nevertheless, in 2004 seed treatment with TMX and Chinook resulted in a significantly lower proportion of fodder radish seedlings with severe pest damage at one site. A similar, but non-significant trend was observed at the other site. However, treatment with TMX was found to significantly reduce seed abundance and yield of fodder radish in 2004. The most likely explanation for this was the incorrect dosing of the larger fodder radish seeds compared with oilseed rape during off label application of this product. This was confirmed in the following year when this treatment resulted in the highest seed abundance and yield at one site. Further experiments are required to test the effectiveness of these seed treatments either in years where pest pressure is lower, or under conditions where the abundance of pest species can be more carefully controlled.

10.5 Do some pesticide treatment regimes have serious detrimental effects on non-target invertebrate species?

Previous research has shown that removing arable land from production and creating wildlife habitat by sowing mixtures of broad-leaved plants is a highly effective means of providing habitat for invertebrates (e.g. Meek et al., 2002; Pywell et al., 2004; 2005). Recent research confirms that these Wild Bird Seed mixtures provide important habitat for a wide diversity of invertebrates (Pywell et al., 2007). The susceptible brassica species, (e.g. kale, mustard and fodder radish) are known to be an important source of pollen and nectar, and are food plants for a wide range of phytophagous invertebrates. Consequently WBS mixtures may provide an important additional source of invertebrate prey for farmland birds during the critical breeding season. Frequent application of broad-spectrum insecticides to control pest species may therefore pose a significant risk non-target invertebrates and indirectly to bird breeding success. Further research is required to quantify the potential value of WBS mixtures for providing habitat and food resources for breeding birds on farmland.

The effects of pesticide application on non-target invertebrates were very consistent across all three experiments in this study. There were very few significant detriment effects of either post-emergence insecticide spray or insecticide seed treatment on soil surface active invertebrates as recorded by pitfall trapping. This concurred with results of the Defra-funded Talisman and Scarab projects (Young et al., 2001). This group of invertebrates are dominated by predatory species (Standen, 2000; Meek et al., 2002) which are unlikely to be affected by the systemic insecticide seed treatments. They also contain many highly mobility species, such as carabid beetles, which are able to rapidly re-colonise the plots following insecticide application. Finally, this may reflect the ability of these invertebrates to shelter under the crop canopy and in soil gaps, thus avoiding direct contact with the insecticide. In contrast, application of foliar insecticide in the

summer had a large number of significant negative effects on epigeal invertebrates. Many of the groups affected were phytophagous, such as bugs (Hemiptera) and weevils (Coleoptera: Curculionide). These groups are much less mobile and rely on the sown species as either food or habitat.

10.6 Is it possible to recommend cost-effective and environmentally sustainable pest management regimes for wild bird seed mixtures managed under the agri-environment schemes?

Research has shown that the brassica component of WBS mixtures support the highest densities and widest range of bird species (Henderson, Vickery and Carter, 2004). However, our results show that in some years pest pressure on these species is so high that highly intensive and environmentally unsustainable management with insecticides may be required to establish and maintain these species. The following approach is therefore recommended as a means of minimising the risk of severe pest damage and maximising the seed yield from brassica-based WBS mixtures: i) vary the composition of seed mixtures between patches and always include a wide range of pest resistant ‘insurance’ species; ii) avoid sowing too early (April – May recommended) into cold, dry or otherwise poor-quality seed beds; iii) avoid establishment adjacent to oilseed rape; iv) rotate the location of the patch every 1-2 years to avoid build up of pests; iv) dress the seeds of susceptible species (fodder radish, kale and linseed) with a combined insecticide and fungicide to reduce flea beetle damage in spring; and vi) avoid summer pesticide application if possible to reduce the risk of significant damage to non-target invertebrate populations.

11. Acknowledgements

This study was funded by a commission (BD1623) from the Department for Environment, Food and Rural Affairs. The authors are grateful to William White (Widmere Farm, Marlow) and John Sargent (Hill Farm, Little Wittenham) for managing and hosting the experiments. We also thank Carole Freeland for data input and Tim Sparks for advice on statistical analysis.

12. Tables and figures

Table 7. Details of the seed mixture sown in Experiments 1-3.

a) Experiment 1.Species % kg ha-1

Echinochloa esculenta White Millet 35 2.6Linum usitatissimum Linseed 35 2.6Raphanus sativus Fodder

Radish 15 1.1Chenopodium quinoa Quinoa 15 1.1Total 100 7.5 b) Experiment 2.Species % kg ha-1

x Triticosecale Wittmack Triticale 70 28Brassica oleracea L. Acephala group Kale 15 6Chenopodium quinoa Quinoa 15 6Total 100 40 c) Experiment 3.Species % kg ha-1

Echinochloa esculenta White Millet 25 2.5Linum usitatissimum Linseed 25 2.5Raphanus sativus Fodder

Radish 20 2.0Chenopodium quinoa Quinoa 30 3.0Total 100 10.0

Table 8. Effects of spring and summer pest control measures on the abundance and pest damage of fodder radish and linseed seedlings in a) May and b) June in an annual Wild Bird Seed mixture (Expt. 1). Means with the same letter in the same column are not significantly different (P> 0.05).

a) May censusFodder radish Linseed

Seedlings m-2 % damage


Treatment Minor Moderate Severe Minor Moderate Severe1. Nil 66.4 13.3 63.3 23.3 80.0 1.7 55.0 43.32. Toppel 79.5 26.7 71.7 1.7 97.3 11.7 43.3 45.03. Fastac 86.1 16.7 76.7 6.7 90.4 6.7 48.3 45.04. Mavrik 99.2 5.0 85.0 10.0 100.3 6.7 56.7 36.7Anova F3,12

0.91ns 0.55 ns 0.47 ns 2.89 ns1.16 ns 0.42

ns0.60 ns 0.28 ns

b) June censusFodder radish Linseed



Treatment Minor Moderate Severe Minor Moderate Severe1. Nil 67.7 1.7b 33.3 65.0a 65.9b 0.0b 100.0a 0.02. Toppel 76.0 21.7ab 66.7 11.7b 89.6a 0.0b 100.0a 0.03. Fastac 74.4 58.3a 33.3 0.0b 98.9a 13.3a 86.7b 0.04. Mavrik 77.1 48.3a 38.3 8.3b 81.3ab 0.0b 100.0a 0.0Anova F3,12 0.20 ns 8.26** 2.19 ns 9.89*** 10.37*** 5.86* 5.86* -

ns = no significant difference; * = P < 0.05; ** = P < 0.01.

Table 9. Effects of spring and summer pest control measures on the mean percentage cover of sown and unsown species in an annual Wild Bird Seed mixture (Expt. 1). Means with the same letter in the same column are not significantly different (P> 0.05).

Fodderradish Linseed Quinoa

Whitemillet

Unsownspecies

Bareground

Treatment1. Nil 18.2b 7.7b 40.7a 1.5ab 21.6ab 20.12. Toppel 39.0a 19.0a 24.0b 0.8b 19.4b 15.73. Fastac 24.7ab 22.5a 29.0ab 1.0ab 21.7ab 20.34. Mavrik 24.0ab 17.5a 27.0ab 1.9a 27.9a 22.4Anova F3,12 6.01** 8.11** 3.87* 3.92* 4.10* 1.03 ns


Table 10. Effects of spring and summer pest control measures on the seed production m-2 of a) fodder radish and b) linseed in an annual Wild Bird Seed mixture (Expt. 1). Means with the same letter in the same column are not significantly different (P> 0.05).

a) Fodder radish

Treatment No. plants

No. Mature pods

Seeds per pod

Total seeds

Dry seed weight (g)

1. Nil 48.8 173.1b 3.7 647.3b 5.6b2. Toppel 53.3 300.0a 4.4 1322.0a 11.5a3. Fastac 48.5 231.2ab 4.4 1020.4ab 7.9ab4. Mavrik 44.5 213.1ab 4.3 916.2ab 8.2abAnova F3,12 0.99 ns 3.61* 3.21 ns 4.39* 3.58*

b) Linseed

Treatment No. stems No. Mature pods

Seeds per pod

Total seeds

Dry seed weight (g)

1. Nil 72.5b 513.3 118.3 4000.1 19.22. Toppel 93.9ab 664.8 149.9 5317.8 25.73. Fastac 104.5a 663.2 174.7 4974.4 25.54. Mavrik 80.8ab 516.3 230.0 4137.8 18.3Anova F3,12 3.65* 2.30 ns 1.58 ns 2.01 ns 2.96 ns

ns = no significant difference; * = P < 0.05.

Table 11. Effects of spring and summer pest control measures on the dry weight and seed production m-2 of sown and unsown species in an annual Wild Bird Seed mixture (Expt. 1). Means with the same letter in the same column are not significantly different (P> 0.05).

Treatment No. stems Millet

Dry seed weight (g) Millet

No. stems Quinoa

Dry seed weight (g) Quinoa

Dry weight (g) unsown species

1. Nil 25.3 1.4 41.6a 18.7a 49.22. Toppel 18.1 1.1 25.3b 7.0c 34.43. Fastac 28.8 2.1 41.3a 10.9bc 35.84. Mavrik 17.3 1.6 36.8a 12.5ab 37.0Anova F3,12 3.22ns 0.50ns 10.26*** 13.97*** 0.50ns

ns = no significant difference; *** = P < 0.001.

Fig. 4. Effects of spring and summer pest control measures on a) the mean (± SE) density of sown species m-2 and b) mean (± SE) seed weight m-2.

a) Density of sown species m-2

0

50

100

150

200

250

1. Nil 2. Toppel 3. Fastac 4. Mavrik

No. p

lant

s m

-2 Fodder radish

Linseed

Quinoa

Millet

b) Seed weight m-2

0

10

20

30

40

50

1. Nil 2. Toppel 3. Fastac 4. Mavrik

Seed

wei

ght (

g) m

-2

Fodder radish

Linseed

Quinoa

Millet

Table 12. Effects of spring and summer pest control measures on the abundance of ground-dwelling invertebrates from pitfall traps in an annual Wild Bird Seed mixture (Expt. 1). Means with the same letter in the same row are not significantly different (P> 0.05).

