In 2000 and 2001 Orange River levels were higher than normal: associated serious outbreaks of blackfly had a substantial detrimental impact on the local economy. The poor control was attributed to the suspected development of larval resistance to temephos. A long-term solution to blackfly control, through the identification of a suitable replacement to temephos for use during high flow conditions, was proposed. This study, however, failed to identify or register a suitable larvicide for use during high flow conditions. Although permethrin was highly effective against blackfly larvae, it was rejected because of its detrimental impacts on non-target fauna. Various formulations of locally produced dry Bacillus thuringiensis var. israelensis (B.t.i.) were tested, but these were ineffective against blackflies. The study also confirmed that resistance to temephos has developed among Simulium chutteri in the middle and lower Orange River. The feasibility of \"reversing\" the resistance to temephos through the use of the synergist piperonyl butoxide (PBO) was investigated, but the results were not favourable. Furthermore, PBO was highly toxic to blackflies and non-target organisms, and was not recommended for further testing. This means that B.t.i. currently remains the only symptomatic measure of treatment currently applied. Although resistance to B.t.i. has not been reported for blackflies elsewhere in South Africa, there is a need to remain vigilant and to implement an operational strategy that minimizes the risks of resistance developing.

In its natural host, Bacillus thuringiensis, the insertion sequence IS231A is preferentially inserted into the terminal inverted repeats of the transposon Tn4430. Using a novel transposition assay, we demonstrate that the Tn4430 ends behave as insertion hot spots for IS231A in Escherichia coli. Sequence analysis reveals that IS231A insertion sites match the 5\'-GGG(N)5CCC-3\' consensus. However, this consensus is not the only determinant of IS231A insertion specificity. Although both Tn4430 ends have identical sequences, one is strongly preferred to the other and the orientation of insertion into this end is not random. We demonstrate that this preference is determined by the flanking regions of the site. These regions display a conserved periodic organization of their sequence which, by conferring anisotropic flexibility, would induce the DNA to bend in a roughly \'S\'-shaped structure centered on the target consensus. DNA conformation analysis by polyacrylamide gel electrophoresis indeed shows that the preferred target site of IS231A is flanked by DNA segments curved in opposite directions. We present a model in which DNA bendability and curvature would contribute to the positioning of IS231A transposase on the target DNA.