There is a physical basis for all of these standard catastrophes, but it is much slenderer than is usually supposed. For instance, almost every one can tell you how friends of theirs have "brought on congestion of the lungs," or pneumonia, by going without an overcoat on a winter day, or breaking through the ice when skating, or even by getting their feet wet and not changing their stockings; and this single dramatic instance has firmly convinced them that the chief cause of "lung fever" is a chill or a wetting. Yet when we come to tabulate long series of causes, rising into thousands, we find that the percentage in which even the patients themselves attribute the disease to exposure, or a chill, sinks to a surprisingly small amount. For instance, in the largest series collected with this point in mind, that of Musser and Norris, out of forty-two hundred cases only seventeen per cent gave a history of exposure and "catching cold"; and the smaller series range from ten to fifteen per cent. So that, even in the face of the returns, not more than one-fifth of all cases of pneumonia can reasonably be attributed to chill. And when we further remember that under this heading of exposure and "catching cold" are included many mere coincidences and the chilly sensations attending the beginning of those milder infections which we term "common colds," it is probable that even this small percentage could be reduced one-half. Indeed, most cautious investigators of the question have expressed themselves to this effect. This harmonizes with a number of obstinate facts which have long proved stumbling-blocks in the way of the theory of exposure as a cause of pneumonia. One of the classic ones was that, during Napoleon's frightful retreat from Moscow in the dead of winter, while his wretched soldiers died by thousands of frost-bite and starvation, exceedingly little pneumonia developed among them. Another was that, as we have already seen with colds, instead of being commoner and more frequent in the extreme Northern climate and on the borders of the Arctic Zone, pneumonia is almost unknown there. Of course, given the presence of the germ, prolonged exposure to cold may depress the vital powers sufficiently to permit an attack to develop.