Abstract
It is generally agreed that radium and X rays can ring about definite alterations in the blood picture. It has accordingly become the custom to rely on blood counts, and particularly on leucocyte counts, to given warning of danger in those whose work involves the risk of exposure to irradiation. the rationale of this procedure presupposes the possibility of making a sharp distinction between the normal and the abnormal. Now the leucocyte count is not constant, even in a healthy individual, and it is therefore extremely difficult to know whether minor variations have any pathological significance or not, especially as a decision usually has to be made without the help of clinical data. This difficulty raises the fundamental question of whether the leucocyte count lends itself to the detection of small abnormalities, and is a sufficiently delicate indicator to be used for a purpose such as this. The answer obviously depends to a large extent on the possibility of formulating a more precise definition of the normal than is usually necessary for diagnosis.