DEPARTMENT for ENVIRONMENT, FOOD and RURAL AFFAIRS
Research and Development
CSG 15
Final Project Report
(Not to be used for LINK projects)
Two hard copies of this form should be returned to:
Research Policy and International Division, Final Reports Unit
DEFRA, Area 301
Cromwell House, Dean Stanley Street, London, SW1P 3JH.
An electronic version should be e-mailed to [email protected]
Project title
Epidemiology of Rhynchosporium to improve barley risk assessment
DEFRA project code
AR0510
Contractor organisation
and location
Plant Pathogen Interactions Division
Rothamsted Research
Harpenden
Herts
AL5 2JQ
Total DEFRA project costs
Project start date
£ 145,247
01/09/2001
Project end date
28/02/2002
Executive summary (maximum 2 sides A4)
Aims
This 18 month pre-LINK work aims to develop an understanding of the role of barley plant architecture,
weather and fungicides affecting leaf blotch disease spread and epidemic development. Data from detailed
experiments on the effect of fungicide on disease and development of epidemics have been examined and
analysed. Preliminary experiments to examine the effect of plant architecture on disease have been done to
establish the importance of disease escape in this pathosystem. The work provides the basis to assess the risk of
severe epidemics, and thereby improve the targeting of fungicide applications, which will benefit the
environment. By contributing to a more rational use of fungicides this work will promote sustainable barley
production
Introduction and Policy Rationale
Leaf blotch (Rhynchosporium secalis) is the most damaging disease of winter barley in the UK and is a major
reason for fungicide use. The severity of epidemics varies substantially between seasons, regions and crops
(DEFRA winter barley disease survey). In 2000, the estimated value of yield and quality loss due to leaf blotch
was £6.7M in England and Scotland, despite 95% of crops receiving fungicide applications with an average of
1.6 fungicide sprays per crop. Unfortunately, host plant resistance has not proved effective. Better management
of fungicides will result in improved disease control, less risk of fungicide resistance developing, and thus a
more sustainable barley production while benefiting the environment. However, to achieve this there is a basic
need to understand the epidemiology of the disease to allow better decisions to be made and to identify
mechanisms to reduce dependence on fungicides. In the short term, the project aims to improve guidance on the
CSG 15 (Rev. 6/02)
1
Project
title
Epidemiology of Rhynchosporium to improve barley risk
assessment
DEFRA
project code
AR0510
targeting of fungicide use and will also investigate canopy traits that may contribute to disease escape and
provide possible new options to control disease in barley through plant breeding.
Summary of results
The relationship between optimum spray timing and leaf emergence for Rhynchosporium on barley
In wheat, the effective fungicide dose, to control Septoria tritici and Puccinia striiformis, was minimised when
sprays were applied around the time of emergence of the target leaf (Paveley et al., 2000). In contrast, the
observations in this study suggest that for moderate-high Rhynchosporium disease pressure in barley, the most
effective spray time was not at the emergence of the target leaf. Instead effective fungicide dose was
minimised sometime between the emergence of the two leaves below the target leaf. This equates to
approximately 10-20 days before emergence of the target leaf. The mechanism for this difference is not
obvious, but may be related to the importance of inoculum pressure in driving Rhynchosporium epidemics.
Epidemic development.
A strong correlation between disease progress and degree-days (physiological time) suggests that
Rhynchosporium epidemic progress was a function of accumulated temperature. However some of the
fluctuations in amounts of disease could be related to specific weather events such as periods of rain or
temperatures above 20°C. Increases in disease were associated with rainfall 10-20 days earlier. Periods of low
temperature or low rainfall decreased disease on the remaining live leaves. Even with periods of rain there was
little additional disease recorded while mean temperatures were below 0°C. Dry periods were often
accompanied by daytime temperatures above 25°C, unfavourably high for R. secalis, which may also have
played a role in impeding the progress of epidemics. Overall spring barley was less affected by leaf blotch than
winter barley, possibly as a result of more rapid growth.
Early sowing of winter barley may increase the risk of severe leaf blotch. Temperature differences initially
affected the amounts of disease at the rosette pre-extension stage on plants sown at different dates. Early-sown
winter barley emerged in late autumn when conditions were favourable for disease and when the production of
new leaves was relatively rapid, providing an opportunity for a reservoir of disease to be established at the base
of the main shoot and tillers. Late-sown winter barley emerged when temperatures were low and
consequentially less favourable for disease, and when the production of new leaves was slower. There was
therefore a smaller reservoir of disease at the base of the late-sown plants. Spring barley emerged when the
conditions were again favourable for the disease but the rosette stage was short by comparison with that of
winter barley and fewer leaves were involved. For both late-sown winter barley and spring barley there was less
inoculum at the base of the plants to initiate further infections during the period of stem extension.
Analysis of spatial data showed that direction was an important factor affecting the spread of disease from a
point source, with significantly more spread in the direction of prevailing winds. The data also showed how
sowing in mixtures of susceptible and resistant cultivars affects the spatial spread of disease. Spatial spread of
disease was more limited and severity was lower in mixtures of 2:1 and 4:1 resistant to susceptible compared to
susceptible alone.
