Appendiks 5:








Milk powder seed treatment

to control common bunt

(Tilletia tritici) in wheat






Summary is submitted, and if approved by the scientific comity, the full article will be submitted to: Seed treatment - Challenges and Opportunities. Symposium arranged by British Crop Protection Counsil. *)









By










Anders Borgen and Lars Kristensen














*) Note til anden udgave: Status for publikationen er, at en forkortet udgave, hvor også andre resultater indgår er accepteret til symposiet.

Keywords: T.caries, biological control, seed pelleting, organic farming, germination vigour.



Summary

Common bunt (Tilletia tritici) can infect the germinating plant of wheat from spores attached to the seeds or from inoculum in the soil. Coating the seeds with milk powder prior to sowing serves as a nutrient source for microorganisms resting in the soil or seed surface which are antagonistic to the pathogen.

Five years of field trial shows that the reduction of bunt frequency as a result of increasing dose of milk powder follows a non-symmetric sigma-curve, while a side effect on germination vigour are proportional with the dose applied. In the study the effect of milk powder is compared with other organic nutrients like 'Tryptic Soi Broth' and flour of wheat (Triticum aestivum), rye (Secale cereale), oil seed rape (Brassica napus), corn cockle (Agrostemma githago), quinoa (Chenopodium quinoa), maize (Zea mays) and mustard (Sinapis alba). The effect seems to be unaffected by inoculum density and to whether the source of inoculum is via soil or seed.

The maximum reduction of bunt frequency was over 95% in most experiments, but failed to do so in one experiment with a maximum of only 70%.

Milk powder seed treatment can contribute the regulation of common bunt in organic agriculture where other cheaper seed treatment are banned. In order to reduce the risk of loss of germination vigour it is recommended to use milk powder in a low dose in combination with other control measures, rather than using milk powder in a full dose as the only means of control.



Introduction

Common bunt caused by the fungus Tilletia tritici (syn. T.caries) is one of the most potentially devastating seed borne diseases in cereal production, and the main reason for the use of fungicide seed treatments in wheat (Nielsen et al. 1998). In organic agriculture, where use of almost all synthetic pesticides is prohibited, common bunt is a very serious disease (Piorr 1991). In order to reduce or avoid the use of synthetic pesticides, attempts have been made to develop alternative seed treatments against common bunt based on seed pelleting with naturally occurring carbon based compounds like wheat flour or milk powder (Nordin 1991, Becker 1992, Becker and Weltzien 1993, Plakholm 1993, Heyden 1993, 1997, Borgen et al. 1995, Tränkner 1993, 1996, ICARDA 1996, 1997, Winter et al. 1997, Nielsen 1999). These compounds serve as nutrient source for microorganisms in the soil or the seed surface, which can have antagonistic or competitive effect on the development of the pathogen (Becker and Weltzien 1993). Even though milk powder is not as effective as the most effective pesticides in prevention of bunt infection, the effect in most studies is in general >90%. However, seed treatment with milk powder may decrease seed germination rate (Winter et al. 1997).

None of the studies on treatments against common bunt based on naturally occurring organic compounds have in detail studied the effect of different doses, and the effect on germination has not been studied in relation to seed vigour, but only to the percentage of germination in lab under optimal germination conditions, or as field emergence (Becker and Weltzien 1993, Winter et al. 1997). The description of the selectivity of the treatment on the published studies is therefore inadequate to evaluate the optimal dose for practical use.

Seed borne inoculum is the main source for infections by T.tritici, but infections can also be caused by inoculum from the soil (Borgen 2000). Studies of seed treatments with synthetic pesticides have shown that many pesticides have a limited effect on soil borne inoculum even if the effect on seed borne inoculum is good (Nielsen and Nielsen 1994, Wainwright and Morris 1989). Whether this is also the case for milk powder as a seed treatment is unknown.

We aim in the present study to investigate the correlation between the dose of milk powder, its effect on common bunt and its effect on seed germination vigour, and factors which might affect this correlation. We also compare milk powder with other compounds in order to evaluate the potential of these methods to control common bunt.



