African Crop Science Journal African Crop Science Society ISSN: 1021-9730 EISSN: 2072-6589 Vol. 7, Num. 4, 1999, pp. 531-537

African Crop Science Journal, Vol. 7. No. 4,  pp. 531-537, 1999                                                              

THE SIGNIFICANCE AND MANAGEMENT OF MAIZE COB ROTS IN SMALLHOLDER FARMS IN CENTRAL MALAWI

S.J. Kapindu, V.W. Saka, A.M. Julian¹*, R. Hillocks¹ and W.A.B. Msuku
University of Malawi, Bunda College of Agriculture, P.0. Box 219, Lilongwe, Malawi
¹Natural Resources Institute, Chatham Maritime, Chatham, Kent, ME4 4TB. U.K.

* Current address for A.M. Julian: universtiy of Reading, Early Gate, U.K.

Code Number: CS99044

ABSTRACT

The main objective of this study was to describe the importance of cob rots in farmers’ maize fields and reduced food quality through spoilage. During the 1997/98 season disease incidence and severity levels were evaluated in the Local, MH18, NSCM41, and NSCM51 varieties on farmers’ maize from two villages in central Malawi. Farmer- and researcher-selected samples from farmers’ harvests during the 1996/97 and 1997/98 growing seasons respectively were analysed in the laboratory for fungal contaminants. Another experiment was carried out on-station for comparative purposes using the four varieties. Each variety was inoculated with Stenocarpella maydis and Fusarium  moniliforme and control plots were treated with sterile distilled water using the silk-spray method. Stenocarpella macrospora was the most predominant pathogen in both farmer- and researcher-selected small-holder farmers’ maize. Fusarium moniliforme was also commonly isolated, though at low incidence levels. The results indicate that both ‘healthy’ and ‘rotten’ maize are contaminated by almost the same fungal species, but the levels of contamination were generally lower in ‘good’ than in ‘rotten’ maize. Cob rot infections were significantly high for NSCM41 followed by NSCM51, MH18, and lowest in the Local variety, both in the field and on-station trials (P < 0.05). High infection levels were more associated with S. maydis than F. moniliforme inoculations. We conclude that NSCM41 is more susceptible to cob rots followed by  NSCM51, MH18, with the Local variety being most tolerant.

Key Words: Fusarium moniliforme, incidence, Malawi, Stenocarpella macrospora, Zea mays

RÉSUMÉ

L’objectif principal de cette étude était de décrire  l’importance de la pourriture d’epis chez le maïs en terme de perte de recolte et de reduction de le qualité alimentaire à travers la déterioration. Des niveaux de séverité et d’incidence des maladies ont été évalués pendant la saison de 1997/98 dans les variétés: La locale, MH18, NSCM41, et NSCM51 du maïs des agriculteurs de deux villages près du College de Bunda.  Des échantillons selectionnés par les agriculteurs et chercheurs durant les saisons culturales 1996/97 et 1997/98 respectivement ont été analysées au laboratoire pour des contaminants fongiques. Un autre essai a été conduit en station pour des raisons comparatives utilisant les quatre variétés.  Chaque variété à été inoculée avec Stenocarpella  maydis et Fusarium moniliforme  et les parcelles témoins ont été traités avec de l’eau distillée sterile utilisant la méthode silk-spray.  Stenocarpella macrospora était le pathogène le plus prédominant ensemble dans les petites explaitations de maïs sélectionnées par l’agriculteur et le chercheur.  Fusarium moniliforme était aussi communément isolé, mais à de niveaux bas d’incidence.  Les résultats montrent que le maïs "bon" et "pourris" sont contaminés par presque les mêmes espèces fongiques.  Cependant les niveaux de contamination sont générallement les plus faibles dans le maïs "bon" que dans le mais "pourri".  Les infections de pourriture des épis étaient significativement élevées pour NSCM41 suivie par NSCM51, MH18, et les plus faibles chez la variété Locale, ensemble pour les essais en champs et en station (P < 0.05). Les niveaux  d’infection plus élevés étaient significativement associés avec l’ inoculation de  S. maydis plus que celle de F. moniliforme.  Nous concluons que NSCM41est plus sensible aux pourritures des épis suivi de NSCM51, MH18, et la variété locale étant la plus tolérante.