Effects of crop architecture and rain splash on spread of Rhynchosporium
In a small-scale field experiment four winter barley cultivars (Avenue, Maris Otter, Sumo and Vertige) with
contrasting architectures were used to test the hypothesis that canopy architecture affects the likelihood of
disease spread. Crop heights were manipulated using plant growth regulator regimes. There were considerable
differences among the cultivars in disease levels and in yield. Cultivars Avenue and Sumo, with the largest
amount of green leaf area for leaves two and three in the last 3-4 weeks of assessment, produced the highest
yields and Maris Otter and Vertige the least. The two cultivars with highest yields also had the largest average
leaf sizes and the smallest leaf insertion angles. Maris Otter was the lowest yielding cultivar with an average of
11% of leaf 2 and 15% of leaf 3 covered by lesions, while Avenue was best yielding with 7% lesions for leaf 2
and 8% for leaf 3. These results indicate another important difference between the Septoria/wheat and
Rhynchosporium/barley pathosystems. In wheat the extent and duration of the green leaf areas of the flag leaf
CSG 15 (Rev. 6/02)
2
Project
title
Epidemiology of Rhynchosporium to improve barley risk
assessment
DEFRA
project code
AR0510
and second leaf from anthesis onwards are of key importance in determining yield but for barley the flag leaf,
which can be very small in some cultivars, appears to be less important. The effects of R. secalis on the green
leaf areas of the second and third leaves were more important for determining yield in barley.
Technology Transfer
This project aims to provide strategic understanding in preparation for future LINK project proposals, jointly
with industry. Technology transfer was encouraged through meetings with growers and representatives of the
agrochemical industry. The project team also interacted closely with other projects in the UK and overseas on
barley physiology/pathology and related cereal research. The knowledge and results gained in this project were
transferred into the scientific community, to industry, farmers and advisers and to the press through conference
proceedings, project reports, scientific literature and posters, industry press, seminars, workshops and industry
meetings.
Conclusions and Relevance to Policy
This project addresses a central policy objective of DEFRA ‘to contribute to sustainable agriculture through an
integrated approach to disease management’ and contributes to both approaches identified by DEFRA by
rationalising fungicide use and developing new options for disease control.
I. Improving guidance on the targeting of fungicide use
The project has (i) identified the critical periods during which disease management can, or cannot, be justified
to protect source or sink capacity, and (ii) evaluated the effect of different fungicide timings and dosages for
manipulating epidemics in relation to current agronomic practice. A key requirement for the development of
further barley risk assessment procedures is an understanding of the seasonal development of epidemics of
Rhynchosporium secalis, including spread of disease from sources of inoculum up through successive leaf
layers or at different distances from an inoculum source. The project results show important aspects of the
relationship of disease to distance, time and weather conditions and provides an understanding central to
ascertaining disease risk. The work provides a platform for future LINK projects enabling better-informed
decisions about fungicide treatment to be made as a result.
II. Providing support for research into new options to control diseases in cereals
The project has contributed to a methodology that could allow the quantification of disease escape in projects
relating crop architectural traits to disease escape for the barley/Rhynchosporium secalis pathosystem. This
work has contributed to developing a mechanistic understanding of leaf blotch development, which can be used
to inform management decisions and provide the potential for breeding programmes to reduce pesticide
dependence.
CSG 15 (Rev. 6/02)
3
Project
title
Epidemiology of Rhynchosporium to improve barley risk
assessment
DEFRA
project code
AR0510
Scientific report (maximum 20 sides A4)
1. The relationship between optimum spray timing and leaf emergence for Rhynchosporium on barley
Methods
Experimental design, treatments and assessments
Field experiments were conducted in harvest years 1992 and 1993 near Liskeard, Cornwall, and in 1994 near
Taunton, Somerset. Each experiment consisted of 102 plots (not less than 1.9 x 25m) of the Rhynchosporiumsusceptible winter barley variety Fighter at Liskeard, and variety Puffin at Taunton. The sowing dates were 6
October 1991, 12 October 1992 and 19 October 1993. The experimental design was a randomised block with
three replicates. Single fungicide treatments of Sportak Delta (400 g cyproconazole and 60 g prochloraz L-1;
Aventis) and Corbel (750 g fenpropimorph L-1; BASF) were applied as a tank mix at four dose rates (0.25, 0.5,
0.75 and 1.0 of the label rate), using MDM Oxford precision hand held sprayers at 473 KPa in 240 ml water
through flat fan 110o nozzles in 1992, through Lurmark F110-03 nozzles in 1993 and through Teejet XR11002
nozzles in 1994. This was repeated to give eight spray timings. The first spray timing was on 13 March + one
day in each of the three years, and the intervals between each spray were 10-11 days. Each replicate contained
two untreated plots (i.e., zero fungicide dose).
Assessments of R. secalis disease were made in all individual plots at the date of the first planned spray and
every 10 or 11 days thereafter, up to and including GS 75. On each occasion, every leaf layer with green leaf
area remaining was assessed on 10 randomly selected shoots per plot, using disease assessment keys (Anon.,
1972). At each disease assessment date prior to GS 39, shoots were dissected to allow disease scores to be
recorded under the positions that leaves would eventually occupy on the mature plant. Leaf 1 was the flag leaf,
leaf 2 the leaf below, and so on down the shoot.
