14 JOURNAL OF THE SOCIETY OF COSMETIC CHEMISTS rium hygroscopicity is incorrect. We will try to show how the rate of loss of water from a humectant is more important for this purpose than the amount of water held at equilibrium. The amount of humectant used in cosmetic creams is seldom as much as 10% of the total weight of the formula, whereas, the water Figure 7 Time - linu•ee ! ¾eigh• Lose l'ro• S_olutione When •enefer•.e.d rro• an arm. of •8• U.H. to an eta. of • U•H. Glycerin .... Serbitel -- 0 o -10 o -15 \\\\\• MoiSt Loss from •olutl• when •anoforrod from an arm. of 7• R.H. •o an atto. of 6• R.H. 6lyeoran .... SerbStel Fibre 8 content varies from 4:0 to 80%. Considering the most favorable set of conditions in which the amount of humectant is high and the amount of water is low, the concentration of humectant in the aqueous phase is seldom more than 20%. All organic humectant solutions at this concen- tration (20%) will lose water under normal atmospheric conditions since the relative humidity would have to be greater than 95% for no water loss to occur. This means that if the cap is left off sufficiently long, no matter which of the standard hu- mectants that you employ, your cream will lose a major amount of water until equilibrium conditions are attained. Average temperature and humidity conditions in New York (2) for example, during June and July, are 74øF. and 70% R.H. while during December and January the average conditions are 33øF. and 70% R.H. The important difference in summer and winter seasons is that in the summer in- door and outdoor conditions are similar, while in the winter the in- door temperature is about 74øF. (like the summer outdoor tempera- ture) resulting in a relative humidity of 17% indoors. Thus, the relative humidity conditions to which cos- metics are exposed are seldom greater than 70% R.H. and average a great deal less for most of the year. Similar conditions exist throughout most of the East. Even under the most favorable high humidity con- ditions, for the aqueous phase to be in equilibrium with the atmosphere it would of necessity contain almost three parts of humectant to every part of water. It is doubtful if products made with such high con- centrations of polyhydric alcohols or other humectants would be satis- factory from a cosmetic standpoint regardless of the increased cost.

HYGROSCOPIC AGENTS AND THEIR USE IN COSMETICS 15 From these considerations you will see that it is not practical to prepare cosmetics that are in equilibrium with the surrounding atmosphere. Also, it is apparent that existing creams will lose water and that the amounts of humectant that are added do not alter the equilibrium hygroscopicity of the creams ap- preciably. However, it is desirable to retard the loss of water and this immediately becomes a study of rate rather than of equilibrium. It is most interesting that dynamic hy- groscopicity is apparently not a func- tion of equilibrium moisture holding power of humectants. This effect has been illustrated in our recent study of the hygroscopic properties of cosmetic creams. The purpose in these studies was not to determine the equilibrium hygro- scopicity of the cream as a whole but to note the rate of loss of water under standard conditions. Loss in weight was determined on a large number of samples at one time. In our equipment, Fig. 9, as many as 77 separate samples could be ob- served at one time. The samples were placed on a screen tray in a cabinet in which the relative humid- ity and temperature were con- trolled (11). Loose fitting covers were removed simultaneously by a multiple magnet arrangement. By this arrangement creams having various formulas were exposed for lengths of time varying from 5 minutes to 48 hours under chosen conditions. Cessna, et al. (6), found that aqueous solutions of humectants lost Figure 9.--Cap lifting device---multiple magnet frame in raised position over tray. Picture shows three jars with caps removed, suspended from magnets, and three with caps in place. Maximum of 77 samples accommodated. The equipment is shown here installed inside the constant tempera- ture-constant humidity cabinet. weight at a more or less constant rate until the concentration of the solutions approached that of the equilibrium values. Earlier work that we had done indicated that a similar behavior could be expected from O/W cosmetic creams. From these data it was estimated that several days would be required to lose an amount of water sufficient to allow the humectant to exert an influence based on its equilibrium moisture holding power. In 'addi-