An alternative might be to consider adaptive sampling (e.g., see Thompson & Seber 1996), in which additional samples can be added when the target species/size is encountered. This can be of particular merit for surveying rare and patchily distributed organisms, but the approach also has its disadvantages. These include the introduction of biases into conventional estimators so new unbiased estimators are needed, and the potentially unending nature of sampling at particular sites where the target is repeatedly detected. The logistics of sampling are also less straightforward than conventional sampling at predetermined positions.