Neutrinos are the most mysterious particles in the standard model. Many of theirfundamental properties such as their masses, lifetimes, and nature (Dirac or Majorana) are yet to be pinned down by experiments. Currently, the strongest bound on neutrino masses comes from cosmology. This bound is obtained by scrutinizing the gravitational effect of the cosmic neutrinos on the evolution of structure in our universe. However, this bound assumes that the neutrinos from the Big Bang have survived until the present day. In this dissertation, the unstable neutrino scenario is studied in light of current and near-future cosmological experiments. We show that the current cosmological bound on the neutrino masses can be relaxed significantly in an unstable neutrino scenario. We further show that near-future experiments offer the possibility of independently measuring both the masses of the neutrinos and their lifetimes. We consider an elusive scenario in which the cosmic neutrinos decay into invisible radiation after becoming non-relativistic. The Boltzmann equations that govern the cosmological evolution of density perturbations in the case of unstable neutrinos are derived and solved numerically to determine the effects on the matter power spectrum and lensing of the cosmic microwave background (CMB). A Markov-Chain Monte-Carlo (MCMC) analysis is done on the current cosmological data and mock future data to obtain its sensitivity to the neutrino masses and lifetimes. We show that the effect of the neutrino masses on large scale structures is dampened by the decay of neutrinos, which leads to a parameter degeneracy between the neutrino masses and lifetimes inferred from the cosmological data. This degeneracy allows for a significant relaxation of the current cosmological upper bound on the sum of neutrino masses from about 0.2 eV in the stable neutrino case to 0.9 eV in the unstable neutrino scenario. This window is important for terrestrial experiments such as KATRIN which are seeking to independently measure the neutrino masses in the laboratory. We further show that near-future large scale structure measurements from the Euclid satellite, when combined with cosmic microwave background data from Planck, may allow an independent determination of both the lifetimes of the neutrinos and the sum of their masses. These parameters can be independently determined because the Euclid data will cover a range of redshifts, allowing the growth of structure over time to be tracked. If neutrinos are stable on the timescale of the age of the universe, we show that these observations can improve the lower limit on the lifetimes of the neutrinos by seven orders of magnitude, from O(10) years to 2 × 108 years(95%C.L.), without significantly affecting the measurement of the neutrino masses. On the other hand, if neutrinos decay after becoming non-relativistic but on timescales less than O(100) million years, these observations may allow for, not just the first measurement of the sum of neutrino masses, but also the determination of the neutrino lifetime from cosmology.