1. Nil 2. Toppel 3. Fastac 4. Mavrik Anova F3,12

CLASS OLIGOCHAETA 0.0 0.0 0.0 0.0 -CLASS GASTROPODASlugs 0.0 0.0 0.4 0.2 1.83nsLand Snails 0.0 0.0 0.0 0.0 -CLASS ARACHNIDAOrder Araneae 32.6 24.4 23.8 27.0 2.17nsOrder Opiliones 0.8 1.2 0.6 0.2 1.94nsCLASS CRUSTACEAOrder Isopoda 0.6 0.4 0.8 0.8 0.15nsCLASS CHILOPODA 0.2 0.8 0.6 1.0 0.72nsCLASS DIPLOPODA 1.0 1.6 1.0 1.0 0.27nsCLASS INSECTAOrder Collembola 39.0 34.0 38.0 45.0 0.47nsOrder Dermaptera 0.0 0.4 0.2 0.4 1.00nsOrder Hemiptera 0.6 0.2 0.6 0.4 0.44nsOrder Lepidoptera 0.2 0.2 0.0 0.0 1.00nsOrder Diptera (Adults) 182.0 172.0 174.0 194.0 0.80nsOrder Diptera (Larvae) 0.0 0.0 0.0 0.0 -Order Hymenoptera (Ants) 5.2 1.8 7.2 7.2 0.71nsOrder Hymenoptera (Non-Ants) 0.8 0.6 1.6 1.8 0.76nsOrder ColeopteraLarvae 2.2 2.0 2.2 0.6 1.87nsFamily Anthicidae 1.2 0.6 1.4 0.6 0.71nsFamily Apionidae 0.0 0.4 0.0 0.2 1.29nsFamily Byrrhidae 0.0 0.0 0.0 0.2 1.00nsFamily Cantharidae 0.6b 4.0a 3.0ab 1.2ab 5.24*Family Carabidae 36.2 32.8 35.8 35.4 0.17nsFamily ChrysomelidaeFamily Chrysomelidae (Flea Beetles) 392.0a 265.0b 236.2b 284.0ab 4.96*Family Chrysomelidae (Non Flea Beetles) 0.0 0.0 0.0 0.2 1.00nsFamily Coccinellidae 4.0 1.4 1.8 2.0 1.71nsFamily Cryptophagidae 1.6 0.2 0.8 0.2 1.08nsFamily Curculionidae 3.8 4.2 3.6 5.2 0.47nsFamily Elateridae 0.4 0.0 0.2 0.0 0.65nsFamily Hydrophilidae 0.0 0.0 0.0 0.0 -Family Lathridiidae 0.0 0.0 0.0 0.0 -Family Leiodidae 0.0 0.0 0.0 0.0 -Family Nitidulidae 0.2 1.2 0.8 0.4 0.89nsFamily Scarabaeidae 0.0 0.0 0.0 0.4 2.67nsFamily Staphylinidae 44.8 36.0 28.6 34.6 2.05ns

Total invertebrates 750.0 585.4 563.2 644.2 2.89nsNumber of families 11.0 11.4 11.8 11.2 0.66ns


Table 13. Effects of spring and summer pest control measures on the abundance m-2 of canopy-dwelling invertebrates from vacuum sampling in an annual Wild Bird Seed mixture (Expt. 1). Means with the same letter in the same row are not significantly different (P> 0.05).


CLASS ARACHNIDAOrder Araneae 10.4 11.4 6.8 9.9 0.91nsOrder Opiliones 1.6 0.5 0.5 0.0 2.01nsOrder Acarina 4.2 5.2 3.6 1.0 1.11nsCLASS INSECTALarvae 12.0 6.2 17.1 14.0 1.51nsOrder Collembola 5.7 19.7 18.7 14.5 1.39nsOrder Hemiptera (Aphids) 50.4b 190.2a 285.8a 143.4a 15.01***Order Hemiptera (Non-aphids) 7.8 10.4 5.2 4.7 0.95nsOrder Thysanoptera 19.7 14.5 10.9 10.4 1.60nsOrder Lepidoptera 1.0 0.0 0.0 0.5 0.65nsOrder Diptera 18.2 21.3 25.5 18.7 1.55nsOrder Hymenoptera (Ants) 1.6 1.0 1.6 3.6 0.63nsOrder Hymenoptera (Non-Ants) 79.5 128.3 139.3 126.3 3.12nsOrder ColeopteraFamily Apionidae 0.0 0.0 0.5 0.0 1.00nsFamily Cantharidae 4.2 3.1 0.5 1.6 3.01nsFamily Carabidae 2.6 1.6 3.1 2.1 0.13nsFamily Chrysomelidae (Flea Beetles) 61.3a 11.4b 21.8b 20.3b 18.74***Family Chrysomelidae (Non Flea Beetles) 0.0 0.0 0.0 0.0 -Family Coccinellidae 1.6 0.0 0.0 0.0 2.46nsFamily Elateridae 0.0 0.0 0.5 0.0 0.95nsFamily Lathridiidae 1.0 0.5 1.0 0.0 0.62nsFamily Melyridae 0.0 0.0 0.5 0.0 1.00nsFamily Nitidulidae (Pollen Beetles) 725.9a 304.0b 307.6b 408.4b 17.67***Family Oedemeridae 0.0 0.5 0.5 0.0 0.62nsFamily Staphylinidae 7.8c 17.7ab 19.7a 10.4b 9.27**Family Curculionidae 1.0 1.6 0.5 4.2 5.85*

Total invertebrates 1017.4 749.3 871.4 794.0 2.47nsNumber of families 14.8 14.2 14.4 14.0 0.26ns

ns = no significant difference; * = P < 0.05; ** = P < 0.01; *** = P < 0.001.

Table 14. Effects of spring and summer pest control measures on mean percentage cover and abundance of sown and unsown species in a biennial Wild Bird Seed mixture (Expt. 2).

Treatment % cover kale

No. kale plants in

seed

Total no. kale

plants

% cover unsown species

% cover bare

ground1. Nil 43.5 5.5 8.0 23.7 17.52. Toppel 41.1 3.1 5.9 28.9 20.73. Fastac 46.7 2.9 9.2 20.6 18.34. Mavrik 28.8 2.4 5.6 37.3 24.5Anova F3,9 1.25ns 0.78ns 2.99ns 2.04ns 0.60ns

ns = no significant difference.

Table 15. Effects of spring and summer pest control measures on the mean seed yield (g m-2) of kale in a biennial Wild Bird Seed mixture (Expt. 2).

Treatment Total seed weight (g m-

2)Seed weight (g) per plant

1. Nil 27.3 8.52. Toppel 28.2 16.33. Fastac 31.0 12.64. Mavrik 24.2 7.3Anova F3,9 0.42ns 1.81ns

ns = no significant difference.

Table 16. Effects of spring and summer pest control measures on the abundance m-2 of canopy-dwelling invertebrates from sweep sampling in a biennial Wild Bird Seed mixture (Expt. 2). Means with the same letter in the same row are not significantly different (P> 0.05).


CLASS ARACHNIDAOrder Araneae 4.8a 0.3c 0.5bc 1.8b 17.70***CLASS INSECTAOrder Hemiptera 0.3 2.0 0.0 0.0 0.76nsOrder Thysanoptera 1.5 1.3 0.3 0.8 0.88nsOrder Diptera 13.0a 3.8ab 3.0b 5.0ab 4.02*Order Hymenoptera (Ants) 0.0 0.0 0.0 0.0 -Order Hymenoptera (Non-Ants) 24.8a 3.0b 2.5b 3.3b 14.39***Order ColeopteraFamily Carabidae 0.0 0.0 0.3 0.0 -Family Chrysomelidae (Flea Beetles) 1.5a 0.0b 0.0b 0.0b 6.99**Family Chrysomelidae (Non Flea Beetles) 0.0 0.0 0.0 0.0 -Family Curculionidae 10.3a 0.3b 0.3b 0.8b 29.69***Family Lathridiidae 0.0 0.0 0.0 0.3 1.00nsFamily Nitidulidae (Pollen Beetles) 197.8a 7.0b 13.8b 9.0b 39.83***

Total invertebrates 254.0a 17.5b 20.5b 21.0b 35.70***Number of families 6.8a 3.5b 3.8b 4.8ab 5.79*


Table 17. Effects of spring and summer pest control measures on the mean density Fodder radish and Linseed plants m -2 recorded in 2004 and 2005. Means with the same letter in the same column are not significantly different (P> 0.05).

a) Marlow2004 2005

No. Fodder radish plants m-2 No. Linseed plants m-2 No. Fodder radish plants m-2 No. Linseed plants m-2

1. Spring treatment May Jun Sep May Jun Sep May Jul Aug May Jul AugNil 6.75 6.42ab 4.33a 2.00 0.00 0.00 8.75 6.00 6.17 1.25 0.00 0.00TMX 3.25 4.25b 1.50b 7.58 0.00 0.00 12.00 7.00 7.83 4.17 0.00 0.00Chinook 7.58 7.83a 5.83a 8.17 0.00 0.00 10.00 5.67 5.67 5.67 0.17 0.00Mavrik 6.08 6.92a 4.33a 8.33 0.00 0.00 5.33 3.25 3.52 0.67 0.08 0.00Anova F-value 3,14 2.49ns 3.65* 4.08* 2.28ns - - 2.16ns 1.70ns 1.48ns 0.83ns 1.36ns -2.Summer treatmentNil 5.67 6.33 4.50 7.04 0.00 0.00 10.83 6.46 6.59 1.88 0.00 0.00Mavrik 6.17 6.38 3.50 6.00 0.00 0.00 7.21 4.50 5.00 4.00 0.13 0.00Anova F-value 1,14 0.18ns 0.00ns 1.25ns 0.27ns - - 3.62ns 1.35ns 2.25ns 2.03ns 1.10ns -3.InteractionsSpring×summer 3,14 0.03ns 3.52* 1.68ns 0.54ns - - 0.70ns 1.70ns 0.34ns 0.83ns 0.15ns -

b)LittleWittenham2004 2005

No. Fodder radish plants m-2 No. Linseed plants m-2 No. Fodder radish plants m-2 No. Linseed plants m-2

1. Spring treatment Jun Jul Oct Jun Jul Oct Jun Aug Oct May Jun SepNil 11.18 9.27 8.50 11.43 6.93 10.33 5.43 3.68 3.76 0.17 0.00 0.00TMX 9.84 7.67 8.00 11.59 10.84 11.33 10.17 4.09 2.92 1.42 0.25 0.00Chinook 13.34 10.93 13.50 11.93 8.67 6.83 5.92 1.87 1.76 0.51 0.00 0.00Mavrik 10.93 8.43 9.33 9.84 5.35 6.50 4.92 1.83 1.67 0.00 0.00 0.00Anova F-value 3,14 0.49ns 0.75ns 1.75ns 0.14ns 1.61ns 0.24ns 1.32ns 1.07ns 1.25ns 2.00ns 3.08ns -2.Summer treatmentNil 11.01 8.64 10.33 10.59 7.72 8.00 6.34 2.72 2.55 0.50 0.00 0.00Mavrik 11.63 9.51 9.33 11.80 8.17 9.50 6.88 3.02 2.51 0.54 0.13 0.00Anova F-value 1,14 0.09ns 0.29ns 0.28ns 0.24ns 0.06ns 0.16ns 0.07ns 0.07ns 0.00ns 0.01ns 3.08ns -3.InteractionsSpring×summer 3,14 1.99ns 3.26* 0.66ns 1.66ns 1.76ns 0.34ns 0.92ns 0.20ns 0.46ns 0.19ns 3.08ns -


