Occurrence of clubroot and Plasmodiophora brassicae Wor. races in Finland

. Examination of clubroot in cruciferous vegetables in Finland in 1974—1978 revealed the disease in 81 % of the 101 communes inspected. The disease was most common in southern and central Finland, but was also discovered in the northern parts of the region in which cruciferous crops were cultivated (66 —67° N. lat.). Clubroot was found in 65 % of the 375 plant samples collected. It occurred in 68 % of the samples of the most commonly cultivated vegetable, cabbage (56 % of the material), in 63 °Jo of the cauliflower samples (22 % of the material), in 56 % of the samples of other cole species (13 % of the material) and in 64 % of the samples of cruciferous root crops (10 % of the material). P. brassicae race determinations were performed on 90 samples. The classification system of Williams (1966) was applied. The races that were isolated were 1,2, 3,4, 6 and 7. Race 2 was by far the most common, being found in 32 communes; races 3,4, 6 and 7 were each found in 9 —12 communes; race 1 only in one commune. No clear differences in the occurrence of the races in the various parts of the country could be observed. A comparison is made between Williams’ and the ECD (Buczacki et al. 1975) classification systems. In addition, the pathotypes in clubroot material from Norway and Iceland were determined.


Introduction
The records on the occurrence of clubroot disease in Finland date back to the 1860s (Jamalainen 1936). It is assumed that the disease came to Finland from the east, from Russia (Rainio 1930). In his studies on the cause of crucifer hernia, Woronin (1878) (Rainio 1930, Jamalainen 1936. In the 1930 s the susceptibility of cruciferous vegetable cultivars and wild Brassicaceae plants to the fungus Plasmodiophora brassicae Wor. was also studied at the Research Centre. In addition, experiments for chemical control of the disease were carried out (Jamalainen 1936). The efficiency of new pesticides was studied in 1944(Linnasalmi 1948, 1952, 1959Linnasalmi and Tiittanen 1960;Murtomaa and Uoti 1972; annual reports of the Department of Plant Pathology of the Agricultural Research Centre 1960Centre -1977. A preliminary study on P. brassicae races in Finland had been carried out in the years 1971-1972 (Linnasalmi and Palonen 1974).
In this study, the occurrence of clubroot and the pathotypes of P. brassicae in cruciferous vegetables in Finland in 1974-1978 were examined. In addition, the pathotypes in clubroot material sampled in Norway and Iceland were determined.
The study is a part of the Nordic clubroot project, NKJ project 27 1974NKJ project 27 -1977, of the Nordic Joint Committee for Agricultural Research: Breeding for clubroot resistance, Plasmodiophora brassicae Wor. races, and the efficiency of new pesticides.

Collection of the basic material
Most of the material for studying the distribution of clubroot was collected by inspecting cruciferous vegetable fields throughout Finland during the growing season, from June to October. A small part of the material, about 2 %, was sent in by agricultural research stations, advisers of agricultural information organizations and farmers. The inspections chiefly focused on large cole and root crop farms, but a considerable number of small farmers' crucifer fields were also inspected. The research material collected as described above can be assumed to give a general picture of the distribution of clubroot disease and the P. brassicae races in the period under study in the 1970 s in Finland.
Samples of galls from the various cultivars affected with clubroot were taken for the analysis of P. brassicae races. Propagation of P. brassicae club materialand preparation of the inoculum The young galls of the original samples were washed thoroughly with running water. They were crushed to prepare a water suspension, which was used to inoculate a steamed (at 100°C for 1 h) soil-peat medium. From the samples for race analysis additional club material was grown on the original plant species and, if possible, on the original cultivar or on a cultivar known to be susceptible to P. brassicae, such as cabbage cv. Ditmarsk and cauliflower cv. Erfurter. To check and ensure the viability of the inocula, the black mustard (Brassica nigra (L.) W.D.J. Koch) breeding line Sv. 72-6842 of the Swedish Seed Association, Sweden, which was very susceptible to clubroot, was also used as a host plant both during the. propagation of club materials and later in identification tests of P. brassicae races.
During the initial stage of the study, the inoculation of the host was repeated two or three times as five-week growth periods. As the sampling technique improved, the club material from the first propagation could be used as basic material to grow callus cultures and small-club material. The material was stored at -lB°C.
The callus cultures were prepared from young galls, I -2 1 -2 cm long, which were surface-disinfected with ethanol (C2H S OH, 94 %) and mercuric chloride (HgCl 2 , 1 %) and rinsed with distilled water. Small pieces, ca. 2 mm in diameter, were cut from the inside of the galls, treated with ethanol (99.5 %) for a few seconds and rinsed with distilled water. The pieces were placed in Erlenmeyer flasks or test tubes on an agar medium. The nutrient solution used was a modification of the solution of Murashige and Skoog (1962), in which 3-indoleacetic acid had been replaced Isolation and identification of races with 1-naphthylacetic acid. Coconut milk (100 ml/1) was added because our tests had shown that it increased the growth rate of callus tissue considerably. The cultures were incubated in the dark at ca. 22°C. Small pieces of callus tissue were transferred every 7 to 10 days to a fresh culture medium, the passages totalling three to four. The callus material was stored at -5°C. Small-club inoculum material was made direct from young fresh or frozen ( -lB°C) galls, which were washed with running water and rinsed with distilled water.
For preparing the inoculum, the callus and small-club materials were crushed mechanically, suspended in distilled water and filtered through a nylon filter cloth. The filtrate was centrifuged (2400 g, 7 min.) three to four times. Before each test, the suspension used was diluted to a concentration of 10 8 spores per ml, using a hemocytometer.

Method
The isolation and identification of races was carried out according to the system of Williams (1966), a method based on selecting four Brassicaceae plants as differential hosts and studying their resistance or susceptibility to P. brassicae.
The use of genetically uniform, homozygous differential host material is a prerequisite for the reliability of the race determinations. The seed material used in this study was Possible host reactions to infection by races of Plasmodiophora brassicae. for 1 h. Seeds of differential hosts were sown in 9 x 9 cm pots of thermosetting plastic, 10 seeds/pot, four replicates. Each pot had a plastic tray of its own. Inoculation with the spore suspension, 10ml/pot, was performed after the sowing and the seeds were covered with a 0.5 cm layer of the growing medium. Throughout the growing period, the plants were watered by pouring water into the trays to keep the moisture even and to prevent the possibility of cross-contamination. The growing period was from four to six weeks. The galls were well developed, the average size being 0.5 to 1.0 cm 3 . They were light in colour and usually contained a great number of spores.

Differential
To conclude the test, the roots were washed with running water and rinsed with distilled water. The degree of clubroot infection of each plant was assessed on a scale of o-3, 0 -3, and the clubroot index was calculated according to Williams (1966): clubroot index lQxn,+ 60xn 2+ 100xn 3 n o +n, + n 2 +n 3 n 0 = no clubs n, = a few small clubs on the secondary roots n 2 = considerable clubbing on the lateral roots n 3 = severe clubs on the primary and secondary roots This was the primary analysis. The clubs from the replicates of each differential host were combined to make a spore suspension, which was used to re-inoculate all four differential host plants, in accordance with the following scheme: If the result of this cross-testing was similar to that of the primary analysis with respect to both degree of infection and clubroot index, the race determination could be considered to be reliable. If the result was not similar, the cross-testing was repeated three to four times. 2 Comparison with the ECD classification system Comparison with the ECD classification system, developed by the International Clubroot Working Group (Buczacki et al. 1975) 2 Purified and identified race material was sent to the Scandinavian members of thisproject from 1974 onwards to be used in their resistance breeding work. Since 1980 the type isolate material has been deposited in the race bank at the Swedish University of Agricultural Sciences, Department of Resistance Biology, Alnarp, Sweden. was performed with some of our P. brassicae isolates (totalling 12) classified by Williams' 1). Thus the study revealed that clubroot was quite common all over Finland. Although the locations inspected were distributed over the country, the number of samples from the various areas differed so much that it is impossible to draw far-reaching conclusions about regional differences in the occurrence of the disease. It can, however, be noted that clubroot was fairly common in the old, densely populated farming areas, Uusimaa (N), Varsinais-Suomi (Ab) and Etela-Hame (Ta), and also in Etela-Savo (Sa) and Etela-Karjala (Ka). The disease was more severe in eastern Finland than in western Finland. This was possibly partly caused by cultivation techniques, mainly by the fewer opportunities for crop rotation on the farms in eastern Finland.
In these areas many farmers had to give up the production of cruciferous vegetables altogether because of clubroot disease in the 1960 s and '7os.
It also turned out that clubroot was often more common, and even more severe, on smaller than on larger farms, one probable reason being the limited opportunities for crop rotation. This was evident, for example, in gardens around population centres, wherevar-   Heikinheimo and Raatikainen (1971).
ious cole species and cruciferous root crops had been grown in the same fields for many years.

Clubroot in different plant species
Cruciferous vegetables are cultivated almost throughout Finland up to 66-67°N. lat., but the main production takes place in the southern part of the country. In the 1980 s, the areas of cabbage and cauliflower remained roughly the same, totalling some 1000 ha. The area of swede decreased markedly, from almost 2000 to 400 ha. The cultivation of Chinese cabbage increased to 600 ha. The total area of the other cole species was some 80 ha and that of radish and turnip totalled some 40 ha. The data on the occurrence of clubroot in differentplant species are presented in Table 2. Clubroot was found in 68 % of the cabbage samples, in 63 % of the cauliflower samples, in 56 % of the samples of other cole species, and in 64°7o of the root crop samples. The distribution and number of samples from different plant species roughly indicate the frequency of cultivation of these crops.
Plasmodiophora brassicae races and their occurrence As in practice it was not possible, within the limits of the study, to carry out a race determination of every clubroot sample, the samples for tests were chosen to represent as many communes as possible in each biological prov-  ince, altogether 69 communes, or 84 % of the communes where clubroot was found. Most determinations were performed on club samples of the crucifers cultivated most frequently: cabbage and cauliflower. Some additional determinations were made on samples taken from less common species. The number of P. brassicae isolates totalled 90.

Races
The races isolated were 1,2, 3,4, 6 and 7 (Appendix 1). The identification results for races 1,2, 4 and 7 were clear. The test results for races 3 and 6 showed some deviation from the classification scheme. Several isolates from both of these races showed slight contamination in Badger Shipper, although according to the scheme this differential host should have been resistant to these races.
As the identification tests with certain isolates of other races also showed more variation in the clubroot indices in Badger Shipper than in other differential hosts, it seems that the seed material of Badger Shipper may have had some genetic heterogeneity. It is also possible that the isolate materials classified as races 3 and 6 included some pathotypes that were variants of the main races. The possibil-ity of mutation in the differential hosts or in the purified P. brassicae race isolates must also be taken into account.
A preliminary report on the occurrence of P. brassicae races determined in the present study was presented at the Brassica conference 1981 (Linnasalmi and Toiviainen 1981).

