**Acknowledgements**

factors [115]. The concentration and severity of FUM produced by *Fusarium* spp. varies with meteorological conditions, genotype and location [19]. In general, favourable conditions for *F. verticillioides* infection include high temperatures [56], drought stress [56, 116] and insect damage stress [56]. A mathematical simulation of the growth of *F. graminearum* and *F. verticillioides* in maize ears was developed; however, the model only simulates fungal growth and not mycotoxin accumulation [117]. A preliminary model developed in the Philippines and Argentina identified four weather periods near silking as critical to FUM accumulation at harvest [19]. This model accounted for 82% of the variability of total FUM across all locations in 2 years of study, but did not consider meteorological conditions during grain maturation when FUM are synthesised.

A risk assessment model (FUMAgrain) developed for FUM contamination of maize grain in Italy gives an initial risk alert at the end of flowering based on meteorological conditions [118]. A second alert follows at kernel maturation following assessments of grain moisture, European corn borer damage and FUM synthesis risk. FUMAgrain could simulate FUM synthesis in maize accounting for 70% of the variation for calibration and 71% for validation. The importance of meteorological conditions at flowering and the growth of *F. verticillioides* and FUM synthesis during grain maturation was emphasised as the most important factors contributing to FUM contamination [118]. Another model consistently identified mean maximum temperature and minimum humidity as driving variables in the colonisation of maize kernels by fumonisin-producing *Fusarium* spp [99]. Furthermore, *Fusarium* colonisation of grain and fumonisins were related to prevailing weather conditions during early post-flowering and dough stage of grain development, respectively [99]. A prediction model using variables such as cultivar, climate, management practice, soil type, phenological stages of the host plant and pathogen variation would be advantages in identifying areas with potentially dangerous levels of fungal contamination and associated mycotoxin production, enabling them to imple-

Food and feed crops are consistently threatened by mycotoxigenic fungi and compound their infection by depositing toxic metabolites, including mycotoxins. Preharvest management of mycotoxin contamination is vital to maintaining contamination levels below economically feasible and legislated thresholds. Planting genotypes with enhanced host resistance is considered the most practical, affordable and environmentally sound method of controlling mycotoxigenic fungi and their mycotoxins. However, integrating resistant varieties with good agricultural practises such as crop rotation, chemical/biological control and other strategies that optimise plant production by minimising stressors may further reduce the risks associated with mycotoxin contamination. Resistance to mycotoxigenic fungi exists and has been identified in appropriate breeding materials but such resistance needs to be introduced in high-yielding and locally adapted hybrids. To date, conventional breeding has not been able to introgress disease and/or mycotoxin resistance into important staple crops like maize. Therefore, further research is required into factors with a greater efficacy to reduce mycotoxigenic fungi and

mycotoxins preharvest as resistant varieties are being developed.

ment mycotoxin management strategies.

48 Mycotoxins - Impact and Management Strategies

**3. Conclusion**

The South African Maize Trust and the National Research Foundation (NRF) of South Africa (Thuthuka; South Africa—Kenya Research Partnership Programme Bilateral); the MAIZE Competitive Grants Initiative, International Maize and Wheat Improvement Centre (CIMMYT), and CGIAR, the National Commission for Science, Technology and Innovation (NACOSTI) of Kenya; the Agricultural Research Council of South Africa are all acknowledged for funding.