Table 18. Effects of spring and summer pest control measures on estimated amount of pest damage of Fodder radish in 2004 and 2005. Means with the same letter in the same column are not significantly different (P> 0.05).

a) Marlow

2004 20051. Spring treatment Minor Moderate Severe Minor Moderate SevereNil 6.0 29.1 46.2a 12.8 28.0 39.4TMX 19.8 18.1 18.4b 11.6 40.7 29.4Chinook 26.9 33.2 21.2b 8.4 28.8 47.5Mavrik 11.1 28.6 22.7ab 3.6 19.1 34.2Anova F-value 3,14 1.77ns 1.10ns 4.52* 0.52ns 1.89 ns 1.06 ns2.Summer treatmentNil 20.7 26.1 23.0 11.9 31.6 37.5Mavrik 11.2 28.4 31.2 6.2 26.7 37.9Anova F-value 1,14 1.88ns 0.15ns 1.86ns 1.00ns 0.57ns 0.00ns3.InteractionsSpring×summer 3,14 1.11ns 0.80ns 0.78ns 0.51ns 0.12ns 0.08ns

b) LittleWittenham

2004 20051. Spring treatment Minor Moderate Severe Minor Moderate SevereNil 0.0 20.1 71.5 0.0 10.9 50.5TMX 3.9 13.6 65.8 0.0 6.8 59.9Chinook 6.6 32.1 55.0 4.2 4.2 47.9Mavrik 3.4 25.7 47.9 2.5 2.8 48.8Anova F-value 3,14 1.39ns 1.79ns 1.65ns 0.89ns 1.14ns 0.22ns2.Summer treatmentNil 4.6 27.2 57.8 0.0 2.9 56.0Mavrik 2.4 18.6 62.4 3.3 9.5 47.6Anova F-value 1,14 0.93ns 2.16ns 0.31ns 2.37ns 3.95ns 0.51ns3.InteractionsSpring×summer 3,14 0.63ns 3.06ns 0.79ns 0.89ns 0.24ns 0.87ns


Table 19. Effects of spring and summer pest control measures on the mean percentage cover of sown and unsown species recorded in 2004 and 2005. Means with the same letter in the same column are not significantly different (P> 0.05).

a) Marlow2004 2005


Whitemillet

Unsownspecies

Bareground

Fodder radish Linseed Quinoa

Whitemillet

Unsown species

Bare ground

1. Spring treatmentNil 7.09a 0.00 3.44 0.79 30.31 32.88 14.92 0.00 13.09 5.30 32.21 34.58aTMX 2.02b 0.00 2.82 0.71 36.29 28.60 17.54 0.003 15.33 3.61 20.71 44.35bChinook 12.13a 0.00 3.06 0.67 25.56 28.29 9.17 0.00 16.31 3.67 28.67 41.88abMavrik 10.17a 0.00 6.18 0.77 27.04 25.67 5.04 0.00 17.33 3.94 26.31 47.42bAnova F-value 3,14 11.93*** - 1.18ns 0.21ns 1.54ns 0.66ns 1.91ns 1.00ns 0.20ns 1.19ns 0.62ns 4.05*2.Summer treatmentNil 7.55 0.00 3.54 0.44 30.23 28.91 13.26 0.00 17.90 3.58 26.17 39.40Mavrik 8.15 0.00 4.20 1.03 29.38 28.81 10.07 0.001 13.14 4.67 27.78 44.72Anova F-value 1,14 0.02ns - 0.05ns 2.66ns 0.20ns 0.00ns 0.84ns 1.00ns 1.33ns 0.98ns 0.05ns 3.74ns3.InteractionsSpring×summer 3,14 0.66ns - 1.79ns 1.62ns 1.59ns 1.54ns 0.51ns 1.00ns 0.34ns 3.52* 0.61ns 2.81ns

b) Little Wittenham2004 2005


Whitemillet

Unsownspecies

Bareground

Fodder radish Linseed Quinoa

Whitemillet

Unsown species

Bare ground

1. Spring treatmentNil 23.23 2.98 14.94 4.25 31.76 24.69 11.88 0.00 4.11 10.60 28.02 44.25TMX 36.73 2.17 14.69 3.17 13.93 32.15 9.23 0.04 5.40 6.88 36.71 42.94Chinook 24.67 2.12 13.04 4.13 27.40 29.85 6.02 0.00 4.26 10.81 34.31 44.69Mavrik 21.00 0.61 17.96 2.85 33.84 25.81 5.11 0.00 4.94 13.23 46.25 30.61Anova F-value 3,14 2.22ns 2.53ns 0.12ns 1.15ns 1.56ns 0.54ns 1.74ns 1.00ns 0.02ns 0.82ns 2.86ns 2.37ns2.Summer treatmentNil 25.18 1.95 14.86 3.02 27.11 30.44 9.74 0.00 4.59 10.48 36.41 38.30Mavrik 27.64 1.99 15.46 4.18 26.35 25.82 6.38 0.02 4.77 10.28 36.24 42.94Anova F-value 1,14 0.43ns 0.03ns 0.12ns 2.55ns 0.0ns 0.70ns 1.60ns 1.00ns 0.06ns 0.19ns 0.00ns 1.02ns3.InteractionsSpring×summer 3,14 1.18ns 0.15ns 1.63 1.44ns 0.51ns 0.05ns 0.30ns 1.00ns 0.07ns 0.97ns 2.08ns 0.81nsns = no significant difference; * = P < 0.05.

Table 20. Effects of spring and summer pest control measures on the mean seed production in Fodder radish m-2 recorded in 2004 and 2005. Means with the same letter in the same column are not significantly different (P> 0.05).

a) Marlow2004 2005

No. Mature pods m-2

Seeds per pod

Seeds per m2

Dry seed weight (g m-2)

No. Mature pods m-2

Seeds per pod

Seeds per m2


1. Spring treatmentNil 170.7a 5.4 1177.5a 7.8a 384.9 4.1ab 2076.6ab 10.2TMX 45.3b 5.4 243.7b 1.9b 489.3 5.1a 2779.4a 13.1Chinook 273.2a 6.0 1667.3a 13.6a 253.5 3.4ab 1297.8ab 6.4Mavrik 199.2a 5.9 1220.3a 10.5a 50.8 3.1b 174.6b 1.1Anova F-value 3,14 12.08*** 2.08ns 8.57** 9.44*** 2.99ns 3.64* 3.68* 2.32ns2.Summer treatmentNil 200.0 5.8 1294.8 10.5 304.3 3.8 1619.7 8.0Mavrik 140.2 5.6 859.7 6.4 285.0 4.1 1544.5 7.4Anova F-value 1,14 5.21* 0.72ns 4.19ns 5.36* 0.02ns 0.39ns 0.05ns 0.05ns3.InteractionsSpring×summer 3,14 2.98ns 0.65ns 1.79ns 2.37ns 0.83ns 0.45ns 1.02ns 0.33ns

b) Little Wittenham2004 2005

No. Mature pods m-2

Seeds per pod

Seeds per m2


No. Mature pods m-2

Seeds per pod

Seeds per m2


1. Spring treatmentNil 24.9 6.2a 3617.5a 27.0a 76.2 3.1 385.9 2.6TMX 20.0 5.0b 927.0b 8.7b 59.7 4.0 308.3 2.3Chinook 25.6 6.4a 3701.7a 28.6a 10.0 2.1 49.0 0.4Mavrik 23.5 5.9ab 5482.3ab 32.4ab 11.9 2.0 57.1 0.4Anova F-value 3,14 1.60ns 4.51* 4.25* 4.32* 1.94ns 1.86ns 1.68ns 1.68ns2.Summer treatmentNil 400.1 5.7 2740.0 23.0 88.0 3.0 186.3 1.3Mavrik 458.6 6.1 4124.2 25.4 80.4 2.6 213.9 1.6Anova F-value 1,14 0.04ns 0.30ns 0.13ns 0.03ns 0.01ns 0.76ns 0.03ns 0.05ns3.InteractionsSpring×summer 3,14 1.60ns 5.84** 5.70** 6.66** 0.33ns 1.16ns 0.55ns 0.02nsns = no significant difference; * = P < 0.05; ** = P < 0.01; *** = P < 0.001.

Table 21. Effects of spring and summer pest control measures on the abundance of Flea beetles from sticky trapping in an annual Wild Bird Seed mixture in 2004 (Expt. 3). Means with the same letter in the same row are not significantly different (P> 0.05).

a) Marlow

1. Spring treatment 2. Summer treatment3.

Interactions Nil TMX Chinook Mavrik F3,14 Nil Mavrik F1,14 F3,14

Flea Beetles May 71.8b 82.2ab 93.9a 57.8c 16.97*** 72.9 79.9 4.41ns 2.26nsJune 0.1 0.2 0.0 0.1 0.83ns 0.2 0.0 4.85* 0.14ns

Pollen Beetles May 39.7 45.2 41.5 40.7 0.13ns 42.3 41.3 0.00ns 0.22ns

June 1.2 1.7 1.2 1.0 0.35ns 1.5 1.0 2.20ns 0.31ns

b) Little Wittenham


Interactions Nil TMX Chinook Mavrik F3,14 Nil Mavrik F1,14 F3,14

Flea Beetles Early June 11.9a 13.1a 9.4ab 5.8b 7.34** 10.2 9.9 0.03ns 0.74nsLate June 0.5ab 0.6a 0.3ab 0.1b 3.20* 0.6 0.2 8.94** 0.36ns

Pollen Beetles Early June 0.8 0.9 1.7 0.5 2.31ns 1.2 0.7 2.72ns 1.43ns

Late June 18.5 14.9 20.9 19.0 0.70ns 17.2 19.5 0.05ns 0.29ns


Table 22. Effects of spring and summer pest control measures on the mean abundance of soil surface active invertebrates as estimated by pitfall trapping in an annual Wild Bird Seed mixture in 2004 (Expt. 3). Means with the same letter in the same row are not significantly different (P> 0.05).

a) Marlow

Order / family 1. Spring treatment 2. Summer treatment3.

Interactions

Nil TMXChinoo

k Mavrik F3,14 Nil Mavrik F1,14 F3,14

CLASS OLIGOCHAETA 0.0 0.2 0.2 0.2 0.33ns 0.1 0.2 0.33ns 1.22nsCLASS GASTROPODA 0.3 0.5 0.5 0.0 0.65ns 0.3 0.4 0.04ns 1.39nsCLASS ARACHNIDAOrder Araneae 47.2 52.2 55.0 38.3 2.58ns 46.9 49.4 0.17ns 0.56nsOrder Opiliones 0.2 0.2 0.3 0.0 1.17ns 0.1 0.3 1.78ns 0.58nsCLASS CRUSTACEAOrder Isopoda 1.8 0.3 0.5 0.7 0.66ns 0.7 1.0 0.18ns 0.26nsCLASS CHILOPODA 0.2 0.3 0.0 0.3 1.35ns 0.2 0.3 0.37ns 2.33nsCLASS DIPLOPODA 4.3 3.7 5.3 4.8 0.42ns 6.1 3.0 8.85* 2.47nsCLASS INSECTAOrder Collembola 158.3 150.0 201.7 145.0 0.89ns 179.2 148.3 1.59ns 0.05nsOrder Dermaptera 0.3 0.2 0.0 0.2 0.39ns 0.3 0.1 0.57ns 1.17nsOrder Hemiptera 1.7 1.3 1.5 2.5 0.43ns 1.3 2.2 1.06ns 1.02nsOrder Lepidoptera 0.3 0.0 0.0 0.2 1.22ns 0.1 0.2 0.33ns 0.33nsOrder Diptera (Adults) 26.0 25.5 24.8 24.8 0.23ns 29.6 21.0 7.13* 0.86nsOrder Diptera (Larvae) 0.2 0.0 0.5 0.2 0.95ns 0.3 0.1 1.38ns 0.19nsOrder Hymenoptera (Ants) 4.5 0.7 0.2 2.5 1.03ns 2.0 1.9 0.10ns 0.41nsOrder Hymenoptera (Non-Ants) 5.3 3.0 2.5 3.5 1.14ns 3.8 3.4 0.56ns 3.88*Order ColeopteraColeoptera Larvae 1.2 1.3 1.3 1.8 0.18ns 1.4 1.4 0.05ns 0.70nsFamily Anthicidae 0.8 2.3 2.0 1.0 0.75ns 1.6 1.5 0.13ns 1.53nsFamily Apionidae 0.0 0.2 0.0 0.0 1.00ns 0.0 0.1 1.00ns 1.00nsFamily Cantharidae 0.2 1.0 0.8 0.7 2.04ns 0.6 0.8 0.12ns 2.95nsFamily Carabidae 9.7 9.2 11.2 9.2 0.26ns 10.8 8.8 0.39ns 0.44nsFamily Chrysomelidae (Flea Beetles) 659.7 738.0 889.2 735.0 0.73ns 751.5 759.4 0.06ns 0.09nsFamily Chrysomelidae (Non Flea Beetles) 0.5 0.2 0.0 0.5 1.25ns 0.1 0.5 3.35ns 2.57nsFamily Coccinellidae 3.0 2.3 3.5 3.7 0.38ns 3.1 3.2 0.00ns 1.34nsFamily Cryptophagidae 31.8 42.7 22.8 33.7 1.76ns 28.1 37.4 0.21ns 0.30nsFamily Curculionidae 2.0 3.3 0.8 2.8 2.76ns 2.1 2.4 0.13ns 2.27nsFamily Elateridae 0.2 0.2 0.2 0.3 0.21ns 0.2 0.3 0.21ns 0.78nsFamily Lathridiidae 0.3 0.8 0.7 0.5 0.25ns 0.7 0.5 0.37ns 1.60nsFamily Leiodidae 0.3 0.3 0.0 0.0 1.34ns 0.0 0.3 3.91ns 1.34nsFamily Melyridae 0.0 0.0 0.0 0.0 - 0.0 0.0 - -Family Nitidulidae (Pollen Beetles) 2.5 4.8 3.5 2.8 2.00ns 3.4 3.4 0.00ns 0.97nsFamily Scarabaeidae 0.0 0.3 0.0 0.3 1.49ns 0.2 0.2 0.00ns 0.00nsFamily Scraptiidae 0.0 0.0 0.0 0.0 - 0.0 0.0 - -Family Silphidae 0.0 0.0 0.0 0.0 - 0.0 0.0 - -Family Staphylinidae 45.7 54.7 55.8 48.8 1.19ns 53.5 49.0 0.92ns 0.80ns

Total invertebrates 1008.5 1099.7 1284.8 1064.3 0.89ns 1127.8 1100.8 0.28ns 0.16nsNumber of families 16.3 17.8 15.8 17.2 1.79ns 16.4 17.2 1.28ns 1.32ns


b) Little Wittenham

Order / family 1. Spring treatment 2. Summer treatment3.

Interactions

Nil TMXChinoo


CLASS OLIGOCHAETA 0.0 0.0 0.3 0.2 1.60ns 0.2 0.1 0.44ns 2.77nsCLASS GASTROPODA 0.2 0.0 0.0 0.0 1.00ns 0.0 0.1 1.00ns 1.00nsCLASS ARACHNIDAOrder Araneae 124.7a 99.8ab 108.3a 60.5b 7.33** 98.5 98.2 0.62ns 0.91nsOrder Opiliones 6.8 4.0 6.2 5.7 1.44ns 5.8 5.6 0.21ns 1.79nsCLASS CRUSTACEAOrder Isopoda 0.3ab 0.5ab 0.2b 1.3a 3.82* 0.5 0.7 0.88ns 0.95nsCLASS CHILOPODA 1.7 2.8 1.5 2.5 0.44ns 2.3 2.0 0.04ns 1.96nsCLASS DIPLOPODA 12.2 22.7 15.3 41.0 0.45ns 31.7 13.9 1.42ns 3.46*CLASS INSECTAOrder Collembola 130.0 125.0 110.0 130.0 0.58ns 125.0 122.5 0.07ns 3.20nsOrder Dermaptera 0.0 0.0 0.0 0.7 4.91* 0.3 0.1 0.96ns 0.96nsOrder Hemiptera 1.3 0.7 1.0 2.0 1.53ns 1.5 1.0 1.68ns 0.43nsOrder Lepidoptera 0.5 0.7 0.0 0.5 2.14ns 0.2 0.7 6.12* 4.08*Order Diptera (Adults) 23.8 32.7 31.0 40.5 1.77ns 25.3 38.8 2.35ns 2.45nsOrder Diptera (Larvae) 0.0 0.0 0.0 0.0 - 0.0 0.0 - -Order Hymenoptera (Ants) 0.8 2.2 1.3 0.5 0.34ns 1.3 1.1 0.33ns 0.32nsOrder Hymenoptera (Non-Ants) 6.3 3.5 5.7 10.3 0.47ns 3.4 9.5 3.79ns 0.90nsOrder ColeopteraColeoptera Larvae 4.0 2.0 2.7 2.0 0.99ns 2.3 3.1 0.24ns 0.70nsFamily Anthicidae 11.0a 1.5b 7.5a 7.5a 9.95*** 7.2 6.6 2.39ns 3.91*Family Apionidae 0.0 0.0 0.0 0.0 - 0.0 0.0 - -Family Cantharidae 0.2b 1.5ab 2.3a 1.5ab 4.19* 1.5 1.3 0.32ns 1.25nsFamily Carabidae 37.0 31.0 40.0 73.0 0.37ns 54.1 36.4 0.73ns 1.32nsFamily Chrysomelidae (Flea Beetles) 111.2 108.8 106.7 89.8 1.18ns 124.8 83.4 6.74* 8.30**Family Chrysomelidae (Non Flea Beetles) 0.2 0.2 0.2 0.0 0.29ns 0.1 0.2 0.29ns 1.07nsFamily Coccinellidae 4.8 4.5 7.7 5.5 0.70ns 6.3 5.0 0.99ns 1.16nsFamily Cryptophagidae 0.7 2.0 1.3 0.8 0.80ns 1.6 0.8 0.88ns 0.50nsFamily Curculionidae 1.5 2.8 2.0 3.0 0.76ns 3.3 1.4 7.28* 2.51nsFamily Elateridae 1.8 0.8 0.0 0.3 1.05ns 0.6 0.9 0.03ns 1.27nsFamily Lathridiidae 0.2 0.3 0.0 0.3 0.78ns 0.2 0.3 0.21ns 0.21nsFamily Leiodidae 0.2 0.0 0.0 0.0 1.00ns 0.1 0.0 1.00ns 1.00nsFamily Melyridae 0.0 0.0 0.0 0.2 1.00ns 0.1 0.0 1.00ns 1.00nsFamily Nitidulidae (Pollen Beetles) 0.0 0.0 0.2 0.2 0.62ns 0.1 0.1 0.00ns 1.24nsFamily Scarabaeidae 0.0 0.0 0.0 0.0 - 0.0 0.0 - -Family Scraptiidae 0.0 0.2 0.0 0.0 1.00ns 0.0 0.1 1.00ns 1.00nsFamily Silphidae 0.0 0.0 0.2 0.0 1.00ns 0.0 0.1 1.00ns 1.00nsFamily Staphylinidae 53.2 57.0 52.5 45.5 0.22ns 56.8 47.3 1.59ns 2.86ns

Total invertebrates 534.5 507.2 504.0 525.3 0.20ns 554.6 480.9 4.02ns 7.97**Number of families 17.2 16.5 16.2 18.0 0.81ns 17.0 16.9 0.01ns 2.08ns


Table 23. Effects of spring and summer pest control measures on the abundance of soil surface active invertebrates as estimated by pitfall trapping in an annual Wild Bird Seed mixture in 2005 (Expt. 3). Means with the same letter in the same row are not significantly different (P> 0.05).

a) Marlow


Interactions

Nil TMXChinoo


CLASS OLIGOCHAETA 0.3 0.0 0.3 0.7 1.10ns 0.3 0.4 0.32ns 0.32nsCLASS GASTROPODA 0.5 0.0 0.2 0.3 1.23ns 0.0 0.5 7.40* 1.23nsCLASS ARACHNIDAOrder Araneae 44.0a 32.3b 28.3ab 28.2b 6.79** 32.3 34.2 0.79ns 6.38**Order Opiliones 0.5 0.0 0.2 0.0 1.81ns 0.1 0.3 0.76ns 2.85nsOrder Pseudoscorpionida 0.0 0.0 0.2 0.2 0.78ns 0.1 0.1 0.00ns 1.56nsOrder Acarina 200.2 139.8 188.0 280.5 0.94ns 227.0 177.3 0.05ns 0.63nsCLASS CRUSTACEAOrder Isopoda 4.0 0.7 2.7 3.0 1.55ns 1.8 3.4 1.75ns 0.37nsCLASS CHILOPODA 0.5 0.7 0.7 1.2 0.37ns 1.0 0.5 1.24ns 1.16nsCLASS DIPLOPODA 15.5 11.8 15.8 5.0 2.33ns 13.3 10.8 0.08ns 0.19nsCLASS INSECTAOrder Collembola 610.8 384.0 402.2 682.2 1.44ns 439.0 600.6 1.90ns 0.85nsOrder Protura 0.8 0.3 0.2 0.0 0.46ns 0.5 0.2 0.39ns 1.19nsOrder Dermaptera 0.3 0.0 0.0 0.0 1.00ns 0.2 0.0 1.00ns 1.00nsOrder Thysanoptera 1.2 1.5 0.2 1.2 1.65ns 0.7 1.3 1.92ns 1.73nsOrder Orthoptera 0.2 0.0 0.0 0.2 0.78ns 0.1 0.1 0.00ns 1.56nsOrder Siphonaptera 0.3 0.0 0.0 0.0 1.00ns 0.2 0.0 1.00ns 1.00nsOrder Hemiptera 2.5a 1.7ab 0.2b 0.5ab 3.75* 1.8 0.7 1.78ns 2.77nsOrder Lepidoptera 1.2 0.5 0.5 0.2 0.55ns 1.0 0.2 3.67ns 0.89nsOrder Diptera (Adults) 1954.8 1823.3 1981.5 2103.0 0.42ns 2038.1 1893.3 1.12ns 1.00nsOrder Diptera (Larvae) 0.7 1.2 0.7 1.2 0.05ns 1.0 0.8 0.48ns 1.61nsOrder Hymenoptera (Ants) 1.7 25.7 1.8 1.5 0.65ns 2.4 12.9 0.52ns 0.13nsOrder Hymenoptera (Non-Ants) 29.2 16.3 20.5 24.7 1.09ns 20.8 24.6 1.04ns 2.05nsColeoptera Larvae 6.0 2.2 5.2 4.3 1.13ns 5.4 3.4 1.95ns 1.40nsFamily Alexiidae 1.2 0.0 0.2 0.3 1.42ns 0.6 0.3 0.37ns 1.12nsFamily Anobiidae 0.0 0.0 0.0 0.0 - 0.0 0.0 - -Family Anthicidae 0.5 0.7 0.7 0.8 0.71ns 0.8 0.6 0.49ns 0.71nsFamily Apionidae 0.3 0.3 0.5 1.3 0.73ns 0.3 0.9 1.69ns 0.58nsFamily Cantharidae 0.3 0.5 0.5 1.0 0.21ns 0.8 0.4 0.29ns 1.13nsFamily Carabidae 24.0 10.5 12.2 11.8 2.42ns 16.3 12.9 1.26ns 0.65nsFamily Choleridae 0.0 0.0 0.3 0.0 1.00ns 0.2 0.0 1.00ns 1.00nsFamily Chrysomelidae (Flea Beetles) 459.5 602.7 725.0 490.8 2.27ns 628.3 510.8 1.90ns 2.97nsFamily Chrysomelidae (Non Flea Beetles) 0.3 0.0 0.3 0.7 0.66ns 0.3 0.4 0.52ns 0.71nsFamily Coccinellidae 0.8 0.7 1.2 0.2 0.57ns 0.9 0.5 0.44ns 0.09nsFamily Cryptophagidae 36.2 13.7 24.3 27.7 0.51ns 26.8 24.2 0.29ns 0.73nsFamily Curculionidae 3.2 1.5 3.5 4.0 1.03ns 3.7 2.4 1.16ns 0.45nsFamily Elateridae 0.5 0.2 0.2 0.5 0.45ns 0.2 0.5 1.36ns 0.03nsFamily Lathridiidae 0.2 1.5 0.0 0.0 0.79ns 0.0 0.8 1.49ns 0.79nsFamily Leiodidae 0.0 0.0 0.0 0.0 - 0.0 0.0 - -Family Melyridae 0.0 0.0 0.0 0.0 - 0.0 0.0 - -Family Nitidulidae (pollen beetles) 0.3 0.0 0.5 0.0 3.15ns 0.1 0.3 3.15ns 1.28nsFamily Oedemeridae 0.0 0.0 0.2 0.0 1.00ns 0.0 0.1 1.00ns 1.00nsFamily Phalacridae 0.0 0.0 0.0 0.0 - 0.0 0.0 - -Family Ptiliidae 0.0 0.0 0.0 0.0 - 0.0 0.0 - -Family Scarabaeidae 0.2 0.0 0.0 0.2 0.62ns 0.1 0.1 0.00ns 1.24ns

Family Scolytidae 0.0 0.0 0.2 0.0 1.00ns 0.1 0.0 1.00ns 1.00nsFamily Scraptiidae 0.0 0.0 0.0 0.0 - 0.0 0.0 - -Family Silphidae 0.0 0.0 0.0 0.2 1.00ns 0.1 0.0 1.00ns 1.00nsFamily Staphylinidae 123.8 100.8 83.5 101.0 0.20ns 115.1 89.5 1.08ns 0.66ns



b) Little Wittenham


Interactions

Nil TMXChinoo


CLASS OLIGOCHAETA 1.5 1.2 2.2 2.5 0.47ns 1.8 1.9 0.20ns 0.54nsCLASS GASTROPODA 1.8 0.7 0.8 0.7 1.44ns 0.9 1.1 0.00ns 1.45nsCLASS ARACHNIDAOrder Araneae 47.0 43.0 51.7 44.6 1.13ns 45.7 47.4 0.37ns 1.04nsOrder Opiliones 8.0 6.8 3.5 2.2 1.38ns 6.0 4.3 2.07ns 1.70nsOrder Pseudoscorpionida 0.0 0.0 0.0 0.2 1.00ns 0.1 0.0 1.00ns 1.00nsOrder Acarina 47.7 36.5 26.8 23.1 0.94ns 35.5 31.6 0.23ns 0.61nsCLASS CRUSTACEAOrder Isopoda 1.2 3.5 1.2 0.5 2.87ns 1.4 1.8 0.70ns 0.46nsCLASS CHILOPODA 11.0 21.3 8.3 13.5 1.69ns 22.3 4.8 3.35ns 0.02nsCLASS DIPLOPODA 33.8 12.0 18.7 7.5 0.94ns 6.0 30.0 2.31ns 0.31nsCLASS INSECTAOrder Collembola 571.3 504.8 549.7 718.2 0.28ns 497.8 674.2 1.73ns 0.47nsOrder Protura 0.0 0.0 0.2 0.0 1.00ns 0.0 0.1 1.0ns 1.00nsOrder Dermaptera 0.7 1.0 1.8 1.5 1.22ns 1.5 1.0 2.13ns 0.11nsOrder Thysanoptera 53.5 2.7 0.7 2.5 1.30ns 27.9 1.8 0.76ns 0.20nsOrder Orthoptera 0.0 0.0 0.0 0.2 1.00ns 0.1 0.0 1.00ns 1.00nsOrder Siphonaptera 0.0 0.0 0.0 0.0 - 0.0 0.0 - -Order Hemiptera 15.5ab 14.0b 23.2a 11.1b 6.72** 14.7 17.2 0.63ns 1.50nsOrder Lepidoptera 0.7 0.0 0.2 0.5 1.58ns 0.2 0.5 1.55ns 1.58nsOrder Diptera (Adults) 93.2 94.8 105.8 102.0 0.50ns 89.7 108.2 3.31ns 1.60nsOrder Diptera (Larvae) 1.0 2.7 0.8 11.8 1.54ns 2.1 6.1 0.38ns 0.30nsOrder Hymenoptera (Ants) 6.7 3.3 3.3 3.2 0.72ns 3.1 5.2 0.17ns 0.67nsOrder Hymenoptera (Non-Ants) 33.3 26.5 30.8 26.1 0.34ns 25.3 33.1 1.87ns 2.76nsColeoptera Larvae 4.0a 0.7b 2.7a 2.9a 5.47* 2.5 2.7 0.08ns 1.70nsFamily Alexiidae 0.0 0.0 0.0 0.0 - 0.0 0.0 - -Family Anobiidae 0.0 0.0 0.2 0.0 1.00ns 0.1 0.0 1.00ns 1.00nsFamily Anthicidae 10.7a 1.2b 8.0a 15.9a 8.30** 7.2 10.7 0.83ns 2.50nsFamily Apionidae 2.8 1.2 1.5 0.6 0.25ns 1.3 1.8 0.35ns 1.97nsFamily Cantharidae 5.5 1.0 4.0 2.4 2.34ns 2.8 3.7 0.68ns 0.73nsFamily Carabidae 75.2 50.0 56.0 58.8 0.23ns 60.6 59.3 0.08ns 0.99nsFamily Choleridae 0.5 0.7 0.5 1.2 0.43ns 0.9 0.5 0.31ns 2.02nsFamily Chrysomelidae (Flea Beetles) 568.2 698.8 851.2 657.1 1.31ns 547.7 839.9 6.25* 0.36nsFamily Chrysomelidae (Non Flea Beetles) 0.2 0.0 0.7 0.6 1.11ns 0.5 0.3 0.50ns 0.08nsFamily Coccinellidae 1.3 1.3 0.5 0.2 1.18ns 0.5 1.2 1.14ns 0.84nsFamily Cryptophagidae 378.7 481.5 464.2 393.0 0.43ns 382.3 476.4 1.10ns 0.10nsFamily Curculionidae 1.3 2.2 1.8 2.1 0.34ns 1.1 2.6 5.11* 1.18nsFamily Elateridae 0.0 0.0 1.7 0.0 1.00ns 0.8 0.0 1.00ns 1.00nsFamily Lathridiidae 0.3 0.3 0.2 0.0 0.56ns 0.3 0.2 0.06ns 0.86nsFamily Leiodidae 0.0 0.3 0.0 0.0 1.00ns 0.2 0.0 1.00ns 1.00nsFamily Melyridae 0.0 0.0 0.0 0.0 - 0.0 0.0 - -Family Nitidulidae (pollen beetles) 0.0 0.3 0.5 0.0 2.19ns 0.2 0.3 0.10ns 0.10nsFamily Oedemeridae 0.0 0.0 0.0 0.0 - 0.0 0.0 - -Family Phalacridae 0.3 0.0 0.2 0.2 0.52ns 0.0 0.3 3.87ns 0.52nsFamily Ptiliidae 0.2 0.0 0.0 0.3 1.22ns 0.1 0.2 0.33ns 0.33nsFamily Scarabaeidae 0.0 0.0 0.0 0.0 - 0.0 0.0 - -Family Scolytidae 0.0 0.0 0.0 0.0 - 0.0 0.0 - -Family Scraptiidae 0.0 0.0 0.0 0.0 - 0.0 0.0 - -Family Silphidae 0.0 0.0 0.0 0.0 - 0.0 0.0 - -Family Staphylinidae 136.3 124.8 164.5 103.4 0.69ns 120.5 144.1 0.25ns 0.22ns



