The disease common bunt (Tilletia tritici syn. T.caries) has become a more
frequent problem in Europe during the last 15 year. In organic farming,
common bunt causes serious problems for seed producers, and many seed lots
are discarded due to contamination with Tilletia tritici. Due to the biology of
Tilletia tritici, with spores loosely attached to the surface of the grain, there is a
risk that spores can be disseminated via grain handling equipment, including
combine harvesters. Preventive efforts should focus on the role of grain
handling equipment, and how the fungus life cycle can be broken. We have
investigated the dissemination of spores via the combiner after harvesting
infected fields. We conducted six trials over three years, counting the number of
spores in up to seven emptyings of the combiner. Up to 4 emptyings with non-contaminated grains are needed in order to achieve an acceptable low level of
contamination. Using Denmark as a case, we discuss several important factors
influencing disease spread via harvest equipment. It is concluded that one easy
preventive strategy is to avoid using the first 4 emptyings of the combiner. This
method should always be combined with analysis of seeds for spore load.
The fungal disease common bunt (Tilletia tritici (Berk.) Win 1775 syn. Tilletia caries (D.C.) Tul. 1847) became an increasing problem in wheat in Denmark and many other European countries in the late 1980's. It has been argued that the presence of common bunt usually is related to missing or insufficient seed treatment (Stapel et al. 1976), including use of farm produced seeds (Nielsen 1996). However, the development and use of more effective fungicides in conventional agriculture during 1990's has not improved the situation. Another cause for the increasing levels of common bunt may be the increased area of winter wheat sown, contributing to a general increase in inoculum pressure. The area for winter wheat production in Denmark increased by almost 8 fold during the past 30 years, now comprising 25% of the total agricultural area, and approximately 44% of the area grown with cereals (Landbrugsraadet 1999). Random samples of certified winter wheat seeds tested by the Danish seed testing authorities in 1989, 1992 and 1993 showed low but significant presence of contamination in 14% of the samples. Thus, the inoculum is present and poses a potential risk.
In organic farming, common bunt is one of the most important cereal diseases, due to its ability to build up to devastating levels within 1-2 years (Borgen et al. 1992). On the basis of consultations with Danish organic seed producers, we estimate, that approximately about -½ of certified organic cereal seeds batches are rejected due to problems mainly with seed pathogens. Of these problems, common bunt is without doubt the largest, accounting for around half of the rejected seed batches. During the past 10 years, many studies have been conducted on alternative ways to control common bunt in organic farming (Becker & Weltzien 1993, Borgen & Davanlou 2000, Borgen et al. 1994, Heyden 1999, Kristensen & Borgen 2000, Piorr 1991, Spiess & Dutscke 1991, Winter et al. 1994, Winter et al. 1998), but so far there are no reliable large scale and organically appropriate control methods. There are difficulties in procuring enough organic seeds throughout Europe (Wyss & Viethaler 2000), and in some countries conventional and fungicide treated seeds are used organic farming. EU-regulations demand that organic seeds are to be used after 2004. Therefore, there is urgent need for initiatives which can facilitate the conversion of this part of the organic agricultural practice, and due to the principles of organic farming the emphasis should be on seeking for preventive measures.
At harvest sori containing teleutospores of Tilletia tritici (bunt balls) are
crushed in the combine harvester, and spores are disseminated to the surface of
the healthy seeds within the machine. Spores are loosely attached to the surface
of the seeds, leading to contamination of seed handling equipment. When
contaminated seeds are sown, the disease is spread from one growing season to
the next. Thus, the combiner and other seed handling equipment are crucial
factors in the spread of Tilletia tritici. The role of agricultural machinery is
mentioned by several authors (e.g. Nielsen & Nielsen 1994, Wiese 1991, Nielsen
& Jensen 1989), but to our knowledge no quantitative information on the role of
seed handling equipment on the spread of the disease is available. We conducted
a study to investigate the spread of Tilletia-spores from infected fields to clean
fields via combining equipment.
Materials and methods
In the period 1993 to 1995 six experiments were conducted at the Royal
Veterinary and Agricultural University Research Farm in Taastrup, Denmark. A
14 feet wide Dronningborg combiner harvester with a loading capacity of
approximately 2500 kg grain was used. A highly infected winter wheat area
previously used for experimental purposes with common bunt and infection
levels above 50% (experiment 1993, 1994a and 1995a,b; Table 1), or naturally
infected fields (experiment 1994b and 1995c; Table 1) served as source of
inoculum. After harvesting the infected field the combiner was used to harvest a
clean field, i.e. a field established from certified seeds treated with approved
efficient seed dressings. In these fields the infection level is close to zero.
Because of the limited area available the harvester was only loaded to half full
capacity before emptying. Approximately 10 to 15 kg samples of grain were
collected continuously throughout the emptying procedure into a clean bag. The
bag was sealed, and the procedure repeated from 4 to 7 times. After the
sampling in the field, the contents of the bags were thoroughly mixed and a
subsample taken for analysis of Tilletia-spores. Analysis of samples are carried
out by the Danish Seed Testing Station, using the ISTA haemacytometer
standard method (Kietrieber 1984). Results are presented as number of spores
per gram of grain.
Since it cannot be expected that this approach can reduce the infection level to zero we used a segmented model to the data, i.e. emptyings with clean seed will result in reduction in number of spores, but the number of spores will reach a stable level which is >0. In order to determine the number of emptyings necessary to reach a stable low level, data was fitted to a segmented model by PROC NLIN in SAS (SAS Institute, USA). All data on spore counts were log-transformed to stabilize variance and to fit equations. The model used to describe data was:
log (Y) = a + bx if x < xm
log (Y) = k if x > xm
where Y = (log-transformation of spore number + 1), a-parameter is the intercept or the spore load in infected fields, b-parameter is the relative effect of each emptying, and x is the number of emptyings. xm denotes the number of emptyings necessary to achieve maximum possible reduction in spore load.
xm = (k-a)/b
Parameters of the predicted model are presented including 95% confidence-limits. xm is calculated from the parameters, and a corresponding interval of xm is
presented (Table 2).
In 1995 (a+b) the same infected experimental area was used. In the other
experiments different fields are used as the source of inoculum before harvesting
a clean field. As expected the number of spores found decreased each time the
machine was emptied and refilled with clean seeds (Table 1). The initial level of
contamination seems to have an impact on both the degree and the duration of
contamination in the following emptyings (Table 1).
Table 1: Occurrence of spores (number of spores per gram of grain) of Tilletia tritici in different emptyings of the combiner after harvesting an infected field, emptying no. 0 = infected field.
|Emptying no. 0||167,250||-||-||385,000||720,500||120,000|
- no sample taken.
* not used in the analysis.
The statistical model fit the data quite well (Figure 1, Table 2). First,
observations from each sub-experiment was fitted to the model, and only 1994b
and 1995a differed significantly in the k-parameter, which mainly was caused by
a very exact and low estimate of the k-parameter in 1995a. All sub-experiments
were used in the analysis, but 1994a emptying no. 2 was considered extreme and
removed from the dataset.
Figure 1: Log(no. of spores pr. gram grain + 1) as a function of number of emptyings of the
combiner. Solid line: Predicted according to model. Dashed lines: 95% confidence limits
according to model.
Table 2: Estimated parameters, backtransfromed spore numbers, and corresponding xm- values.
|Parameter||log-value||±95% conf.limits||spore numbers
95% conf. interval
Between 2 and 4 emptyings were sufficient to reach a stable low level between 20-125 spores per gram grain (Table 2, Figure 1). In no case were any spores observed after 5 emptyings. However, some precautions have to be taken. First, 'emptyings' is not a very exact measurement, and secondly, it should be noted that the combiner was only half loaded each time. Other factors, such as type of equipment, size and adjustment, as well as moisture content of grains and bunt balls might influence the results. It is possible that 'pockets' of spores in the combiner can be released later, for example after stop-start procedures. However, the results presented show a significant reduction in spore load with emptying/filling procedures, even though there will be some spores present in the combiner. In 1994 neither of the experiments reached the 0-level, even though the initial infection level was not that high (Table 1). Perhaps more emptyings were needed or perhaps the 'clean' field was not completely clean. In 1995 samples from the clean field were taken before the combiner was brought in from the infected field, and those samples showed that the clean field was not contaminated by spores before the experiment. Our initial harvest was all in heavily infected fields, and the results indicate that the initial level of contamination has an influence on the contamination level in the subsequent emptying. Since initial contamination levels would not be that high in practice, we conclude that a maximum of 4 emptyings would be satisfactory in harvesting 'normal low-level contaminated' crops.
A range of different factors influence the epidemics and the level of common bunt attack. When preventive strategies are discussed, it is important to consider which factors of the disease cycle are the most important. Especially in organic farming, were no treatment possibilities are available, preventive efforts must be considered carefully. We judge the following factors to be of major importance:
Use of seeds that are not treated and not analysed
Farm border crossing use of grain handling equipment.
Frequency of susceptible crop/ size of winter wheat area.
