O-18/O-16 ratios vary systematically in the 1,000 m section of interlayered metasediments and granitoids at Lizzies Basin, the deepest structural level exposed in the East Humboldt Range metamorphic core complex. In the lower 300 m of the section (the Lower Zone) deltaO-18 is homogeneous and low (+6.6 to +8.8 in silicate rocks, +8.7 to +12.1 in marbles). A detailed oxygen isotope profile across an individual marble layer within the Lower Zone has a similar range in deltaO-18 (+9.5 to +11.9) with the highest values preserved in the middle of the layer. In the upper 700 m of the section (the Upper Zone) metasediments have been less strongly O-18 depleted. The deltaO-18 values are higher and more heterogeneous and in a profile across a marble layer of similar thickness, rise steeply from marginal values of +12 to core values of +23. Quartz from silicate metasediments throughout the Upper Zone ranges from +11 to +13 and is thus fairly homogeneous, particularly where detailed profiles have been measured adjacent to the O-18-rich marble layers. Covariation of O-18/O-16 and C-13/C-12 ratios in marbles suggests that the isotopic composition of these elements has been altered by exchange with infiltrating, water-rich, CO2-H2O fluids with mantle-like isotopic composition. In most cases, marble cores preserve protolith deltaC-13 values and these vary systematically throughout the section according to structural level, including some exceptionally C-13-rich values up to +12. This range may reflect stratigraphic variation in deltaC-13 of the Proterozoic sedimentary protoliths of these rocks. The Upper/Lower Zone boundary, and the contrast in isotope systematics to either side, has been previously explained either as an impermeable barrier to fluid flow, or as an infiltration front (Wickham and Peters 1990). The second alternative has been tested using a numerical model in which low-O-18 aqueous fluid flows vertically upward through an alternating sequence of monomineralic quartz (deltaO-18 = +12) and calcite (deltaO-18 = +22) layers in a regime in which the oxygen isotopic composition is controlled by advection, diffusion (in the fluid), and fluid/solid exchange. Solutions to the relevant transport equations indicate that the abrupt change in the average deltaO-18 value of the layers at the Upper/Lower Zone boundary can be reproduced by the model with a uniform porosity throughout the system, but the observed contrast in the shapes of the marble layer profiles in the Upper and Lower Zones cannot be reproduced under these conditions. However, if the calcite layers have two orders of magnitude lower porosity in comparison with the quartz layers and exchange within them is diffusion-dominated (as opposed to advection-dominated in the quartz) the contrasting shapes of the marble profiles in the two zones are reproduced, as well as the Upper/Lower Zone discontinuity. The range of conditions that generates both an infiltration front (at the appropriate scale) and contrasting marble profiles (at the appropriate scale) is quite narrow but requires a volatile flux that could be generated by plausible volumes of mantle-derived magma crystallizing at depth beneath the area, providing support for this mechanism as a viable agent of O-18 depletion in the deep-level rocks of this terrane.