Regional distribution
The regional distribution of the races is shown in Table 1 and on the map in Fig. 2. The communes from which the samples were taken and data on the original host are given in App. 1. Of the six races isolated, race 2 was by far the most common. It was found in 32 communes, i.e. in 46 % of the communes with clubroot. Races 3,4, 6 and 7 were each found in about ten communes (9 -12 communes, i.e. 14-17 %); race 1 occurred only once. Two different races were found in only a few communes: races 2 and 3 in Nurmijarvi (N) and Joutseno (Sa), races 2 and 6 in Haukivuori (Sb), races 6 and 7 in Hartola (Ta), Kangasala (Ta) and Oulu (Ob). On the basis of this study, no clear prevalence of races can be demonstrated in different parts of Finland. All races were distributed fairly evenly over the country, with the exception of the  , 1974-1978. rare race 1. However, race 4 seems to show some prevalence in the western parts of the country, while race 3 occurs mostly in the east.
In the preliminary studies conducted in 1971 -1972 on the occurrence of the P. brassicae races in Finland (Linnasalmi and Palonen 1974), the races 2,4, 6 and 7 were isolated. In two test places the races that were found in 1972 were, according to the present study, still the same, viz. race 4 in the test field of the Institute of Plant Pathology in Tikkurila (Vantaa) and race 7 in the test field of the Institute of Horticulture in Piikkio. Ten years earlier, in 1964, samples from these localities were sent to Prof. P.H. Williams. He identified race 7 from Piikkio and race 2 from Tikkurila (Williams 1966).
Of the cabbage cultivars grown in Finland, (cf. App. 2) only the Norwegian Resista and Respla (Weisaeth 1977) are partly resistant to some P. hrassicae races. These cultivars were accepted for marketing in Norway in 1973 andin Finland in 1975 In some farms where these cultivars were grown in field sectors situated next to each other, severe contamination and consequently weaker growth of Blatopp were evident, whereas Resista and Respla were only sparsely and slightly contaminated, and headed well.
The occurrence of P. brassicae races determined according to Williams (Table 3). A report on the occurrence of races 1, 7 and 9 in Norway, in the Trondelag region, is given in the publication of Linnasalmi and Weisaeth (1978). On the basis of our new sample material (cf. data given above), it would seem that race 4 is more common than the others elsewhere in Norway. There were some differences between the Finnish and Norwegian race spectra: race 9 was not found in Finland, and races 3 and 6, which are fairly common in Finland, were not found in the Norwegian material. However, data on races 3, 5 and 6, as well as races 1 and 9, are presented in some earlier studies on the Norwegian race spectrum (Weisaeth 1972). Two cabbage samples (Weisaeth's breeding lines) from Iceland were analysed. Only race 7 was found in both samples (Linnasalmi and Weisaeth 1978). Sweden in connection with breeding work for clubroot resistance in cruciferous oil crops, R. JOnsson investigated the occurrence of P.
brassicae races in Skane, in southern Sweden.
Using the method of Williams, JOnsson (1971 concluded that in the populations collected from various localities there was the possibility of occurrence of several P. brassicae races, and definite occurrence of race 15 in one case. Using the callus technique, JOnsson isolated races 1,2, 3,4, 6 and 7 (JOnsson 1981).

Comparison between the Williams' and ECD classification systems
In this study with the P. brassicae race ma- terial classified according to Williams' sys-been reported from Canada by Chiang and tem (1966), a comparative test series was carried out using the ECD classification system (Buczacki et al. 1975). Pure isolates of races 1,2, 3,4, 6 and 7 (12 in total) were chosen for the tests. The results are shown in Table 3.
In the Brassica campestris group, theresistance and susceptibility of all differential hosts to the various races are the same, code number 16, with exception of one race 4 isolate, for which the code is 17 because of the susceptibility of differential host 01.
In the Brassica napus group, the differential hosts show more diversity in their reactions to the different races, but there is also variability with regard to isolates of the same race. For example, the codes for race 7 are 00, 02 and 03. Number 03 is also the code of one of the isolates of race 2. For the isolates of race 2, the reactions of the ECD differential hosts are to a large extent borderline cases between susceptibility and resistance, and therefore the code can be either 03 or 19. The differential hosts of the group do not serve to separate races 1 and 4. All are 100 % susceptible, code 31.
In the Brassica oleracea group, some differences were found. The code numbers are 12, 30 and 31. Code 12 applies only to race 1, but code number 30 to both races 3 and 6, and code 31 to races 2, 4 and 7.
In addition to the studies mentioned above, determinations of P. brassicae pathotypes by the ECD system have been reported in the following publications: Toxopeus (1974) and Toxopeus and Jansen (1975) from the Netherlands, Heyn (1981) from the FRG, Jones et al. (1982 a) from the UK, Naiki et al. (1984) from Japan and Lammerink (1986) from New Zealand.
In comparing the advantages and disadvantages of Williams' method with the ECD method for isolating and classifying P. brassicae pathotypes, it can be concluded that Williams' system is a more time-saving method. By the cross-testing technique described previously (p. 418), it was possible to detect the main pathotypes in a large area (approx. 3 000 ha) by means of only four differential hosts. On the other hand, it is evident that with such a restricted set of differential hosts it is not possible to identify all possible pathotypes and pathotype variants with certainty.
The ECD system, with its fifteen differential hosts, is much more laborious and the requirement for growing space is many times greater than with Williams' method. Even more serious is the fact that in our comparative test series, which was carried out with very clear and pure race isolates classified according to Williams' system, the ECD results showed uncertainties and contradictions, as can be seen in Table 3 and in the report of the results.
Soon after introduction of the ECD system, other workers also began to draw attention to the uncertainties of the pathotype determinations obtained by the method. The studies of Tinggal and Webster (1981) showed that P. brassicae populations identified by the ECD method and assumed to be pure isolates in fact contained several pathotypes. Similarly, Dixon et al. (1981) and Jones et al. (1982 a, b) found mixtures of pathotypes in their collections. Toxopeus et al. (1986) mention uncertainties in coding P. brassicae populations by the ECD system. Worth mention is also the critical assessment of the limitations of the ECD method by Crute et al. (1980).