Tilletia tritici is usually considered to be predominately seed borne (Weise 1991), and infections from soil only occurring under special climatic conditions in cases where wheat are grown without intercrops in a field previously grown with infected wheat (Yarham 1993). During the past few years it has been realized that Tilletia tritici can survive in, and infect from soil after several years without susceptible crops in the area as well (Borgen 2000). This puts even higher pressure on organic farming, because even if clean seeds are used, infection can still occur. In conventional agriculture this problem is less pronounced because the fungicides normally used (e.g. Sibutol LS 280) have effects on soil-borne infection as well (Nielsen 1998). However, possibilities for soil-borne infection emphasizes the need for attention on preventive measures in organic farming, since soil borne indection can serve as the primary source of inoculum, introducing infection to the system.
Use of seeds that are not treated and not analysed may also serve as a source of inoculum. This os often, but not necessarliy, connected to the use of home produced seeds. In Denmark it is estimated that approximately 15% of winter cereal crops are established on home produced seeds ( Nielsen et al. 1998). This production take place outside the seed certification scheme, and is therefore it is less likely that the seeds are subjected to control. Even in the conventional system, the home produced seeds might not be regularly subjected to seed treatment with fungicides to the same extent as within the certification scheme. Use of home produced seeds can cause problems in two ways: If contaminated seeds are used, the farmer can have serious problems with quality and yield loss within few years of production. If the farmer in some way 'share' harvesting equipment, either by neighbour-agreements or by professional machine pool support, spores can be disseminated, and home produced seeds can contribute to the maintenance of the common bunt life cycle and the general infection load. On the other hand, if the farm system is not infected, home production in a closed cycle using own seeds and combiner can be an advantage, since it minimizes the risk of introducing inoculum. Whatever seed propagation system used seed analysis should always be made to monitor the level of infection.
In Denmark approximately 15% of the cereals are harvested by professional machine pools. Neighbour based harvest cooperation has a significant extent as well, so we would estimate that about 50% of the Danish cereals are harvested by farm border-crossing equipment (Groth pers.comm.). This trend is supported by the prices of combine harvesters. Today a combine harvesting equipment have to harvest at least 200-250 ha/year to be profitable (Groth pers.comm.). In Denmark the average farm size was is 44 ha in 1997 (Anon. 2000), suggesting that a combiner have to cross farm borders. These structural and economical relations may play an important role in the maintenance and spread of common bunt.
The winter wheat area in Denmark has increased by almost 8-fold during
the past 30 years (Landbrugsraadet 1999). From an 'everything-else-equal' point
of view this means that use of home produced seeds and/or fields with soil
infection in combination with use of farm border-crossing harvest equipment,
will increase the probability for an undesirable connection of infected fields to
fields used for seed production. The combine harvester is an important factor
tying together the four above mentioned factors for attack and spread of
common bunt. Harvest machinery thus provides a focal point for preventive
methods. The role of shared harvesting machinery suggest that there is an
interaction between common bunt problems in organic and conventional
Our results show that spores of T. tritici can be spread from infected fields to
clean fileds, and indicate that 4 emptyings with clean seeds passing through a
combiner previously harvesting an infected field are sufficient to reach a
significant reduction in spore contamination. Starting with very heavy
contaminations, spore load will then be from 20-125 spores per gram seed.
Thus, this strategy does not ensure that the problem disappear, but it decreases
the probability and speed of further problems. The 'cleaning' of the combiner
might as well take place in non-susceptible crops, and would then not restrict the
use of harvested seeds. If there is knowledge on previous harvest of infected
fields, further efforts to prevent contamination might be necessary. We therefore
expect this strategy to be most relevant first of all in cases where harvest are
used for seeds, and when no information is available on previous harvest
infection level or combine harvester route. In those cases it can be considered a
cheap and easy preventive strategy to avoid the first 4 emptyings. Subseeding
post-field contamination in transport facilities, bags, storage facilities etc. off
course should be avoided as well. It is important to underline that this preventive
strategy does not replace seed analysis, which should always be carried out in
organic wheat production.
We thank the technical staff at The Royal Veterinary and Agricultural University
Research Farm for being helpful and patient when we did sampling from the
combine harvester. This study was possible through grants from the Danish
Ministry of Food, Agriculture and Fisheries, Strukturdirektoratet. We are
grateful for all support.
Becker, J. & H.C.Weltzien (1993) Bekämpfung des Weizensteinbrandes (Tilletia caries (D.C.)Tul. & C.Tul.) mit organischen Nährstoffen. Zeitschrift für Pflanzenkrankheiten und Pflanzenschutz 100(1):49-57.
Borgen, A. (2000) Perennial survival of common bunt (Tilletia tritici) in soil under modern farming practice. Zeitschrift für Pflanzenkrankheiten und Pflanzenschutz. 107(2):182-188
Borgen, A. & M. Davanlou (2000) Biological control of common bunt in organic agriculture. In: "Nature Farming and Micirobial Applications" (H.L. Xu , J. F. Parr and H. Umemura, eds.). Haworth Press Inc., New York, pp 159-174. Thematic issue of Journal of Crop Production.