Mots Clés: Fusarium moniliforme, incidence, Malawi, Stenocarpella macrospora, Zea mays

INTRODUCTION

Maize (Zea mays L.) forms the staple carbohydrate source for the majority of the population in Southern Africa, especially in Malawi, Zimbabwe, South Africa, and Zambia. However, production is constrained by a number of factors including poor soil fertility, rainfall, and diseases. These diseases include cob rots. Maize cob rots are caused by a fungal complex including Fusarium spp., Stenocarpella spp., Nigrospora spp., and Aspergillus spp. (Flett, 1992; MacDonald and Chapman, 1997). They are important for two reasons. First, because heavy infestations directly result in grain spoilage, significantly reducing both the yield quantity and quality. Second, the cob rot fungi produce mycotoxins, which have been linked with mycotoxicoses and carcinomas of humans and domestic animals (Marasas et al., 1988; Rabie et al., 1993; Julian et al., 1995; Castelo et al., 1998).

Ngwira (1990, unpubl.) reported a 10% yield loss due to cob rots in Malawi. In fact, cob rots have been ranked among the top three important maize diseases and fourth in distribution in Malawi (Anonymous, 1996, unpubl.). Varietal reactions to the disease have been observed (Ngwira, 1990 unpubl.); however, little documented information exists on the relative susceptibility of individual maize varieties commonly grown on smallholder farms in Malawi. Furthermore, the reactions of these varieties to the major cob rot pathogens, i.e., Stenocarpella and Fusarium species as reported by Ngwira (1990, unpubl.) have not yet been tested.

This study therefore aimed, at determining fungal contaminants of farmer-selected maize; assessing the incidence and severity of cob rots on small holder farmers’ fields; identifying the fungi associated with maize cob rots; and determining  the reaction of maize to inoculations with F. moniliforme and S. maydis.

MATERIALS AND METHODS

Fungal contaminants from farmer-selected maize .The study was conducted during the 1996/97 growing season on farmers’ fields at two sites in Mitundu Extension Planning Area (Chiphe and Makwenda) situated south east of Lilongwe, Central Malawi. Twenty-two farmers were randomly selected from each site. Soon after harvesting, each farmer was asked to select randomly two samples of ‘healthy’ and ‘rotten’ maize comprising 10 cobs from different hybrids. Each sample was shelled and combined and were analysed in the laboratory at the Natural Resource Institute (NRI) in the United Kingdom. From each composite sample, 25 grains were surface sterilised for 3 minutes in 10% Sodium hypochlorite (Na0Cl) solution and rinsed in sterile distilled water. The grains were then plated on quarter strength Potato Dextrose Agar (Merck) plus 0.01g/l streptomycin sulphate (Merck). Plates were incubated in the dark at 25° for 5-7 days. Thereafter, Fusarium colonies were sub-cultured on Simple Nutrient Agar plus 0.01g/l Streptomycin sulphate; pieces of sterile filter paper were added to the plates to encourage sporulation. To further encourage sporulation, the plates were incubated at 25°C under fluorescent and black-light lamps of a 12hr photo-period for 7-10 days. Fungi were identified 7-30 days after incubation, depending on the species.

The pathogen abundance and distributions in the studied fields defined as the proportion of plated grains, and the proportion of farms infected by a particular fungus, respectively, were calculated using the formulae;

(i) Pathogen abundance (%) = z x 100 / N
where;
z = Number of plated grains from which a fungus species was isolated
N = Total number of plated grains (25)
(ii) Pathogen distribution on farms (%) = y x 100 / N                                               
where;
y = Number of samples from which a fungus species was isolated
N = Total number of samples analysed

Maize varietal reactions to cob rots in small-holder farms. The 1996/97 trial as described above was repeated with a few modifications on 29 small-holder farms at the two sites in the 1997/98 growing season. It was a factorial experiment arranged in a nested design with each farmer as a replicate, each site as a main plot, and variety as a sub-plot with four levels (Local, MH18, NSCM41, NSCM51); and grain quality was a sub-sub-plot with two levels (‘healthy’ and ‘rotten’). The plots were farmer-managed. Soon after harvesting, variable cob numbers with a maximum of 500 were sampled randomly from each sub-plot and rated on a 0-5 scale (0 for  no visible infection, 1 =1-25%, 2 = 26-50%, 3 = 51-75%, 4 = 75-99%, and 5 = completely ‘rotten’ cobs). Out of these cobs, sub-samples of 25 ‘rotten’ and 5 ‘healthy’ cobs were randomly sampled for fungal laboratory analysis. The difference in sample size arose in order to maintain farmers’ cooperation during the study, as many expressed serious doubts over the reasons behind sampling the intended number of ‘healthy’ cobs (25) from each variety. The severity of the cob rots disease on each plot was expressed as a disease index (DI) (Ullstrup, 1949);

DI = (0.25n1 +0.5n2 + 0.75n3 + 0.99n4 + n5) x 100 /N
where n = number of cobs in classes 1-5, N = total number of cobs evaluated per heap.

