Search

Within-tree larval and progeny adult populations of Dendroctonus frontalis Zimm. attacking loblolly pine, Pinus taeda L., were investigated in order to define the distribution of larvae within sample units (100 cm2 disks of bark) and to describe the functional relationship of both larvae and progeny adults to the infested portion of the tree bole.Larvae and progeny adults were sampled over the infested bole of the 50 trees. Two separate measures of progeny adults were obtained. The first measure was taken just prior to emergence from the tree and the second measure was obtained by rearing the insect until emergence from the sample disks was completed.Larvae were found to be uniformly distributed within the total disk. The functional relationship of the larvae and the two measures of progeny adults to the normalized height on the infested bole was described by the two parameter nonlinear model y = Ax(1−x)eBx.Highest larval density along the infested portion of the tree bole occurred just below the center and tapered abruptly towards the base and gradually towards the top of the tree. The density of progeny adults on the other hand was greatest just above the center portion of the infested bole and tapered abruptly towards the top and gradually towards the base of the tree. Both measures of progeny adults produced curves of virtually the same shape. However, the density of progeny adults was substantially less in the samples which were reared indicating that mortality agents continued to act on the population until emergence from the tree.

The within-sample distributions of gallery length (GL) and egg (E) density as well as their functional relationships to the infested bole were studied in an epidemic population of the southern pine beetle in southeast Texas.A least-squares linear regression analysis through the origin showed an average of 1.59 eggs per centimeter of gallery. GL accounted for 81% of the variation in E and thus is useful for estimating egg numbers. The density of attacking adults is unsatisfactory for predicting E.GL and E are uniformly distributed within but not among the 100-cm2 sample disks at a given height. The functional relationship of both variables to the infested bole is adequately described by the model y = (A+Bx)exp(Cln(x−x2)), where y = GL or E per 100 cm2, x = the normalized height on the infested bole, and A, B, and C are parameters to be estimated for each set of data.