I have said that the historic prejudice is essentially right and the catastrophic prejudice essentially wrong. Of course, there are exceptions. A man may be run over in the street for reasons that we cannot discover to be connected in any possible way with his previous history; a man gets a burn, gets a broken leg, is hit by a missile in an air raid over London or Paris, in ways that are essentially catastrophic. And yet even in the field of accidents, industrial accidents for instance, the more we study, the more we find that injuries are not wholly accidental. The whole of science is the attempt to prove that nothing is an accident, that everything comes out of previous causes. The percentage of accident in the so-called "accidental" injuries decreases as we study industrial accidents. (_a_) They happen at certain hours of the day more than at other hours of the day: if they were really accidental this would not be so. (_b_) They happen on certain days of the week, especially Mondays, for obvious reasons. (_c_) They happen especially to greenhorns, to the newcomers, who have not learned how to avoid them. One of the expenses incidental to hiring new help is the expense of accidents. Thus these events turn out to have a good deal of law and reason, a good deal in the history of the individual (alcoholism?), and the nature of the industrial process (speeding up?) which helps to explain them. By eliminating such causative factors, we may prevent some accidents. The remedies that we apply fit the type of trouble; _in so far as the trouble is accidental or catastrophic, the remedy is mechanical_; in so far as the thing is historical and continuous, the remedy cannot be mechanical. When a man breaks his leg we put on a splint; that is mechanical. But if he is in a low state of health and the fracture won't unite, we have to do something non-mechanical, physiological, psychological. We may have to get him into a different state of nutrition or even into a better state of mind before his tissues will heal. Our job, then, in taking histories--that is, in finding out how things happen that lead up to disease or misfortune--should begin by writing down _the thing for which the patient comes_--headache, cough, emaciation, poverty, desertion, unemployment. This is the "presenting symptom"; it should always be the first thing written down in our history, not in terms of medical diagnosis such as asthma or anemia, but in the form of a complaint. Our attempt is first to put that down, to get a starting-point, and then to weave that into a chain of evidence which we call a history. That history makes it possible to make a diagnosis and to plan treatment. _The network of events_