Only when a phenomenon includes conditions as yet undefined, can we compile; we must learn, therefore, that we compile statistics only when we cannot possibly help it.
Translated by Henry Copley Greene, An Introduction to the Study of Experimental Medicine