Symptomise leaves of banana plant were collected from the BOH Plantation Limited in Tiko subdivision of the South-West Region of Cameroon. Samples were put in appropriate bags, labelled and transported to the Research Unit of Phytopathology and Agricultural Zoology of the Faculty of Agronomy and Agricultural Sciences of the University of Dschang for fungi isolation.

The symptomized leaves collected from the field were washed thoroughly in tap water and cut into small fragments of about 2 mm². These fragments were then disinfected in a sodium hypochlorite solution at 2% for 2 minutes and rinsed three times with sterile distilled water. The fragments were plated in Petri dishes containing 20 ml of Potato Dextrose Agar (PDA) medium amended with chloramphenicol (1 g/l) to prevent bacterial contamination and incubated at 24˚C ± 2˚C [14] .

After 2 to 3 days of incubation, the growing mycelium was sub-cultured on fresh PDA medium until pure cultures were obtained and maintained in the refrigerator at 4˚C. Morphological identification of fungal isolates was carried out based on the cultural characteristics and with the help of identification key of mycology [15] .

Frequency occurrence of isolation of each fungus was calculated using the following formula:

F = N F N T × 100

where F represent the frequency of occurrence (%) of a fungus, NT is the total number of all fungi isolated and NF is the specific number of fungus isolated.

Fresh leaves of Cymbopogon citratus and Allium ampeloprasum were collected in Dschang locality. These leaves were disinfected separately with a sodium hypochlorite solution at 2%, rinsed with sterile distilled water to remove any impurities, chopped into small fragments using a sterilized knife and dried in darkness for one week. When fully dried, the samples were grinded to powder using an electric grinding machine (trade of machine). Thereafter, using cold solvent extraction method [12] , 100 g of each processed samples of A. ampeloprasum and C. citratus were macerated in 500 ml of each solvent (sterile distilled water and ethanol) in a bottle for 48 hours at room temperature. After 48 hours, the mixture was filtered using cheese cloth followed by Whatmann filter paper N˚. 1.

The different filtrates (aqueous and ethanol) of the two plants were poured into sterilized stainless steel trays (plates) and dried in an electric oven at temperatures of 40˚C. The plant extracts were transferred into labeled sterile bottles and stored at 4˚C in a refrigerator pending utilization for the antifungal activity tests [16] .

Three fungi; Pseudocercospora fijiensis, Pseudocercospora musicola and Pestalosis sp. were selected for this test. The choice of these three fungi was related to high frequencies during the inventory of fungi associated with banana leaves. The evaluation of the inhibition of radial growth of the different fungi was done using the agar dilution method on Potato Dextrose Agar medium. The ethanol extracts were dissolved in Dimethyl Sulphoxide (DMSO). The effect of DMSO was pre-tested to ensure it zero influence on the growth of the selected fungi [12] . For both extracts the concentrations used were 5, 10 and 15 mg/ml. 1 ml of each extracts at different concentrations were incorporated into 19 ml of PDA and poured into sterilised Petri dishes of 90 mm diameter and allowed to solidify in a fume hood which was lighted by flame from a Bunsen burner.

After solidification, mycelia discs (5 mm) of 10-days old growing in pure cultures of each fungus were transferred to the Petri dishes with the help of a cork borer of 5 mm in diameter [11] . PDA plates mixed with fungicide (Dithane) at the recommended dosage of the producer (1 mg/ml) served as the positive control while distilled water served as the negative control. The Petri dishes were incubated at 24˚C ± 2˚C until there was a 100% growth of negative control. This test for In vitro evaluation of plant extracts was done in a complete randomized design with 3 replicates and the following parameters were evaluated:

➢ Percentage of inhibition

The effectiveness of the extract was recorded in terms of percentage inhibition (PI), which was calculated according to the following formula:

P I = ( D T − D ) D T × 100

where DT is the growth diameter of the negative control Petri dish and D is the growth diameter of the supplementary Petri dish of plant extracts. The EC₅₀ and EC₉₀ were obtained by transforming percentage inhibitions in probity.