Analysis
Four leaf layers (from the flag leaf downwards) were analysed from each experiment. The area under each
disease progress curve (AUDPC) was calculated from the original disease severity data in Genstat, using the
trapezoidal rule. AUDPC values, dose and spray time for each leaf layer in each experiment were used to
estimate the parameters θ, σ, k, and μ using the equation:

  t  2
AUDPC  UT   
 2 2

e
 1  e


 k .dose

i) using FITNONLINEAR in GENSTAT (Paveley et al., 2000). Fitted response surfaces were plotted, with
spray timing, dose and AUDPC in the x, y, and z dimensions, respectively. The parameter μ locates the
centre of the valley of the surface plots in which the AUDPC is minimised, i.e., the optimum spray time.
Contours of the response surfaces, at a range of percentage control levels, were plotted to illustrate the
change in effective dose with spray timing.
Results
The highest levels of disease were seen in 1992, when leaf 1 reached a maximum of 40% disease, whereas in
1993 and 1994, the maximum disease on leaf 1 was 25% and 2.5%, respectively. Foliar diseases other than
Rhynchosporium were at zero or low levels and were unlikely to have interfered with epidemic development of
the target disease.
Values for the area under the disease progress curve, calculated from the original disease severity values, were
examined for all doses and spray times. The surface plots (e.g., Fig 1, 1993) indicated that the largest
differences were between the untreated and the low (0.25) dose values, compared to relatively small differences
CSG 15 (Rev. 6/02)
4
Project
title
Epidemiology of Rhynchosporium to improve barley risk
assessment
DEFRA
project code
AR0510
between the different dose levels (0.25, 0.5, 0.75 and 1.0). When tested by ANOVA, the AUDPC values in
each of the three years from all treated plots were significantly different from those from untreated control plots
(p< 0.005). AUDPCs from doses 0.25 – 1.0 were not significantly different from each other (p< 0.005), but
were significantly different for the different spray times (p< 0.005).
The optimum timing of sprays in each year varied according to the disease severity. In 1992 and 1993, when
there was moderate-high disease (e.g., up to 35% disease on L2), the most effective time to spray for control on
a particular target leaf layer appeared to be at the time of emergence of the leaf below, or two leaves below
(Table 1). In 1994, when there was low disease (<6% on L2), the optimum spray time for a leaf layer was at
Fitted of
AUDPC
dose-response
by spray
timing
surfaces
the time
emergence
of that
leaf
layer.
for leaves four to one, Starcross 1993
Leaf 4
Leaf 3
Leaf 2
Leaf 1
Figure 1 Fitted AUDPC dose-response by spray timing surfaces for leaves four to one, Starcross, 1993.
Table 1. Optimum spray times, expressed as leaf emergence times for leaves below, for control of
Rhynchosporium on target leaves above.
Target leaf layer for control of Rhynchosporium
L4
L3
L2
L1
L5
L4
L4-L3
L2
Leaf layers to be sprayed 1992
1993
L6
L5
L5-L4
L2-L1
1994
L5-L4
L3
L2
L1
CSG 15 (1/00)
5
Project
title
Epidemiology of Rhynchosporium to improve barley risk
assessment
DEFRA
project code
AR0510
Discussion
In wheat, the effective fungicide dose, to control Septoria tritici and Puccinia striiformis, was minimised when
sprays were applied around the time of emergence of the target leaf (Paveley et al., 2000). In contrast, the
observations in this study suggest that for moderate-high Rhynchosporium disease pressure in barley, the most
effective spray time was not at the emergence of the target leaf. Instead effective fungicide dose was
minimised sometime between the emergence of the two leaves below the target leaf. This equates to
approximately 10-20 days before emergence of the target leaf. The mechanism for this difference is not
obvious, but is presumably related to the importance of inoculum pressure in driving Rhynchosporium
epidemics. Greater understanding of the basic epidemiology of Rhynchosporium is necessary to elucidate the
mechanism more clearly. In the meantime, progress towards reducing fungicide use might be achieved by
timely and accurate prediction of seasonal disease risk. For example, it would be helpful to distinguish the low
disease year (1994) in this study from the moderate (1993) and high (1992) years. Reliable schemes for the
prediction of this disease are not currently available. However, the data mining approach used by Pietravalle et
al., (2002) to develop an early warning scheme for disease in wheat, might be applied to formulate a similar
prediction scheme for Rhynchosporium.
2. Temporal and spatial development of epidemics.
Introduction
The symptoms of barley “leaf blotch” are irregular lozenge-shaped pale brown lesions with darker brown edges
(Ozoe, 1956). Lesions merge when the disease is severe and whole leaves become brown, dry and shrivelled.
Severe epidemics develop when there is rapid infection of newly unfolded leaves during stem extension,
resulting in yield losses of up to 30-40% but more usually 1-10% in the UK (Jenkins & Jemmett, 1967).
