Optical spectroscopy is a valuable tool for online monitoring of a variety of processes. Ultraviolet−visible (UV−vis) spectroscopy can monitor the concentration of analytes as well as identify the speciation and oxidation state. However, it can be difficult or impossible to employ UV−vis-based sensors in chemical systems that are very dark (i.e., have a high optical density), requiring exceedingly short path lengths (for transmission approaches) or an effective means of backscattering (for reflectance approaches). Examples of processes that produce highly absorbing solutions and that would benefit significantly from the diagnostic potential of optical sensors include used nuclear fuel recycling and molten salt systems with high concentrations of dissolved uranium. Utilizing an attenuated total reflectance (ATR) UV−vis approach can overcome these challenges and allow for the measurement of solutions orders of magnitude more concentrated than transmission UV−vis. However, determining ideal sensor specifications for various processes can be timeconsuming and expensive. Here, we evaluate the ability of a novel ATR−UV−vis probe to measure very concentrated solutions of Co(II) and Ni(II) nitrate as well as organic dyes (methylene blue, acid red 1, and crystal violet). This sensor design provides a modular method for exploring possible "path lengths" by altering the length of the ATR fiber that was submerged within solution during spectral measurements. Measurements within the ATR sensor cell were compared to measurements gathered by transmission UV−vis of samples within a commercially available 1 cm optical cuvette. The ATR−UV−vis probe was capable of measuring absorbance of solutions with a chromophore concentration 600 times greater than that in the 1 cm cuvette. Advanced data analysis in the form of multivariate curve resolution (MCR) was used to analyze the speciation of methylene blue over a large concentration range. The application of this novel ATR−UV−vis probe to the investigation of dark solutions is a promising avenue for use in online monitoring of nuclear processes.