Discussion
Regardless of the classification system employed, one of the most important requirements for thereliable determination of pathotypes is that a genetically uniform inoculum material can be obtained. One way to achieve this is the single spore technique starting from the resting spores of the P. brassicae fungus. Difficulties have been encountered in developing the technique. In the studies of Buczacki (1977), Tinggal and Webster (1981) and Jones et al. (1982 b), with ECD population materials, in which Brassica napus and B. campestris varieties were used as test plants, the results have not been very promising; the infection rate remained low, 20-30 % at best (Tinggal and Webster). Moreover, Tinggal and Webster found that when two single spore isolates from ECD populations were tested further, roughly one half gave a result that corresponded with the original code, whereas four new races differentiated from one of the populations and two new races from the other.
The single spore technique can obviously be improved, but judgingby what is known thus far about the multistage internal life cycle of P. brassicae and its development in the host plant (Ingram and Tommerup 1972, Ingram 1978, and about the microstructure of the fungus as revealed by electron microscopy (Dekhuijzen 1979, Ikegami et al. 1978, Buc-zacki et al. 1979, it may be difficult to achieve a homozygotic fungus material. Theoretically there are many possibilities for different recombinations of differential pathogenicity genes (cf. Crute et al. 1980 with refer-ences, Tinggal and Webster 1981).
The significance of the genetic properties of the host plants used in the P. brassicae infection studies began to receive attention in the 19505. Macfarlane (1955) concluded that in certain cases the heterogeneity of the host plant population can cause variation in the infection results. In the 1960 s Williams (1966), among others, stressed the importance of genetically uniform differential host materials in pathotype determinations. Since our preliminary work on P. brassicae races (Linnasalmi and Palonen 1974), our aim has been to use homozygotic seed material of the differential hosts (cf. p. 417). This criterion was not always taken into account prior to the 1980s. Difficulties have been encountered in the production of homozygotic test plant lines by conventional methods. According to Crute et al. (1980), heterogeneity is apparent in some differential host lines of the B. oleracea group, possible in the B. campestris group and less likely only in the B. napus group, because the species is strongly inbreeding.
However  (Keller and Armstrong 1977, 1979, B. oleracea var. gemmifera (Ockendon 1984) and B. oleracea var. capitata (Chiang et al. 1985). The rapid advance of plant molecular biochemistry and genetics offers new opportunities for studying questions of resistance to diseases. One of the new possibilities already in sight is the application of gene technology to modify the genom of a plant directly as desired. It remains to be seen how long it will take before the prerequisites exist for applying these techniques in the breeding of clubroot resistant cruciferous vegetable cultivars.
Acknowledgements. This study was partly funded by the National Research Council for Agriculture and Forestry of the Academy of Finland, which we acknowledge with gratitude. We express our sincere thanks to the organizations and persons who have given valuable help in our work. We are especially grateful for the skilful technical assistance of Ms. Kirsti Nieminen. Appendix 1. Clubroot (Plasmodiophora brassicae Wor.) races in Finland, 1974Finland, -1978  Scientific nomenclature according to Zander Handworterbuch der Pflanzennamen (Encke et al. 1981