Sources of conidia initiating leaf blotch epidemics are thought to be infected crop debris on the soil surface or
infected seed (Polley, 1971; Stedman, 1982). Rhynchosporium secalis can survive over winter on the debris and
sporulate for up to 340 days. PCR analysis can detect the disease in young barley up to 14 days (incubation
period) before lesions start to appear on leaves. Successive cycles of disease spread conidia from older leaves to
healthy young leaves during the growing season. Spread of the disease is associated with rainfall rather than
wind as the conidia are embedded in mucilage (Skoropad, 1959; Stedman, 1980). Rain drops falling on infected
plants can splash numerous conidia to carry the infection to other leaves and plants (Stedman, 1980). Regular
periods of rainfall sustain the disease by splash dispersal (Fitt et al., 1986) and by providing long periods of leaf
wetness favouring infection and sporulation (Rowe, 1979). Gradients of dispersal influence the rate at which
patches of disease increase in both size and intensitey. Colder winters also favour the disease, possibly because
the primary source of inoculum is depleted more slowly than in mild winters (Skoropad, 1966). Both winter and
spring barley are susceptible to leaf blotch, but winter barley is generally affected more severely (Lester, 1966),
particularly when sown early (Stedman, 1982).
Methods
Experimental design, treatments and assessments
i) Crops of winter barley (cv. Maris Otter) and spring barley (cv. Apex) were grown at Rothamsted over three
successive seasons, 1985-88, at sites where barley had not been grown for several years (Davis, 1990). In the
1985-1986 experiment winter barley cv. Maris Otter was sown on October 21st 1985 and in 1986 spring
barley cv. Apex was sown on May 6th. In the 1986-87 experiment early winter barley was sown on
September 25th and late on November 10th. Inoculum was introduced into all plots except the control by
scattering chopped straw from crops previously infected with R. secalis; no fungicides were applied. Two
plots were used for each inoculated treatment and one for control. Fifty randomly selected plants in each plot
were marked with numbered stakes after second leaf expansion and individual leaves were colour tagged. At
least 25 marked plants remained in each plot at the end of the season. Leaf blotch severity was recorded as
percentage of leaf area covered by lesions on each leaf. Epidemic progress was measured at the leaf level and
recorded weekly throughout the life of the crop, while meteorological data were recorded at a station about
100 m from the experimental sites.
CSG 15 (1/00)
6
Project
title
Epidemiology of Rhynchosporium to improve barley risk
assessment
DEFRA
project code
AR0510
ii) The development of leaf blotch (Rhynchosporium secalis) foci were studied at Rothamsted in 1984 in plots
sown either with a pure stand of of the susceptible spring barley cultivar Apex or in mixtures of Apex and the
resistant cultivar Koru. Mixture ratios were either 1:2 or 1:4 of the susceptible:resistant cultivars. Each
treatment was replicated three times and the nine plots were arranged in a randomised block design. The 9 x
9m plots were sown with undressed seed at a rate of 157 kg/ha and row spacing of 11cm. Blocks were
separated by 26m of crop (cv Atem) sown at the same rate but treated with triadimenol plus fuberidazole (as
Baytan). The whole experiment was sprayed with ethirimol (as Milgo) on 6th June to prevent the
development of powdery mildew. Meteorological measurements were made at a weather station less than
100m from the plots. On 16th May five pots each containing 10 barley seedlings (cv Apex) grown in the
glasshouse and infected with R. secalis, were placed at the centre of each plot. The pots were replaced at
fortnightly intervals with similarly infected seedlings. In order not to spread inoculum by disturbing plants,
disease assessments were carried out from a flat trolley running on a ladder supported horizontally above the
crop. One end of the ladder was supported by a post at the centre of the plot, the other end on a trestle at the
plot edge, enabling the ladder to be swung radially over the plot to encompass all directions. Assessments
were made along eight radii: N, NE, E, SE, S, SW, W and NW. In the first and second assessments, 10 plants
were examined in an area 10 x 30cm at set distances from the pots at the plot centre; in subsequent
assessments, when more disease was present, five plants were examined in each 10 x 30cm rectangle.
Disease on each leaf of the main stem was estimated using a standard key (Anon., 1972). The assessments
were carried out on 15th June, 24th June and 16th July at distances up to 200cm by 25cm increments along
each radius.
Analysis of treatments and effects.
The spatial and temporal aspects of disease development within a crop and during the season were investigated,
based on detailed analysis of disease incidence and severity data from individual leaf layers taken at regular
intervals. Epidemic development was compared using a modelling approach and the differences between
epidemics were assessed to understand how the disease spreads. The data were also linked to weather factors,
especially temperature and rainfall.
Results
Winter and spring barley 1985-1986
The life span of individual leaves and the intervals between the appearance of one leaf and the next differed
considerably within and between plants in winter barley, but were more rapid and less variable in spring barley.
After rain, R. secalis spread onto the first two expanding leaves in late November and further lesions were
recorded at low incidences throughout the winter (Figure 2). During April/May (c. day 130) growth was rapid
and temperatures were relatively high. There was a period of rain in early April, with a sharp increase in leaf
blotch by day 163. A period of hot dry weather in June was associated with a check in epidemic progress.
