Abstract
As shown below, the point-centred quarter method described by Cottam and Curtis (1956) has been found consistently to measure trees whose average diameter is larger than that of the stand when there is a fair range of size classes present. If the area bounded by the right bisectors of the straight lines joining the centre of a tree to each of its neighbours is called the area potentially available (a.p.a.) of that tree (Brown, 1964), then by definition that tree will be the closest tree to any point within its a.p.a. Measurements of stands containing a range of size classes show that small trees tend to have smaller a.p.a.'s than large trees, even where there is no noticeable grouping of small trees. Thus, randomly located points have proportionately more chance of being located in the a.p.a.s of large trees than in those of small trees. In the point-centred quarter method, this effect is reduced because four trees are measured from each randomly located point, but it is still present, as shown by the following table:

This publication has 3 references indexed in Scilit: