This theory of causation, confessedly provisional and imperfect as it is, helps us to harmonize the other known facts about the disease; it has already greatly improved our treatment and given us a foothold for attacking the problem of prevention. For instance, it has long been known that rheumatism was very apt to follow tonsillitis or other forms of sore throat; indeed, many of the earlier authorities put down tonsillitis as one of the great group of "rheumatic" disturbances, and persons of rheumatic family tendency were supposed to have tonsillitis in childhood and rheumatism in later life. Not more than ten or fifteen per cent of all cases gave a history of tonsillitis; but since we have broadened our conception of infection and begun to inquire, not merely for symptoms of tonsillitis, but also for those of influenza, "common colds," measles, whooping-cough, and the like, we reach the most significant result of finding that forty to sixty per cent of our cases of rheumatism have been preceded, anywhere from one to three weeks before, by an attack of some sort of "cold," sore throat, catarrhal fever, cough, bronchitis, or other group of disturbances due to a mild infection. Further, it has long been notorious that when a rheumatic individual "catches cold" it is exceedingly apt to "settle in the joints," and, if these cases happen to come under the eye of a physician, they are recognized as secondary attacks of true rheumatism. In other words, the "cold" may simply be a second dose of the same germ which caused the primary attack of rheumatism. This brings us to the widespread article of popular belief that rheumatism is most commonly due to cold, exposure, chill, or damp. Much of this is found on investigation to be due to the well-known historic confusion between "cold," in the sense of exposure to cold air, and "cold," in the sense of a catarrh or influenza, with running at the nose, coughing, sore throat, and fever, a group of symptoms now clearly recognized to be due to an infection. In short, the vast majority of common colds are unmistakably infections, and spread from one victim to another, and this is the type of "cold" which causes the majority of rheumatic attacks. The chill, which any one who is "coming down" with a cold experiences, and usually refers to a draft or a cold room, is, in nine cases out of ten, the rigor which precedes the fever, and has nothing whatever to do with the external temperature. The large majority of our cases of rheumatism can give no clear or convincing history of exposure to wet, cold, or damp. But popular impression is seldom entirely mistaken, and there can be no question that, given the presence of the infectious germ, a prolonged exposure to cold, and particularly to wet, will often prove to be the last straw which will break down the patient's power of resistance, and determine an attack of rheumatism.