The atmospheric trend of methane isotopic ratios since the mid- 20th century has been reconstructed from Antarctic firn air. High volume air samples were extracted at several depth levels at two sites in East Antarctica. Methane concentration and its 13C/12C and D/H ratios were determined by gas chromatography, mass spectrometry, and infrared spectroscopy. A firn air transport model was applied to reconstruct past atmospheric trends in methane and its isotopic composition. By subsequent application of an atmospheric model, changes in methane sources and OH sink compatible with the past atmospheric trends are explored. In step with increasing methane mixing ratios, 513C increased by -1.7%0 over the last 50 years. These changes mainly reflect a shift in relative source strength toward the heavier anthropogenic methane source, such as biomass burning and methane of nonbiological origin. The 5D (CH4) showed a period of decline between the 1950s and 1975, followed by a gradual increase of 0.55%0/y, also toward the heavier anthropogenic source. Dependent on possible changes in the OH sink, to which 5D of methane is very sensitive, the inferred isotopic trends of 513C and 5D over the last 50 years constrain the relationship between natural and anthropogenic sources over the last century. The observed 5D minimum around 1975 suggests that the slowing down in the methane source growth took place during this period.

We have previously reported initial investigations of a novel micronucleus assay which utilizes a 3-dimensional human skin model EpiDerm™ supplied by MatTek Corporation (Ashland, MA). The advantage of this in vitro system is that it might potentially replace second-tier animal studies (or possibly first-tier non-specific traditional in vitro assays) in the genotoxicity evaluation of many materials, especially cosmetics ingredients. If so, it could then satisfy the requirement for non-animal testing of cosmetics ingredients after 2009 imposed by the European Seventh Amendment to the Cosmetics Directive. An important advantage of this assay system is that materials can be applied topically to the surface of the in vitro human skin model in a fashion similar to probable human exposure. In addition to previously reported positive responses with mitomycin C, vinblastine sulfate, MNNG and methyl methane sulfonate, four non-genotoxins (as shown by lack of micronucleus induction activity in an in vivo rodent skin model) have now been shown to induce no response in the reconstituted skin model. The non- genotoxins (4-nitrophenol, 1,2 epoxydodecane, trichloroethylene, and 2-ethyl-1,3-hexanediol) were tested up to levels of approximately 50% survival. The model has been further characterized by measuring expression of xenobiotic metabolizing enzymes. Although many enzymes were constitutively expressed, 3-methylcholanthrene and â-naphthoflavone strongly increased expression of CYP1A1 and CYP1B1, and slightly enhanced expression of CYP2C19, CYP2D6, CYP3A4 and CYP3A5. Overall, the in vitro model showed an enzyme expression pattern very similar to normal human skin. It appears that this model may be very useful for detecting of micronuclei induction in a structurally complex tissue of human origin.