Predicting the Dosage of Fumigant Needed to Eradicate Insect Pests from Stored Products

Samples of wheat weighing 10 g and containing up to 100 pre-imaginal stages of Sitophilus granarius either 30 or 29-31 days old were exposed in a metal chamber to a concentration of 10 mg/l of methyl bromide for periods of between 1 and 6 h. Probit regression lines were fitted to the results for a series of generations of the same stock of weevils and differences of LD50were attributed to the difficulty of obtaining populations of preimaginal weevils in similar developmental stages. The estimate of LD99obtained by extrapolating these lines was compared for three generations with the values obtained by exposing 320-g samples containing about 2500 weevils. The probit regression lines for the large samples were obviously curved upwards so that high doses of fumigant gave larger kills than predicted by the extrapolated straight lines previously calculated. A dose of 52.5 mg h/l of methyl bromide killed all the 6000 weevils exposed to it, and nineteen survived out of more than 40 000 exposed to 50.1 mg h/l. The longevity and fertility of the surviving individuals was impaired so that, for instance, after exposure to 34.7 mg h/l, weevils yielded a generation only 5% the size of that yielded by controls. This next generation was fully fertile. When 97% of a population was killed by a fumigation at 41.7 mg h/l of methyl bromide, the LD50of this next generation was raised by about 25%. The observed curvature of the probit lines is consistent with that which would result from the mixture of two or more homogeneous populations with different susceptibilities regardless of the slopes of the probit lines of these populations.