A new statistical model is proposed for the molecular mass distributions (MMD) of polymerized anions in silicate melts. The model is based on the known distribution of Q(n) species in the MeO-Me2O-SiO2 system. In this model, chain and ring complexes are regarded as a random series of Q(n) structons with various concentrations of bridging bonds (1 <= n <= 4, Q(0) corresponds to SiO44-). This approach makes it possible to estimate the probability of formation of various ensembles of polymer species corresponding to the general formula (SiiO3i+1-j)(2(i+1-j)-), where i is the size of the ion, and j is the cyclization number of intrachain bonds. The statistical model is utilized in the STRUCTON computer model, which makes use of the Monte Carlo method and is intended for the calculation of the composition and proportions of polyanions at a specified degree of polymerization of silicate melts (STRUCTON, version 1.2; 2007). Using this program, we simulated 1200 MMD for polyanions in the range of 0.52 <= p <= 98, where p is the fraction of nonbridging bonds in the silicon-oxygen matrix. The average number of types of anions in this range was determined to increase from three (SiO44-, Si2O76-, and Si3O108-) to 153, and their average size increases from 1 to 7.2. A special option of the STRUCTON program combines MMD reconstructions in silicate melts with the formalism of the Toop-Samis model, which enables the calculation of the mole fraction of the O2- ion relative to all anions in melts of specified composition. It is demonstrated that, with regard for the distribution and average size of anion complexes, the concentration of the O2- ion in the MeO-SiO2 system is characterized by two extrema: a minimum at 40-45 mol % SiO2, which corresponds to the initial stages of the gelenization of the polycondensated silicate matrix, and a maximum, which is predicted for the range of 60-80 mol % SiO2.