The soil my coflora of an Acacia karroo Community in the Western Transvaal

A survey of the composition and distribution of the soil mycoflora of an Acacia karroo Community in the Potchefstroom area was undertaken. A total of 858 sporulating cultures representing 76 genera and 144 species was recovered from this soil. The majority belong in the Fungi Imperfecti and only a limited number of Zygo­ mycetes and Ascomycetes and no Oomycetes or Basidiomycetes were recorded. Members of the genera Penicillium and Aspergillus were the most abundant. The greatest concentration of individuals and species occurred in the surface layers and a rapid decrease in numbers was noticeable with increasing depth. The nature of this flora suggests a close correlation with the natural plant cover and the existing ecological conditions.

During the past decades various aspects of the soil mycoflora have been extensively studied in many parts o f the world and a voluminous literature has accumulated (Alexander, 1970;Barron, 1968;Doeksen & Van der Drift, 1963;Parkinson & Waid, 1960;Gilman, 1959;Burges, 1958;Chesters, 1949;Domsch & Gams, 1970, etc.).As a result a fairly clear picture of this flora has emerged and in recent years the suggestion that it is of a uniform and cosmopolitan nature (Waksman, 1916) has been criticized and partly rejected in favour of the view that the soil mycoflora is usually closely correlated with certain ecological factors including the natural plant cover (Eicker, 1974;M orrall & Vanterpool, 1968;Park, 1965;Clark, 1965;Garrett, 1963;O rpurt & Curtis, 1957;Christensen et al., 1952;Tresner et al., 1954).
In South Africa the fungus flora of soils under a natural vegetational cover has so far received little attention and because of the great diversity of climate, soil type and vegetation a vast and interesting field remains to be investigated.Eicker (1969Eicker ( , 1970Eicker ( , 1974) ) published a comprehensive list o f fungi inhabiting two forest soils in Zululand and the soil of the opensavanna o f the Transvaal.
This paper is a report on a study which was under taken to determine the composition and distribution of the soil fungus population of an Acacia karroo Community in the Western Transvaal.The Acacia karroo Community is typical o f the savanna o f the more arid western portion of the huge inland plateau of the southern sub-continent commonly referred to as the High Veld.It is dominated by Acacia karroo Hayne and in this so-called Thorn Veld the Acacia trees appear singly or in groups and are variously scattered leaving large open spaces covered mainly by grass species, herbaceous plants and shrubs o f different sizes.The following are the more common species found in the area studied: The community concerned is situated on the northern boundary o f the present campus of the University of Potchefstroom.The annual precipita tion in this area is c. 600 mm and occurs mainly in the form of rain of which more than 80% falls during the months October to M arch.The soil tem perature 15 cm below ground level varies and fluctuates between an annual summer maximum of c. 23 °C and a winter minimum of c. 16 °C.The soil is a typical red-brown loam consisting o f the weathered products of the underlying diabase.

METHODS
In order to obtain a composite and representative soil sample of the community 10 sampling sites covering the entire area were randomly selected.To expose the profile a trench measuring 90 x 60 cm was excavated at each site and from its sides soil samples were taken at preselected depths of 0 -2 ,5 cm, 15 cm and 30 cm.Samples were collected by means of a tapered and sharpened metal tube which could be driven into the sides o f the trench.Two cores were taken from each level o f the ten sampling sites.These were placed in separate sterile containers and even tually combined in a large glass container, thoroughly mixed and homogenized and subsequently sieved.From this composite sample 25 g portions were removed for further analysis.Peptone-dextrose agar containing rose bengal and streptomycin (M artin, 1950;Johnson et al., 1959) was used as isolation medium.
The isolation methods employed were all standard procedures used in soil mycological investigations and included the dilution-plate (Waksman & Fred, 1922), soil-plate (Johnson & M anka, 1961;Warcup, 1950) and the soil washing (Gams & Domsch, 1967;Parkinson & Williams, 1961) techniques.Since factors such as the method and media employed in this type of investigation obviously have certain limitations and selective influences it is not claimed that the list compiled from existing data is a complete inventory o f this mycoflora.

RESULTS
During this survey a total of 858 sporulating isolates was obtained and preserved in pure culture.An analysis of these isolates gave the following distribu tion: The individual species are listed in Table 1.