Spring barley emerged about May 5th 1986 (c. day 180 on winter barley scale) and rapid growth occurred, with
mean temperatures above 10°C and a period of rain. Extensive leaf blotch developed after about 14 days on the
first leaf but the next few leaves seemed to ‘grow away’ from the infection in spite of suitable temperatures and
rainfall. After heavy rainfall at the end of May there was an increase in disease by June 11 (day 220) followed
by drier hotter weather with little infection of flag leaves by the final assessment on July 7 (day 246). At the end
of the season, there was less disease on spring barley than on winter barley and the rates of progress of the
epidemics appeared to be different. However, when disease severity was plotted against physiological time
(degree-days above 0°C), the rates of progress appeared similar.
CSG 15 (1/00)
7
Epidemiology of Rhynchosporium to improve barley risk
assessment
Project
title
DEFRA
project code
AR0510
a)
Rainfall mm
Mean temperature °C
20
20
10
10
0
0
Temperature (°C)
Rainfall (mm)
30
-10
b)
% disease on live leaves
10
winter barley cv. Maris Otter
spring barley cv. Apex
5
0
0
50
100
150
200
250
Days from 20 November 1985
Figure 2. Epidemics of leaf blotch on winter and spring barley (inoculated with infected crop debris) during
1985-1986. (a) Rainfall and mean temperature during 1985-86 growing season (b) Mean percentage area
affected by leaf blotch on all living leaves
Early-sown and late-sown winter barley 1986-87
The early-sown barley emerged on October 10th 1986, about 14 days after sowing and by the end of a mild wet
October (Figure 3a) there were three leaves on main shoots with some tillering. Disease was apparent on the
inoculated plots by early November, but not on the uninoculated control. In mid-November, when the late-sown
barley was emerging, there was a marked decrease in temperature and plant growth rates on all plots declined.
There were more leaves on the early-sown barley than on the late-sown barley and consequently more disease
on the inoculated early-sown plots (Figure 3b). Disease severity increased until December on the inoculated
early-sown plots (Figure 3a and 3b) but the uninoculated control and late-sown plots showed little disease.
Conditions in March were mild and new lesions appeared on all plots, with proportionately more on the earlyCSG 15 (1/00)
8
Epidemiology of Rhynchosporium to improve barley risk
assessment
Project
title
DEFRA
project code
AR0510
sown plots. The weather then became dryer with a sharp increase in temperature to a max of 28°C in April.
There were few new lesions and rapid production of new leaves so that plants ‘grew away’ from the disease.
Plants grew rapidly after the increase in temperature in March and by June the number of leaves produced on
main shoots were similar in all plots, though the height of the flag leaf was 85 cm in early-sown barley
compared to 75 cm in late-sown barley. The flag was leaf 14 on the inoculated early-sown and late-sown barley
but on the uninoculated plot the flag was leaf 15.
a)
Rainfall mm
Mean temperature °C
20
10
10
0
0
Temperature (°C)
Rainfall (mm)
20
-10
b)
% disease on live leaves
June 26
early sown
early sown control
late sown
20
April 22
Dec-24
10
0
0
50
100
150
200
250
Days from 10th October 1986
Figure 3. Epidemics of leaf blotch on uninoculated early-sown (control), inoculated early-sown and late-sown
winter barley during 1986-1987. (a) Rainfall and mean temperature (b) Mean percentage area affected by leaf
blotch on all living leaves
CSG 15 (1/00)
9
Project
title
Epidemiology of Rhynchosporium to improve barley risk
assessment
DEFRA
project code
AR0510
Analysis of spatial data showed that direction was a significant factor with more spread of disease in the
direction of prevailing winds (Figure 4). The data also showed how sowing in mixtures of susceptible and
resistant cultivars affects the spatial spread of disease. Spatial spread of disease was more limited and severity
was lower in mixtures of 2:1 and 4:1 resistant to susceptible compared to susceptible alone.
Disease Gradient Analysis
2
3
Direct.: SW
4
5
Direct.: W
-3
-7
Direct.: S
Direct.: SE
Direct.: NE
Direct.: NW
logit disease
-3
-7
-3
-7
Direct.: E
Direct.: N
-3
-7
2
3
4
5
log distance
Figure 4. Example of disease gradient from point source of inoculum. This assessment made on 15th June was
made 30 days after introducing a point source of inoculum as infected potted ssedlings. There is significantly
more spread in the direction of prevailing winds.
Discussion
The results show important aspects of the relationship of disease to distance, time and weather conditions and
will provide an understanding central to ascertaining disease risk. The strong correlation between disease
progress and degree-days in 1985-86 and 1986-1987 suggests that epidemic progress was a function of
accumulated temperature. However some of the fluctuations in amounts of disease were probably related to
specific weather events such as periods of rain or temperatures above 20°C. In particular, in the autumn of
1986 there were high incidences of disease in early-sown winter barley that developed when there were
frequent periods of rain and mean temperatures between 5° and 15°C. In the winter months, temperatures
below 5°C apparently limited the amount of disease that developed on all the winter barley crops. Even when
there were periods of rain in 1986 and 1987 there was little additional disease recorded while mean
temperatures were  0°C.
