The pathogenicity of 52 isolates from several fungus species was studied for the false spider mite Brevipalpus phoenicis. In addition, the main stages during the course of infection by Hirsutella thompsonii, by far the most virulent pathogen, were studied by means of light and electron microscopy. Adult mites were confined to arenas prepared with citrus leaves in acrylic dishes containing agar-water. Conidial suspensions containing 10(8) conidia/ml were applied, except for H. thompsonii, where a concentration of 10(7) conidia/ml was used. The H. thompsonii isolates caused higher mortality, with indices higher than 90%. Observations under the scanning electron microscope (SEM) were performed at 0, 6, 12, 24, 48, 72, and 120 h after application of a H. thompsonii suspension containing 10(7) conidia/ml. Twenty-four hours after inoculation, H. thompsonii conidia were observed attached to the mite's integument. The conidia germinated and penetrated through the base of the setae on the hysterosoma. Colonization occurred after 48 h, as evidenced by mortality. Conidiogenesis occurred after 120 h, with the development of mycelium and conidiophores emerging from the posterior and anterior parts of the mite.

The virulence of Hirsutella thompsonii (Fischer) to Brevipalpus phoenicis (Geijskes) was evaluated in laboratory, grown on complete and solid culture media (MC-S); complete and liquid culture media (MC-L); rice (APC) and powdered rice (APC-SM). Adults were confined to arenas prepared with citrus leaves in acrylic dishes containing water-agar. Conidial suspensions were prepared at different concentration (3.2 x 10(5) to 1 x 10(7) spores/ml) and applied on mites to establish the table curve-response on fourth day. For field evaluation, adults were maintained in arenas prepared with fruits which were placed in plants. In this test, four treatments were tried: H. thompsonii cultured on rice (APC) at two concentrations (20 kg/ha and 10 kg/ha), H. thompsonii produced by liquid fermentation (MC-L) (5 L/ha) and control (sterile water). Adult survival, number of eggs and nymphs per fruit were observed 10 and 20 days after the fungus application. The lowest LC25 value calculated was from pathogen produced in MC-S (1.9 x 10(5) conidia/ml).The LC25 values calculated to APC and APC-SM did not differ statistically. The LC25 values to MC-L and MC-S were 1.9 x 10(6) infective cells/ml and 2.2 x 10(5) conidia/ml. In the field, concentration and time to death differed between treatments and control. The applications resulted in reduction of adult survival and number of eggs.

Em especial, ao professor Dr. Sérgio Batista Alves pela confiança depositada em mim e pela amizade. Obrigada por todos os grandes e valiosos ensinamentos, mas em especial, aquele de que nunca podemos desistir perante as dificuldades da vida. São valores que sempre levarei comigo. Aos colegas do laboratório de Patologia e Controle Microbiano de Insetos, em especial à Solange Ap. Vieira, pela amizade e companheirismo. Aos professores do Departamento de Entomologia da Esalq/USP. Em especial, agradeço ao Dr. Ítalo Delalibera Jr. pelas sugestões e auxílio na elaboração e correção do abstract, ao Dr. Celso Omoto pelo convívio durante os períodos de monitoria, ao Dr. Sinval Silveira Neto pelo apoio e valiosos ensinamentos e à Dra. Marinéia de Lara Haddad e pelo auxílio no entendimento das análises estatísticas. Aos engenheiros agrônomos Dr. Rogério B. Lopes e Dr. Everaldo B. Alves, pela amizade e por todas as sugestões dadas durante o planejamento e execução deste trabalho. Ao colega Gabriel M. Mascarin pelo auxílio na realização das análises estatísticas e pela amizade. Ao colega Giuliano Pauli pela ajuda na realização dos experimentos de campo e amizade. À minha grande amiga Dra Tatiele A. B. Fenilli, por todos os anos de convivência, por estar ao meu lado nos momentos de alegria e de dificuldade.