310 JOURNAL OF THE SOCIETY OF COSMETIC CHEMISTS ing into the surfactant phase with that of the spacings. Figs. 12 and 13 show the changes in the water content caused by its migration into the surfactant phase with a lapse in time. In the case of water only, the migration to the surfactant phase was as much as 6 per cent. In case of monosodium L-glutamate monohydrate, the amount of migration decreased with an increase in concentration. This corresponds well with the decreasing trend of affinity between the surfactant and water. The water content became constant rapidly and did not increase over a prolonged period. The relationship of this water content to the spacings and the stability of the gels is in- dicated in Table IV. As is evident from this Table, with the increase of water migrating to the surfactant, the spacing becomes much wider. The spacing obtained with 40 per cent aqueous solution of monosodium L-glutamate monohydrate was exactly the same as that of the surfactant itself (Sunsoft O-30B). No changes in the structure of the surfactant was observed when the solubilized water was less than 3 per cent. The larger the variation of the spacings, the poorer the gel formation became. The effect of pH on the stability of the gel can also be explained by the variation of water content and the spacings as shown in Figs. 14 and 15. As the pH of water increased, the water solubilized in the surfactant increased, and at the same time the spacings also increased rapidly. From the above results and discussions, it is possible to draw the structural model indi- cating the mechanism for the stabilization of the gels by the amino acids as is shown in Fig. 16. The upper models indicate the overall view, and the lower models show a magnified view. The surfactant, though it is in the liquid state, has a lameliar structure as seen in view (1) of Fig. 16. In the case of the aqueous solution in the absence of amino acids, a large amount of water is solubilized in the hydrophilic part of the surfactant having an orderly form and its original structure becomes disordered and loosened due to the widening of its spacings. Under such conditions, water particles coalesce easily, which indicates instability. The same explanation can be made in the case of KSCN and urea, which change the structure of the surfactant. This is shown in view (2) of Fig. 16. View (3) of Fig. 16 indicates the case when amino acid is added. Since the amount of water to be solubilized in the hydrophilic part of the surfactant is limited and does not induce any change of structure, a concentrated and tight inter- facial atmosphere is established around the particles. As a result, coalescence of the particles hardly occurs thus, maintaining stability. As an important fact to support such a structural model, we succeeded in taking the photograph of the gel by EM (as is seen in Fig. 17). The sample shown was the gel obtained with Sunsoft O-30B and a 40 per cent aqueous solution of monosodium L-glutamate monohydrate mixed in the ratio of 1:4. Apparently, the water particles surrounded by the surfactant phase in the lamellar structure can be seen. GEL-EMULSIFICATION METHOD The relationship between the stability of the gels and that of w/o emulsions made by using the gels has also been noted. Hardness of the gels was influenced by the shear stress given at the time of preparation. When the hardness of the gels was changed, the

WATER-IN-OIL EMULSIONS 311 hardness of the w/o cream using such gels also changed depending on the former change. As was previously described, the gel obtained by Sunsoft O-30B and the amino acid was stable however, it is necessary to observe the changes in the cream when an unsta- ble gel is used. Table V and Fig. 18 indicate the results obtained when an unstable gel, such as obtained between Sunsoft O-30B and water in the absence of an amino acid, was used. In comparing these results with the case where stable gels were used in the preparation of creams, remarkable differences were recognized as to viscosity, hard- ness, stability, and size of the emulsion particles. Table VI indicates the relationship of the X-ray diffraction patterns of the surfactants to it• function in gel-formation and the properties of the w/o emulsions using these gels. As is evident from this Table, it is readily understood that the surfactants having clear X-ray diffraction patterns can readily form gels and the creams obtained were stable while those having indistinct X-Ray diffraction patterns did not form gels and produced unstable creams. These facts can also be confirmed by other studies. As was mentioned previously, the amino acid functions to prevent the expansion of the spacings of the surfactanr when an aqueous solution of amino acid is added to the surfactant phase. It was also found that amino acids possessing this property produced a good stable gel. Furthermore, the relationship between these properties and the stability of w/o emulsions were also studied. The results are shown in Table VII. There is a correlation between the stability of the gels and the stability ofw/o emulsions namely, the better the stability of the gels, the better the w/o emulsion obtained. Furthermore, the stability of the gels, which have been studied on the standard sample by using squalane as the main constituent of the oil, must be changed with the polarity of the oil. Figure 19 shows the influence of a component in the oil phase on the stability of the w/o emulsion. When squalane only was used as the oil phase, the destruction of the gel was not observed and the resulting w/o emulsion had a high viscosity of greater than 100,000 cps, but as the mixing ratio of glycerol tri-2-ethylhexanoate became higher, the viscosity dropped rapidly and finally, no emulsification occurred which was accompanied by a total destruction of the gel in the oil phase. In a nonpolar oil, such as squalane, there was no structural change of the surfactant however, in a polar oil, such as glycerol tri-2-ethyl- hexanoate, the structure disappeared completely. Therefore, as the polar components in the oil phase increased, the emulsion became increasingly unstable. Thus, in the gel- emulsification method, it is advisable to use nonpolar oils and waxes in the oil phase, such as, squalane, liquid paraffin, and microcrystalline wax, etc. These facts have been proven to be identical in practical formulations of cosmetic creams as well as in the simple systems studied. In the gel-emulsification method, the drawings of the emulsion when the oil is added can be assumed to be as is shown in Fig. 20. In the case of nonpolar oils, the water particles surrounded by several layers of surfactant disperse in the oil phase. On the contrary, in the case of polar oils, it may be assumed that the orientation of the surfactant is loosened and its adsorption at the interface are hindered, resulting in an unstable emulsion. The above results may be summarized as follows. By initially forming a stable gel and maintaining conditions to maintain gel stability, a stable w/o emulsion having high vis-