Abstract
A sample of size N is drawn at random from a population of animals of various species. Methods are given for estimating, knowing only the contents of this sample, the number of species which will be represented r times in a second sample of size λN; these also enable us to estimate the number of different species and the proportion of the whole population represented in the second sample. A formula is found for the variance of the estimate; when A>2, this variance becomes in general very large, so that the estimate is useless without some modification. This difficulty can be partly overcome, at least for AOur Mutual Friend, to an entomological sample, and to a sample of nouns from Macaulay's essay on Bacon.

This publication has 0 references indexed in Scilit: