Observations on the accuracy of point counting including a description of a new graticule

Abstract

The accuracy of point counting as a histometric technique depends very much on the number of points counted, and the number necessary varies greatly with both the level of accuracy required and the proportion the relevant tissue occupies in the section. It is shown that the number of "hits" is a more useful measure than the total number of points counted, and a simple rule derived for determining the number of hits necessary to achieve any necessary degree of accuracy. A method of progressive estimation of accuracy when attempting to measure the proportion of a tissue that is unevenly distributed is also described. The point-counting graticules commonly used, with 25 points, are inadequate with tissue proportions below 10%. A new graticule design, with 100 points optimally distributed, and usable as a 25-point array if required, is illustrated.