Progress in understanding telomere replication and its relationship to senescence has been hampered by the intrinsic variations in telomeres and the stochastic nature of senescence onset. The work in our lab relies on our ability to circumvent and experimentally dissect the causes of this heterogeneity. Our strategy is to manipulate and track single telomeres in individual cells and experimentally evaluate the effects on proliferation potential of single cell lines to build a mathematical model of senescence. We therefore set up a microfluidics-based live cell imaging assay to study replicative senescence in single Saccharomyces cerevisiae cell lineages after telomerase inactivation. We found that most lineages undergo an abrupt and irreversible transition consistent with a mathematical model in which the first telomere reaching a short critical length triggers the onset of senescence. Other cells exhibit transient checkpoint-dependent cell cycle delays followed by normal cell cycles before senescence. We have now shown that these two pathways to senescence correspond to two kinetically and mechanistically distinct, age-dependent processes underlying non- terminal and terminal senescence arrest. Taking advantage of our novel FinalCut system, we also investigated the proliferation potential of telomerase-negative cells with critically short telomeres of different defined lengths. Our results are consistent with a model whereby a small portion of cells may enter senescence in a probabilistic manner as the length of the shortest telomere decreases. Our results thus directly demonstrate that the length of the shortest telomere is a major determinant of entry into senescence, but a hidden parameter may also contribute to the onset of senescence upon telomerase inactivation. However, when the shortest telomere reaches a critical length, all cells immediately arrest. In addition, we found that at this critical length, the shortest telomere can be stably maintained without signs of fusions or degradations. This suggests that telomeres can be both critically short and functional with respect to telomere protection.