Abstract
Methods are presented for sampling and constructing life tables for Scolytus ventralis LeConte with the objective of comparing the histories of populations inhabiting individual trees. A bark area of 72 sq. in. was an effective sample unit. Within-tree variation was minimized by taking two sample units from two vertical strata at each sampling date. Trees were serially sampled several times during development of the insect.Population tables were constructed for each sampling date and life tables were developed from the complete series of population tables. The precision of the life table estimates varied between 10% and 40% of the mean and generally decreased as population density diminished during development. A sampling design is presented for increasing the precision of life tables, and methods are given for determining the number of trees required for estimating the mean density when studying populations occupying large areas.