Fusarium species from Marion and Prince Edward Islands: sub-Antarctic

J.P. Rheeder*, P.S. van Wyk 1 and W.F.O. Marasas Research Institute for Nutritional Diseases, South African Medical Research Council, P.O. Box 70, Tygerberg, 7505 Republic of South Africa and 1 Department of Plant Pathology, University of the Orange Free State, Bloemfontein, 9300 Republic of South Africa 1 Present address: Department of Microbiology, University of the Orange Free State, P.O. Box 339, Bloemfontein, 9300 Republic of South Africa

Marion Island (and similarly Prince Edward Island) experiences low temperatures (annual mean 5.1°C), high rainfall (>2 500 mm per annum) and a high incidence of galeforce winds (Schulze 1971). Both islands consist of two distinct lava types , a grey preglacial and a black postglacial eruption. The influence of geology on vegetation is through variation in microclimate and soil drainage, rather than through chemical differences in the lava types (Huntley 1971). A comprehensive review of the the chemical composition of the soils and the plant ecology of Marion Island is given by Smith (1977Smith ( , 1978. Manuring by birds and seals markedly enhances soil microbial populations by adding nutrients and possibly energy sources to the soils (Steyn & Smith 1981 ;Grobler et al. 1987).
Fusarium spp. have been recorded in the subarctic region from tundra and glaciated soils in Alaska (Stoner 1981), and from soils and plant roots in Iceland (Kommedahl et al. 1975). In the subantarctic, however, only brief reference is made to fungi in general on Marion and Prince Edward Islands (Joubert 1971;Steyn & Smith 1981) . Mercantini et al. (1989) studied the distribution of keratinophilic and other fungi on the Antarctic Continent , but found no Fusarium spp. According to Corte & Daglio (1963), Tubaki is the only person to refer to a Fusarium sp. isolated from the Antarctic. To our knowledge, Fusarium spp. have not previously been recorded in the subantarctic region. This paper reports on the first intensive survey of Fusarium spp. on these subantarctic islands. The frequency of Fusarium spp . was determined in plant debris associated with various biotically and nonbiotically influenced soil environments . Isolations were also made from necrotic leaves of Pringlea antiscorbutica R.Br.

Soil sampling sites and procedures
During April-May of 1985 and 1986 soil samples (± 500 g/sample) were collected from Marion ( Figure 1) and Prince Edward (Figure 2) Islands. A description of each sampling site is summarized in Table 1  metres apart, depending on the terrain and vegetation) taken from the top 10 cm of soil and then bulked. Plant residues and rocks were removed from the soil surface before the subsamples were collected with a hand trowel. The samples were placed in polyethelene bags and taken to the station on Marion Island .

Extraction of plant debris from soil
Soil samples were treated as described by Nelson et al. hundred pieces of debris from each soil sample were plated out (10 debris pieces/petri dish) and incubated at 10°C for 7 days and then at 25°C for another 7 days. The fungal colonies that developed were transferred to plates of potato-dextrose agar to which 30 mg dm-3 streptomycin had been added (PDAS). Cultures of Fusarium spp . isolated on PDAS were single-spored, transferred to potato-dextrose agar (PDA) and carnation leaf agar (CLA) (Fisher et al. 1982) and identified according to Nelson et al. (1983) . Representative isolates were deposited in the culture collection of the Medical Research Council (MRC), Tygerberg.
Isolation from Pringlea leaves Leaves of the indigenous plant Pringlea antiscorbutica (Brassicaceae), which had dark necrotic lesions, were collected near the meteorological station. These leaves were dried in a heated (18-25°C) room, sealed in a polyethelene bag and returned to South Africa . Small sections of the necrotic leaves were plated on SF A (25 leaf pieces) and also on water agar (25 leaf pieces) to which 30 mg dm-3 streptomycin had been added. These plates were incubated at 10°C for 7 days and then at 25°C for another 7 days. The colonies that developed were treated similarly to those from the debris plates.

