High replication error rates strongly limit the length of sequences that can transmit reliable information. However, when we consider that selection acts on the phenotype, the extremely large degeneracy between genotype and phenotype spaces confers robustness (in the form of increased molecular neutrality) to RNA populations. An RNA neutral network is a connected ensemble of all RNA sequences in the genome space folding into the same minimum free energy secondary structure. Each sequence occupies a node and two nodes are connected if the corresponding sequences differ in only one nucleotide. Therefore, a population of sequences can move, through mutations, on such networks without seeing its functionality affected. In this work we study analytically and numerically the properties of RNA neutral networks that affect robustness: its areas of maximal neutrality against mutations and the minimum free energy associated to the folded state of each sequence. The topological properties of such networks determine the time required to attain maximally neutral states and the diversity of sequences in the population at that state. When information on the energy associated to each sequence is included, topology also affects the survivability of the populations under temperature fluctuations. In summary, we have obtained detailed analytical results characterizing the evolution of populations in simple networks, with applicability to natural RNA networks. References: [1] J. Aguirre, J. M. Buldú and S. C. Manrubia, Evolutionary dynamics on networks of selectively neutral genotypes: Effects of topology and sequence stability. Submitted. [2] E. van Nimwegen, J. P. Crutchfield and M. Huynen, Neutral evolution of mutational robustness, PNAS 96, 9716 (1999). [3] P. Schuster, Prediction of RNA secondary structures: from theory to models and real molecules. Rep. Prog. Phys. 69:1419 (2006).