RESULTS

The aqueous extracts from the five plant species were assayed for their ability to suppress the growth of fungal plant pathogens In vitro.

The ability of the aqueous extracts to suppress the fungi varied between the weed species and over the incubation period. The weeds A. conyzoides, O. corniculata and P. debilis suppressed the fungal mycelial growth by 71 to 86% after three days incubation (Table 1). A. conyzoides successfully controlled the growth of R. solani, A. niger and P. theae by at least 70% after three days of incubation (Fig. 1).

The suppression of mycelia growth decreased with the number of days of incubation. The extract of A. conyzoides was active up to 9 days and it was particularly effective against R. solani (Table 1).

The extract from A. conyzoides was also sterile filtered and incorporated in the PDA medium to determine the heat stability of the fungal inhibitory factor. There was no significant difference between the sterile filtered and autoclaved extracts (Fig. 2).

The extract of O. corniculata was effective against A. niger. It was moderately active against R. solani and P. theae three days after incubation. The activity rapidly declined by nine days. The extract of P. debilis inhibited the growth of P. theae by 72% after three days of incubation, which declined thereafter. A similar effect was produced by A. conyzoides. P. debilis was also moderately active against R. solani and A. niger, 3 days after incubation.

The extracts of V. cinerea and D. triflorum suppressed the mycelial growth by 20-30% only which rapidly declined.

Table 1:	Inhibition of mycelial growth (%) by the water extracts of five weed species on some plant pathogenic fungi
Values followed by similar letters down the columns are not significantly different (P = 0.05)

Fig. 1:	Mycelia inhibition of four fungi by aqueous water extract of Ageratum conyzoides after three days of incubation. Vertical bars represent ±SE.

Fig. 2:	Mycelia inhibition by autoclaved and filter sterilized extracts of Ageratum conyzoids on Rhizoctonia solani. Bars with similar letters are not significantly different (P=0.05)

Whereas of the four fungi tested three were inhibited by one or more of the five plant species, B. theobromae was particularly resistant to all the extracts.

DISCUSSION

Antifungal compounds are broadly classified by Kuhn and Hargreaves (1987) into active and constitutive compounds. Active compounds are only expressed during a fungal infection, while constitutive compounds are already present in healthy plants prior to infection. In this experiment since only healthy plants were chosen for extraction, the antifungal activity observed can be regarded as constitutive.

Of the five weed species, A. conyzoides was the most effective against the R. solani, A. niger and P. theae. It inhibited the growth of mycelium by at least 70%. Almost all members of the family Asteraceae (to which the genus Ageratum belongs) are known to contain sesquiterpene lactones, which are cytotoxic while A. conyzoides contains hydrocyanic acid, coumarin and an alkaloid (Lewis and Elvin-Lewis, 1977). The latter compounds also probably inhibit grazing of the plant by herbivores. The active compounds in A. conyzoides are heat stable (at 121°C, 15 psi) as there was no significant difference from the filter sterilized samples of the same aqueous extract.

The extract from other species were active against specific fungi. Although A. conyzoides was active against R. solani, A. niger and P. theae, it hardly inhibited the growth of B. theobromae. In fact none of the extracts were able to suppress this fungus. O. corniculata was effective against A. niger and moderately against R. solani and P. theae. The extract of P. debilis suppressed the growth of P. theae but was moderately active against R. solani. The different degrees of activity by the extracts and the specificity of the extract to the fungus precludes the possibility that the inhibition of fungal growth is due to alterations in the culture medium environment as may be possible through changes in the pH, osmotic pressure etc. as suggested by Kuhn and Hargreaves (1987). In a similar study, aqueous extracts from the weeds Ranunculus asiaticus, Chenopodium murale, Inula viscosa and Solanum nigrum inhibited the fungi Penicillium digitatum, Sclerotinia sclerotiorum and Verticillium dahliae (Qasem and Abu-Blan, 1995).

Present study was restricted to aqueous extracts from the mature shoots of weed species. Thus only the water-soluble compounds would be present in the extract. Plant tissues contain many products of secondary metabolism. It is possible that some of these metabolites suppress fungal growth. Their concentration in the plant tissues could vary with developmental stages and environmental conditions (e.g., rainy or arid climates). The synthesis of these secondary metabolites could also be a function of the genotype, which offers a potential for selection.

The active aqueous extracts suppressed fungal growth for three days after incubation and thereafter gradually declined. With A. conyzoides the antifungal activity was maintained up to 9 days. The decline and difference of activity could be due to breakdown of the active components or even metabolizing the fungi to non-toxic products. The latter is possible if the initial inoculum is in excess of the concentration of the antifungal compound or compounds. Qasem and Abu-Balan (1995) also reported a similar decline of the antifungal effect with incubation period of weed extracts. They suggest this may be due to transformation of these materials to non-toxic forms or the loss of volatile inhibitors. Chaturvedi et al. (1987) reported a similar decline in antifungal effect from loss of volatile inhibitors.

Similar studies with crude plant extracts were reported for controlling post harvest losses in fruits and seeds. In tomato, fruit rot caused by A. niger was reduced when the fruits were treated with leaf extract of Argemone mexicana (Saxena and Saxena, 1990) and the latex of Euphorbia hirta (Sinha and Saxena, 1989). In fruit-rot of grapes caused by Phomopsis species satisfactory control was achieved with Eucalyptus and Aegle marmelos leaf extracts (Arun, 1988). Aqueous leaf extracts of Cycas revoluta and Thuja orientalis reduced the fungal incidence in stored rice seeds (Kumar, 1990).

The constitutive antifungal compounds present in healthy plant tissues belong to a broad range of structural classes such as phenols, isoflavones, terpenoids, alkaloids, triterpenoid saponins etc. (Harborne and Ingham, 1978; Mansfield, 1983; Schoenbeck and Schloesser, 1976).

An implication of this study is to investigate the effect on soil fungi by incorporating the weeds, especially Ageratum, in the soil. The degradation of the plant tissues can be expected to similarly inhibit fungal growth. Angus et al. (1994) studied the incorporation of Brassica plants in the soil. The breakdown of Brassica tissues in the soil and the subsequent build up of isothiocyanates controlled pathogenic fungi.

This study shows the existence of specific inhibition of fungi by chemical compounds present in aqueous extracts from certain weeds. It is necessary to increase the efficacy of extraction by polar and non-polar organic solvents and eventually characterize their chemical nature. It should enable the improvement of their stability by alteration of the compound and provide non-hazardous pesticides to control specific fungi.

ACKNOWLEDGMENTS

The authors wish to thank Mr. R.B. Hapukotuwa for collection and preparation of the material.