=The Loss of Heat by the Skin= is in three different ways. First, by _conduction_, when the skin comes in contact with anything cooler than itself; secondly, by _radiation_ into space; and thirdly, by _evaporation_ of the perspiration. The last cause produces a considerable reduction of temperature, even when the perspiration is not so abundant as to be visible, but is in the form of insensible perspiration. The losses by these different sources vary in amount; when one is increased, another is diminished, by way of compensation. Thus, in very cold weather, the amount of radiation and conduction of heat are increased; but evaporation greatly decreases, and the diminished loss of heat in this respect counter-balances in some degree the increased loss by radiation and conduction. When the external warmth is considerable, increased evaporation occurs; while when the weather is cold, the cutaneous arteries contract, and less blood goes to the skin, and so the loss of heat is diminished. In most climates, however, this action of the skin requires to be supplemented by some kind of clothing. =Requisites of Dress.=—1. The first and most important requirement is that _clothing should maintain a uniform and equable temperature_ in all parts of the body. In hot climates clothes are required in order to protect the body from external heat. In this country, they are required to prevent the too rapid escape of heat from the body. For both these purposes, dress must be of a non-conducting material, in order not to encourage transfer of heat into or from the surrounding atmosphere. The loss of heat by the skin may be prevented by interfering with radiation or conduction of heat, or with evaporation from its surface. Radiation of heat from the skin is prevented by clothing, the dress taking the place of the skin as a radiating surface. The amount of radiation from the dress will depend on the rapidity of conduction of heat from the skin. The amount of conduction and of radiation of heat will vary considerably with the _material_ and _colour_ of the dress. As regards =conductivity=, the two extremes are represented by linen and fur. It is found that if the conducting power for heat of linen = 100, then that of wool = 50 to 70. This partly explains why woollen goods are so much warmer than linen. We shall find that there is another explanation in the relative hygroscopic properties of the two materials. As regards =radiation= of heat, in one experiment it was found that while a piece of linen took 10½ minutes to cool, a corresponding piece of flannel took 11½ minutes.