In the second phase of history-taking, then, which begins after we have listened appreciatively but quietly to the patient's own version--usually catastrophic and full of fanciful theories--we lead him by questions (but not by "leading questions") along the paths which will open up a full view of the trouble, medical or social, which has been suggested to us by the patient's first statements. Suppose, for instance, one happened to know of an extraordinarily rare but curable disease, one symptom of which the patient had mentioned, "My hair comes out by handfuls." One would go on to ask, "Do you feel warmer or colder than usual this winter?" Then, "The expression of your face is not notably changed, is it, so that your friends comment on it?" "Is your skin drier or moister than usual?" "Does your tongue bother you in any way?" "Is your mind more or less active than usual?" Thus one would confirm or refute the suggestion of the disease called myxoedema, a suggestion which was given to us by the patient's first complaint--rapid loss of hair. Given one symptom in a known group, one can trace out the others as the anatomist who finds a single fishbone may be able to reconstruct imaginatively the whole fish. I said just now that we must not ask "leading questions." If we do, we can make a patient of a very suggestible type of mind say anything. If you ask him whether he has any symptom whatsoever he may obligingly say "yes." The way to avoid this is to put our questions in the negative: "You have no headache at all, have you?" "You do not cough?" "You never spit blood?" By these negatives we can get at the positive symptoms if they are present. Schedules of questions to be used in history-taking may be medical or social. Some of the social question-lists are suggested in later chapters of this book. A masterly account of social questioning is contained in Miss Mary E. Richmond's "Social Diagnosis" (published by the Survey Associates, New York, 1917). I wish now to illustrate the methods to be used by social workers in questioning patients about their symptoms so as to assist the doctor in his diagnosis. _Pain: How long?_ For a day, a month, a year, six years? Very chronic pains are seldom serious but seldom curable. Headache that has lasted years either has no cause known to medical science, or else it means neurasthenia. In either event it is apt to be stubborn. A headache that has lasted only a day, and did not occur before, cannot possibly be due to migraine. This suggests how the length of time that a pain has lasted is very important in diagnosis. The patient will often say, "I have always had it"; but to this we should oppose a pretty strong cross-examination. The patient usually means that he has had it off and on throughout an indefinite period. We ask him then, "When did you first have it?" and then, "How much of the time--half the time, a quarter of the time, for one day a week or one day a month?"