Summary
When examining field samples collected from a river bed the routine process of counting the animals present is exceedingly laborious.
To shorten the amount of work involved the larger animals are separated off by sieving, forming a micro-sample of the small animals and then only a sub-sample, consisting of one eighth of the micro-sample by volume, is actually counted.
The sampling errors introduced by this time saving device are analysed and 95% confidence intervals obtained for the numbers in the micro-sample.
This is done for a wide range of practical conditions. In this particular study the method has been used with considerable success on samples taken from marginal vegetation, stony runs and sandy or muddy bottoms.
This paper is published with the joint permission of the South African Council for Scientific and Industrial Research and the Transvaal Provincial Administration.