Materials and methods

The experiments with seed treatments can be grouped into several parts:

1) Dose-response studies with milk powder,

2) Comparison with other compounds

3) Interaction between sowing time and effect of milk powder,

4) Interaction between inoculum density and effect of milk powder,

5) Interaction between type of milk powder and its effect, and

6) effect on soil borne inoculum.

1) Dose-response experiments

The effect of different doses of milk powder was tested by applying increasing doses of a milk powder and water mixture to spore contaminated seeds. Milk powder was mixed with inert water in a ratio of 1:1,5. This thick fluid liquid was applied to 100 g seed in a spinning while seed dresser (Hege 11). After treatment seeds were dried in the open air at room temperature, and where more than 2 ml water was added, the treatment was divided into a series of treatments with each application of no more than 2 ml water. The seeds were dried between each application in a cold air stream. After treatment the seeds were stored at 5 Cº. Samples were taken out for field test, which took place 2-6 days after seed treatment. Germination tests were conducted 1-3 month later.

Germination tests were done as a cold sand test in plastic plates containing 1.5 kg sand with water (65ml H2O/kg quartz sand). Sowing depth was 1.5 cm and temperature was 10 Cº. The emergent number of seedlings were counted every day days after first emergence. The emergence was counted the following 6 days in 1996, 3 days in 1997 and 5 days in 1998. The number of replicates were one in 1996, 3 in 1997 and in 1998.

In the experiment 1995-6 whole milk powder was used, in 96-7 skimmed milk powder was used and the same probe of skimmed milk powder was used in 1997-8 after storing in a closed box at 5 Cº. The doses and results are presented in Figure 2 for the effect on bunt frequence and the effect on germination is presented in Figure 3.

In 1996 the cultivar 'Yacht' and in 1997 and 1998 'Pepital' was used. Before treatment the seeds were contaminated with 5 g spores per kg seeds, which in the three years resulted in a contamination between 1.7 to 2.0 mill spores per gram seeds when tested by the ISDA haemocytometer method (Keitreiber 1984). The cultivar 'Pepital' belong to the group of varieties with very high susceptibility to common bunt in Denmark, while 'Yacht' belongs to the intermediate group with some degree of undescribed resistance (Nielsen 1999).

Field trials were at Højbakkegård, an experimental farm of the Royal Veterinary and Agricultural University located 18 km east of Copenhagen, Denmark (55º40' N, 12º18' E, 28 m above mean sea level). The soil type is moreanic sandy loam, and common bunt had not been recorded in the trial areas beforehand.

In the Field trials, treatments were sown in rows of each 1.25 m in 10 replicates. Total plant number in these trials was 2000 on average in each treatment. After heading the total number of ears and the number of diseased ears were counted based on visible macro-symptoms. The experiment is presented in Figure 2-4.

2) Comparison with different organic compounds

In the year 1996-7 other compounds were included in the experiment for comparison with milk powder. The other compounds used were: Wheat flour (Triticum aestivum), rye flour (Secale cereale), oil seed rape flour (Brassica napus), corn cockle flour (Agrostemma githago), quinoa flour (Chenopodium quinoa), maize flour (Zea mays) and mustard flour (Sinapis alba). These compounds were used as whole seed flours except mustard and rape which were used without their seed coat. Also the refined growth media 'Tryptic Soi Broth' was tested. All compounds were added in a dose of 30 g per kg seed with water as the only binding material. Seeds treated with Tryptic Soi Broth were extremely sticky after application, and it was not possible to use a regular sowing machine for this product. The seeds in this treatment were therefore sown by hand instead. All other treated and not treated seeds were sown with sowing machine. All other experimental design was as described for the dose-response experiment. The experiment is presented in Table 1.

In a separate experiment in the season 1994-5 the effect of three different types of milk powder tested. A seed lot of winter wheat (Triticum aestivum) cultivar 'Kosack' was contaminated with spores of T. tritici at a dose of 5 g per kg of seed, which resulted in a spore contamination of 2.291.000 spores/g when tested by the ISDA haemocytometer method (Keitreiber 1984). Spores was collected from the same cultivar the same year. Butter milk powder, skimmed milk powder and whole milk powder was compared by dissolving the powder in water (1:1.5). The seeds was treated with this thick fluid milk solution in a spinning wheel seed dresser at a dose of 20 ml/kg.

