The diamondback moth, Plutella xylostella (L.) (Lepidoptera: Plutellidae), is a major pest of Brassica crops in many parts of the world. Because of its ability to develop resistance to virtually all major groups of insecticides, including Bacillus thuringiensis Berliner (Bt), much attention has therefore been given to biological control using parasitoids. South Africa has an abundance of parasitoids attacking this pest. Cotesia plutellae (Kurdjumov) (Hymenoptera: Braconidae) is the most abundant larval parasitoid of P. xylostella in South Africa. In East Africa, its role in biological control of P. xylostella is insignificant, and the most abundant parasitoid there is Diadegma mollipla (Holmgren) (Hymenoptera: Ichneumonidae), a larvalpupal parasitoid. In South Africa, however, D. mollipla is out-competed by C. plutellae. Total parasitism of P. xylostella in East Africa rarely exceeds 15%, therefore there is a need to introduce more effective and heat-tolerant parasitoids of P. xylostella to that region. This study was therefore initiated to examine the potential of C. plutellae and D. mollipla as biological control agents by studying certain aspects of their biology in the laboratory, as well as the suitability of C. plutellae for introduction into East Africa. Biological aspects studied were: (i) host instar preference, fecundity, and searching efficiency of C. plutellae and D. mollipla at different host and parasitoid densities; (ii) effects of temperature on parasitism of P. xylostella by C. plutellae and D. mollipla, and on their rates of development and emergence. In addition, the role of parasitoids in controlling P. xylostella on unsprayed cabbage plots was investigated. Both C. plutellae and D. mollipla preferred to attack second and third instar hosts than fourth instars in choice and no-choice tests. However, D. mollipla attacked more fourth instar hosts than C. plutellae. Cotesia plutellae laid mainly female eggs in second and third instar hosts than in fourth instars, whereas D. mollipla laid more female eggs in fourth instar hosts than in second and third instar hosts. Diadegma mollipla was more fecund [82.57 ± 32.87, (mean ± s.d.) than C. plutellae (42.13 ± 12.2), and was long lived (7.13 ± 3.69 days) compared to the latter (5.23 ± 2.7 days). An increase in host density resulted in the reduction in the area of discovery (a) and the killing power (K) for both parasitoids. No significant differences were detected between the searching efficiency (t = -1.42NS, d.f. = 48, P < 0.001) of the two parasitoids. An increase in parasitoid density also resulted in a decline in searching efficiency, but not the killing power, of both parasitoids. Cotesia plutellae completed development at all temperatures tested (21-33°C), whereas D. mollipla completed development at temperatures from 18-30°C, and C. plutellae had a lower threshold for development (8.14°C) compared to D. mollipla (10.23°C). At all tested temperatures, the generation time for C. plutellae was shorter compared to D. mollipla. The possible reasons for the dominance of C. plutellae over D. mollipla in the field are: shorter generation time, high production of female progeny in younger hosts, low interference among searching females, and relatively wide thermal tolerance. The role of parasitoids in regulating diamondback moth, Plutella xylostella (L.) (Lepidoptera: Plutellidae), populations was studied for two years (February 2000-January 2002) on unsprayed cabbage fields at Brits, North West Province, South Africa. Cabbage seedlings were transplanted in three consecutive times each year. Cabbage infestations by P. xylostella larvae and pupae, and their parasitoids, were monitored at weekly intervals. The flight activity of P. xylostella male moths was monitored using sex-pheromone traps. Trap catches indicated that the moths were active throughout the year. The flight activity of the moths corresponded with infestations on the crop. Trap catches and infestation levels were generally low from December to August and high from September to November. Eight hymenopteran parasitoids were reared: the larval parasitoids Cotesia plutellae (Kurdjumov) (Braconidae) and Apanteles halfordi (Ullyett) (Braconidae); the larval-pupal parasitoids Oomyzus sokolowskii (Kurdjumov) (Eulophidae) and Diadegma mollipla (Holmgren)(Ichneumonidae); the pupal parasitoid Diadromus collaris (Gravenhorst) (Ichneumonidae); and the hyperparasitoids Eurytoma sp. (Eurytomidae), Mesochorus sp. (Ichneumonidae), and Pteromalus sp. (Pteromalidae). Cotesia plutellae was the most abundant parasitoid of P. xylostella followed by O. sokolowskii, D. collaris, A. halfordi and D. mollipla. Parasitism of P. xylostella larvae was high reaching 100% on several occasions during late spring to end of autumn (November-May) each year. Parasitism was lower (<50%) in winter and early spring (June-September). Apanteles halfordi was absent in winter but re-appeared in spring. Parasitism of P. xylostella pupae by D. collaris was high only during spring (September-October). Hyperparasitism was generally low except when primary parasitoids were abundant in spring (September-November) and summer (December-February) when up to 25% of P. xylostella larvae and C. plutellae cocoons yielded hyperparasitoids. The role of other biotic and abiotic mortality factors on the population dynamics of P. xylostella is discussed.

The susceptibility of a population of diamondback moth, Plutella xvlostella L (Lepidoptera: Plutellidae), collected from Montgomery County, Virginia, and its ability to develop resistance to Bacillus thuringiensis was evaluated. The field population of diamondback moths was found to be susceptible to B. thuringiensis. Selection pressure at 153 I.U. of B. thuringiensis endotoxin per ml for nine generations did not cause any significant difference in mean mortality of third instar diamondback moths although there was a trend towards lower mortality, there was significant negative linear correlation between generation and mean mortality of diamondback moth larvae (P = 0.003, r²= 0.73). LC₅₀ increased from 264 I.U. of B. thuringiensis endotoxin per ml in generation I to 514 LU/ml in generation IX. The interrelationship of B. thuringiensis to diamondback moth and its primary parasitoid, Diadegma insulare Cress (Hymenoptera. Ichneumonidae), was studied by determining: the differential response of third instar diamondback moth, parasitized and unparasitized, to B. thuringiensis, and the ability of D. insulare to discriminate between B. thuringiensis-treated and untreated hosts. There was no significant difference (P > 0.05) between mean mortality of parasitized and unparasitized larvae at each of the three concentrations consisting of 154, 334, and 2237 I.U. of B. thuringiensis endotoxin per ml. The regressions of the response of parasitized and unparasitized larvae, however, were highly significant (P = 0.0001). The LC₅₀s of parasitized versus unparasitized larvae were 373 and 176 I.U./ml B. thuringiensis endotoxin, respectively. Female D. insulare did not discriminate between B. thuringiensis-treated and untreated hosts. The percentage of D. insulare females emerging from B. thuringiensis-treated larvae (41.4%) was not significantly different from that of untreated larvae (32.0 %). Mean mortality of third instar diamondback moth subjected to B. thuringiensis endotoxin at 153, 334, and 2237 I.U./ml were not significantly different at temperatures of 15 and 20 °C, but were significantly lower than that at 30 °C. The effects of B. thuringiensis endotoxin residues on leaves under room conditions [27 ± 1 °C, RH 27 ± 72%, and 8:16 (L:D)] were not significantly different at 2 and 192 hours after treatment. / Master of Science