Table 24. Effects of spring and summer pest control measures on the abundance of epigeal invertebrates as estimated from sweep net sampling in an annual Wild Bird Seed mixture in 2004 (Expt. 3). Means with the same letter in the same row are not significantly different (P> 0.05).

a) Marlow


Interactions

Nil TMXChinoo


CLASS ARACHNIDAOrder Araneae 0.50 0.00 0.50 0.33 2.24ns 0.50 0.17 4.48* 2.24nsOrder Opiliones 0.00 0.00 0.00 0.00 - 0.00 0.00 - -CLASS INSECTALarvae 7.67 3.67 9.67 4.50 1.24ns 11.58 1.17 14.60** 0.03nsOrder Hemiptera (Aphids) 14.17 9.50 4.67 1.83 1.52ns 3.00 12.08 3.19ns 1.02nsOrder Hemiptera (Non-aphids) 8.67 5.50 6.83 4.83 0.99ns 9.25 3.67 11.94** 5.83**Order Thysanoptera 4.67 2.33 1.83 2.67 2.11ns 3.67 2.08 1.47ns 0.25nsOrder Diptera (Adults) 4.50 1.17 3.67 4.00 2.94ns 4.08 2.58 0.15ns 1.41nsOrder Hymenoptera (Non-Ants) 14.00 20.17 14.33 19.67 0.25ns 16.17 17.92 0.48ns 0.44nsOrder Neuroptera 0.00 0.00 0.00 0.00 - 0.00 0.00 - -Order ColeopteraFamily Cantharidae 0.00 0.00 0.00 0.17 1.00ns 0.08 0.00 1.00ns 1.00nsFamily Chrysomelidae (Flea Beetles) 4.67 3.17 3.50 2.83 0.78ns 5.58 1.50 5.54* 0.49nsFamily Coccinellidae 0.33 0.17 0.00 0.33 1.03ns 0.33 0.08 2.52ns 1.03nsFamily Cryptophagidae 0.00 0.00 0.00 0.00 - 0.00 0.00 - -Family Curculionidae 2.00 1.83 1.83 1.33 0.22ns 2.17 1.33 4.97* 1.66nsFamily Elateridae 0.00 0.00 0.00 0.00 - 0.00 0.00 - -Family Lathridiidae 0.00 0.00 0.00 0.00 - 0.00 0.00 - -Family Melyridae 0.17 0.17 0.17 0.00 0.33ns 0.17 0.08 0.33ns 1.22nsFamily Nitidulidae (Pollen Beetles) 36.17 52.50 44.83 36.17 0.23ns 51.50 33.33 1.14ns 0.58nsFamily Oedermeridae 0.00 0.00 0.00 0.17 1.00ns 0.00 0.08 1.00ns 1.00nsFamily Staphylinidae 0.00 0.00 0.00 0.17 1.00ns 0.08 0.00 1.00ns 1.00ns

Total invertebrates 97.50 100.17 91.83 79.00 0.15ns 108.17 76.08 0.45ns 0.24nsNumber of families 9.00 7.67 8.17 8.83 1.07ns 9.25 7.58 7.87* 0.37ns


b) Little Wittenham


Interactions

Nil TMXChinoo


CLASS ARACHNIDAOrder Araneae 0.17 0.33 0.00 0.00 0.66ns 0.25 0.00 1.79ns 0.66nsOrder Opiliones 0.00 0.00 0.00 0.33 1.00ns 0.17 0.00 1.00ns 1.00nsCLASS INSECTALarvae 1.17 1.33 0.83 1.17 0.18ns 1.00 1.25 0.43ns 1.66nsOrder Hemiptera (Aphids) 11.00 10.83 10.67 10.00 0.19ns 9.33 11.92 1.57ns 0.71nsOrder Hemiptera (Non-aphids) 9.17 6.67 7.17 7.17 0.23ns 11.08 4.00 14.79** 0.60nsOrder Thysanoptera 2.00 1.17 6.00 5.33 3.71ns 4.83 2.42 2.65ns 0.39nsOrder Diptera (Adults) 12.67 7.33 13.50 9.33 1.78ns 11.25 10.17 0.07ns 0.63nsOrder Hymenoptera (Non-Ants) 20.67a 8.17b 10.83ab 7.50ab 3.59* 17.08 6.50 14.35** 3.49*Order Neuroptera 0.00 0.00 0.17 0.33 1.97ns 0.17 0.08 0.54ns 3.41*Order ColeopteraFamily Cantharidae 0.00 0.67 0.67 0.00 1.99ns 0.50 0.17 0.90ns 1.35nsFamily Chrysomelidae (Flea Beetles) 4.33 3.00 2.33 2.33 0.37ns 5.92 0.08 25.90*** 0.14nsFamily Coccinellidae 1.00 2.17 1.00 0.83 0.89ns 2.50 0.00 24.34*** 0.89nsFamily Cryptophagidae 0.00 0.33 0.00 0.00 1.00ns 0.00 0.17 1.00ns 1.00nsFamily Curculionidae 1.00 0.50 1.00 0.33 0.73ns 1.33 0.08 10.69ns 0.56nsFamily Elateridae 0.00 0.00 0.00 0.17 1.00ns 0.00 0.08 1.00ns 1.00nsFamily Lathridiidae 0.33 0.33 0.00 0.00 1.34ns 0.33 0.00 3.91ns 1.34nsFamily Melyridae 0.00 0.00 0.00 0.00 - 0.00 0.00 - -Family Nitidulidae (Pollen Beetles) 10.33a 1.83b 15.00b 19.17b 9.15*** 13.92 9.25 2.16ns 1.12nsFamily Oedermeridae 0.00 0.00 0.00 0.00 - 0.00 0.00 - -Family Staphylinidae 0.00 0.17 0.00 0.17 0.62ns 0.08 0.08 0.00ns 1.24ns

Total invertebrates 73.83 44.83 69.17 64.17 1.56ns 79.75 46.25 7.10* 0.65nsNumber of families 7.83 8.00 8.33 8.33 0.28ns 9.67 6.58 42.22*** 0.94ns


Table 25. Effects of spring and summer pest control measures on the abundance of epigeal invertebrates as estimated from sweep net sampling in an annual Wild Bird Seed mixture in 2005 (Expt. 3). Means with the same letter in the same row are not significantly different (P> 0.05).

a) Marlow


InteractionsNil TMX Chinook Mavrik F3,14 Nil Mavrik F1,14 F3,14

CLASS ARACHNIDAOrder Araneae 0.33 0.17 0.17 0.00 0.75ns 0.17 0.17 0.00ns 0.75nsOrder Opiliones 0.00 0.00 0.00 0.00 - 0.00 0.00 - -CLASS INSECTALarvae 7.50 4.50 3.33 5.67 1.83ns 7.25 3.25 14.24** 0.03nsOrder Hemiptera (Aphids) 40.17 26.83 13.00 35.17 1.33ns 33.08 24.50 0.67ns 1.11nsOrder Hemiptera (Non-aphids) 2.00 0.67 2.50 4.67 0.84ns 4.42 0.50 10.31** 1.04nsOrder Thysanoptera 10.33ab 5.50ab 1.67b 14.50a 4.41* 12.08 3.92 12.85** 3.52*Order Diptera (Adults) 11.00 3.17 2.83 5.67 0.69ns 8.50 2.83 2.93ns 0.74nsOrder Hymenoptera (Non-Ants) 12.67 6.67 5.67 9.33 0.49ns 12.17 5.00 9.33** 3.05nsOrder Neuroptera 0.00 0.00 0.00 0.00 - 0.00 0.00 - -Order Lepidoptera 0.00 0.00 0.17 0.00 1.00ns 0.08 0.00 1.00ns 1.00nsOrder ColeopteraFamily Cantharidae 0.17 0.00 0.00 0.00 1.00ns 0.08 0.00 1.00ns 1.00nsFamily Carabidae 0.00 0.00 0.17 0.00 1.00ns 0.08 0.00 1.00ns 1.00nsFamily Chrysomelidae (Flea Beetles) 9.33 3.17 2.00 4.33 1.34ns 7.08 2.33 13.40** 0.95nsFamily Coccinellidae 0.67 0.67 0.33 0.83 0.44ns 0.92 0.33 1.97ns 0.35nsFamily Cryptophagidae 0.00 0.00 0.00 0.00 - 0.00 0.00 - -Family Curculionidae 5.17 2.17 1.33 2.50 1.20ns 4.50 1.08 7.64* 0.01nsFamily Elateridae 0.00 0.00 0.00 0.00 - 0.00 0.00 - -Family Lathridiidae 0.17 0.00 0.17 0.00 0.62ns 0.08 0.08 0.00ns 1.24nsFamily Melyridae 0.00 0.00 0.17 0.00 1.00ns 0.08 0.00 1.00ns 1.00nsFamily Nitidulidae (Pollen Beetles) 114.50 59.00 45.50 90.00 0.20ns 111.33 43.17 14.21** 1.05nsFamily Oedermeridae 0.17 0.00 0.00 0.00 1.00ns 0.08 0.00 1.00ns 1.00nsFamily Phalacridae 0.17 0.00 0.00 0.17 0.62ns 0.17 0.00 1.87ns 0.62nsFamily Ptiliidae 0.00 0.00 0.00 0.00 - 0.00 0.00 - -Family Scraptiidae 0.17 0.00 0.00 0.00 1.00ns 0.08 0.00 1.00ns 1.00nsFamily Staphylinidae 0.00 0.33 0.00 0.33 3.11ns 0.00 0.33 9.33** 3.11ns

Total invertebrates 214.50a112.83ab 79.00b

173.17ab 3.62* 202.25 87.50 16.22*** 1.40ns

Number of families 9.17 8.50 8.17 8.50 0.40ns 10.00 7.17 18.18*** 1.24ns


b) Little Wittenham


Interactions

Nil TMXChinoo


CLASS ARACHNIDAOrder Araneae 0.67 1.33 0.50 0.67 0.17ns 0.92 0.67 0.01ns 1.63nsOrder Opiliones 0.00 0.00 0.00 0.00 - 0.00 0.00 - -CLASS INSECTALarvae 4.00 0.67 0.67 0.50 1.90ns 1.92 1.00 0.40ns 0.18nsOrder Hemiptera (Aphids) 6.50 3.50 9.33 8.83 0.58ns 5.92 8.17 0.14ns 0.57nsOrder Hemiptera (Non-aphids) 8.83 5.83 7.17 3.00 1.25ns 9.00 3.42 3.42ns 0.53nsOrder Thysanoptera 10.17 8.00 11.33 1.00 2.39ns 13.17 2.08 7.59* 1.15nsOrder Diptera (Adults) 8.67 10.50 12.00 8.67 0.33ns 12.58 7.33 4.69* 1.02nsOrder Hymenoptera (Non-Ants) 21.67 21.50 30.33 18.00 0.59ns 29.08 16.67 1.76ns 3.84*Order Neuroptera 0.00 0.17 0.17 0.50 0.74ns 0.25 0.17 0.06ns 0.49nsOrder Lepidoptera 0.00 0.00 0.33 0.00 1.00ns 0.17 0.00 1.00ns 1.00nsOrder ColeopteraFamily Cantharidae 0.00 0.33 0.17 0.17 1.10ns 0.17 0.17 0.00ns 3.29nsFamily Carabidae 0.00 0.00 0.00 0.00 - 0.00 0.00 - -Family Chrysomelidae (Flea Beetles) 39.50 14.67 6.17 3.50 3.25ns 26.50 5.42 3.82ns 0.92nsFamily Coccinellidae 0.67 0.50 0.67 0.00 0.84ns 0.83 0.08 4.35ns 0.78nsFamily Cryptophagidae 0.67 0.33 0.17 0.17 0.46ns 0.42 0.25 0.12ns 0.46ns