The incidence of infected cobs was calculated using the following formulae;
Cob rots incidence  = 100 (x / N)
where;
x = the number of infected cobs
N = Total Number of cobs in heap

Each sample was then analysed in the laboratory for fungal contaminants as described earlier.

Maize varietal reactions to Fusarium moniliforme and Stenocarpella maydis. A factorial split-plot experiment, with 2 factors,  arranged in a RCBD and replicated six times was carried out on-station at Bunda College in Central Malawi in the 1997/98 season. Maize variety acted as a main unit with four levels (Local, MH18, NSCM41, NSCM51) and fungal inoculum as a sub-unit with three levels (F. moniliforme, S. maydis, and Sterile Distilled Water - control). The maize was planted on 72 sub-plots of land at a spacing of 90 cm between rows and 30 cm between plants (one plant per station). Each sub-plot consisted of four maize rows of 5 plants each.

Single spore isolates for the inocula were obtained from rotten maize cob samples collected during the 1996/97 season. Inoculum preparation was done by filling 20 ml vials containing sterile distilled water with ‘healthy’ maize grains that were autoclaved for 20 minutes. After cooling, the vials were inoculated with 1 x 10³/ml spore suspensions of the two pathogens from actively sporulating pure culture plates and incubated at 25ºC for two days. Thereafter, the colonised grains were plated on PDA (Merck), incubated for 7-10 days under 12 hr light/dark for F. moniliforme and 10-14 days for S. maydis. After sporulation, the plates were flooded with sterile distilled water, and scraped. The spore suspension was adjusted to 2.5 x 10⁵ conidia spores/ml for S. maydis and 1 x 10⁵ macro-conidia spores/ml for F. moniliforme by adding sterile distilled water. Fifteen millilitres and 5 ml of S. maydis and F. moniliforme spore suspensions, respectively, were sprayed to the cob silks of ten plants in the two middle rows of each sub-plot. The silk-spray methods described by Klapproth (1991) and Warren (1978) were used for maize inoculation with S. maydis and F. moniliforme, respectively.

Data analysis. To satisfy the assumption of normality in analysis of variance, pathogen abundance values were transformed to the natural logarithm. To account for the effect of sample size differences between ‘healthy’ and ‘rotten’ maize on pathogen abundance in the 1997/98 season, sample size values (5 and 25) for each of the observations were weighed using the weight command of SAS (SAS Institute, 1987). Data were subjected to Analysis of Variance following General Linear Model procedures of SAS. The Least Significant Difference (LSD) test, at 5% probability level, was used to compare groups of means. Only untransformed means are reported, and  LSD values are provided only where variance ratios were significant.

RESULTS

Fungal contaminants from farmer-selected’ maize. During the 1996/97 season 11 different fungi were identified from the 44 samples selected by farmers. Overall, S. macrospora was the most common, with a mean pathogen abundance and farm distribution of (75.2%; 65.9%) followed by F. graminearum (72%; 4.5%), F. intermedium (56%; 4.5%), S. maydis (55%; 4.5%), F. moniliforme (44.6%; 15.9%), and Nigrospora spp. (28.0%; 11.4%). Other less abundant contaminants included F. oxysporum , F. sacchari, F. solani, Penicillium spp., and Colletotrichum graminicola. Abundance of the fungi did not vary significantly (P = 0.18) between the two sites.

 The effect of grain quality (‘healthy’ and ‘rotten’) on the abundance of fungus species was not significant. However, the occurrence of S. macrospora (73.0%, 59.0%) and S. maydis (9.0%, 0.0%) were higher in ‘rotten’ maize (Table 1) while those of F. moniliforme (14.0%, 18.0%) and Nigrospora spp. (0.0%, 23.0%) were higher in ‘healthy’ maize. The contaminants were mostly isolated from ‘rotten’ maize except for Nigrospora spp. which was not isolated from ‘rotten’ samples. Stenocarpella maydis was not isolated from ‘healthy’ maize.