In spring, once mean temperatures were above c. 5°C, the largest increases in amounts of disease could
generally be related to periods of several days with rain c. 15-20 days earlier. In dry weather, infected leaves
CSG 15 (1/00)
10
Project
title
Epidemiology of Rhynchosporium to improve barley risk
assessment
DEFRA
project code
AR0510
senesced more rapidly and the largest decreases in amounts of disease on the remaining live leaves generally
occurred c. 15-20 days after a dry period. Dry periods were often accompanied by daytime temperatures above
25°C, unfavourably high for R. secalis, which may also have played a role in impeding the progress of
epidemics.
Temperature differences initially affected the amounts of disease at the rosette pre-extension stage on plants
sown at different dates. Early-sown winter barley emerged in late autumn when conditions were favourable for
disease and when the production of new leaves was relatively rapid. Thus there was an opportunity for a
reservoir of disease to be established at the base of the main shoot and tillers. Poor growth and a reduction of
tillering in barley crops affected with R. secalis is observed commonly (Jenkyn et al., 1989; Ozoe, 1956). When
conditions are favourable for R. secalis, particularly at the rosette stage, a reduction in barley growth rate could
also contribute to the reservoir of disease. Late-sown winter barley emerged when temperatures were low and
consequentially less favourable for disease, and when the production of new leaves was slower. There was
therefore a smaller reservoir of disease at the base of the late-sown plants. Spring barley emerged when the
conditions were again favourable for the disease but the rosette stage was short by comparison with that of
winter barley and fewer leaves were involved. For both late-sown winter barley and spring barley there was less
inoculum at the base of the plants to initiate further infections during the period of stem extension.
Canopy density in barley crops is also an important factor when considering conditions for disease development
and a lower canopy density can provide less favourable conditions (Mayfield & Clare, 1984). Differences in
canopy density may also have affected the rates at which leaf blotch spread up the plants in these experiments.
In the coldest period in the 1986-87 winter, there was a loss of winter barley plants with fewer plants lost in the
well-established early-sown winter barley than the late-sown barley. The difference in canopy density (not
measured here) may partly explain why the amount of disease was greater on early rather than late-sown barley
in the 1986-87 season, particularly on later leaves.
3. Effects of crop architecture and rain splash.
Methods
Experimental design, treatments and assessments
The spread of the pathogen Rhynchosporium secalis by rain splash in relation to crop architecture was
investigated in a small-scale field experiment at Rothamsted in 2002. The experiment will be repeated in 2003.
i) Four winter barley cultivars (Avenue, Maris Otter, Sumo and Vertige) with contrasting architectures were
used to test the hypothesis that canopy architecture affects the likelihood of disease spread. In addition four
different plant growth regulator regimes, using Moddus and Terpal, were used to manipulate crop height
within cultivars. Moddus was applied at early stem extension time to shorten the lower stem and Terpal later
to shorten the upper stem
ii) The experiment involved regular measurements of leaf layer heights, lengths and widths of leaves and leaf
insertion angles. Average number of plants per square meter and no of tillers per plant were determined.
Tagged plants in each plot were measured at least weekly and destructive samples were taken from a separate
area of each plot to assess disease development and inoculum spread.
iii) Regular disease assessments (at least weekly) were made for each leaf layer to assess disease progress and
transfer; the heights of the highest lesions in each plot were also recorded.
Results
This experiment has provided important basic information confirming the role of splash dispersal, the effect of
contrasting crop architectures on disease development and a basis for planning future field experiments. PCR
diagnostic carried out on ten plants from each plot showed that Rhynchosporium secalis was present in all plots,
but not visible on 11/02/2002. Lesions became visible in all 48 plots by 22/02/2002. Periods of heavy rain
during stem extension then spread the disease rapidly through the leaf layers causing an epidemic on all plots.
Analysis of results for 2002 shows that there may be important differences between the Septoria/wheat and
Rhynchosporium/barley pathosystems. In wheat the extent and duration of the green leaf areas of the flag leaf
CSG 15 (1/00)
11
Epidemiology of Rhynchosporium to improve barley risk
assessment
Project
title
DEFRA
project code
AR0510
and second leaf from anthesis onwards are of key importance in determining yield but for barley the flag leaf,
which can be very small in some cultivars, appears to be less important. The effects of R. secalis on the green
leaf areas of the second and third leaves may be more important for determining yield in barley (Figure 5).
There were considerable differences among the cultivars in disease levels and in yield. Cultivars Avenue and
Sumo, with the largest amount of green leaf area for leaves two and three in the last 3-4 weeks of assessment,
produced the highest yields and Maris Otter and Vertige the least. The two cultivars with highest yields also
had the largest average leaf sizes. Maris Otter was the lowest yielding cultivar with an average of 11% of leaf 2
and 15% of leaf 3 covered by lesions, while Avenue was best yielding with 7% lesions for leaf 2 and 8% for
leaf 3. PGR effects in 2002 were small, possibly because dose was too low and/or not applied at the optimum
time because of wet or windy weather. Table 2 summarises some of the results for 2002.
