186 JOURNAL OF THE SOCIETY OF COSMETIC CHEMISTS Figure 3. Concentration of lanolin/marker in trilaminar membrane of cell envelope. lanolin and/or marker. The dimensions we see certainly agree well with Swanbeck's measurements. This trilaminar structure with a central protein band has been disputed, but lipid layers with a spacing of 50 to 80 fix were reported by Friberg et al. (10), who also showed that interaction of other lipids with the bilayers could affect the spacing. Other work reported by Ward and du Reau (l 1) also showed an alteration in spacing by the solubilization of oleyl alcohol into the bilayers and consequent increase in water capacity. It is possible in our case, therefore, that incorporation of applied substance into the lamellae has altered the spacing of lipid layers from that reported by Friberg et al. Trilaminar structures can also be detected in Figure 4, and here too, as in Figure 3, some of them have become detached from the cell surface. Trilaminar membranes have been previously reported by Odland and Holbrook (12) and Elias (13), and are now an accepted structural feature. In our case it appears that although most of the natural intercellular lipids have been lost during preparation of the specimens (except perhaps in Figure 10), the lamellae covalently bound to the corneocyte surfaces appear to have survived in places, although some have been detached from the cell surface. In an attempt to overcome this loss of lipids, further samples were prepared using a freeze- substitution technique that has been reported to fix and retain better, at low temper- ature, the natural lipids. This does not appear to have been so in our case, however, with most intercellular spaces still showing absence of natural lipids. Nevertheless, some interesting features are visible. Figure 5 shows that lanolin (grey) has penetrated here into the intercellular spaces down as far as the stratum granulosum. Such stained material was not visible in the untreated

MOISTURIZING FUNCTION OF LANOLIN 187 ß ! ! Figure 4. Lanolin/marker incorporated in membranes attached to, and detached from, the skin surface. control (Figure 6), where the intercellular spaces appear white. There also appears to be a heavy concentration of electron-opaque material at a junction between the corneocytes (arrowed) in Figure 5 (perhaps a desmosome remnant), and more evidence of this is reported in Stage 3. STAGE 3 Outline and methodology. Although this was basically a repeat of Stage 2, Section 2, freeze substitution was used throughout in the preparation of specimens, and there was a change in the lead-doping technique. Instead of using lead lanolate, doping was carried out with lead oleate, which has better lipid solubility and is easier to wash free from residual inorganic lead salts. Examination of thin films of the lead-oleate-doped lanolin by electron microscopy showed it to be quite homogeneous, with no dense particles. Tissue specimens were not stained, so that electron-opaque areas indicate the presence of lead oleate and presumably, therefore, of lanolin, unless selective migration of lead salt into other lipids has occurred, against expectation. Results. Figure 7 confirms the penetration of lanolin deep into the stratum corneum. It also confirms findings reported in Stage 1, Section 2, of the existence of a structure with high lead and presumably lanolin content forming a layer on, or near, the surface of some corneocytes bounding intercellular spaces, more clearly seen in an enlargement of the area marked 'L' in Figure 8. Similar layers on or close to, and parallel with, corneocyte surfaces are very clear in Figure 9 and 10, where signs of a lameliar structure