Table 1. Abundance and distribution in farms of six fungi associated with cob rots from 22 farms in Central Malawi, 1996/97 season

*Grain quality effect on pathogen abundance was not significant according to LSD (P>0.05)

Occurrence of cob rots on small-holder farms. In the 1997/98 season, there were no significant differences in severity and incidence of cob rots between the two sites. However, incidence and severity of cob rots varied significantly among the four varieties tested. Overall, the highest levels of cob rots were found in NSCM41 followed by NSCM51 and MH18; the Local variety had the least incidence and severity (Table 2).

Table 2. Incidence and severity of maize cob rots  at  two sites in Central Malawi, 1997/98 season

*Incidence and severity of cob rots among varieties were significant at both sites.  Their overall levels were also significant according to LSD (P<0.05)

Twenty-three different fungi associated with cob rots were found in the samples selected by the researcher, but with no significant difference between the two sites. Stenocarpella macrospora was predominant in both samples (Table 3). Fusarium moniliforme, Aspergillus spp., Rhizopus spp., and Nigrospora spp. were also isolated from many fields, but were less abundant than S. macrospora.

The effect of grain quality  ('healthy’ and ‘rotten’) on the abundance of fungus species was significant (P = 0.0001). Table 3 shows relatively higher isolation incidence levels for S. macrospora and Aspergillus spp. in ‘rotten’ maize than in the ‘healthy’. On the other hand, F. moniliforme, Fusarium spp., Rhizopus spp., and Nigrospora spp. levels were higher in ‘healthy’ than ‘rotten’ maize.

Table 3. Pathogen abundance in healthy and rotten maize cobs and frequency of recovery of six fungi associated with cob rots from 29 farms in Central Malawi, 1997/98 season

*Grain quality effect on pathogen abundance was significant according to LSD (P=0.0001)
**Differences in abundance among pathogens were significant according to LSD (P=0.0001)

Maize varietal reactions to Fusarium moniliforme and Stenocarpella maydis. There were significant differences in cob rot incidence and severity (P = 0.0001) levels  among the four inoculated maize cultivars. Overall, NSCM41 was more susceptible than the other three varieties (Table 4). In general, maize reacted differently to F. moniliforme and S. maydis inoculations. Higher infection levels were associated with S. maydis inoculations than F. moniliforme (Table 5). The reaction of maize to F. moniliforme inoculation (67.4%; 23.3%), was not significantly different from the control (64.9%; 20.0%).

Table 4. Incidence and severity of cob rots in one local maize cultivar and three released maize hybrids in an on-station experiment  during the 1997/98 season in Central Malawi

DISCUSSION

In 1996/97 season, there was high incidence of S. macrospora , F. graminearum, and F. intermedium in the ‘rotten’ samples (Table 1) while in the 1997/98 season,  S. macrospora and Aspergillus spp. were more prevalent (Table 3). In contrast, the abundance and distribution in fields of principal contaminants were low in ‘healthy’ maize (Tables 1 and 3). Furthermore, the occurrence of fungal isolates did not differ between the two villages clearly indicating the uniform distribution of these pathogens.

These results also indicate that ‘rotten’ and ‘healthy’ grains are infested by almost the same fungal pathogens. This is in line with the findings of  Desjardins et al. (1998) who observed extensive colonisation of symptomatic and symptomless grains with F. moniliforme. According to Julian et al. (1995) infection by Fusarium species, particularly F. moniliforme, can be asymptomatic, and may easily be overlooked. On the other hand, MacDonald and Chapman (1997) reported that Stenocarpella spp. and F. graminearum infections on maize grain exhibit obvious discolouration which makes grains visually distinguished as ‘rotten’. These pathogens (cob rot fungi) produce mycotoxins, some of which are carcinogenic (Marasas et al., 1988). Furthermore, these toxic metabolites have good thermal stability (Pittet, 1998). This implies that ‘healthy’ maize may also contain high levels of mycotoxins produced by such fungi. As a consequence, consumption of ‘healthy’ maize in the long run might be dangerous.  The study also found S. macrospora to be the most predominant pathogen in cob rots in Central Malawi, followed by F.  moniliforme (Tables 1 and 3).

One way of reducing incidence of cob rots is by growing resistant varieties. In this study four cultivars were tested in farmers fields as well as on-station. The Local variety was relatively tolerant to both natural infections and inoculations with F. moniliforme and S. maydis, NSCM41 being the most susceptible. The differences in susceptibility of the four varieties correspond to the earlier results of Schaafsma et al. (1993) who reported consistently higher cob rots severity for P3737 than the P3790 cultivar. In another study by  Klapproth (1991), the incidence of infected ears varied considerably between the MBS613 and B73Ht lines.