Yield (85% DM) vs sum of AUGLPC
9
Yield (85%DM) tonnes/ha
8
7
6
y = 0.997Ln(x) + 1.2707
R2 = 0.7753
5
G-85%
Log. (G-85%)
4
3
2
1
0
0
100
200
300
400
500
600
700
800
Sum of AUGLPC (Leaf 2+3)
Figure 5. The relationship between yield and green leaf area of leaves 2 and 3 during 27th May to 24th June
2002. AUGLPC is area under green leaf area progress curve.
Table 2. Summary of barley cultivar measurements and comparisons 2002
Leaf length
Leaf insertion angle
Green leaf area
% leaf affected by disease
Senescence rate
Yield/1000g weights
Leaf layer height
Avenue
+
+
+
+
Sumo
+
+
+
-
Vertige
+
+
+
-
Maris Otter
+
+
+
+
The experiment will be repeated in 2003. These and other results will be crucial to planning an investigation of
the role of rain and host structure and will underpin and facilitate the proposed LINK project, which will form
the basis of a disease risk assessment model.
CSG 15 (1/00)
12
Project
title
Epidemiology of Rhynchosporium to improve barley risk
assessment
DEFRA
project code
AR0510
Related Research
The value of epidemiological information, including the influence of crop growth and canopy architecture on
disease progress, has been shown in work on Septoria tritici on winter wheat (Lovell et al. 1997). Recent
developments in our understanding of the concepts of splash dispersal (Lovell et al. 2002; Pietravalle et al.
2001) and the role of the sexual stage in the S. tritici/wheat pathosystem will provide a useful basis for
investigating and understanding disease progress in the R. secalis/barley pathosystem. Recent work at ADAS
on the determinants of fungicide spray decisions and prediction of effective fungicide doses through
observation of leaf emergence for the wheat/Septoria pathosystem have provided a platform for similar work on
barley/R. secalis (Paveley et al. 1997; Paveley et al. 2000). The pre-LINK project has made use of the
frameworks developed for the Septoria/wheat pathosystem. This project complements an HGCA/SEERAD
project ‘Winter barley reference cropping to provide an improved understanding of growth and yield
formation’ to provide a ‘Barley Growth Guide’ for growers, and an HGCA project ‘Appropriate fungicide
doses on winter barley’. The relationship between disease and yield loss in winter barley has been shown to be
as variable as in wheat (Gaunt & Wright 1992; Wright & Gaunt 1992). These two projects will provide
understanding of the processes of yield formation in barley and the impact of disease on those processes. This
should allow assessments of disease risk to be integrated with assessments of the likely impact of that disease
on yield and grain quality, in order to rationalise fungicide use.
Forward View
Further work is required on the biology and epidemiology of leaf blotch (R. secalis) on winter barley:Barley risk analysis and improved targeting of fungicide
Objectives
 Understand the interactions of weather with epidemic development of leaf blotch.
 Provide early risk assessment of the likelihood of a severe epidemic of leaf blotch based on weather
conditions early in the growing season (WindowPane analysis developed at Rothamsted).
Barley crop improvement to provide natural resistance to leaf blotch and minimal inputs
Objectives
 Determine the plant (cultivar) characteristics that favour ‘escape’ from disease.
 In collaboration with other divisions and functions at Rothamsted, and other research institutes, combine
research results and optimise the plant characters that are most effective in providing ‘escape’ from disease,
weed suppression and also minimise pesticide requirements.
 Define an ideotype for winter barley (the traits of an ideal variety) requiring minimal inputs and provide
specific requirements for a crop genetic improvement programme.
Early epidemic establishment and possible role of a sexual stage in R. secalis
Objectives
 Investigate how R. secalis is able to rapidly invade experiments where barley has not been grown for
several years and with fairly even distribution through the crop – ascospores, infected seed?
 Investigate whether natural wild grasses (e.g. Alopecurus pratensis, Arrhenatherum elatius Holcus lanatus,
Elytrigia repens wild Hordeum spp. etc.) are providing a green bridge for R. secalis or an alternative host
for a potential sexual stage
 Carry out spore trapping from crop debris, through autumn and winter - PCR analysis, look for possible
ascospores.
 Investigate possible ‘mating types’ in R. secalis. Compare ‘mating type’ gene sequences with those of
closely related species known to have a sexual stage.
 Determine the key period for the development of R. secalis epidemics in winter barley
- between crop emergence and early stem extension as in wheat/Septoria tritici?
CSG 15 (1/00)
13
Project
title
Epidemiology of Rhynchosporium to improve barley risk
assessment
DEFRA
project code
AR0510
Selected Publications, papers and posters
ADAS, 2003, ‘The relationship between optimum spray timing and leaf emergence for Rhynchosporium on
barley’, in preparation.
Henman, DF, Davis, H, Fitt, BDL, 2002. Development of leaf blotch (Rhynchosporium secalis) epidemics on
barley. Paper for BCPC Pests and Diseases 2002, Brighton.
Henman, DF, Davis, H, Fitt, BDL, 2002. Development of leaf blotch (Rhynchosporium secalis) epidemics on
barley, Poster for 2002 BCPC conference, Brighton.
Henman, DF, Davis, H, Fitt, BDL, 2002. Development of leaf blotch (Rhynchosporium secalis) epidemics on
barley, Poster for Rothamsted Research day.
