Erynia neoaphidis Remaudiere and Hennebert (Zygomycetes: Entomophthorales) is an obligate pathogen of invertebrates, especially aphids, and has therefore been studied as a possible biological control agent for a number of years. However, a number of important physiological and ecological questions regarding optimal conditions for conidial production and transmission 0 f the fungus through an aphid population had to be answered. This thesis investigated some of these aspects. Solid and liquid media were used to culture the fungus, and E. neoaphidis was cultured on a fully defined medium for the first time. A sporulation monitor and digital image analysis was used to quantify conidial production from E. neoaphidis biomass produced in vivo and in vitro. This was a completely novel method and is useful for gathering data on large numbers of conidia, 50 that size distributions can be constructed and the physiological status of the conidia inferred from this. E. neoaphidis infected aphid cadavers produced more, smaller conidia when grown in vitro. Biomass harvested from exponential growth phase in fed batch culture produced significantly more conidia than biomass harvested from any other growth phase although further work on the nutritional requirements of E. neoaphidis in vitro is required. The duration of the conidial discharge was also greatest from biomass harvested at the exponential phase and therefore. biomass harvested from the exponential phase should be used if the fungus is to be applied as a control agent. E. neoaphidis biomass kept at low humidity during simulated winter conditions produced infective conidia after 24 weeks, indicating that mycosed cadavers may act as a reservoir to infect the next season's hosts. Pesticides adversely affected the growth and production of conidia by E. neoaphidis, with herbicides having the least deleterious effects, and therefore being most compatible in an integrated pest management program. Laboratory and field studies were used to assess the transmission of E. neoaphidis through aphid populations. Position of the inoculum on the host plant affected the primary transmission of the fungus through aphid populations in the laboratory and in the field, and secondary transmission of the fungus in the laboratory. It is therefore important to apply the fungus to where it will maximally spread. There was some evidence for effects of host and inoculum density on the transmission of the fungus, especially in the laboratory, indicating that, in practice, the fungus is unlikely to spread rapidly through low densities of aphids and therefore to achieve control of such populations, a high inoculum density may be required. There was also very Iittle transmission of the fungus via aphid vectors to susceptible aphid populations on different host, although as a general observation, vectoring of conidia by the wind may be very important. The smaller conidia produced by in vivo biomass may be vectored more easily by wind than the large conidia produced in vitro.

The optimisation of spore production by the potential fungal biological control agent for aphids, Erynia neoaphidis Remaudiere and Hennebert (Zygomycetes: Entomophthoraceae) was studied. The fungus was able to grow in semi-defined Frynia medium (SDEM) containing glucose, yeast extract, mycological peptone, and 0.02% oleic acid buffered to a pH 6. Oleic acid was fungicidal at 0.1 % (v/v) while 0.02% (v/v) oleic acid was the optimum for radial grovvth. Plugs cut 5-10 mm from the margin ofa colony produced more conidia than plugs cut 13-20 mm from the colony margin. Renewed grovvth continued through two subcultures on solid SDEM lacking yeast extract (SDEML YE), and SDEM lacking mycological peptone (SDEMLMP). The continued growth was attributed to the carry over of nutrient in the inoculum. Growth was supported on SDEMNH4S04 when ammonium sulphate was used as the nitrogen source instead of mycological peptone suggesting that the fungus could obtain the growth factors it required from yeast extract. When chitin was added to SDEM in insoluble powder form instead ofglucose (SDEMC 1 & SDEMC2), the absence of a clearing zone around the developing colony suggested that chitin was not metabolised by E. neoaphidis. Biomass grown on SEMA and on SDEMDG (containing double the original concentration ofglucose 3 2grl), resulted in production of fewer conidia oflarger volume compared to SDEMDMP containing double and half the original concentration of mycological peptone (SDEMHP), SDEM containing halfthe original concentration ofglucose (SDEMHG). Increasing the glucose to double the original concentration resulted to an increase in biomass. Erynia neoaphidis grown on aphid cadavers produced many, smaller conidia. Mycelial mats harvested from biomass grown in fed-batch liquid fermenter culture in SDEMDG at the end ofthe exponential phase and placed on water agar discharged conidia at a rate of 6,700 conidia mm -2 h-1which persisted for approximately 3 days. When E. neoaphidis was subcultured onto SDEM from SEMA medium, the colony growth rate increased on the second subculture on SDEM where more lipases and aminopeptidases were detected at higher concentrations using the API ZYM system. This shows that attenuation might have taken place by either a phenotypic or genotypic (eg mutation) change or both when E. neoaphidis was grown on SDEM from SEMA medium. Growth in GASP medium resulted in the production of more biomass and a delay in the onset of decline phase compared to cultures grown in SDEM. Fewer enzymes were detected at a lower concentration in cultures grown in GASP compared to cultures grown in SDEM, this difference might be more likely to relate to the balance of nutrients and the fact that GASP medium is more similar in composition to the nutrients found in the haemocoel of an aphid. Based on this research. It is recommend that E. neoaphidis be grown in SDEM liquid cultures containing 32 grl glucose instead of 16 grl glucose. Biomass for field applications should be harvested at the end ofthe exponential growth and mycelial mats made. The mycelial mats should be maintained at high relative humidity and can be expected to discharge conidia for 3 days.