Reports by various authors indicate that the husk cover plays an important role in promoting infection of maize varieties by cob rot fungi. For example, Warfield and Davis (1996) observed increased infections in maize varieties with open cob tips than in those which were closed. The high susceptibility of NSCM41 to cob rots is probably due to its open tips (data not shown). This may make it easier for insects such as stalk-borers to penetrate through the silk channel opening infection loci for the fungi. Therefore, breeders should as much as possible try to breed varieties with covered cob tips.

The inoculation with S. maydis resulted in significantly higher visible infections than with  F. moniliforme (Table 5). However, F. moniliforme infects maize both asymptomatically and symptomatically depending on where the spore is placed; this does not occur with S. maydis (Sutton, 1981; Flett, 1992). Therefore, the observed difference in infections between the two pathogens may be attributed to this behaviour, and not necessarily that S. maydis is more pathogenic than F. moniliforme.

AKNOWLEDGEMENTS

The authors are indebted to the Rockefeller Foundation for funding this work, to Dr. A. Chirembo for assisting in data analyses, and to the farmers at Chiphe and Makwenda for their cooperation.

REFERENCES

Castelo, M.M., Summer, S.S. and Bullerman, L.B. 1998. Occurrence of fumonisins in corn-based food products. Journal of Food Protection 61:704-707.

Desjardins, A.E., Plattner, R.D., Lu, M. and Claflin, L.W. 1998. Distribution of fumonisins in maize ears infected with strains of Fusarium moniliforme that differ  in fumonisin production. Plant Disease 82:953-958.

Flett, B.C. and van Rensburg, J.B.J.  1992. Effect of Busseola fusca on the incidence of maize ear rot caused by Fusarium moniliforme and Stenocarpella maydis. South African Journal of Plant Soil 9:177-179.

Julian, A.M., Wareing, P.W., Philips, S.I., Medlock, V.F.P., MacDonald, M.V. and del Rio, L.E. 1995. Fungal contamination and selected mycotoxins in pre- and post- harvest maize in Honduras. Mycopathologia 129: 5-16.

Klapproth, J.C. 1991. Evaluation of four inoculation techniques for infecting corn ears with Stenocarpella maydis. Plant Disease  75:1057-1060.

MacDonald, M.V. and Chapman, R. 1997. The incidence of Fusarium moniliforme on maize from Central America, Africa and Asia during 1992-1995. Plant Pathology 46:112-125.

Marasas, W.F.O., Jaskiewicz, K., Venter, F.S. and van Schalkwyk, D.J. 1988. Fusarium moniliforme contamination of maize in human oesophageal cancer areas in Transkei. South African Medical Journal 74:110-14.

Pittet, A. 1998. Natural occurrence of mycotoxins in foods and feeds. An updated  review. Revue de Medecine Veterinaire 149:479-492.

Rabie, C.J., Kellerman, T.S., Krick, N.P.J., van der Westhuizen, G.C.A. and de Wet, P.J. 1985. Toxicity of Diplodia maydis in farm and laboratory animals. Food Chemistry and Wet  P.J. Toxicology 23:349-353.

SAS,1987. Proprietary Software Release 6.04. SAS Institute Inc., Cary, NC 27512-8000, U.S.A. Licenced to FH BINGEN FB LANDBAU, Site 19040001.

Schaafsma, A.W., Miller, J.D. , Savard, M.E.  and Ewing, R.J. 1993. Ear rot development and mycotoxin production in corn in relation to inoculation method, corn hybrid, and species of Fusarium. Canadian Journal of Plant Pathology 15:185-192.

Sutton, J.C. 1981. Epidemiology of wheat head blight and maize ear rot caused by Fusarium graminaerum. Canadian Journal of Plant Pathology 4:195-209.

Ullstrup, A.J. 1949. A method of artificial epidemics of Diplodia ear rot. Phytopathology 39:91-100.

Warfield, C.Y. and Davis, R.M. 1996. Importance of husk covering on the susceptibility of corn hybrids to Fusarium ear rot. Plant Disease  80:208-210.

Warren, H.L. 1978. Comparison of normal and high lysine maize inbreds for resistance to kernel rot caused by Fusarium moniliforme. Phytopathology 68:1331-1335.

©1999, African Crop Science Society