Family Curculionidae 10.67 9.00 6.00 2.17 0.74ns 9.58 4.33 4.32ns 1.04nsFamily Elateridae 0.00 0.00 0.00 0.00 - 0.00 0.00 - -Family Lathridiidae 0.50 0.67 1.00 0.33 0.14ns 0.58 0.67 0.00ns 2.19nsFamily Melyridae 0.00 0.00 0.00 0.00 - 0.00 0.00 - -Family Nitidulidae (Pollen Beetles) 21.50 12.00 3.00 3.83 1.28ns 15.75 4.42 3.80ns 2.17nsFamily Oedermeridae 0.00 0.00 0.00 0.00 - 0.00 0.00 - -Family Phalacridae 0.00 0.00 0.33 0.00 1.00ns 0.17 0.00 1.00ns 1.00nsFamily Ptiliidae 0.00 0.00 0.33 0.00 1.00ns 0.17 0.00 1.00ns 1.00nsFamily Scraptiidae 0.00 0.00 0.00 0.00 - 0.00 0.00 - -Family Staphylinidae 0.00 0.33 0.00 0.00 4.00* 0.17 0.00 4.00ns 4.00*

Total invertebrates 134.00 89.33 89.83 51.33 1.38ns 127.42 54.83 9.98** 2.05nsNumber of families 8.83 9.83 9.00 8.00 0.80ns 9.75 8.08 3.93ns 1.60ns


13. Technology transfer activities

Between April 2004 and March 2006 the results of this project were presented at more than 50 meetings attended by over 1500 people comprising farmers, project officers, agronomists, the agrochemical industry, the organic sector, and food processors & packers. In addition, two specialist training courses were held for Defra project officers and specialists, and four two-day intensive training courses were held for 84 farmers. In total the attendees were responsible for managing 697,000 ha of farmland.

14. Publications

PublicationsCarvell, C. (2004) Bumblebees – bringing back the farmland buzz. Fact sheet number 1. Centre for Ecology and Hydrology, Farmed Environment Company Ltd.Nowakowski, M. (2004) Agri-environment: a vision of the future. Crop Production Magazine, April 2004.Nowakowski, M. (2004) First taste of cross-compliance,. Crop Production Magazine, June 2004.Nowakowski, M., & Meek, W.R.M. (2004) The buzz of biodiversity down on the farm. Conservation Land Management, Autumn 2004, 4-7.Pywell, R.F. (2004) Review Report: BD1623: Environmentally sustainable techniques to establish and manage wildlife seed mixtures (WM1), and pollen and nectar seed mixtures (WM2). Defra’s R&D on Farmland Conservation & Biodiversity.Nowakowski, M. (2006) Increasing wildlife in the farmed countryside: some good news. In: Arable crop protection in the balance: Profit and the environment, pp15.1-15.11, HGCA, London

Science presentationsDecember 2003 - Presentation to farmers and food-processors as part of the Conservation Grade Initiative entitled: ‘The Importance of a Biodiverse Countryside.’ December 2003 - Seminar presentation to BBSRC Silsoe Research Institute entitled: ‘The Enhancement of Biodiversity on Farmland.’November 2003 – Presentation at LEAF Conference: Profitable Farm Practical Conservation, ‘Balancing viable agriculture with an enhanced environment’ March 2004 - Presentation to the University of Liverpool MSc in Restoration Ecology entitled: ‘The Enhancement of Biodiversity on Farmland.’November 2004 – Presentation of results at Defra’s R&D on Farmland Conservation & Biodiversity, 11-12 November 2004, Warwick.March 2005 - Presentation to the University of Liverpool MSc in Restoration Ecology entitled: ‘The Enhancement of Biodiversity on Farmland.’April 2005 - RDS Ecology Specialist Conference, AxbridgeApril 2005 - Butterfly Conservation Conference, Southampton September 2005 - British Ecological Society Annual meeting, University of Hertfordshire, January 2006 - EASY Project Symposium, Wageningen, NetherlandsFebruary 2006 - Crops Magazine / Farmed Environment Conference, LintonMarch 2006 - RDS Science and Stewardship Conference, OxfordMarch 2006 - University of Liverpool MSc in Restoration Ecology entitled: ‘The Enhancement of Biodiversity on Farmland’

15. Future research

Further research is required into the following aspects of managing wild bird seed mixtures (EF2), pollen and nectar (EF4) under the agri-environment schemes:1) Pest-resistant seed crop species: the current project provides a useful framework within which to investigate the pest-resistance of a range of seed crop species, together with their seed yield and value to birds.2) Hybrid EF2 and EF4 seed mixtures: future research should seek to develop dual purpose mixtures of seed crops which also provide good habitat for a wide range of invertebrates, including bumblebees and butterflies. These habitats would also provide useful forage of insectivorous birds in the summer. 3) Techniques to prolong the longevity of EF4: results from other research (BD1624) suggest the abundance of key forage species, Trifolium pratense and T. hybridum, declines markedly after 3 years. Future research needs to investigate different management techniques and seed mixtures to increase longevity of this habitat.

16. References

Al-Doghairi, M.A. 1999. Pest management tactics for western cabbage flea beetle. (Phyllotreta pusilla Horn) on brassica crops. Unpublished PhD thesis, Colorado State University.

Alford, D.V. 1975. Bumblebees. Davis-Poynter, London.Andersen, C. L., Hazzard, R., Van driesche, R. and Mangan, F. X. 2006. Alternative Management Tactics for

Control of Phyllotreta cruciferae and Phyllotreta striolata (Coleoptera: Chrysomelidae) on Brassica rapa in Massachusetts. Journal of Economic Entomology, 99, 803-810.

Anon. 1999. StatXact 4 for Windows user manual. Cytel Software Corporation, Cambridge.Anon. 2005. Entry Level Stewardship Handbook. London: Defra.

Anon. 2006. UK Biodiversity Action Plan Targets Review - Habitat Targets 1. http://www.ukbap.org.uk/library/BRIG/TargetsReview06/HabitatTargets_1.xls Peterborough: Biodiversity Reporting and Information Group.

Banaszak, J. 1980. Studies on methods of censusing the numbers of bees (Hymenoptera: Apoidea). Polish Ecological Studies 6, 355–366.

Barlet, E., Willimas, I.H. 1991. Factors restricting the feeding of the cabbage stem flea beetle (Pyslliodes chrysocephala). Journal of Economic Entomology, 85, 594-600.

Biesmeijer, J.C., Roberts, S.P.M., Reemer, M., Ohlemüller, R.,Edwards, M., Peeters, T., Schaffers, A.P., Potts, S.G., Kleukers, R., Thomas, C.D., Settele, J., Kunin, W.E. 2006. Parallel Declines in Pollinators and Insect-Pollinated Plants in Britain and the Netherlands. Science, 313, pp. 351 - 354

Bignal, E.M. 1998. Using an ecological understanding of farmland to reconcile nature conservation requirements, EU agriculture policy and world trade agreements. Journal of Applied Ecology, 35, 949–954.

Bradbury, R. B., Browne, S. J., Stevens, D. K., Aebischer, N. J. 2004. Five-year evaluation of the impact of the Arable Stewardship Pilot Scheme on birds. Ibis 146 Suppl., 2, 171-180.

Bullock, J.M. 1996. Plant Competition And Community Dynamics. The ecology and management of grazing systems (eds J. Hodgson & A. W. Illius), pp. 69-100. CAB International, Wallingford.

Carvell, C., Meek, W.R., Pywell, R.F., Goulson, D., Nowakowski, M. 2007. Comparing the efficacy of agri-environment schemes to enhance bumblebee abundance and diversity on arable field margins. Journal of Applied Ecology, 44, 29-40.

Carvell, C., Roy, D.B., Smart, S.M., Pywell, R.F., Preston, C., Goulson, D. 2006. Declines in forage availability for bumblebees at a national scale. Biological Conservation, 132, 481-489.

Cook, S. Murray, D.A. & Williams, I.H. 1999. Pollen beetle, Meligethes aeneus fabricius, incidence in the composite hybrid winter oilseed rape, synergy. In: N. Wratten and P.A. Salisbury (eds) New Horizons for an Old Crop. Proceedings of the 10th international Rapeseed Congress, Canberra, Australia, 1999.

Defra. 2002. The Countryside Stewardship Scheme Handbook. Department for Environment, Food and Rural Affairs, London, UK.

Defra. 2005a. Environmental Stewardship: Entry Level Stewardship Handbook. Department for Environment, Food and Rural Affairs, London, UK.

Defra. 2005b. Environmental Stewardship: Higher Level Stewardship Handbook. Department for Environment, Food and Rural Affairs, London, UK.

Dicks, L.V., Corbet, S.A., Pywell, R.F. 2002. Compartmentalization in plant-insect flower visitor webs. Journal of Animal Ecology 71, 32-43.

Dietrick, E.J. 1961. An improved backpack motor fan for suction sampling insect populations. Journal of Economic Entomology, 54, 394–395.

Ester, A., de Putter, H., van Bilsen, J.G.P.M. 2003. Filmcoating the seed of cabbage (Brassica oleracea L. convar. Capitata L.) and cauliflower (Brassica oleracea L. var. Botrytis L.) with imidacloprid and spinosad to control insect pests. Crop Protection, 22, 761-768.

Feber, R.E., Smith, H., Macdonald, D.W. 1996. The effects on butterfly abundance of the management of uncropped edges of arable field. Journal of Applied Ecology 33, 1191-1205.

Fleishman, E. Murphy, D.D., Brussard, P.F. 2000. A new method for selection of umbrella species for conservation planning. Ecological Applications, 10, 569–579.