Borgen, A., L. Kristensen & P. Kølster (1994) Bekæmpelse of hvedens stinkbrand uden brug of pesticider. 11 Danske Planteværnskonference 1994. Sygdomme og skadedyr. pp. 149-158.
Borgen, A, C.Markussen & L.Kristensen (1992) Frøbårne svampesygdomme på byg og hvede i økologisk landbrug - udbredelse og betydning i 1990. Den Kgl. Veterinær og Landbohøjskole, Institut for Jordbrugsvidenskab/Plantebiologi. Msc.-Thesis. 137pp.
Groth, E. Personal communication. Cheif advisor, Landsforeningen af Danske Maskinstationer.
Heyden, B. (1999) Laufende Versuche und Ergebnisse der Vegetationsperiode 97/98. Saatgut. Mitteilungen aus der Arbeit des Johanna und Carl Graf Keyserlink-Institutes 14:15-21
Keitreiber, M. (1984) ISTA Handbook on Seed Health Testing. Working Sheet no. 53. Bundesanstalt für Pflanzenbau, Wien, Austria.
Kristensen, L. & A. Borgen (2000) Water based thermotherapy to control common bunt (Tilletia tritici) in wheat (Triticum aestivum) in organic agriculture. in prep.
Landbrugsraadet. AgroFact.dk.[online] Landbrugsraadet, Copenhagen 09.03.1999. Available at http://www.agrofact.dk/. [Cited 14.07.2000]
Nielsen, B.J. (1998) Frøbårne sygdomme i korn. In: Pesticidafprøvning 1997, Landbrugsafgrøder. (L.N. Jørgensen & P.E. Jensen (eds.). DjF-Rapprot nr. 1 1998, pp. 59-79.
Nielsen, B.J. (1996) Udsædsbårne sygdomme i korn. In: Pesticidafprøvning 1995, Landbrugsafgrøder. SP-rapprot nr. 1 1996, pp. 75-85.
Nielsen, B.J., A.Borgen, C.Scheel og G.C.Nielsen (1998) Strategies for controlling seed-borne diseases in cereals and possibilities for reducing fungiside seed treatments. The Brighton Conference - Pest and Diseases 1998, p.893- 900
Nielsen, G.C. & J.P. Jensen (1989) Markens sygdomme og skadedyr. Det kgl Danske Landhusholdningsselskab, Frederiksberg.
Nielsen, B.J. & G.C. Nielsen (1994): Stinkbrand og jordsmitte. 11 Danske Planteværnskonference 1994. Sygdomme og skadedyr pp. 89-103.
Piorr, H.P. (1991) Bedutung und Kontrolle saatgutüberertragbarer Schaderreger an Winterweizen im Organischen Landbau. Hohen Landwirtschaftlichen Fakultät de Rhenischen Friedrich-Wilhelms-Universität zu Bonn. 149 pp.
Plantedirektoratet (1996) Rapport fra Arbejdsgruppen vedrørende undersøgelse af mulighederne for reduktion af brugen af bejdsemidler til sædekorn. Bilag 9.
Spiess, H. & Dutscke (1991): Bekämpfung des Weizensteinbrandes (Tilletia caries) im biologisch-dynamischen Landbau unter experimentellen und praktischen bedingungen.Gesunde Pflanzen 43(8):264-270
Stapel, C., J.Jørgensen & J.E.Hermansen (1976) Sædekornets sygdomme i Danmark. Tidsskrift for Landøkonomi 3:187-283
Wiese, M.W. (1991) Compendium of Wheat Diseases. APS Press.
Winter, W., I. Bänziger, H. Krebs & A. Rüegger (1994) Warmwasserbehandlung von Weisensaatgut. Agrarforschung 11-12:492-495
Winter, W., I. Bänziger, H. Krebs and A. Ruegger (1998) Getreidebrände und Gerste-Streifenkrankheit sanft behandeln. Agrarforschung 5(1):29-32
Wyss, E & C. Viethaler (2000). Final report on the international conference on biodiversity and organic plant breeding. 2nd and 3rd December 1999 in Frick Switzerland. [online]. FiBL, Research Institute of Organic Agriculture. Frick, Switzerland. 38 pp., May 2000. Available at http://www.biogene.org/e/e-indexx.html [Cited 14.07.2000]
Yarham, D.J. 1993: Soilborne spores as a source of inoculum for wheat bunt
(Tilletia caries). Plant Pathology 42:654-656