Treated seeds were sown with a regular plot-sowing machine in plots of 1.25 x 4 m covering 10 rows. Plots were sown in 4 randomized replicates, and the two outer rows of each plot was diagnosed for bunt infection after heading by counting the total number of ears and the number of diseased ears based on visible macro-symptoms. The total number of diagnosed ears in these trials was 2000 on average in each treatment. Germination tests were done in 3 replicates as a cold sand test as described above. The emergent number of seedlings were counted every day for 6 days after first emergence.

The experiment is presented in Table 2

3) Interaction between sowing time and effect of milk powder

The influence of sowing time on the effectiveness of milk powder was investigated the year 1994-5. Field trial and germination test were conducted as described in the experiment testing the effect of different types of milk powder as described above.

Butter-milk powder was applied to the seeds in the year 1995 by applying 5 ml/kg inert water in a spinning-wheel seed-dresser (Hege 11) followed by application of as much milk powder as could adhere to the wet seeds. This treatment was repeated until 130 g milk powder had been applied per kg seed. The three sowing times were in middle of October, end of October and middle of November, which is quit late for Danish conditions. Seed treatment was conducted one week before each sowing. The experiment is presented in Table 3.

4) Interaction between inoculum density and effect of milk powder

The influence of inoculum density on the effect of milk powder against common bunt was tested in the season 1993-4. Seeds of the cultivar 'Kosack' was contaminated with either the high dose of 5 g bunt spores per kg seed or with a low dose of 0,5 g bunt spores per kg seed. A medium dose was made by dipping seeds of the high contamination in water in order to wash of spores. Dry seeds were then applied 70 ml water per kg, and the seeds were put into a box with instant skimmed milk powder and shaken. By this treatment about 50-100 g milk powder adhered to the wet seeds per kg.

Treated seeds were sown with a regular plot-sowing machine in plots of 1.25 x 4 m covering 10 rows. Plots were sown in randomized blocs with 4 replicates, and half of each plot was diagnosed for bunt infection after heading as described above. The total number of diagnosed ears in these trials was 4223 on average in each treatment. The experiment is presented in Table 3.

5) Effect of milk powder on common bunt from soil borne inoculum

The experimental area for this experiment had the previous year been grown with the cultivar 'Kosack' which was contaminated with bunt spores and resulted in an infection level of 156 infected plants per m2. This crop was harvested with a normal combine harvester leaving chopped straw and spores on the ground. After harvest the area was left for one month where it was ploughed and sown with milk powder treated and un-treated seeds.

Seed of the cultivar 'Husar' were treated with 50 gram skimmed milk powder per kg seed as described for the test with other compounds. The seeds was found free from spore contamination when tested by the ISDA haemocytometer method (Keitreiber 1984). Treated and not treated seeds were sown in four replicates in blocs with systematical placing. Each plot had an average of 949 plants and all plants in all plots were diagnosed for bunt infection after heading. The experiment is presented in Table 5.

Statistical analysis

Data for both the disease frequencies and germination tests are binomially distributed. For this reason, and in order to make the best of the data material, the data was analysed by a Generalized Linear Model (PROC GENMOD in the software SAS ver.6.12) or by a Generalized Linear Mixed Model (the GLIMMIX macro in the software SAS ver.6.12) in cases where random effects occured. Multiple comparison was done by the LSMEANS /pdiff procedure in GLIMMIX or CONTRAST statement in GENMOD. Presented results are invers logit to the estimated effects. Mean Germination Time (MGTtotal) is defined as the time for 50% germination of the sown seeds in the cold sand test, and is calculated from the parameter estimates generated by the model.



Results

Both the effect on bunt infection and the side effect on seed germination vigour were highly influenced by the dose of milk powder applied presented in Figure 1, 2 and 3. The negative side effect on germination vigour was proportional to the amount of milk powder applied, while the bunt infection follows a non symmetric sigma curve. The selectivity of the treatments are affected both of the year and of the dose applied. In 1996, the doses of 44 g/kg and higher significantly increased Mean Germination Time of the seeds (p<0.05) while in 1997 and 1998 only the doses of 80 g/kg and higher significantly increased Mean Germination Time(p<0.05).

The different tested compounds had different effects on bunt frequency and germination vigour as presented in Table 1. Mustard flour and Tryptic Soi Broth had the largest effect bunt infection, but at the same time the application resulted in a significant increase in Mean Germination Time. Maize flour had no effect on neither bunt infection or Mean Germination Time. Flour of rye, oil seed rape and wheat was less effective than milk powder while flour of corn cockle and quinoa could not be distinguished from milk powder in the effect on bunt infection.



Table 1: Frequency of common bunt as affected by seed treatment with 3% (by weight) of different compounds. Disease rate tested in field trials; germination speed tested in lab.
Treatment % infected plants 95% confidence interval
Control 27,09 25,97 28,24
Not contaminated 0,67 0,19 2,33
Milk powder 6,98 4,85 9,97
Wheat flour (Triticum aestivum) 1495 11,92 18,59
Maize flour (Zea mays) 27,88 25,02 30,92
Mustard flour (Sinapis alba) 0,38 0,06 2,22
Tryptic Soi Broth 2,45 1,18 5,05
Corn cockle flour (Agrostemma githago) 9,03 6,24 12,90
Rye flour (Secale cereale) 12,01 857 1659
Quinoa flour (Chenopodium quinoa) 5,78 3,56 9,27






Three types of milk powder were tested, as presented in Table 2. The differences were not significant at the dose tested.



Table 2: Comparison of three different types of milk powder.95% confidence interval in brackets. Numbers with different letters are statistically different (p<0.05). Germination vigour was not statistically affected by the treatments.
Bunt frequency %
Control 52.89 (50.8-55.0) A
Water 54.28 (48.9-59.5)A
Skimmed milk powder 48.64 (43.5-53.9)BA
Butter milk powder 43.17 (38.1-48.5)B
Whole milk powder 44.97 (39.9-50.2)B






Application of milk powder to the seeds had a high effect on the bunt infection both at high and low level of spore contamination and at different the sowing times, as presented in Table 3 and 4. The efficacy at the latest sowing time in middle of November is lower than in the end of October as were the frequency of bunt infections. However, the bunt frequency in the treated plots was low at all sowing times, and it is not possible to conclude whether the differences is a result of environmental conditions or other reasons.





Table 3. Field trail with effectiveness of milk powder as affected by initial contamination level. 95% confidence interval in brackets.
Inoculum density level bunt frequency untreated seeds bunt frequency seeds treated with milk powder Percent reduction
High 40.19% (36.9-43.6) 2.08% (1.3-3.4) 94,5 %
medium 7.57% (5.9-9.6) 0.21% (0.05-0.9) 97,1 %
Low 1.37% (0.8-2.5) 0.02% (0.0-2.0) 98.6%














Table 4: Comparison of the effect of milk powder at three different sowing times. Field trail conducted 1994-5. The effects was significant of both sowing time (p=0.0345) and treatment (p<0.0001) and the interaction between treatment and sowing time (p<0.0001) . Milk powder treatment increased the time for 50% germination from 7.18 to 8.01 days (p<0.0001)
Sowing time Bunt frequency without milk powder Bunt frequency with milk powder reduction
Mean 95% confidence interval Mean 95% confidence interval
Beginning October 52.89 (50.82 - 54.96) 2.17 (1.03 - 4.51) 95.9%
End October 35.89 (31.09 - 40.99) 0.91 (0.29 - 2.83) 97.5%
Beginning November 25.45 (21.17 - 30.25) 3.87 (2.22 - 6.68) 84.7%




Milk powder treatment in the dose of 50 g/kg of uncontaminated seeds reduced the infection of common bunt by 92.6% when caused by soil-borne inoculum (Table 5).



Table 5: Effect of seed treatment with milk powder on common bunt coursed by soil borne inoculum. Field trail conducted 1996-7. 95% confidence interval in brackets.
Frequency of common bunt (95% confidence interval)
Control 0.71 % (0.49-1.04)
Skimmed milk powder 0.06 % (0.02-0.22)




Discussion

The experiments have shown that the development of common bunt is highly influenced by the dose of milk powder used for seed treatment prior to sowing. To achieve a maximum control with milk powder a dose of 70-80 g/kg milk powder should be applied, but at this dose the reduction in germination vigour is significant. Winter et al. (1997) found reductions between 86% to 100% in 9 experiments with milk powder with the dose of 80 g/kg. In 5 of these experiments two doses were tested, and they did not find increasing effect on bunt infection when increasing the dose from 80 g/kg to 160 g/kg, but the side effect on germination was significant at both doses, highest at the highest dose. Becker and Weltzin (1993) found increasing effect on bunt frequency from 94,1% at 80 g/kg to 98,3% at 160 g/kg without any reduction in field emergence. The experiment shows that the increase in Mean Germination Time is proportional to the dose applied (Figure 2). Nielsen (1998) reduced bunt frequency by 75% at a dose of 80 g/kg with an insignificant tendency to reduced field emergence.

The selectivity of the treatment with milk powder is decreasing with increasing dose (Figure 3), a known phenomenon for many pesticides. This indicate that the dose of milk powder used for control of common bunt should be reduced as much as possible.

From the different compounds tested in the limited dose of 30 g per kg seed, mustard flour had the largest effect on bunt and was significantly higher than most other compounds at the same dose, as found by Spie and Dutschke (1991) who were also able to control bunt with mustard flour. The effect of mustard flour compared with other compounds may be related to the metabolic chemicals in the mustard flour. Also Tryptic Soi Broth gives a large decrease in bunt frequency, but also here combined with a significant increase in Mean Germination Time compared with the less effective compounds.

Flour of corn cockle reduced the bunt frequency by 66.5% compared with control which is close to the effect of milk powder at the same dose. Heyden (1997) showed a reduction of 53,7% with a similar design with corn cockle. Also flour of quinoa had an effect of the same level. Both quinoa and corn cockle contains saponins, which influences the tension of the water surface. This fact may influence the rate of water uptake of both spores and seeds and thereby the infection.

The differences between the effects of the different compounds tested can be explained both physically and chemically. The compounds have different solubility in water and cover the grains with different uniformity. The selectivity of the different compounds differ, but it is not possible to conclude whether the differences are effects of the compounds them selves or of the doses of the compound, since the selectivity are affected by the effectivity of the treatments. We have chosen to study milk powder in further detail because of the relatively good effect and good water solubility and binding ability (Becker 1992, Tränkner 1993, ICARDA 1997).

Tränkner (1996) write that the efficiency of milk powder is highest at low contamination levels. This effect was investigated in our experiment as well. As presented in Table 3 we found a tendency of increased efficiency, but this effect was not significant in our experiment.

There was a significant difference in the effectiveness of milk powder between the years. In 1998 the maximum control of bunt was much lower than in the other years and not adequate for commercial cereal production in cases of high contamination of bunt spores in the seed lot. Also Winter et al. 1994 had one experiment (out of 9 conducted) with a reduction of only 86%. The reason for these unstable effectivity was investigated in the other experiments.

The infection level was very high in 1998, but based on the result of the experiment with contamination levels it is not likely that the differences in infection levels can account for the whole difference in efficiency of the milk powder. In 1998 an uncontaminated control was not included, and it can therefor not be excluded that soil infection could occur in the area. Bunt frequency caused by naturally soil infection is normally low (Borgen 2000), and it is not likely that bunt infection from soil borne inoculum exclusively can account for the limited effect in this year. Moreover, the experiment with soil borne inoculum indicate that milk powder also has an effect on soil infections and Nielsen (1998) reduced bunt frequency by 94% at a dose of 80 g/kg. Table 4 indicate that the effect of milk powder on seed borne inoculum can be influenced by environmental conditions. It is most likely that this is also the case for soil borne inoculum. We assume therefore that changing environmental conditions between the years causing the very high infection level in 1998, and maybe in combination with soil borne inoculum in the area is the main reason for the differences between the years.

The mechanism of interaction between the tested compounds and T.tritici is not studied in this investigation. Tubeuf (1902) and Hahne (1925) were the first to show that sugar reduced the germination of bunt spores. Rabien (1928) found that this effect could be explained by the decreased pressure of oxygen likely to be caused by the growth of different microorganisms in a sugar solution. It is likely that the reduction in bunt frequency and the reduction in germination vigour can be explained by a reduced availability of oxygen, since milk powder has a high content of sugar and other nutrients favourable for growth of various microorganisms. Becker and Weltzien (1993) have shown an increase in the growth of different bacteria (Aerobic spore forming bacteria, Pseudomonas ssp, Bacillus ssp) in milk powder, and the same bacteria also had a reducing effect on bunt spore germination in vitro.

Seed treatment with synthetic pesticides is still possible in conventional agriculture. In most situations these pesticides are more effective and cheaper than milk powder treatments. However, increasing environmental taxes are put on pesticides in many countries and price relations may therefore be different in the future. Organic farmers and consumers reject the use of the most effective synthetic pesticides against common bunt and seek alternatives in the control against this and other seed borne diseases. Knowledge about alternatives to pesticides are therefor relevant and important even if their commercial development is not favourable for conventional farming for the time being.

As the seed germination vigour decreases proportionally with the amount of milk powder applied and the bunt infection decreases rapidly at low doses of milk powder, the selectivity of milk powder is high and a relatively good control of the disease can be obtained with a limited negative side effect. On the other hand, in a highly contaminated seed lot it is necessary with a high dose of milk powder to give adequate control. At the high dose of milk powder the selectivity of the treatment is low, and the treatment is likely to result in an unacceptable decrease of seed germination vigour. If the explanation for the effect of milk powder as argued is a competition for oxygen due to the development of the microbial flora on the seed surface, it is not likely to be possible to improve the selectivity of the treatment, if the control exclusively shall relay on milk powder in combination with naturally occurring microorganisms.

Combining the milk powder treatment with an active biological control agent improves the effect of the milk powder in the control of common bunt as compared with milk powder or biocontrol when used alone, and this can be achieved without increasing the negative side effects on germination vigour (Becker and Weltzien 1993, Borgen and Davanlou 2000). Tränkner (1993) concluded that milk powder should be used alone, and recommended the use of higher doses of milk powder rather than to combine milk powder with biological control agents. Based on the coherence between selectivity and dose we conclude that the dose of milk powder should be reduced as much as possible. The potential future for milk powder in the control of common bunt we believe, lies in an integrated control strategy, by combining limited doses of milk powder with active biocontrol and, where possible, resistance and other measures.



Acknowledgment

We thank Hanne Bay Christensen and Susanne Olsen for using most of their summers in the experimental fields diagnosing bunted wheat plants. We thank Kristian Kristensen for statistical support and our tutors for feed back in the discussions.















Cited Literature

Becker, J. 1992:Untersuchungen zur bekämpfung des Weizensteinbrandes (Tilletia tritici (Bjerk.) Wint.) mit nährstoffreichen organischen Substanzen und Mikroorganismen. Dissertation. Institut für Pflanzenkrankheiten der Rheinischen-Friedrich-Wilhelms-Universität, Bonn. 132 pp.

Becker, J and 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. 170(2):182-188

Borgen, A, L.Kristensen and P. Kølster 1995: Bekæmpelse af hvedens stinkbrand uden brug af pesticider. Control of common bunt without use of pesticides. Proceedings from the 12th Danish Plant Protection Conference, SP-report 4:149-158.

Borgen, A. og M. Davanlou 2000: Biological control of common bunt in organic agriculture. Journal of Crop Production 3(5):159-174. Also printed in: "Nature Farming and Microbial Applications", Ed: H.L. Xu , J. F. Parr og H. Umemura. Haworth Press Inc., New York. ISBN: 1-56022-082-1 / ISBN:1-56022-083-X , page 159-174.

