Methods of Estimating the Population of Insects in a Field

COTTNTS on the occurrence of an insect were seoured to make clear a valid and efficient method of estimating the population of the insect in an area. The theory of sampling, as developed by Neyman (1934), was applied to this problem. It was desired to know: first, what form observations should take in sampling; secondly, how good are the results of stratification, or control of regional varia-bility; and, thirdly, how accuracy varies when various fractions of the total area are sampled. The present study should supply a general method of sampling to be applied in experimental work or in surveys. Details in connexion with the method would, presumably, vary according to the insect, to the type of crop under investigation, to the number and size of samples possible, and to the importance of damage to the crop. The problem investigated was that of estimating the total number of insects in a single field. This problem differs to some extent from that of estimating the population in an area such as a county. REVIEW OF LITERATURE A considerable amount of investigation on the best method for sampling in agronomic work has been carried out. Some of that work, pertinent to the present problem, is disoussed below with two investigations on the technique of sampling for insects. The usage of Wishart Clapham (1929) may first be noted. To these workers unite are the ultimate parte of a sample, that is, the smallest area from which yield has been examined; sampling-units are the parte of a sample which are located independently and at random within the area to be sampled. Each may consist of one or many units; a sample is the aggregate of sampling-unite taken from the area. Clapham (1929) made a study of various methods of sampling cereals from a plot. This work showed, first, systematic arrangement of sampling-units to give an invalid estimate of chance variability and so random drawing to be necessary, secondly, the variability of estimates to be much smaller when samples were drawn from within subplots than when drawn from the plot in general, and thirdly, drawing throughout the plot to be superior to drawing