➢ Nature of toxicity of plant extracts

The evaluation of the toxicity of plant extracts consisted of seeing if the mycelia growth, where complete inhibition was observed, was accompanied by a fungistatic or fungitoxic activity. In this case, the explants of the mycelium where complete inhibition was observed on the supplementary PDA medium on plant extracts were retaken and placed aseptically on a none fungicide or plant extract containing PDA medium in a sterilized hood lighted by a Bunsen flame. After 10 days of re-incubation, at a temperature of 24˚C ± 2˚C, the activity of plant extracts was considered as fungistatic if there was mycelia regrowth of the fungal pathogen and fungitoxic if there was no mycelia growth of the pathogen.

Data collected on the percentage inhibition, EC₅₀ and EC₉₀ were subjected to the analysis of variance (ANOVA) using SPSS software version 22.0 and the mean values were separated using Duncan Multiple Range Test (DMRT) at 5% probability.

Nine fungal species identified were associated with banana leaves (Figure 1). Amomg these species, the most frequent fungus was Pseudocercospora fijiensis with an isolation frequency of 51.58%. This was followed by Pestalopsis sp. and Pseudoscercospora musicola with isolation frequencies of 15.47% and 12.03% respectively. The least common fungal species were Verticillium theobromae, Deightoniella torulosa and Trichoderma harzianum with isolation frequencies of 0.29%, 1.72% and 2.29% respectively.

The morphological and microscopic aspects varied with respect to the different fungi (Figure 2). Pseudocercospora musicola showed faint pink coloured mycelia which became faint purple as it grew to a 10 days old pure culture. P. fijiensis showed whitish mycelia which became darkish as it grew to a 10 days old pure culture. Both species under the light microscope, showed elongated conidiophores which are greyish with septate hyphae. Fusarium oxysporium was characterised by abundant aerial mycelium which were whitish and hyaline in colour. Microscopically, macro and micro conidia were observed with macro conidia slightly curved with tapering ends. Trichoderma harzianum, showed numerous mycelia that were initially whitish which later as the conidia grows, formed scattered blue-green, dark green to yellow green patches. Microscopically, its conidiophores are numerous, branched, ovoid in shape and occurred in clusters. Verticillium theobromae showed cream white mycelia with smooth and septate hyphae conidiophores under the microscope with some ramified, some simple and straight.

The nature of activities of the plant extracts depended on the plant and the fungus (Table 5). The aqueous extract of C. citratus showed a fongitoxic effect at 15 mg/ml on the growth of Pseudocercospora fijiensis and Pestalopsis. Meanwhile the ethanol extract of A. ampeloprasum had a fungistatic effect on the development of Pestalopsis sp. at 15 mg/ml.

This study identified 9 fungal species associated with Musa acuminata leaves such as Pseudocercospora fijiensis, P. musicola, Fusarium oxysporum, Pestalopsis sp., Cladosporium sp., Rhizoctonia solani. The presence of this diversity of fungi could be due to the fact that Musa acuminata leaves constitute an important source of carbohydrate for these fungi. Species like P. fijiensis (51.58%), P. musicola (12.03%), and Pestalosis sp. (15.47%) had the highest frequency compared to the other fungal species isolated. These three species are generally reported to cause significant damage to the leaves, pseudostems, fruits and of course to the entire banana plant. These results are similar to those of [7] , who reported that Pseudocercospora fijiensis is a very formidable species for a wide range of hosts, attacking the foliage, stems and fruits of its hosts. Cladosporium sp., R. solani, D. torulosa, T. harzianum and V. dahlia were relatively less important with respect to their low isolation frequencies. These fungi are however harmful to the banana plant and have also been reported as pathogenic in some fruits including mango, apple, citrus and grape in other parts in the tropics [9] [17] . Several reports showed the implication of the genus Pseudocercospora and Fusarium as being responsible for the major economic losses in the production of banana [18] . The origin of banana plant infection especially on the leaves by fungi is difficult to determine as it is difficult to predict the potential spread of the pathogens to and from neighboring areas. These results are also similar to

*F: fungitoxic, f: fungistatic and /: none 100% of inhibition on the radial growth of fungus.

those reported by [19] , on isolation and pathogenicity evaluation of postharvest fungal of some fruits in Cameroon.

In this study, we investigated the antifungal activities of Cymbopogon citratus and Allium ampeloprasum extracts against Pseudocercospora fijiensis, P. musicola, and Pestalopsis sp. Our results demonstrated that C. citratus and A. ampeloprasum extract at all tested concentrations had greater overall depressive effect on the radial growth of these three fungi strains than the negative control. The plants used could have contained inhibiting compounds or substances that could have influenced the growth of the fungal species tested. Indeed, the work of [20] have shown, some plants contain compounds with antifungal properties (alkaloids, sterols, terpenoids, flavonoids, anthraquinone phenols, saponins or tannins), hence their use sometimes in traditional medicine. These results are also similar to those of [21] who showed that Moringa oleifera L. seed extract had antifungal activities against fungi associated with banana plant which included; Fusarium oxysporum, F. solani, Alternaria solani A. alternate, Rhizoctonia solani, Sclerotium rolfsii and Macrophomina phaseolina.

Radial growth of the fungi species tested was influenced by the concentration applied as well as by the type of extract. The radial growth of the different test pathogens decreased with increasing concentrations, which may suggest that high concentration of extracts is more fungicidal than low concentrations Similar results of the antifungal activity of some plant extracts against post-harvest fungi in avocado fruit have already been reported by [22] . Different radial growths were observed at the same concentrations. These differences in the activity of plant extracts towards these fungi could be attributed to their active ingredients. On the other hand, according to [23] , it should be mentioned that the antifungal activity of plant extracts is often very much linked to simultaneous actions of their constituents.

The results obtained with Cynbopogon citratus extracts were in accordance with those obtained by [12] [24] who showed that these extracts had antifungal activities on the development of Colletotrichum kahawae responsible for anthracnose of coffee berry and Colletotrichum musae and Aspergillus niger responsible for banana fruit rot respectively. Similarly, those obtained with Allium ampeloprasum extract are similar to the results obtained by [25] who showed that the A. sativum and A. ampeloprasum extracts inhibited the development of Alternaria triticina, causal agent of leaf blight in wheat and Magnaporthe oryzae, causal agent of blast disease in rice.

The EC₅₀ and EC₉₀ values were heterogeneous. Ethanol extracts were significantly more effective than aqueous extracts. The difference in efficacy between ethanol extracts and aqueous extracts from the tested plants could be due to the fact that ethanol may have extracted compounds with more effective antifungal properties than distill water that may have extracted the majority of compounds with less effective antifungal properties. This hypothesis is similar to the hypothesis of [26] [27] who reported that organic solvent extraction is more effective on antimicrobial activity than water.

Explants on Petri dishes with 100% inhibition were transplanted onto extract-free PDA medium to demonstrate fungistatic or fungitoxic activity. The fungistatic activity observed with Allium ampeloprasum extracts are similar to that reported by [11] who showed that Erigeron floribundus and Euphorbia hirta extracts had fungistatic activity on the fungi associated with avocado. In addition, the results are also similar to those of [28] , reported that Cupressus lusitanica extracts had fungistatic activity on the growth of Phytophtora colocasiae, the agent responsible for taro blight. The fungitoxicity of Cymbopogon citratus extract on the growth of the pathogens was in accordance with that of [25] , who showed the fungitoxicity activity of these extracts on the development of Alternaria triticina, causal agent of leaf blight in wheat.