Rothamsted, 2003, ‘The effects of crop architecture and rainsplash on the spread of leaf blotch
(Rhynchosporium secalis)’, in preparation.
Presentations
Progress report presented to DEFRA at Wheat Self Defence day, London April 2002.
Annual review presented to DEFRA, London, November 2002.
Barley crop architecture experiments featured in farm visit at ARIA day, Rothamsted, June 2002.
Barley and Rhynchosporium resistance at ARIA/HGCA Fungicide resistance workshop, Newbury, April 2003.
Articles
Bock, CH, Cutting risk of resistance, Farmers Guardian, May 10 2002.
Other references
Anon, Manual of Plant Growth Stages and Disease Assessment Keys. Lion House, Alnwick, Northumberland:
MAFF (Publications).
Davis, H, 1990. Studies of the biology and epidemiology of Rhynchosporium secalis (leaf blotch on barley),
London, University of London, PhD thesis.
Fitt B D L; Creighton N F; Lacey M E; McCartney H A (1986). Effects of rainfall intensity and duration on
dispersal of Rhynchosporium secalis conidia from infected barley leaves. Transactions of the British
Mycological Society, 86, 611-618
Gaunt, RE, Wright, AC, 1992. Disease yield relationship in barley .2. contribution of stored stem reserves to
grain filling. Plant Pathology 41, 688-701.
Genstat, 2002. Genstat for Windows. Release 6.1. Sixth ed. Oxford: VSN International Ltd.
Jenkins, JEE, Jemmett, JL, 1967. Barley Leaf Blotch. National Agricultural Advisory Service Quarterly Review
75, 127-132.
Jenkyn, JF, Stedman, OJ, Dyke, GV, Todd, AD, 1989. Effects of straw inoculum and fungicides on leaf blotch
(Rhynchosporium secalis), growth and yield of winter barley. Journal of Agricultural Science 112, 8595.
Lester E (1966). Cereal Disease 1. Leaf blotch of barley. The Journal of the Institute of Corn and Agricultural
Merchants, 14, 16-17.
Lovell, DJ, Parker, SR, Hunter, T, Royle, DJ, Coker, RR, 1997. Influence of crop growth and structure on the
risk of epidemics by Mycosphaerella graminicola (Septoria tritici) in winter wheat. Plant Pathology 46,
126-138.
Lovell, DJ, Parker, SR, Van Peteghem, P, Webb, DA, Welham, SJ, 2002. Quantification of raindrop kinetic
energy for improved prediction of splash-dispersed pathogens. Phytopathology 92, 497-503.
Ozoe, S, 1956. Studies on the Rhynchosporium scald of barley and its control. Bulletin of the Shimane
Prefectural Agricultural Institute 1, 115-122.
Mayfield, AH, Clare, BG, 1984. Effects of common stubble treatments and sowing sequences on scald disease
(Rhynchosporium secalis) in barley crops. Australian Journal of Agricultural Research 35, 799-805.
Paveley, ND, Lockley, KD, SylvesterBradley, R, Thomas, J, 1997. Determinants of fungicide spray decisions
for wheat. Pesticide Science 49, 379-388.
CSG 15 (1/00)
14
Project
title
Epidemiology of Rhynchosporium to improve barley risk
assessment
DEFRA
project code
AR0510
Paveley, ND, Lockley, D, Vaughan, TB, Thomas, J, Schmidt, K, 2000. Predicting effective fungicide doses
through observation of leaf emergence. Plant Pathology 49, 748-766.
Pietravalle, S, van den Bosch, F, Welham, SJ, Parker, SR, Lovell, DJ, 2001. Modelling of rain splash
trajectories and prediction of rain splash height. Agricultural and Forest Meteorology 109, 171-185.
Pietravalle, S, Vandenbosch, F, Shaw, MW, Parker, SR, 2002. Towards an early warning system for winter
wheat disease severity. Paper read at BCPC Pests and Diseases 2002, at Brighton.
Polley, RW, 1971. Barley leaf blotch epidemics in relation to weather conditions with observations on the
overwintering of the disease on barley debris. Plant Pathology 20, 184-190.
Rowe J (1979). The use of cultivar trials in assessing the incidence of cereal disease in England and Wales.
Septoria nodorum in wheat and Rhynchosporium secalis in barley. Annals of Applied Biology, 93, 247255.
Skoropad W P (1959). Seed and seedling infection of barley by Rhynchosporium secalis. Phytopathology, 49,
533-622.
Skoropad W P (1966). Sporulating potential of Rhynchosporium secalis on naturally infected leaves of barley.
Canadian Journal of Plant Science, 46, 243-247.
Stedman O J (1980). Observations on the production and dispersal of spores, and infection by Rhynchosporium
secalis. Annals of Applied Biology, 95, 163-175.
Stedman, OJ, 1982. Epidemiology of barley leaf blotch. Spore production of Rhynchosporium secalis on barley
straw. Rothamsted Report for 1981 Part 1, 194-195.
Wright, AC, Gaunt, RE, 1992. Disease yield relationship in barley .1. yield, dry-matter accumulation and yieldloss models. Plant Pathology 41, 676-687.
CSG 15 (1/00)
15