4.3 Biological activity of pesticides

The purpose of applying a pesticide is to achieve a biological effect on the target pest. This effect is often described by scientists as a response and it is dose dependent - which usually means that the higher the dose, the more individuals in a population of organisms will be affected (and ultimately killed).

The population in question could be the target pests, but also unintentionally exposed human beings or other non-target organisms (beneficial or harmless animals and plants).

This is assessed in laboratory experiments called bioassays, where response is measured over a range of doses (different quantities of pesticide [AI] delivered individually to target organisms).

Fig. 4.3.a Origin of the sigmoid dose-response curve from the normal distribution curve accumulated on a 0-100% scale. The doses are on a logarithmic scale (without which the ‘S’ curve would be highly asymmetrical).

Fig. 4.3.b From this relationship, statistics such as the LD50 can be derived.

Described on a graph, the response is non-linear (i.e. not in a straight line), but usually in the form of a sigmoid (‘S’ shaped) curve - see Fig.4.2.

The first diagram shows that this sigmoid curve has been derived from the normal distribution - the bell-shaped curve that describes natural variability which is widespread in living organisms (e.g. the height of people, the weight of cocoa pods, the ability of animals to withstand drought).

By analysis of this dose response line, an estimate can be made of the median lethal dose or LD₅₀ of a pesticide to a group of organisms (i.e. the exact dose which would kill 50% of a test population of pests).

The LD₅₀ is derived from the dose-response curve and represents the dose at which 50% of test organisms (such as pests) are killed.

In practical experiments, there is often considerable variability in measured mortality at different dose rates and statistical methods (called logit or probit analyses) are used to determine LD₅₀ as accurately as possible.

Other levels of response can be used such as LD₁₀ and LD₉₀ (i.e. the 10% and 90% level of control respectively), but LD₅₀ is most commonly used since it represents the point at which the dose can be estimated most accurately.

In some bioassays, the pesticide is not administered directly to the target, so the true dose applied to a given individual is not known.

Different dosages (see section 6.1) may have been applied (e.g. different rates of surface deposit from various concentrations of pesticide mixtures), in which case the median lethal concentration or LC₅₀ will be quoted.