20 JOURNAL OF THE SOCIETY OF COSMETIC CHEMISTS time on q/• is analogous ,to the concentration exponent of bactericides aqueous solution, and is called the phase-volume coefficient. Our early experiments with E. coli showed that a further factor was of greater significance in determining the activity than had hitherto been presumed. A quantity of liquid paraffin was included within a 100 ml, volume of inoculated aqueous phenol solution, the necessary adjustments being made to restore both the aqueous phase concentration and ,the number of organisms/ml to that of the original aqueous solution. The. two-phase system was slightly but significantly more antibacterial than, the aqueous solution. We have shown (36) tl•at a further increase in the oil:water ratio considerably increased the activity (Fig. I). The only difference between the aqueous solution and the two-phase systems was 2.8 2.0 ' Figure 1. I I•atio oiL:water /•-0.2 n -1'0 o - 5'0 x-l,0.O X o 0 0.1 Pheno[ in aqueous 'phase, I Log ø/o w/v ) ' ß ACtivity against E. coli of liquid paraffin--phenol--water mixtures

THE ACTIVITY OF AiNTIBACTERIALS IN TWO-PHASE SYSTEMS 21 the volume of oil contained within the latter and the oil/water interface created by the latter. Photographs show that a proportion of the organ- isms are adsorbed at the interface and, indeed, probably help to stabilize it. The o[w systems containing bacteria were shaken intermittently, and it is possible that organisms which at one moment were in the bulk aqueous phase were at another adsorbed at the aqueous side of the interface. It may also be deduced from interfacial tension measurements and the Gibb's adsorption equation, that the concentration of phenol at the inter- face is greater than in the bulk of the aqueous phase. What is uncertain is precisely how the phenol molecules and the organisms are aligned at• the interface, but it may be, assumed that the molecules at the interface penetrate the aqueous phase to a smaller depth than do the cells. It is therefore probable that part of the bacterial surface at the interface is in contact with a higher concentration of phenol, but this is certainly not true for the whole cell. A complicating factor is the concentration of phenol in the oil phase which is always related to that of the aqueous phase (Table II), and may behave as a reservoir for the latter. Our knowledge of all the factors operating to control the activity of pre-. serratires in two-phase systems is undoubtedly far from complete. Never- theless, the role played by the interface in determining the antimicrobial activity of a system seems too real to be ignored and has, in fact, been underlined by all the subsequent experiments we have carded out. An analysis of 400 experiments arranged in a randomized block design to determine the importance of the concentration of bactericide in the aqueous phase, and of the oil:water ratio in determining the antibacteria! Table III Analysis of variance of extinction-time data for E. coli in phenol/liquid paraffin/water dispersions at 25 ø Sums of Source of variation squares øF Variance F P 6 Between oil-water 404730.48 6 67455.08 409.94 0.001 ratios (R) 358 Between aqueous phase 5 phenol concentrations (P) 1945585.19 5 389117.04 52364.73 0.001 358 30 Interaction (RP) 865299.78 30 28843.33 175.286 0.001 358 Residual 58910.19 358 164.55 rota• 3274525.64 399 '