Abstract. The study of Langmuir circulation has a strongly interdisciplinary history; the first half of this work is a brief and eclectic review of this. Much of the research has been motivated by interest in the biology and chemistry of the mixed layer. These, in turn, are sensitive to details of typical particle trajectories; i.e., to high-order statistics of the flow such as time and space lagged covariances. In contrast, descriptions of the mixed layer have progressed from means (mixed layer velocity, temperature, and depth) to variances only recently. With the development of new observation techniques, and of complex numerical models, building upon equations for the development of Langmuir circulations developed by A. Craik and S. Leibovich in the late 1970's, it appears that descriptions of these high-order statistics may be attainable. However, systematic discrepancies appear in the observations that remain to be explained: e.g., the magnitude of nearsurface velocities associated with Langmuir circulation varies over a factor of 4 between different wind events for apparently similar conditions, yet is well behaved within each individual event. The last part of this work dwells on some recent observations and the nature of this unexplained variability.