DISCUSSION
In the final collection (Table 1.) the Zygomycetes, Ascomycetes and Fungi Imperfecti are represented by 7, 8 and 61 genera and 8, 12 and 124 species respec tively.No Basidiomycetes or Oomycetes were recovered from this soil.
The absence o f members of the Oomycetes should probably be ascribed to unfavourable environmental conditions including low soil moisture and high summer temperatures often experienced in this area.Eicker (1969Eicker ( , 1974) ) was also not able to isolate any Oomycetes from two soil types in Zululand and only obtained a single species o f Saprolegnia from an open-savanna soil.
The Zygomycetes seem to be fairly well established in this habitat and are represented by the following species: Absidia cylindrospora, Circinel/a sp., Cunninghamella echinulata, Gongronella butleri, Mortierella alpina, Mucor circinelloides, Rhizopus arrhizus and Rhizopus sp.With the exception of Rhizopus the zygomycetous genera recovered are each represented by only one species.The total number of isolates (37) for this group indicates that this soil is relatively sparsely populated by Zygomycetes.The most com mon genus is Absidia (13 isolates) followed by Rhizo pus, Gongronella and Cunninghamella with 9, 5 and 4 isolates respectively.Circinella is represented by 3, Mortierella by 2 and Mucor by only a single isolate.
Investigators studying the mycoflora of different soil types from various localities have found that the zygomycetous population can vary considerably.On the whole cool, moist forest soils rich in nutrients and energy sources seem to support a much larger and more varied flora than poor and arid, sandy soils.Many workers who have studied microfungi of soils of more or less arid, sandy areas have indicated a general paucity of mucoraceous forms (W ohlrab & Tuveson, 1965;Kuehn, 1960;Nicot, 1960), Paine (1927) also found that mucors are less prevalent in open pasture land than in forests.Eicker (1969Eicker ( , 1974) ) and Yung & Stenton (1964) reported fairly large numbers of Zygomycetes from various soils.Gochenauer & Whittingham (1967) and Gochenauer & Backus (1967) investigating the microfungi of certain mesic forest and alluvial soils with coarse-grained texture and low humidity reported that the most im portant features of the populations of these habitats were the rareness of members of the Mucorales and Moniliales, other than Penicillium, in the dry sandy soil as contrasted with the abundance of these types in the mesic soil.This phenomenon is ascribed by them to the high surface temperatures, low moisture content and the limited energy sources available.They consider these habitats to be too rigorous for the survival of the generally hyaline and delicate types as opposed to the dematiaceous forms and many Sphaeropsidales which were recovered more abun dantly.The results of this investigation conducted in a more or less semi-arid region, indicate that the zygomycetous flora is not particularly varied and rather poorly represented in number of recordings.This seems to agree with the general pattern for the soils from many arid regions.
Only a relatively small number of Ascomycetes were recorded from this habitat.W arcup (1951b) pointed out that the more generally used techniques, such as the soil dilution method, produce only a few Ascomy cetes while methods where the soil is partially sterilized by heat results in the isolation of more and a greater variety of forms.By means of the more conventional methods applied during this survey an aggregate of 14 isolates representing 8 genera were obtained.Only one specimen of each of the following species was recorded: Auxarthron umbrinum, Chaetomium pachypodioides, C. robustum, C. baineri, Herpotrichia striatispora, Microascus cinereus, Neocosmospora vasinfecta, Westerdykella sp., Thielavia sepedonium and Sordaria fimicola.Chaetomium olivaceum and Chaeto mium sp. were represented by two isolates each.O f the 14 isolates 11 originated from the upper horizon while 3 were recovered from deeper levels.
Regarding the vertical distribution of soil inhabiting fungi the available evidence (Eicker, 1970;Yung & Stenton, 1964;Warcup, 1951a;Waksman, 1944;Burges, 1958;etc.)indicates a general decrease in both num ber of individuals and species with increasing depth.In some reported cases particular fungi are inclined to colonize certain horizons more readily than others.W arcup (1951a) found that forms like Gymnoascus, Cylindrocarpon radicicola and Paecilomyces were more abundant in the lower regions of horizon A and seldom occurred near the soil surface.On the whole the vertical distribution of the fungi in the Acacia soil follows this pattern.The surface layer produced 676 isolates against 158 and 24 for the 15 cm and 30 cm layers respectively.Only two species were convincingly more abundant in the deeper horizons, i.e.Penicillium adametzi and P. spinulosum, while 19 were confined to these sub-surface layers.
From a taxonomic study of the various isolates it was possible to erect 6 new genera and establish 11 new species (Papendorf & Von Arx, 1966;Papendorf, 1967aPapendorf, , 1967bPapendorf, , 1969aPapendorf, , 1969b;;Papendorf & Du Toit, 1967;Van der Aa, 1967;Papendorf & Upadhyay, 1969;Stolk, 1969.).The presence of this relatively large number o f previously unknown fungi in this particular habitat seems to support the view, referred to earlier, that the soil mycoflora is often closely correlated with certain ecological factors and the natural vegetation.