Fox, R. Warren, M.S., Harding, P.T., McLean, I.F.G., Asher, J., Roy, D., Brereton, T. 2001. The State of Britain’s Butterflies. Butterfly Conservation, CEH and JNCC, Wareham.

Free, J.B. 1993. Insect Pollination of Crops, 2nd edn. Academic Press, London.Gibson C. W. D., Brown, V.K. 1991. The nature and rate of development of calcareous grassland in Southern

Britain. Biological Conservation, 58, 297-316. Goulson, D. 2003. Bumblebees Their Behaviour and Ecology. Oxford University Press, Oxford, 253 pp.Hawkins, R.P. 1958. A survey of late flowering and singlecut red clover seed crops. Journal of National Institute of

Agricultural Botany, 8, 450-461. Henderson, I.G., Vickery, J.A., Carter, N. 2004. The use of winter bird crops by farmland birds in lowland

England. Biological Conservation, 118, 21-32. Horton, D.R., Capinera, J.L. 1987. Effects of plant diversity, host density, and host size on population ecology of

the Colorado potato beetle (Coleoptera: Chrysomelidae). Environmental Entomology, 16, 1019-1026.Kevan, P.G. 1975. Forest application of the insecticide Fenitrothion and its effect on wild bee pollinators

(Hymenoptera: Apoidea) of Lowbush Blueberries (Vaccinium spp.) in southern New Brunswick, Canada. Biological Conservation, 7, 301-309.

Kevan, P.G. 1991. Pollination: keystone process in sustainable global productivity. Acta Horticulturae. 288, 103-110.

Lamb, R.J. 1984. Effects of flea beetles, Phyllotreta spp. (Chrysomelidae: Coleoptera), on the survival, growth, seed yield and quality of canola, rape and yellow mustard. The Canadian Entomologist, 116, 269-280.

Lamb, R.J. 1989. Entomology of oilseed Brassica crops. Annual Review of Entomology, 34, 211-229.Luff, M.L. 1996. The use of carabids as environmental indicators in grasslands and cereals. Annals Zoologici

Fennici, 33, 185-195.

http://www.ukbap.org.uk/library/BRIG/TargetsReview06/HabitatTargets_1.xls

Meek, W. M., Loxton, R., Sparks, T. H., Pywell, R. F., Pickett, H. & Nowakowski, M. 2002. The effect of arable field margin composition on invertebrate biodiversity. Biological Conservation, 106, 259-271.

Meek, W.R., Carvell, C, Pywell, R., Nowakowski, M., Sparks, T., Loxton, D., Skidmore, P, Bell, D., Croxton, P., Warman, E., Löbel, S., Walker, K. J. 2003. The BUZZ Project Year 1 Technical Report 2002. CEH Project number CO1157, 50pp.

Mela, T. 2003. Red clover grown in a mixture with grasses: yield, persistence and dynamics of quality characteristics. Agriculture and Food Science in Finland, 12, 195-212.

Morris, M.G. 2000. The effects of structure and its dynamics on the ecology and conservation of arthropods in British grasslands. Biological Conservation, 95 , 129–142.

Palaniswamy, P., Wise, I. 1994. Effects of neem-based products on the number and feeding activity of a crucifer flea beetle, Phyllotreta cruciferae (Goeze) (Coleoptera: Chrysomelidae), on canola. Journal of Agricultural Entomology, 11, 49-60.

Payne, R., Murray, D., Harding, S., Baird, D., Soutar, D. 2002. GenStat® for Windows(tm) (6th Edition) VSN International.

Peach, W. J., Siriwardena, G. M., Gregory, R. 1999. Long-term changes in over-winter survival rates explain the decline in Reed Buntings Emberiza Schoenichus in Britain. Journal of Applied Ecology, 36,798-811.

Pollard, E., Yates, T.J. 1993. Monitoring Butterflies for Ecology and Conservation. Chapman and Hall, London.Prigge, E.C., Bryan, W.B., Goldman-Innis, E.S. 1999. Early- and late-season grazing of Orchardgrass and

Fescue hayfields overseeded with Red Clover. Agronomy Journal, 91, 690-696. Prŷs-Jones, O.E., Corbet, S.A. 1991. Bumblebees. Naturalists' Handbooks 6. The Richmond Publishing Co. Ltd.,

Slough.Putnam, G. 1977. Response of four Brassica seed crop species to attack by the crucifer flea beetle, Phyllotreta

cruciferae. Canadian Journal of Plant Science, 57, 987-989.Pywell R.F., Bullock, J.M., Tallowin, J.B., Walker, K.J., Warman, E.A., Masters, G. 2007. Enhancing diversity of

species-poor grasslands: an experimental assessment of multiple constraints. Journal of Applied Ecology, 44, 81-94.

Pywell, R. F., Warman, E. A., Sparks, T. H., Greatorex-Davis, J. N., Walker, K. J., Meek, W. R., Carvell, C., Petit, S., Firbank, L. G. 2004. Assessing habitat quality for butterflies on intensively managed farmland. Biological Conservation, 118, 313-325.

Pywell, R.F., Bullock, J.M., Hopkins, A., Walker, K.J., Sparks, T.H., Burke, M.J.W., Peel, S. 2002. Restoration of species-rich grassland on arable land: assessing the limiting processes using a multi-site experiment. Journal of Applied Ecology 39, 294-309.

Pywell, R.F., Bullock, J.M., Roy, D.B., Warman, E.A., Rothery, P. 2003. Plant traits as predictors of performance in ecological restoration schemes. Journal of Applied Ecology 40, 65-77.

Pywell, R.F., Shaw, L., Meek, W.M., Turk, A., Shore, R.F. 2007. Do wild bird seed mixtures benefit other taxa? Aspects of Applied Biology, 81, 69-76.

Pywell, R.F., Warman, E.A., Carvell, C., Sparks, T.H., Dicks, L.V., Bennett, D., Wright, A., Critchley, C.N.R., Sherwood, A. 2005. Providing foraging resources for bumblebees in intensively farmed landscapes. Biological Conservation 121, 479–494.

Pywell, R.F., Warman, E.A., Hulmes, L., Hulmes, S., Nuttall, P., Sparks, T.H., Critchley, C.N.R., Sherwood, A. 2006. Effectiveness of new agri-environment schemes in providing foraging resources for bumblebees in intensively farmed landscapes. Biological Conservation, 129, 192-206.

Ryan, B.F., Joiner, B.L., Ryan T. 2000. MINITAB Handbook, Fourth Edition, Brooks Cole, Florence KY.Sakanoue, S.2005. Demography of red clover seeds in mixed-sown meadows. Journal of Agricultural Science,

143, 193-197.Siriwardena, G. M., Baillie, S. R., Crick, H. Q. P., Wilson, J. D. 2000. The importance of variation in the breeding

performance of seed-eating birds in determining their population trends on farmland. Journal of Applied Ecology, 37, 128-148.

Smart, S.M., Firbank, L.G., Bunce, R.G.H., Watkins, J.W. 2000. Quantifying changes in abundance of food plants for butterfly larvae and farmland birds. Journal of Applied Ecology 37, 398-414.

Smith, R. S., Pullan S., Shiel, R. S. 1996. Seed Shed in the Making of Hay From Mesotrophic Grassland in a Field in Northern England: Effects of Hay Cut Date, Grazing and Fertilizer in a Split-Split-Plot Experiment. Journal of Applied Ecology, 33, 833-841.

Sokal, R. R. and Rohlf, F. J.. 1995. Biometry, 3rd ed. W.H. Freeman and Company.Stace, C. 1997. New flora of the British Isles. Cambridge University Press, Cambridge.Standen, V. 2000. The adequacy of collecting techniques for estimating species richness of grassland

invertebrates. Journal of Applied Ecology, 37, 884-893.Steffan-Dewenter, I., Potts, S.G., Packer, L. 2005. Pollinator diversity and crop pollination services are at risk.

Trends in Ecology and Evolution, 20, 651-652. Stevenson, M. 2007. The contribution of English agri-environment schemes to biodiversity action plan targets for

arable land. Aspects of Applied Biology, 81, 333-340.Stewart, K.E.J., Bourn, N.A.D., Thomas, J.A. 2001. An evaluation of three quick methods commonly used to

assess sward height in ecology. Journal of Applied Ecology, 38, 1148-1154.Stoate, C., Henderson, I. G., Parish, D. M. B. 2004. Development of an agri-environment scheme option: seed-

bearing crops for farmland birds. Ibis 146 Suppl., 2, 203-209.

Tallowin, J.R.B., Smith, R.E.N., Pywell, R.F., Goodyear, J., Martyn, T. 2002. Use of fertiliser nitrogen and potassium to reduce soil phosphorous availability. In: Frame, J. (Ed.), Conservation Pays? Reconciling Environmental Benefits with Profitable Grassland Systems. In: British Grassland Society Occasional Symposium, vol. 36. British Grassland Society, Reading, pp. 163–166.

Thomas, J.A., Telfer, M.G., Roy, D.B., Preston, C.D., Greenwood, J.J.D., Asher, J., Fox, R., Clarke, R.T., Lawton, J.H. 2004. Comparative losses of British butterflies, birds, and plants and the global extinction crisis. Science, 303, 1879-1881.

Vickery, J. A., Bradbury, R. B., Henderson, I. G., Eaton, M. A., Grice, P. V. 2004. The role of agri-environment schemes and farm management practices in reversing the decline of farmland birds in England. Biological Conservation, 119, 19-39

Warren, M.S. 1992. The conservation of British butterflies. In: Dennis, R.L.H. (Ed.) The Ecology of Butterflies in Britain. Oxford University Press, Oxford, pp. 246-274.

Wells, T.C.E., Cox, R. 1993. The long-term effects of cutting on the yield, floristic composition and soil nutrient status of chalk grassland. Final report to the Nature Conservancy Council. Centre for Ecology and Hydrology, Huntingdon, UK.

Williams, I. H., J. B. Free, 1978. The feeding and mating behaviour of pollen beetles (Meligethes aeneus Fab.) and seed weevils (Ceutorhynchus assimilis Payk.) on oilseed rape (Brassica napus L.). Journal of Agricultural Science, Cambridge, 91, 343–348.

Winfield, A. L. 1992. Management of Oilseed rape pests in Europe. Agricultural Zoology Reviews, 5, 51-95

Young, J.E.B., Griffin, M.J., Alford, D.V. & Ogilvy, S.E. (eds) 2001. Reducing agrochemical use on the arable farm: the TALISMAN and SCARAB projects. London, Defra.

References to published material9. This section should be used to record links (hypertext links where possible) or references to other

published material generated by, or relating to this project.

Documents

General enquiries on this form should be made to:randd.defra.gov.uk/Document.aspx?Document=BD1623... · Web viewAll analyses were undertaken using Minitab 14 statistical software