Fusarium species isolated from plant debris
The frequency of isolation of each Fusarium sp. from each plant debris sample is given in Table 2. Only three species of Fusarium were isolated from the islands' soils and are listed with their respective sections as in Nelson et al. (1983): F. merismoides Corda in the section Eupionnotes, F. acuminatum Ell. & Ev. in the section Gibbosum, and F. reticula tum Mont. in the section Discolor. A total of 432 Fusarium isolates were recovered from the 27 p lant debris samples . The most frequently isolated species from soils from both islands was F. merismoides, comprising 73 % of the total isolates from Marion Island and 82% from Prince Edward Island. The species totals for F. acuminatum and F. reticulatum are relatively low for both islands and both were isolated with similar frequencies, F. acuminatum being sliglitly more prevalent-than F. reticulatum .

Isolations from necrotic Pringlea leaves
The only fungus isolated from the Pringlea leaf pieces was F. reticulatum. This species was isolated from 12 of the 50 plated necrotic leaf pieces.

Discussion
All three Fusarium species isolated from plant debris during the present survey are soil-inhabiting fungi that exist either as saprophytes and/or as non-aggressive plant pathogens. F. merismoides [syn. F. episphaeria (Tode) Snyd. & Hans. pro parte] (Nelson et al. 1983) is common in soil and in polluted water or sludge (Booth 1971), but was not recorded during a recent survey of South African soils (Maras as et al. 1988). It was found to be common in Icelandic soils (Kommedahl et al. 1975) . S.-Afr.Tydskr. Plantk., 1990,56(4) (Nelson et al. 1983) occurs worldwide as a soil saprophyte , is closely associated with a variety of plants (Booth 1971) and is known to have a specific plant association with legumes (McMullen & Stack 1984) . This species was most frequently isolated from grass roots in Iceland, but not as frequently from the soil (Kommedahl et al. 1975). F. reticulatum [syn. F. rose urn (Lk.) emend. Snyd. & Hans. pro parte] (Nelson et al. 1983) could also have been present amongst the fungi isolated by Kommedahl et al. (1975) from Icelandic soils and plant roots . F. acuminatum and F. reticulatum were both recorded from South African soils by Marasas et al. (1988).
F. merismoides is the only one of the three fungi found at fjaeldmark areas (F6 , F22), which are characterized by bare expanses of rock and scoriae and dominated by the cushion plant Azorella selago Hook. f. Fjaeldmark is the primary habitat on the island from which all other plant communities are derived through ecological succession (Smith 1987). Our results suggest, therefore, that F. merismoides could be the initial colonizing Fusarium sp. of plant debris. F. acuminatum and F. reticulatum appear to follow as the area becomes more vegetated with various other plant species and is frequented by seals, penguins or other birds. This would agree with the suggestion made by Stoner (1981) that Fusarium spp . can be used as important indicators of the "Number of isolates obtained from 100 debris pieces from each site stages of ecological succession and community development.
It appears that the presence of animal and bird activity enhances the prevalence of the abovementioned three Fusarium species, which supports the findings of Steyn & Smith (1981) with respect to microbial populations in Marion Island soils. The prevalence of F. acuminatum and F. reticulatum was highest at a burrowing bird habitat (F4), while F. merismoides was also isolated at high frequencies at this location. On a grass slope frequented by nesting albatross (F8), the incidence of F. merismoides and F. acuminatum was relatively high. The incidence of F. merismoides was at its highest at a seal wallow area (FlO). Two well-vegetated locations in close proximity to animal or bird activity (F24, F26) also had high Fusarium populations. Conversely, those environments that were not biotically influenced and were sparsely vegetated (fjaeldmark) had some of the lowest Fusarium frequencies recorded. This is supported by the low Fusarium counts from a Blechnum-slope (F3) and an inland mire (FS), both with no animal activity. The Fusarium population at FS could also, however, have been influenced by the waterlogged soil conditions. The possibility exists that F. reticula tum caused the necrotic lesions on the Pringlea leaves, but this has not been proven by pathogenicity tests.