In this paper we present a theoretical study of the physical properties of solvated electrons in ammonia based on the Copeland-Kestner-Jortner model, which incorporates short-range interactions uia a first solvation layer and long-range interactions via polaron modes. We have studied bound-bound and bound-continuum optical transition emphasizing the problem of line shapes in absorption and emission. The total energy of the ground and excited states and its dependence on nuclear configuratiens was handled by three successive approximate calculations: (a) a temperature dependent potential including short -range radial displacements; (b) a temperature independent potential incorporating both radial and angular short-range displacements; (c) a multidimensional potential surface including both short-range and long-range (polaron) nuclear displacements. The calculated line shapes in absorption for a single solvent configuration include major contributions from short-range radial displacements and from the polaronl modes. The energy and line shape for the 2p -* ls emission band is predicted. A general formula is presented for photoionization cross section including the contribution of all medium modes and in this caise the role of the polaron modes is crucial. In our previous paper,2 we advanced a model for the solvated electron in polar fluids which took into account the strong short-range interactions of the electron and the first coordination layer solvent molecules as well as the long-range interactions with the bulk medium. This model was capable of yielding quantitative information on the properties of solvated electrons in ammonia as well as providing qualitative data on excess electrons in other polar solvents. In the latter cases we did not attempt a detailed study. Rmxntly, Fueki, Kevan and Christoffersen, in particular, have applied a similar model to study solvated electrons in water and al~ohols.3~,~ Although many questions remain concerning the trends observed in wideby differing solvents4 and solvent mixtures,4-' these are predicted well enough by our model to allow us to consider an entirety different set of problems, namely the details of radiative processes in which the state of the electrlon changes. In this paper we will use metal-ammonia solutions as a representative system for the study of these processes. We expect the same general behavior in other polar fluids except that the relative rates of the various processes could change in different solvent systems. We will begin with a display of the latest calculations on our model2 as well as an improved version. These new results provide a better insight into the physical properties of the solvated electron. We shall focus attention on the radiative processes of the solvated electmn and, in particular, the observed absorption line shapes for bound-bound and for boundcontinuum transitions as well as the yet unobserved emission line shape and the photoionization profile. The study of optical line shapes provides a starting point for the understanding of nonradiative processes of the solvated electron, such as electron capture from the conduction band to form the localized ground state, or the reverse process of thermal ionization of the ground state and of excited states. In general, the nonradiative transition probability can be expressed in terms of a generalized line shape function in the limit of zero frequency. However, in view of the special nature of the problem, where the localized excess electron wave function is strongly dependent on the (short-range and polaron type long-range) nuclear coordinates, the theory of the optical line shapes presented herein requires a gross modification before it can be applied for the elucidation of the nonradiative decay processes of the solvated electron. 11. Calculations