Random signal variation as a cause of systematic microdensitometric error

A high noise/signal ratio in microdensitometry, owing, for example, to the use of a very small measuring spot or light at an extreme of the spectrum, decreases the precision (increases the variability) of results. This can be compensated for by making more measurements, but under some circumstances random signal variations can also cause systematic errors that are not so easily corrected. The present article explains how such errors can arise when measuring either very pale or very dense objects and suggests methods to detect and avoid them. The principles described are applicable to many types of microdensitometer, not only the Vickers flying-spot instruments used in this investigation.