PENETRATION OF SKIN--DEAD AND ALIVE 239 wetted. The wall of the sebaceous gland is dominantly oil permeable and the applied "oil" will reach the cutis vera and the lymphatic system, where it can then eventually enter the blood stream. The other route depends on the ability of the oil to spread over a water- wet surface. If the protein be dry, then the oil, while able to wet the surface, will tend to pile up as a layer of sensible thickness and it will not be able to flow through interstices of infinitely small dimensions, especially if the surface is covered with a film of adsorbed air and the oil is viscous. Under these conditions the oil would find great difficulty in passing through the top layers of the epidermis. On the other hand, massage would be of considerable help, but actual contact with the protein, because of the air surface, would not be at all good unless the oily material was of low viscosity and low surface tension. Although air-dry keratin contains 11 or more per cent of moisture, this water is "bound" and, indeed, is defined as that water which almost seems to be part and parcel of the protein molecule. It is, by definition, not free to dissolve any water-soluble material. In the case of gelatin, it has been shown by various methods that the ':bound" water is of the order of 30 per cent calculated on the total weight, or 1.00 gm. of anhydrous gelatin can bind 0.50 gm. of water. 7 When one realises that hair can only imbibe a total of about 33 per cent of water when in equilibrium with water-saturated air (100 per cent R.H.) it is quite certain that no unbound water is available under conditions where the skin surface is free of sweat. It is clear that the use of water-free oils or vehicles is not a satisfactory method of inunction if penetration is to be accomplished. This is all the more appreciated if the literature is surveyed. Time and again, the water- containing bases produce best penetration of the intact skin, and it was certainly instanced in the experiment already described when using white oil containing an ester of nicotinic acid. (In this instance it was deemed undesirable to use a water-containing base since the ester in the course of time hydrolyses and the preparation loses much of its potency). If then it is necessary to employ a water-free base, the epidermal surface must be anointed with water if penetration of the epidermis is to be achieved. It should be noted that the previous statements refer to water-containing bases in which the "oil" is the continuous phase such as are characterised by hydrous lanolin-containing fats. If these are rubbed on the skin and presuming that the emulsion has been correctly made, they should not throw out water as visible globules. What happens is that as the emulsion is spread as a thin film the water is emitted as vapour through the oil. For example, if a beeswax-mineral oil emulsion be made under conditions which produce a water-in-oil system and this is examined under the microscope as a thin film open to the air, the water globules will be seen to collapse slowly and

240 JOURNAL OF THE SOCIETY OF COSMETIC CHEMISTS crinkle, and eventually disappear, but to leave behind the remnants of the carapace. If the water is deeply coloured with a water-soluble dye and the emulsion applied to the skin, the dye will not colour the skin although the skin will have absorbed water. Because the dye is not soluble in oil it simply can never arrive at the skin surface. This means that oil-insoluble substances are quite ineffective in an oil base. Arising from this, one can consider the mechanism when the base is an emulsion with water as the continuous medium. It is fairly easy to make a "cold cream" so that it is miscible both with oil and with water. When such a cream is anointed on to the skin it will immediately invert into water- in-oil emulsion and it will then func!ion as a reservoir supplying water as vapour to the epidermis, but practically none of the water-soluble substances will reach the skin. If, however, the emulsion consists of fine droplets of oil dispersed by means of a soap, rubbing on the skin will decrease the stability of the emul- sion, especially if the surface be acid, and bring about a continuous but slow coalescence of the oil droplets near the skin surface. The system formed inverts and takes water from the skin surface, the oil continuing to spread over the damp surface of the skin. This oil film being thin can penetrate and function almost like an oleophilic material, passing by diffusion through the oleophilic portions of the mosaic of cellular membranes. Obverse creams (i.e., oil-in-water) which are stable in contact with skin can be employed for holding water-soluble substances. Because they do not invert, they will allow liquid water to remain in direct contact and, provided the water-soluble substance does not set up an electrical potential at the stratum lucidum, it will diffuse into the skin. If the osmotic pressure is greater than that of the serum some delay in penetration will obviously occur. Many excellent emulsions may be made of this character in which anionic detergents are employed. Because these substances are little affected by acids, breaking of the emulsion is at a minimum so that the chances of penetration of oil-soluble substances are somewhat remote. It is true, of course, that some chemi-sorption of the anion will occur, but in the presence of the relatively high concentration of the sulphated alcohol there will be little or no precipitation of the oil globules. Lately, non-ionic emulsifiers have come much to the front. They have many features which make them useful. If they are used to produce water- in-off emulsions, they have a great value in assisting penetration provided always that there is some free water present in the epidermis. They them- selves should be able to supply the necessary water. On the other hand, if used to produce oil-in-water dispersions, they will be almost useless as vehicles for cosmetic use. A good example of this can be demonstrated by dispersing