456 JOURNAL OF THE SOCIETY OF COSMETIC CHEMISTS Powers and Fox (5) studied the effect of various cosmetic ingredients' (including glycerol) on the rate of moisture loss from the skin in an in vivol situation. In their experiments, small desiccator cups containing silica• gel were strapped on the skin, and the rate of water loss measured from an l untreated site in comparison to a site rubbed with anhydrous glycerol. I The authors found that glycerol increased the rate at which water was lostl from the skin. They conclude that glycerol accelerates the diffusion ofl water from moist dermal tissue to the surface, and that this accounts forl the beneficial effect of humectants on skin during dry weather (6). It has been shown that the rate at which moisture diffuses to the surfacel of the skin is governed by a rate regulating barrier membrane (1). Ill Powers and Fox's interpretation of their data is correct, one must concludel that glycerol increases the rate of water diffusion through this barrierl membrane. This does not appear to be a likely action of glycerol. Another possible explanation for these results is that during the processl of rubbing the anhydrous glycerol on the skin, it picked up water from thel air and the weight changes really measured water lost by the skin plus l water lost by the glycerol. An experiment was set up to determine how much water could be taken up from the air by 4.9 cm. 2 surface area of glycerol (equivalent to the sur- face area of the desiccator cups used by Powers and Fox) in short periods of time. Exposing 4.9 cm. 2 of anhydrous glycerol to 65 per cent R.H. for one minute resulted in a moisture uptake of 3.8 mg., 3.0 mg. and 4.5 rag. on successive trials. In a second experiment, about 30 rag. of anhydrous glycerol was spread over an area of 4.9 cm. 2 using a glass stirring rod in a manner analogous to that described by the authors. (This amount ofglycerol corresponds to the amount of glycerol Powers and Fox had covering their one-inch circular area.) After spreading the glycerol for one minute at 65 per cent R.H., the moisture uptake was determined. The results were 3.9 mg. and 3.5 mg. on successive trials. In both experiments, short, wide-mouthed, glass weighing bottles were used in which a circular area of 4.9 cm. 2 was inscribed on the bottom. In the first experiment, glycerol was dropped into the vessel until it formed a circular drop equal to the inscribed area. In the second experiment, 30 rag. of glycerol was placed in the center of the inscribed area and spread with a micro glass stirring rod. These values are of the right magnitude to explain the increased moisture loss observed by Powers and Fox from the glycerol treated site. Thus, they report increased moisture loss on one subject of 3.6 rag., 4.4 rag. and 3.6 rag. during three successive hourly measurements, and on a second subject of 1.7 rag., 1.5 mg., 1.4 rag. and 1.9 rag. during four successive hourly measurements. Since the desiccant cups were removed each hour

ROLE OF GLYCEROL IN SKIN HYDRATION 457 for weighing, this probably gave the glycerol a chance to rehydrate by taking up water from the air. Thus, it is indeed possible that the results obtained by Powers and Fox can be fully explained by water loss from the hydrated glycerol rather than an increase in water loss from the skin. It has recently been reported by Flesch (4) that glycerol plus callus exhibits unusual humectant properties, over and above those which would be predicted from the humectancy of glycerol and callus taken separately. These results have led to the speculation that glycerol penetrates between the keratiu fibers of the horny layer and facilitates the absorption of water. Flesch studied the moisture uptake of glycerol, callus and callus-glycerol mixtures when exposed to 100 per cent R.H. Under such conditions, minor fluctuations in temperature can cause water to condense. Furthermore, the attainment of equilibrium conditions with glycerol is virtually impossi- ble since glycerol is miscible with water in all proportions. In any case, the author's results actually show that the rate at which a mixture of glycerol and callus picks up moisture is greater than would be predicted from the rate at which callus alone and glycerol alone pick up moisture. The fact that Flesch is dealing with rates rather than equilibrium con- ditions is most apparent when one considers that his data show that at the conclusion of this experiment, glycerol had taken up about 130 per cent of its own weight in moisture. At a R.H. in the high 90's, one would expect glycerol to take up well over 900 per cent of its own weight in water (5). When dealing with rates of water uptake, such things as surface area of glycerol and callus, and size and shape of the vessel in which the glycerol and callus are placed become critical. One can readily see how a potentia- tion in rate of water uptake could be achieved in the glycerol callus system if the surface area of the glycerol were increased due to the presence of callus. With this in mind, an experiment was set up to test the hypothesis that the rate of water uptake of a glycerol-callus system was elevated above that which would be predicted by callus and glycerol alone, but that the equilibrium uptake was merely a summation of the water taken up by glycerol plus that taken up by the callus. Into three glass vials, 50 mg. of ground callus (80 mesh) was weighed. To one of these vials was added 50 mg. of anhydrous glycerol and to a second 250 mg. of anhydrous glycerol was added. Into two other identical vials, about 50 mg. and 250 mg. of anhydrous glycerol was weighed. All five vials were then placed into a constant humidity chamber at 97 per cent R.H. and weighed at regular intervals. From the amounts of water taken up by the 50 mg. sample of callus alone and the 50 mg. and 250 rag. samples of glycerol alone, calculations were made as to how much water should be taken up by the two glycerol-callus mixtures, assuming a strictly addi-