334 JOURNAL OF THE SOCIETY OF COSMETIC CHEMISTS RESULTS AND DISCUSSION EMULSION MANUFACTURE Several emulsions (usually five) differing only in polymer content, were prepared at the same time from the same master emulsion. The manufacturing procedure used appeared to be reproducible based on particle size measurements of emulsions prepared on dif- ferent dates. The median particle size for five replicates containing 1% emulsifier (and no polymer) prepared on separate occasions ranged from 4.3 to 4.8 •m with a mean value of 4.52 •m and a standard deviation of 0.22 •m. Figure 2 contains particle size distributions for a series of emulsions containing various concentrations of xanthan gum prepared from the same batch of master emulsion. Values of per cent oversize were quite close to each other and there were no systematic differences. Based on these results, and similar data for other emulsion batches, it appears that emulsions manufactured from the same master emulsion had essentially identical particle size characteristics. Therefore, differences in emulsion properties within a group (containing the same emulsifier concentration) may reasonably be ascribed only to effects of the polymers present. Particle size distributions for two emulsions containing different emulsifier concentra- tions (and no polymer) are shown in Figure 3. As the emulsifier concentration was 11o Ld lOO 9o 8o 7o 6O 50- 40- 30- 20- 10- 0.2 0.4 0.6 0.8 1 1.2 1.4 LOG PARTICLE SIZE (Dr'n) Figure 2. Particle size distribution of a series of emulsion containing different xanthan gum concentrations at 1% (w/w) emulsifier. Xanthan gum concentration: [] 0%, + 0.1%, O 0.2%, /• 0.3%, X 0.4%.

EMULSION STABILIZATION BY GUMS 335 !1o 100 - 90- 80- 70- 60- 50- 40- 30- 20- 10- [] + [] + D + D + 0.2 0.4 0.6 0.8 I I .2 I .4 LOG PARTICLE SIZE (pm) Figure 3. Particle size distribution of two emulsions containing different emulsifier concentrations and no polymer. Emulsifier concentrations: [] 0.25%, + 0.5%. increased, median particle size dropped until a concentration of 3% was reached, at which point there was no further change in median particle size (Figure 4). One difficulty with the particle size distributions was that particles whose diameter was less than two }xm were not counted. If a large number of particles below the instru- ment's threshold are present, the median value determined graphically overestimates the true value. Also, since the distributions were given in terms of particle numbers, the contribution of very small particles is emphasized. It is possible for the major portion of the volume of emulsion to be accounted for by the larger globules, though they are comparatively few in number. Therefore, in addition to determining median particle size values from the distribu- tions, we tabulated the total number of particles as well as the ratio of particles with diameters greater than 20 }xm to the total number of particles counted. The ratio of particles with diameters above 20 }xm was inversely related to emulsifier concentration. As the emulsifier concentration was raised from 0.25% to 1% (Figure 5), the total number of particles increased, reflecting an increase in dispersion. However, at 3% and 5% emulsifier, the particle count was much lower again. This may have been due to a substantial increase in the number of particles below our instrument's threshold or, possibly, an increase in the amount of oil solubilized. This change in total particle