Abstract
Estimates of whitefringed weevil (Graphognathus leucoloma (Boheman)) population density which included all life-history stages present were made in late spring and early summer (October–December), when eggs and larvae weighing less than 3 mg were absent. Analysis of 28 whitefringed weevil population samples from 0.2 ha study plots containing a wide range of population densities revealed that division of the plots into 10 equal-sized strata and proportional allocation of sampling units (10-cm-diameter soil cores) among strata produced small but consistent gains in precision. The relationships of sample variance to mean (s 2 = 1.905 m 1.20) and mean crowding to mean density (m* = 0.39 + 1.37 m) showed that whitefringed weevil populations have an aggregated distribution. These relationships were used to determine the number of sampling units necessary to attain specified levels of precision at different population densities.