Bosch, H.-T. 1993: Bekämpfung des Steinbrands an Weizen (Tilletia caries und Tilletia controversa) im biologish-dynamischen Landbau. Diplomarbeit, Fachhochschule Nürtingen. 50 sider.

Heyden, B 1993: Zum Plizproblem. Mitteilungen aud der Arbeit J. und C.Graf Keyserlingk-Institut. 9:5-32.

Heyden, B 1997: Unsere Versuche in der Vegationsperiode 96/97. Mitteilungen aud der Arbeit J. und C.Graf Keyserlingk-Institut. 12:12-25.

ICARDA 1996: ICARDA Annual Report 1996

ICARDA 1997: ICARDA Annual Report 1997

Kendrick,E.L. & L.H.Purdy 1959: A seedling reaction of wheat indicative of bunt infection. Phytopathology 49:130-132.

Kietreiber, M. (1984). ISTA handbook of seed health testing. Working sheet no. 53. Bundesanstalt für pflanzenbau, Wien, Austria.

Nielsen, B.J. and G.C. Nielsen 1994: Stinkbrand og jordsmitte. Soil infections of common bunt. Proceedings from the 11th Danish Plant Protection Conference, 1994. Pests and diseases p. 89-103.

Nielsen, B.J., A.Borgen, G.C.Nielsen and C.Scheel 1998: Strategies for controlling seed-borne diseases in cereals and possibilities for reducing fungicide seed treatments. The Brighton conference - Pest and Diseases, p.893-900.

Nielsen, B.J., 1998: Frøbårne sygdomme i korn. I: Pesticidafprøvninger 1997. Markbrug, DJF Rapport 1:59-66.

Nielsen, B., S. Christensen and J.O.Bagge 1999: Ny resistens mod frøbårne sygdomme i korn. New resistance against seed borne diseases in cereals. Proceedings from the 14th Danish Plant Protection Conference, DJF-report 10:149-160.

Nordin, K. 1982: Några forsök avseende möjligheten för vanligt stinksot Tillitia caries att angripa vårhvete. Sveriges Landbruksuniversitet, Institutionen för växtskydd. Examensarbeten. 23 p.

Piorr,H.P. 1991: Bedeutung und Kontrolle saatgutübertragbarer Schaderreger an Winterweizen im Organischen Landbau. Ph.D. thesis, Bonn 1991. pp 166.

Plakholm, G, 1993: Zweijährige Überprüfung alternativer Beizverfahren gegen Weizensteinbrand (Tilletia caries [D.C.] Tul.). I: Zerger, U(udgiver): Forschung im ökologischen Landbau. SÖL-Sonderausgabe Nr.42, Verlag C.F.Müller, Bad Dürkheim.

Spie, H and J. Dutschke 1991: Bekämpfung des Weizensteinbrandes (Tilletia caries) im biologisch-dynamischen Landbau unter experimentellen und praktischen Bedingungen. Gesunde Pflanzen, 43(8):264-270.

Tränkner, A. 1993: Kompost und Pflanzengesundheit: Möglichkeiten und Auswirkungen der biologischen Beeinflussung pflanzlicher Oberflächen zur Krankheitsbekämpfung. Ph.D. Bonn Univ. Verlag Dr.Kova, Hamburg.

Tränkner, A. 1996: Biologische Weizensteinbrandbekämpfung - mehrjährige prakstsche Erfahrungen mit der Milchpulverbehandlung. Mitteilungen aus der Biologischen Bundesanstalt für Land- und Forstwirtschaft Berlin-Dahlem. 321:417.

Wainwright, A. and D.B.Morris 1989: Evaluation of triadimenol seed treatments against soil-borne bunt of wheat. Tests of Agrochemicals and Cultivars 10:68-9 (Supplement to Annals of Applied Biology no. 114).

Winter, W, C.Rogger, I. Bänziger, H.Krebs, A.Rüegger, P.Frei, D.Gindrat and L.Tamm 1997: Weizenstinkbrand: Bekämpfung mit Magermilchpulver. Agrarforschung 4:153-156.