sary openings in them—but only because they limit the universal exchange of air in the underlying garments. For protection against the wet from without they are well suited, but they produce another wet on our skin by impeding evaporation. They may be used in wet weather, when accompanied with cold or wind, but never, though wet, when it is warm or calm.
Finally, I have to draw your attention to the relations which the materials of our clothes have to water, by which their functions are considerably altered. They are all hygroscopic; that means that they condense from the atmosphere a certain amount of water. This hydroscopic property, very different in different bodies, increases with the decrease in the temperature of the air, so that all of them condense more water at freezing-point than at higher temperatures. Partly, also, the relative, moisture of the air is of some influence, so that at 68° the hygroscopic body absorbs more water from an air nearly saturated than from a less moist air. As yet we do not know much about our clothing materials in this respect. I have made some preliminary researches, and have found unexpectedly great differences.
I took two equal pieces of flannel and of linen, as representatives of the two most important fabrics made of vegetable and animal fibres, and dried them at 212°, a temperature at which they lose all their hygroscopic water. I put them into well-closed boxes of known weight, and noted the weight of the two together. They were then exposed to the air in places of different temperature, and from time to time put back into the tin boxes, and the weights taken again. By this method it was not difficult to ascertain the relative quantities of hygroscopic water which the flannel and the linen had absorbed. These quantities are tabulated below, as they resulted from different localities, temperatures, and lengths of time, the weight of the linen and flannel being 1,000 grammes each: