The first measurements of ClOOCl in the stratosphere have been acquired from a NASA ER-2 aircraft, deployed from Kiruna, Sweden (68°N, 21°E), during the joint SOLVE/THESEO-2000 mission of the winter of 1999/2000. ClOOCl is detected by thermal dissociation into two ClO fragments that are measured by the well-known technique of chemical conversion, vacuum ultraviolet resonance fluorescence. Ambient ClO is detected simultaneously. Observations of the ratio [ClOOCl]/[ClO]² (estimated uncertainty of ±25%, 1 σ) are used with a time-dependent photochemical model, to test the model representation of the ratios of kinetic parameters J/k^Prod and k^Loss/k^Prod for day and nighttime observations, respectively. Here, k^Prod and k^Loss are the rate constants for ClOOCl production and loss, respectively, and J is the photolysis rate of ClOOCl. The observations are in good agreement with J based upon the 2002 JPL recommended cross sections for ClOOCl [ Sander et al., 2003 ], if the true value of k^Prod is given by either the 2000 JPL recommendation [ Sander et al., 2000 ] or the work of Trolier et al. [1990] . The larger values of k^Prod given by Bloss et al. [2001] and the 2002 JPL recommendation are consistent with the observations only if J is increased by a significant amount. This is accomplished if J is calculated with the larger ClOOCl cross sections measured by Burkholder et al. [1990] . The J values for ClOOCl based on the Huder and DeMore [1995] cross sections are too small, by factors of ∼1.6 to 2.5 for all values of k^Prod, based on the observations. Nighttime results suggest that, for 190 < T < 200 K, the values for K^Eq (the equilibrium constant, equal to the ratio of k^Prod/k^Loss) of Cox and Hayman [1988] and Avallone and Toohey [2001] are in best agreement with the observations.